COVID-19 Disruption, Small Island Economies and Regional Development in Indonesia

COVID-19 Disruption,
Small Island Economies

and Regional Development
in Indonesia
Edited by:
Mansur Afifi
Diswandi
Sulhaini
Tri Mulyaningsih
Djoni Hartono
Budy P. Resosudarmo
IRSA Book Series On Regional Development No. 21

COVID-19 Disruption, Small Island Economies
and Regional Development in Indonesia
Penyunting :
Mansur Afifi
Diswandi
Sulhaini
Tri Mulyaningsih
Djoni Hartono
Budy P. Resosudarmo
Desain sampul :
-----
Tata letak :
Neni Yuliati
Proofreader :
Lilies Achmadi
Diterbitkan oleh :
Perkumpulan Ilmu Regional Indonesia (IRSA)
Sekretariat IRSA, Jl. Cimandiri No. 6 Bandung 40115
West Java-Indonesia
Email : info@irsa.or.id
ISBN : -----------------
Cetakan pertama, Februari 2024
Hak Cipta dilindungi Undang-Undang

TABLE OF CONTENTS
CONTRIBUTORS...................................................................................................................... v
FOREWORD............................................................................................................................... xii
Chapter 1 Implications of COVID-19 Disruption on Small Island Economies
Mansur Afifi, Diswandi, Sulhaini, Tri Mulyaningsih,
Djoni Hartono, and Budy P. Resosudarmo.............................................. 1
PART I. COVID-19 DISRUPTION

Chapter 2 Infrastructure for Mobilities and Citizenship Making: Insights
from Sangihe Islands Indonesia in Times of COVID-19 Pandemic
Naimah Lutfi Abdullah Talib, Ariane Utomo, and Fransiska
Limbengpiah........................................................................................................ 9
Chapter 3 Does Economic Diversity Help Economic Resilience of the
Agglomeration Areas in Java during the COVID-19 Pandemic?
Aloysius Gunadi Brata, Threesya Siburian, Dinar Fandan Sari,
AM. Rini Setyastuti, and Andreas Sukamto............................................ 45
Chapter 4 The Productivity of Urban Farming in the City of Yogyakarta
Catur Sugiyanto, Dion Saputra Arbi, and Ahmad Syahrul Fauzi... 69
Chapter 5 Mental Health Consequences of COVID-19 Pandemic on
Teachers and Caregivers of Early Childhood Education
Ni Wayan Suriastini, Fita Herawati, Ika Yulia Wijayanti,
Indrawan Firdauzi, Setyo Pujiastuti, and Sukamtiningsih.............. 91
PART II. SMALL ISLAND ECONOMIES

Chapter 6 Marine Fisheries Sustainable Livelihoods in Pulau Tiga,
Natuna Islands
Teguh Asandy and Wilmar Salim............................................................. 113
Chapter 7 Designing an Appropriate Renewable Energy System for
Small-Isolated Island Communities
Alan Dwi Wibowo and Andante Hadi Pandyaswargo....................... 139
iii
Chapter 8 The Shift of Ecosystem Services Use in Nusa Lembongan and
Nusa Ceningan due to the COVID-19 Pandemic
Nyoman Priyamanaya and Wilmar A. Salim........................................ 163
Chapter 9 Network of Relationships between Actors in Developing
Bonjeruk National Ecotourism Village: A Case Study
Mega Nisfa Makhroja and Lalu Agung Haris Atmaja........................ 183
Chapter 10 Land Property Rights, Income, and Food Security: Insights
from Ambon Island, Indonesia
Wardis Girsang, Ali Magana, Adelina Siregar, Elia L. Madubun,
and L.O. Kakisina.............................................................................................. 205
Chapter 11 Examining Community Perceptions of PT. Newmont Nusa
Tenggara’s Corporate Social Responsibility (CSR) Initiatives
Baiq Nurul Suryawati, Burhanudin, Zainal Abidin, Muttaqillah,
Isti Fadah, and Intan Nurul Awwaliyah.................................................. 231
PART III. REGIONAL DEVELOPMENT

Chapter 12 Public Financial Governance and Socio-Economic
Performance: Evidence from Local Government in Indonesia
I Kadek Dian Sutrisna Artha, Usman, Dewinta I Sumartono,
and Irwan Amarullah..................................................................................... 257
Chapter 13 Major Determinants of the Diarrhoea Incidence among
Children Under-Five (U5) in Indonesia
Adrian Chrisnahutama and Ni Made Sukartini................................... 279
Chapter 14 Credit Channelling to MSMEs in Indonesia: The Role
of Regional Digital Competitiveness in Pre and During
the COVID-19 Pandemic
Militcyano Samuel Sapulette, Nury Effendi, Vera Intanie Dewi,
Fitri Hastuti, Eva Ervani, and Anhar Fauzan Priyono....................... 325
iv
CONTRIBUTORS
Adelina SIREGAR
Faculty of Agriculture
Universitas Pattimura, Ambon, Indonesia
E-mail: siregar.adelina@gmail.com
Adrian CHRISNAHUTAMA
Faculty of Economics and Business
Universitas Airlangga, Surabaya, Indonesia
E-mail: –
Ahmad Syahrul FAUZI

Universitas Negeri Semarang, Semarang, Indonesia
E-mail: ahmadsyahrulfauzi@mail.unnes.ac.id
Alan Dwi WIBOWO

Universitas Lambung Mangkurat, Banjarbaru, Indonesia
E-mail: alan.dwi@ulm.ac.id
Ali MAGANA
Department of Agriculture, Maluku Province
E-mail: –
Aloysius Gunadi BRATA

Faculty of Business and Economics
Universitas Atma Jaya Yogyakarta, Indonesia
E-mail: gunadi.brata@uajy.ac.id
AM. Rini SETYASTUTI

E-mail: rini.setyastuti@uajy.ac.id
v
Andante Hadi PANDYASWARGO
Waseda Environmental Research Institute
Waseda University, Tokyo, Japan
E-mail: andante.hadi@aoni.waseda.jp
Andreas SUKAMTO
E-mail: andreas.sukamto@uajy.ac.id
Anhar Fauzan PRIYONO
Universitas Padjadjaran, Bandung, Indonesia.
E-mail: anhar.fauzan.priyono@unpad.ac.id
Ariane UTOMO3
School of Geography, Earth and Atmospheric Sciences
University of Melbourne, Melbourne, Australia.
E-mail: ariane.utomo@unimelb.edu.au
Baiq Nurul SURYAWATI

Universitas Mataram, Lombok, Indonesia.
E-mail: nurul.suryawati@unram.ac.id
Budy P. RESOSUDARMO
Arndt-Corden Department of Economics
Australian National University, Canberra, Australia
E-mail: budy.resosudarmo@anu.edu.au
BURHANUDIN
E-mail: –
Catur SUGIYANTO
Universitas Gadjah Mada, Yogyakarta, Indonesia
E-mail: catur@ugm.ac.id
vi
Dewinta I. SUMARTONO
PT Sarana Multi Infrastruktur (SMI)
Jakarta, Indonesia.
E-mail: dewinta@ptsmi.co.id
Dinar Fandan SARI
E-mail: dinarfandan@gmail.com
Dion Saputra ARBI
Universitas Gadjah Mada, Yogyakarta, Indonesia
E-mail: dionsaputra2017@mail.ugm.ac.id
DISWANDI
E-mail: diswandi@unram.ac.id
Djoni HARTONO
Universitas Indonesia, Depok, Indonesia
E-mail: djoni.hartono@gmail.com
Elia L. MADUBUN
E-mail: elmadubun@gmail.com
Eva ERVANI
E-mail: eva.ervani@unpad.ac.id
Fita HERAWATI
SurveyMETER
Yogyakarta, Indonesia.
E-mail: fita@surveymeter.org
vii
Fitri HASTUTI
E-mail: fitri.hastuti@unpad.ac.id
Fransiska LIMBENGPIAH
Independent Social Researcher
Sangihe Island, North Sulawesi, Indonesia.
E-mail: –
I Kadek Dian Sutrisna ARTHA
Jakarta, Indonesia.
E-mail: kadek.dian@ptsmi.co.id
Ika Yulia WIJAYANTI
SurveyMETER
E-mail: ikayuliaw@gmail.com
Indrawan FIRDAUZI
Universitas Jenderal Soedirman, Purwokerto, Indonesia.
E-mail: indrawan.firdauzi@unsoed.ac.id
Intan Nurul AWWALIYAH
Universitas Jember, Jember, Indonesia.
E-mail: –
Irwan AMARULLAH
Jakarta, Indonesia.
E-mail: irwan@ptsmi.co.id
Isti FADAH
Universitas Jember, Jember, Indonesia.
E-mail: –
viii
L.O. KAKISINA
E-mail: leunardkakisina@gmail.com
Lalu Agung Haris ATMAJA
Department of International Relations
E-mail: –
Mansur AFIFI
E-mail: mansurafifi@unram.ac.id
Mega Nisfa MAKHROJA
Department of International Relations
E-mail: mn.makhroja@unram.ac.id
Militcyano Samuel SAPULETTE
College of Business and Economics
Australian National University, Canberra, Australia.
E-mail: militcyano.sapulette@anu.edu.au / iansapulette@gmail.com
MUTTAQILLAH
E-mail: –
Naimah Lutfi Abdullah TALIB
School of Geography, Earth and Atmospheric Sciences
University of Melbourne, Melbourne, Australia.
E-mail: ntalib@student.unimelb.edu.au
Ni Made SUKARTINI
Universitas Airlangga, Surabaya, Indonesia.
E-mail: ni-made-s@feb.unair.ac.id
ix
Ni Wayan SURIASTINI
SurveyMETER
E-mail: suriastini@surveymeter.org
Nury EFFENDI
E-mail: nury.effendi@unpad.ac.id
Nyoman PRIYAMANAYA
School of Architecture, Planning and Policy Development
Institut Teknologi Bandung, Bandung, Indonesia
E-mail: –
Setyo PUJIASTUTI
SurveyMETER
E-mail: setyo@surveymeter.org
SUKAMTININGSIH
SurveyMETER
E-mail: kamtiningsih@surveymeter.org
SULHAINI
E-mail: sulhaini@unram.ac.id
Teguh ASANDY
Bandung Regional Government Office
Bandung, Indonesia
E-mail: –
Threesya SIBURIAN
E-mail: threesyaaprilya@gmail.com
x
Tri MULYANINGSIH
Universitas Sebelas Maret, Solo, Indonesia
E-mail: trimulyaningsih.uns@gmail.com
USMAN
Jakarta, Indonesia.
E-mail: usman@ptsmi.co.id
Vera Intanie DEWI
Faculty of Economics
Universitas Katolik Parahyangan, Bandung, Indonesia
E-mail: vera_id@unpar.ac.id
Wardis GIRSANG
E-mail: wardisgirsang641@gmail.com
Wilmar SALIM
School of Architecture, Planning and Policy Development
Institut Teknologi Bandung, Bandung, Indonesia
E-mail: wsalim@pl.itb.ac.id
Zainal ABIDIN
E-mail: –
xi
FOREWORD
The COVID-19 pandemic has caused a significant and unprecedented disruption

worldwide, affecting businesses, cultures, and people’s lives in a profound way.
As the virus rapidly crossed international boundaries, the complex network
of interconnections that characterizes our globalised society encountered
unexpected obstacles. Following the worldwide crisis, policymakers faced
the challenging tasks of developing effective strategies to reduce the various
repercussions of the pandemic on their countries. Scholars and researchers
are currently engaged in a quest to comprehend and negotiate the complicated
chain of effects resulting from the COVID-19 pandemic. They are exploring and
seeking insights into the many and subtle implications of this global health
crisis.
This book, entitled COVID-19 Disruption, Small Island Economies, and
Regional Development in Indonesia, delves into the unique context of Indonesia,
a sprawling archipelagic nation encompassing a myriad of small islands. In this
volume, we set out to unravel the complexities of the COVID-19’s impact on
the socio-economics and livelihoods of Indonesia’s small islands and regions.
The value of this investigation is found not only in analysing the financial
outcomes, but also in understanding the intricate interweaving of agricultural
activities, health dynamics, and the resourcefulness demonstrated by economic
participants in the face of difficult circumstances.
As we go through the pages of this book, we observe the progressive
story of resilience, adaptation, and inventiveness in the pursuit of sustainable
economic development. The Indonesian government’s support, along with the
creativeness of economic actors, demonstrates the possibility for innovation
and success even in the most challenging of circumstances.
xii
This volume represents more than just an academic pursuit. Indeed, it also
serves as a tribute to the remarkable capacity of the human spirit to adjust, create
new ideas, and persist in the face of challenges. We hope that this investigation of
the effects of COVID-19 on small islands and regional development in Indonesia
will stimulate further discussions, research, and initiatives aimed at achieving a
more resilient and sustainable future.
Bambang Hari Kusumo

Rector of University of Mataram
xiii
xiv COVID-19 Disruption, Small Island Economies and Regional Development in Indonesia
CHAPTER 1
Implications of COVID-19 Disruption

on Small Island Economies
Mansur Afifi, Diswandi, Sulhaini, Tri Mulyaningsih, Djoni Hartono
and Budy P. Resosudarmo
INTRODUCTION
The unprecedented global outbreak of COVID-19 has profoundly disrupted
the world economy. The rapid spread of the virus and its uncertain economic
ramifications have posed challenges for policymakers attempting to craft
suitable responses (McKibbin and Fernando, 2021). While there have been
numerous negative economic outcomes, such as the deceleration of global
economic growth, the consequences have not been uniformed across countries
and regions. Indeed, the impacts have been so diverse that some regions have
observed certain positive effects. One notable example is the reduction in human
mobility, especially via motorised transportation like cars and planes. This has
led to a temporary enhancement of urban air quality in several major cities,
predominantly in developing countries (Pal et al., 2022; Galán-Madruga, 2023).
The diverse impacts of COVID-19 on the socio-economy and livelihoods can
also be observed in Indonesia. This is particularly evident given that Indonesia
is a vast archipelagic nation comprising numerous small islands. As of now,
academic studies assessing the impact of COVID-19 in Indonesia are relatively
accessible (Lewis and Witoelar, 2021). A number of literature not only covers
the issue of economic impacts (Olivia, Gibson and Nasrudin, 2020), but also on
agricultural activities (Halimatussadiah et al., 2022) and health (Resosudarmo
and Irhamni, 2021). However, most reports and academic inquiries focus on the
1
island of Java (Resosudarmo et al., 2021). Limited research has been conducted
on the regions outside of Java, especially on the smaller islands. Among the few
are by Afifi (2021) on COVID-19 impacts in Lombok, Suasih et al. (2021) in Bali,
Teniwut (2021) in Tual, and Girsang et al. (2021) in Ambon.
The Indonesian government, while providing some supports, has
encouraged economic actors to be creative in order to survive in the pandemic
COVID-19 era. Various efforts had been made by economic actors to improve
economic conditions while taking sustainability factors into account in the
framework of sustainable economic development. This book is an attempt to
capture some of the creativities taking place during the COVID-19 Pandemic
period, particularly focussing on what have happened in small islands and
regional cases in Indonesia.
CONTENT OF THIS BOOK

This book is organised into three parts: Part 1, Part 2, and Part 3, which cover
current issues on COVID-19 disruptions, small island economies and regional
development, respectively. Part 1 consists of four chapters—i.e., Chapter 2, 3, 4,
and 5—that discuss issues of mobility, economic resilience, urban farming, and
mental health related to COVID-19 disruptions.
Chapter 2, written by Naimah Lutfi Abdullah Talib, Ariane Utomo, and
Fransiska Limbengpiah, discusses small islands’ mobility during COVID-19
pandemic, using a case study from the Sangihe islands, a group of small islands
located in the north-eastern part of Indonesia. Focusing on the concept of
citizenship infrastructure, the authors argue that people immobility and their
mobility are products of social and political processes, and that they influence
and are influenced by the infrastructure of citizenship in, around, and beyond
Sangihe Islands. Through a mix of qualitative methods (individual in-depth
interviews, household surveys, focus group discussions and community mapping
exercise, and life stories), this chapter provides a comprehensive overview of
the nature of (im)mobilities in Sangihe islands and how COVID-19 pandemic
has reshaped the way people think of and do travel and how it contributes to
small frontier islands people’s perceptions of themselves as “distinct small
island communities” and “as citizens of Indonesia”.
Meanwhile, Chapter 3 investigates the relationship between economic
diversity and economic resilience in agglomeration areas in Java during the
2 COVID-19 Disruption, Small Island Economies and Regional Development in Indonesia

COVID-19 pandemic. In particular, the authors—Aloysius Gunadi Brata, Treesye
Siburian, Dinar Fandan Sari, A.M. Rini Setyastuti, and Andreas Sukamto—argue
that the economic diversity before the economic shock due to the COVID-19
pandemic is a possible factor to explain the different level of economic resilience
in these agglomeration areas. The chapter uses the Gross Regional Domestic
Product (GRDP) of 2019 and 2020 of the 43 regions in six agglomeration areas
in Java to create a local economic resistance index and a set of local economic
diversity indexes based on the Herfindahl-Hirschman Index.
The authors of Chapter 3 then regress the economic diversity indexes on
the resistance index of the local economy to examine their relationship. The
main finding of this chapter is that some economic sectors—such as information
and communication, and financial and insurance activities—support local
economic resistance, while the mining and quarrying sector gives the opposite
impacts. Furthermore, local economies that are more developed tend to be
more vulnerable during the pandemic. The chapter confirms the importance of
diverse economic activities in resisting economic shocks.
Meanwhile, Chapter 4 written by Catur Sugiyanto, Dion Saputra Arbi, and
Ahmad Syahrul Fauzi presents an analysis on the productivity of urban farming
in the city of Yogyakarta, using total factor productivity (TFP) as an estimate.
Based on a sample of 250 urban farmers in all 14 sub-districts in the city of
Yogyakarta, the authors find a variation in productivity levels among different
types of urban farming in the City of Yogyakarta. Fish production emerges as the
most productive, closely followed by urban farming focused on flowers; whereas
livestock and vegetable farming exhibit lower efficiency and productivity. The
authors further suggest that farmers engaged in fish and flower farming must
be provided with additional support and facilities. Simultaneously, training and
knowledge transfer for farmers engaged in livestock and vegetable farming
would be beneficial to improve their productivity.
In Chapter 5, Ni Wayan Surlastini, Fita Herawati, Ika Yulia Wijayanti,
Indrawan Firdauzi, Setyo Pujiastuti, and Sukamtiningsih put forward the issue
of mental health problem as a consequence of COVID-19 pandemic on teachers
and caregivers of early childhood education. Through statistical analysis using
descriptive analysis and logistic regression, they analysed the determinants of
anxiety and depression among 751 teachers and 500 parents in Special Region
of Yogyakarta. They found that the percentage of parents suffering from anxiety
and depression got higher, as compared to teachers.
Implications of COVID-19 Disruption on Small Island Economies 3

Higher education attainment is a protective factor against mental health
problem for both parents and caregivers. Although there is no significant effect
of emergency curriculum proxied by lesson plan on teachers’ mental health,
lesson plan utilisation by parents significantly reduced their probability to
experience mental health symptoms. Furthermore, difficulties in accessing
internet and technology increase teachers’ probability to experience mental
health problem, nonetheless, the availability of technical support helps to
reduce the risk.
Part 2 consists of six chapters—i.e., Chapter 6, 7, 8, 9, 10 and 11—that offer
discussions on development issues of local marine fisheries, renewable energy,
ecosystem services, eco-tourism, land property right, and mining in small
islands. In Chapter 6, Teguh Asandy and Wilmar Salim discuss marine fisheries
as a sustainable livelihood in Pulau Tiga, Natuna Islands. Using qualitative
data gathered from interview, questionnaire, and observations, they assess the
sustainability of fisheries through the lens of Sustainable Livelihood Framework.
They conclude that there needs to be an improvement in financial, social, and
natural capitals, with the help of external parties, to ensure the sustainability of
fisheries in the area. The main source of vulnerability is the seasonal monsoon
that last up to six months a year, which results in price uncertainties and limited
job opportunities. The lower interest of younger generation to be fishermen,
decreasing number of male residents, and decreasing growth rate of potential
fishery products per year also pose a threat to sustainability.
Chapter 7, written by Alan Dwi Wibowo and Andante Hadi Pandyasswargo,
discusses the appropriate renewable energy system in small and isolated
island communities. Using Pulau Burung island—small island village in South
Kalimantan province, Indonesia—as a case study, this chapter compares the
techno-economic performance of centralised versus distributed photovoltaics
(PV) systems by using the Hybrid Optimization Model for Electric Renewable
(HOMER) software. The chapter argues that the centralised system performs
better in terms of net present cost (NPC) and levelised cost of electricity (COE).
However, there are challenges in maintaining such a system, such as: (1) poor
community capacity to maintain the system; (2) unaffordability of battery
replacement and gasoline supplies; and (3) potentially lower transmission and
distribution efficiency. The result suggests that these problems can be addressed
through the creation of a social enterprise directly connected to the renewable
energy system involving the island’s multi- stakeholders.

In Chapter 8, Nyoman Priyamanaya and Wilmar A. Salim highlight the
shifting use of ecosystem services in Nusa Lembongan and Nusa Ceningan
islands, Bali, during COVID-19 pandemic. They argue that COVID-19 pandemic
has had a significant impact on local communities and has given rise to various
socio-economic problems, not less in regions dependant on tourism sector.
Using a combination of quantitative and qualitative data, the authors found
a shift in the use of ecosystem services in both regions—i.e., from cultural
services (tourism) to provisioning services (as indicated by the revival of
seaweed farming)—during the COVID-19 pandemic. This shift occurred quickly
and without major obstacles as the ecosystem had the capacity to do so. The
finding of this chapter serves as a good case study for the development of small
islands, particularly on the need to have an alternative livelihood to the less
resilient tourism that is susceptive to sudden disruptions.
Chapter 9 discusses the network of relationships between actors in
developing Bonjeruk National Ecotourism Village in Central Lombok regency.
Using the concept of the Triple Bottom Line and network of actor relationships
with descriptive qualitative research methods from primary data (observation
and interviews) and secondary data (literature studies), the authors—Mega
Nisfa Makhroja and Lalu Agung Haris Atmaja—explain the interaction between
the government, entrepreneurs, and non-governmental organisations in the
development of Bonjeruk Village as an ecotourism destination. The relationship
between the three actors is explained through three schemes—i.e., power,
interests, and sentiments—that lead to positive and negative interactions. The
analysis provided under this chapter explains why the development of the
village is arguably slow and how a better collaboration should be mandatory
in the long run. Most importantly, this chapter serves as a good case study for
policymakers in planning the development of ecotourism in small islands.
In Chapter 10, Wardis Girsang, Ali Magana, Adelina Siregar, Elia L.
Madubun, and L.O. Kakisina observe the relationship between land property
rights, income, and food security in Ambon islands, Indonesia. In their view,
insecure land property rights can lead to land conversion and environmental
degradation, which will endanger household food security accordingly. A
dataset of 139 households is analysed using SmartPLS3.3.3 to show that land
property rights and household income are the determinants of household
food security. The authors also find that household income is low due to low
productivity, which results from weak communal land property rights. They

suggest the need for farmers and fishermen to access financial loan, so that
the latter can upgrade their technology and attract next generation of farmers.
Furthermore, local communities are encouraged to strengthen their communal
land ownership through their local customary law and reduce dependency on
imported rice in order to increase resilience to external shocks.
Meanwhile, Chapter 11 examines local community perception of the
corporate social responsibility (CSR) initiatives from PT Newmont Nusa
Tenggara, a big mining corporation operating in the Nusa Tenggara Barat,
Indonesia. In this chapter, the authors—Baiq Nurul Suryawati, Burhanudin,
Zainal Abidin, Muttaqillah, Isti Fadah, and Intan Nurul Awwaliyah—employs
Structural Equation Modeling-Partial Least Squares (SEM-PLS) methodology
to evaluate the perception of Sumbawa Barat regency’s residents about their
satisfaction on the company’s CSR program and sustainability as well as
company commitment. Using the data of 500 villagers and 17 industry leaders,
the authors find that there is a relationship between community perspectives
and program sustainability, as well as a relationship between company
commitment and community satisfaction. However, no correlation between
community satisfaction and the lifetime of CSR initiatives is found. This suggests
that the perceptions of the viability of CSR projects do not always correlate with
perceptions of community satisfaction.
Part 3 that is comprised of three chapters—i.e., Chapter 12, 13, and
14—discuss local development issues of public finance, drinking water, and
microcredits, respectively. Chapter 12, written by I Kadek Dian Sutrisna Artha,
Usman, Dewinta I Sumartono, and Irwan Amarullah, investigate the correlation
between public financial governance and socio-economic performance of
Indonesia’s local governments. In particular, the chapter argues that the opinion
from the Audit Board of Indonesia (Badan Pemeriksa Keuangan or BPK) towards
local governments’ financial statements has an impact on local governments’
socio-economic performance—i.e., by using indicators such as income Gross
Regional Domestic Product (GRDP) per capita, Human Development Index
(HDI), poverty rate, and inequality. Using panel data of 514 districts and city
governments in Indonesia from 2015 to 2019, the authors find that regions
receiving better audit rating have better socio-economic performance. Based
on the estimation results, it is recommended that local governments improve
the quality of their financial management, as reflected in a better audit rating
from BPK, to enhance their socio-economic performance.

In Chapter 13, Adrian Chrisnahutama and Ni Made Sukartini look at the
major determinants of diarrhoea incidence among children under-five (U5) in
Indonesia. Using data from Indonesia Demographic and Health Survey (Survei
Demografi Kesehatan Indonesia or SDKI) from 2012 to 2017, the authors find
that higher quality of drinking water is significantly associated with lower
diarrhoea incidence, with piped water having the highest effect. In terms of
sanitation type, a flush toilet with septic tank is associated with lower cases
of U5 diarrhoea. Furthermore, the provision of various handwashing facilities
near households does not correlate with the low number of diarrhoea incidence
among children. Lastly, the proportion of open defecation in the community has a
positive correlation with diarrhoea incidence. This chapter, therefore, highlights
the importance of hygiene practice, preservation of natural environment, and
improvement of clean drinking water and clean sanitation.
Finally, Chapter 14 analyses credit channelling to micro, small and medium
enterprises (MSMEs) before and during the COVID-19 pandemic. In particular,
the authors—Mliitcyano Samuel Sapulette, Nury Effendi, Vera Intanie Dewi,
Fitri Hastuti, Eva Ervani, and Anhar Fauzan Priyono—argue that there is a
correlation between digital competitiveness of a region and the ability of MSMEs
in that region to access loans; and this relationship is likely to be stronger during
the pandemic as digital communication becomes more critical. Using monthly
province-level data from Indonesia, the chapter confirms that the level of digital
competitiveness of a province is significant in determining the amount of loans
MSMEs receive, even more so after the start of the COVID-19 pandemic. The
chapter, therefore, suggests the need for government’s intervention in regions
with low level of digitalisation, especially in two areas: digital literacy (especially
for entrepreneurs) and investments in ICT infrastructures.
REFERENCES
Galán-Madruga, D. (2023). ‘Urban air quality change02/19/24s resulting from
the lockdown period due to the COVID-19 pandemic.’ International Journal
Environmental Science and Technology, 20, 7083–7098. https://doi.
org/10.1007/s13762-022-04464-6
Halimatussadiah, A., R. Edwards, F.R. Moeis and R.F. Maulia. (2022). ‘Agriculture,
development and sustainability in the Covid-19 era.’ Bulletin of Indonesian
Economic Studies, 58:1, 1-30, DOI: 10.1080/00074918.2022.2056935

Lewis, B. and F. Witoelar (Eds.). (2021). Economic Dimensions of COVID-19 in
Indonesia: Responding to the Crisis. Singapore: ISEAS.
McKibbin, W. and R. Fernanco. (2021). ‘The global macroeconomic impacts of
COVID-19: Seven scenarios.’ Asian Economic Papers, 20(2): 1-20.
Olivia, S., J. Gibson and R. Nasrudin. (2020). ‘Indonesia in the Time of
Covid-19.’ Bulletin of Indonesian Economic Studies, 56:2, 143-174, DOI:
10.1080/00074918.2020.1798581
Pal, S.C., I. Chowdhuri, A. Saha, M. Ghosh, P. Roy, B. Das, R. Chakrabortty and M.
Shit. (2022). ‘COVID-19 strict lockdown impact on urban air quality and
atmospheric temperature in four megacities of India.’ Geoscience Frontiers,
13 (6): 101368. https://doi.org/10.1016/j.gsf.2022.101368
Resosudarmo, B.P., T. Mulyaningsih, D.S. Priyarsono, D. Pratomo, and A. Yusuf
(Eds.). (2021). Regional Perspectives of COVID-19 in Indonesia. Bandung:
IRSA Press.
Resosudarmo, B.P. and M. Irhamni. (2021). “Consequences of the COVID-19
Pandemic on Human Capital Development.” In B. Lewis and F. Witoelar
(Eds.). Economic Dimensions of COVID-19 in Indonesia: Responding to the
Crisis. Singapore: ISEAS, 170-189.
Afifi, M. (2021). “Impact of COVID-19 Pandemic on the Economy of West
Nusa Tenggara Province and Government Policy.” In B.P. Resosudarmo, T.
Mulyaningsih, D.S. Priyarsono, D. Pratomo, and A. Yusuf (Eds.). Regional
Perspectives of COVID-19 in Indonesia. Bandung: IRSA Press, 89-108.
Suasih, N.N.R., I.A.N. Saskara and P.Y. Wijaya. (2021). “Communitty Mobilities
in Indoneisa during COVID-19 Pandemic and the Case of Bali.” In B.P.
Resosudarmo, T. Mulyaningsih, D.S. Priyarsono, D. Pratomo, and A. Yusuf
(Eds.), Regional Perspectives of COVID-19 in Indonesia. Bandung: IRSA
Press, 71-88.
Teniwut, W.A. (2021). “COVID-19 Pandemic Risk toward Coastal Local
Economic Welfare of Remote Small Islands Region.” In B.P. Resosudarmo,
T. Mulyaningsih, D.S. Priyarsono, D. Pratomo and A. Yusuf (Eds.). Regional
Perspectives of COVID-19 in Indonesia. Bandung: IRSA Press, 183-200.
Girsang, W., A. Siregar, E. Jambormias and J. Siwalete. (2021). “Correlation
between the COVID-19 Pandemic and Poverty: A Case in Ambon City,
Maluku Province.” In B.P. Resosudarmo, T. Mulyaningsih, D.S. Priyarsono, D.
Pratomo and A. Yusuf (Eds.). Regional Perspectives of COVID-19 in Indonesia.
Bandung: IRSA Press, 183-200.

CHAPTER 2
Infrastructure for Mobilities and

Citizenship Making: Insights from
Sangihe Islands Indonesia in Times
of COVID-19 Pandemic
Naimah Lutfi Abdullah Talib, Ariane Utomo,
and Fransiska Limbengpiah
“Mobilities are never free but are in various ways always channelled, tracked,
controlled, governed, under surveillance and unequal...” (Sheller, 2018)
INTRODUCTION
Why should we care about the perception and experience of travel and
mobilities among small islands’ communities in times of COVID-19? Scholars
have long considered the politics of movement and mobility with a fundamental
understanding that mobility is not apolitical nor given. The ‘new mobilities’
paradigm coined by Sheller and Urry (2006) is helpful in exploring the
circumstances in which mobility can or cannot occur. This approach suggests that
mobility does not simply entail physical movement, but encompasses broader
social, economic, and political contexts at the local, national, transnational, and
global levels (Adey, 2006b; Cresswell, 2010; Pugh, 2013; Sheller and Urry, 2016).
For example, the politics underlying mobility shape the unequal provision of
the infrastructure for mobility. The infrastructure for mobility encompasses
immobile infrastructures that enable such mobilities, such as roads, bridges,
petrol stations, port management systems, airport, subway, and transport
9
technology, and even more broadly, economic conditions and legal systems,
such as passport and visa (Thrift, 2004; Milton et al., 2017). Inequalities in
the distribution of and access to infrastructure for mobility explain why certain
groups/individuals are ‘more’ mobile relative to others.
Scrutinising the social justice aspect of (im)mobility, Sheller and Urry
(2006) further questioned ‘who could move faster? More frequent? And more
comfortable?’ than the rest of the population. Consider the case of statelessness,
where the absence of infrastructure for mobility stemming from discriminatory
border policies results in immobility and containment (Milton et al., 2017).
Scholars have demonstrated how passports, visas, and citizenship status are
important elements of the infrastructure for mobility. Together, they play a
significant role as a marker of one’s legal status, and the absence of these
documents for asylum seekers and refugees signified the politics of mobility at
national and global scale (Milton et al., 2017). In a similar way, case studies on
climate change-induced migration in the Small Islands Developing States (SIDS)
in the South Pacific have suggested that, due to socio-economic and political
constraints both from the origin state and the destination state, the option
of ‘migrating with dignity and by choice’ was limited (Barnett and Waters,
2016; Barnett and McMichael, 2018). Thus, questions around who has access
to infrastructure for mobility could reveal who are (un)able to travel. These
examples demonstrated that mobility is shaped by and construct socio-political
processes, which in turn shape the differentiated ability of individuals and
communities to be mobile (Adey, 2006; Sheller and Urry, 2006, 2016; Cresswell,
2010; Leitner, 2017; Everuss, 2019; Adey et al., 2021). The issue of (im)mobility
is therefore a social justice concern, and a policy problem.1
While there have been discussions on everyday mobilities in urban,
migration and transportation studies (Dodge and Kitchin, 2004; Bissell, 2008,
2010), limited attention has been given to the experience and perception of
small island communities where sea-travelling is often restricted by weather
and seasonal events. We know little how specific geographical environments,
physical terrains, and infrastructural constraints influence people’s experience
and perception of mobility in small island communities. The COVID-19 pandemic
1 This chapter is part of the first author’s (Naimah Lutfi Abdullah Talib) PhD research on ‘Social justice
and the sea: Study of Indonesia’s Sea Toll in North Jakarta and Sangihe Islands’, which explores the
social justice dimensions and everyday lives and practices of the coastal and small island communities
and how the Tol Laut, as a product of political economy and social processes across scales, affecting the
lives of and interacting with the local communities.

has exacerbated the longstanding issue of mobility inequalities faced by small
island communities (Domínguez-Mujica, Parreño-Castellano and Moreno-
Medina, 2022). On the one hand, the challenges of (im)mobilities faced by
small-island communities during the pandemic has implications for their
subjective and physical well-being (Murphy et al., 2020; Salvacion, 2022).2
Moreover, a prolonged and heightened sense of immobility is likely to alter and
contest people’s sense of identity: how they see themselves as an individual,
as part of the islands communities, and national communities (Srebrnik, 2000;
Omondiagbe et al., 2021). This is a pertinent issue when thinking about the
idea, practice, and sense of citizenship in small island regions in an archipelagic
state like Indonesia.
This chapter aims to examine how geographical place and the COVID-19
pandemic have reshaped the way people think of and do travel/ stay put and how
it impacts wellbeing and further sense of identity. Sangihe islands in Indonesia
was selected as the study’ site due to its distinct geographical and geopolitical
location in the context of vast archipelagic state. As a frontier small archipelagic
district, Sangihe is seen as a strategically important place for protecting the
country’s geopolitical interests.3 The archipelago lays in the frontline sea
bordering with neighbouring countries and the Pacific Ocean (see Figure 1).
Sangihe (and Talaud) have been key posts for the Indonesian Navy to secure the
sea in the north-eastern of the country, including to protect the sea space and its
resources, particularly from Illegal, Unreported and Unregulated Fishing (IUPP)
frequently reported to be done by foreign vessels (Nadjib, 2015). Tatoareng
Sea of Sangihe is also regarded as Marine Protected Area contains rich ocean
biodiversity and potentially vast high-carbon storing seagrass fields, which
are not only strategic for advancing the country’s blue economy agenda (the
World Bank, 2021), but also showcasing Indonesia’s achievement in realising
the Kunming-Montreal global biodiversity framework (KKP, 2019, 2023). Thus,
our study of how geographical place and COVID-19 have impacted the Sangihe’s
2 Murphy et al. (2020) examined the effectivity COVID-19 outbreak control through containment
measures, e.g., limiting mobility, across the SIDS in the Caribbean regions. They argued that geographical
nature of small islands provides a “natural” advantage of containment, however, this came with costs
of elevated tensions between public health and economic-social wellbeing. Emphasizing the costs of
immobility, drawing from Geographical Information System (GIS) analysis of Marinduque Island in the
Philippines, Salvacion (2022) argues that being in a small island exposed communities to a larger risk
of COVID-19 transmission due to density in certain areas (town centre) and limited healthcare facility.
3 With these distinct geographical and spatial circumstances, abundant natural resources and mine
sources, rich cultural heritage and practices, Sangihe is increasingly recognised as an important place
by national, regional, and international communities.
Infrastructure for Mobilities and Citizenship Making: Insights from Sangihe Islands Indonesia 11
in Times of COVID-19 Pandemic
communities’ mobilities, wellbeing, and sense of identity, is underpinned by
both social justice and strategic policy rationales.
In the context of Sangihe’s distinct geographical and geopolitical
characteristics, this chapter asks and answers the questions of:
(1) How do communities travel and experience (im)mobilities during the
pandemic?;
(2) What is the role of the state in facilitating such travel?; and,
(3) What is the implication of (im)mobility to wellbeing and identity making?
Methods-wise, this chapter responds to the call for ‘small data’ to
complement ‘big data’ in the study of mobility and transport geography.4 We
conducted a series of remote and in-site fieldwork in Sangihe islands between
November 2021 and April 2022 to investigate how travel-related mobilities
affect people’s wellbeing and sense of identity.5 Given the COVID-19 pandemic
and its impact on mobility, we explored the multidimensional and relational
aspects of the embodied experience of human mobility, especially within the
port areas of Sangihe and on sea transport.
This chapter is structured as follows: the Introduction section lays out the
context of the study, followed by the analytical framework that will be used to
explain the findings. The third section discusses the findings from the fieldwork,
which are divided into three sub-sections, namely typologies of travel and
commute, sea travel and (im)mobilities, and the infrastructure for mobility and
becoming citizens. The discussion is embedded within the last two sub-sections
of findings. In the Conclusion, we draw out how the provision of infrastructure
for mobility is useful to understand citizenship making as an ongoing process
of identifying and maintaining sense of belonging to a nation-state and to
distinct small island communities. Finally, we conclude the chapter with food
for thoughts on how climatic and social uncertainties are part of social reality of
being and living in small islands and urge policymakers to continuously adapt
in providing infrastructure for mobility to facilitate wellbeing and maintain
sense of citizenship among small island communities.
4 Kwan and Schwanen (2016) calls for grounded small data to complement the emerging use of ‘big data’
(e.g., modelling, and statistical prediction on human mobility) in the study of mobility and transport
geography to provide a better understanding of context-specific circumstances, in this case in frontier
small islands.
5 Trialling of household survey’s instrument was conducted between November and December 2021 to
5 individuals; the first author participated through a phone call guided by the third author, who lives in
Sangihe. The first author conducted fieldwork in person, together with the third author, from March to
April 2022 when travel restriction was eased. We followed the COVID-19 spread prevention protocol
suggested by the WHO and the Government of Indonesia’s regulations throughout the fieldwork.

THEORETICAL FRAMEWORK
In this section, first, we introduce the new mobilities paradigm as a ‘beyond
movement’ approach to understand human mobility. Second, we explain how
the concept of the infrastructure of citizenship making, e.g., infrastructural
citizenship, is useful to explain the ongoing processes of place-making and
identity making among small islands residents.
Beyond Movement: Approaching Mobility Through Relational Lens

The concept of and empirical approach to study human mobility have been
constantly evolving. Initially, much of the global scholarships on human mobility,
including in Indonesia, primarily studied mobility through the conceptual lens
of migration as in the works of Lee (1966), Hardjono (1986), Tirtosudarmo
(2009), Adams and Cuecuecha (2010), Haas, Castles and Miller (2019). However,
the literature has since shifted to recognise human mobility as an ontologically
uncertain, dynamic, and complex reality.
The new mobilities paradigm provides a useful critical and heuristic
approach to understand mobility as part of social life, and thus should be
understood within its broader political and social contexts. Sheller and
Urry (2006) elaborate how mobility should be understood beyond physical
movement, which nudges scholar to think towards a more relational approach
in understanding human mobility. This means that while mobility is indeed a
‘physical’ or ‘material’ movement, it involves nuances that shape reasons, bodily
experiences, and the perception of the physical movement itself.
The new mobilities paradigm does not aim to provide a comprehensive
nor solid understanding in a once-for-all manner. Rather, it offers a heuristic
understanding of social reality. This paradigm acknowledges and embraces
uncertainty and contingency of the situations associated with social changes
and dynamics. Of this advantage, the new mobilities paradigm has been used in a
wide array of contexts, including international and internal migration, everyday
mobility/public transport, and mobility at home, such as in the kitchen or living
room (Bissell, 2010; Ahmed, 2019; Adey et al., 2021).
Mobilities are enabled or constrained by broader factors underlying the (un)
availability of transport and legal infrastructures, and human agency to travel
owing to social, economic, technological, cultural, and political circumstances
(Adey, 2006; Sheller and Urry, 2006; Cresswell, 2010; Pugh, 2013). Scholars have
also discussed which groups in the society exhibit the pattern of constrained
mobility consistently. Feminist scholars have provided critiques on mobility
inequality and how gendered norms constructed within patriarchal and
capitalist structures set different bodies and spaces (of freedom, of financial
ability, of stigma-free) and limit the right to- and access to- mobility. As Ahmed
(2014 p.428) puts it, “idealisation of movement, or transformation of movement
into fetish, depends upon the exclusion of others who are already positioned
as not free in the same way.” Cresswell and Uteng (2008) also highlighted the
importance of gender, mobility, and spatiality. Through laying out several cases
of mobility, from everyday daily transport, migration in Europe, and feminist
online networks among motherhoods, Cresswell and Uteng (2008) showcased
the dialectic conversations on the roles of gender, and the pattern of mobility,
on both the everyday and phenomenological gendered mobilities in cities in
Western Europe. This embodied experience of travel, including the unpleasant
micro aggression (e.g., inconvenient, and awkward stare, or hesitant gesture),
constituted the everydayness of embodied traveling for migrants women of
colours in subway, bus, or interstate-rails (Cresswell and Uteng, 2008).
Small islands have its particularities in terms of geographical position,
physical terrains, and social identities (Steinberg, 2013; Arnall and Kothari,
2020; Baldacchino, 2020), which should not be ignored to understand the
complexity and dynamics of human (embodied) mobility. Small island’s
geographical location and physical terrains, namely being surrounded by the sea,
and some being located afar from the ‘mainlands’ where supply and resources
are often derived from, and dependent on the weather situation, contributed to
physical vulnerability, but also agency and adaptability, of living in small islands
(Baldacchino, 2010). A study by Hampton and Jeyacheya (2020) across 12
Small Islands Development States (SIDS) demonstrated that SIDS tend to have a
high dependency towards tourism-economy, which might not be a sustainable
driver of economic development and limiting their capacity to adapt to climatic
changes.6
6 We do not fully agree with the paper’s fatalistic conclusion: SIDS will be in economic recess and
disappear due to sea level rise. Yet, the key point of the paper remains relevant: the lack of options
for economic development are limiting SIDS pathways towards economic, environment, and social
sustainability.

Human Mobility and the Making of Identity
Seeing how it is deeply embedded in the political, economic, and cultural
dimensions of human life, we argue that mobility is closely linked to place-
making and identity making in the small-island communities. We understand
place-making as a set of social processes by which space is made useful and
meaningful. Place-making is distinguished by the unique meaning that human
associated with them and specific uses to which they are put. In the making of
place, place is not necessarily equal to administrative boundary, instead, they can
be fluid, such as “rivers” that are crossing several districts/ suburbs/ countries
(Martin, 2003; Hutchison, 2010; Antonsich, 2018; Akbar and Edelenbos, 2021;
Kärrholm et al., 2023). To further explore the social processes involved in the
meaning and use of space, we must also consider the concept of identity making:
the ongoing social processes by which individual and the society perceived
oneself/us and the “other.” Critical scholars argue that identity making is imbued
by social, political, economic processes, and, influenced by the colonial’s world
view and values. This legacy of colonial’s world view was elaborated through
case studies of India’s lower castes, e.g., Sudra and Paria, which are legacy of
the British colonial era, and gender as well as ‘ability/disability’ binary system
(Crenshaw and Maschke, 1997; Lacapra, 2016; Arnall and Kothari, 2020; Boda,
2022).
Place has a profound space in identity making. One’s sense of “us” and
“the others” are often imbued by one’s perception of place. Scholars argue
that islands, as a place, are dynamic and constantly being constructed and
reinterpreted by humans e.g., humans give meaning to islands, and dialectically
islands give conceptions and feelings to humans (Brinklow, 2015; Arnall and
Kothari, 2020; Omondiagbe et al., 2021). Through their study in Maldives,
Alex Arnall and Umma Kothari (2020) proposed the concept of “making of
islandness” to understand the everyday practices of small island communities
in managing waste to make ‘home’, which was influenced by non-human
factors, such as sea tides, wind, shoreline, in the face of ecological and climatic
changes.7 This exercise of agency is directly related to their identities as small
island communities and how they would like to be perceived by others (p.896).
7 They elaborated that “island is continually being made through the flows and connections afforded
by waste, and by the human and non-human actors that incessantly interact and intermingle with it’
(Arnall and Kothari, 2020 p. 902).
The making of islandness concept is fluid, and recognises individuals and
communities’ agency, which was underpinned by a shared culture, social ties,
and history rooted in the island’s ecosystem, which echoed by other scholars
such as Baldacchino and Kelman (2014) and Gugganig and Klimburg-Witjes
(2021). Thus, we expand the use of this concept to human mobility, while
focusing the analysis on the infrastructure for mobility and how they interact
with identity making.
Human mobility is closely related to citizenship and is neither predetermined
nor static. Rather than viewing citizenship as mere legal status, we understand
citizenship as a dynamic experience that is constantly made, remade, negotiated,
and contested (Marston and Mitchell, 2004; Ho, 2009; Lemanski, 2020). The
social aspects to understand citizenship has received more attention in the past
two decades, including how social relationships, emotions, and social positions
and status play critical roles in making of citizenship (Marston and Mitchell,
2004; Nyamnjoh, 2007; Ho, 2009; Lemanski, 2020; Sultana, 2020). Drawing on
Singaporean migrants in the UK, Ho (2009) illustrates the complex relationship
between emotions, generated through everyday interactions and encounter,
and sense of belonging and acceptance in defining one’s citizenship in place
away from home.8 Furthermore, the provision/or lack of thereof public services
could indicate a state’s acknowledgement/neglect to those populations (Harvey
and Knox, 2015; Anand, Gupta and Appel, 2018), which affects one’s sense of
citizenship, or what is called “infrastructural citizenship” (Lemanski, 2020). To
further illustrate, Sultana (2020) illustrates how the provision of affordable and
clean water to urban settlement in Dhaka through formal/ state-led services
and informal/ private- and mafia-led services are ways of renegotiation of
urban-citizenship.9
8 Ho proposes a concept “emotional citizenship” to explain the social construct of ‘citizenship’ as

everyday identity making and practices; and to complement legal perspective of engagement with
citizenship concept. Ho made an analytical distinction between emotional representation and
emotional subjectivities, which both are interconnected and constitute “citizenship” of Singaporean
migrants in the UK.
9 Sultana (2020) argues that the provision of formal and orchestrated private sector-led drinkable water
services symbolises urban citizenship and acknowledges urban dwellers as Metro Dhaka-jurisdiction
citizens. This study echoed similar studies done by Brian Larkin (2018), and also Nikhil Anand on
‘networked infrastructures’ (2011, 2018) on book titled The Promise of Infrastructure. Anand argues
that provision of water supply services is a marker of state’s acknowledgement towards certain
population in post-Apartheid South Africa municipalities, or what he called as ‘hydraulic citizenship’.

Transportation plays a critical role in shaping and maintaining identity of
islandness; yet some argues that it may pose a threat to their distinct identities.
Circulation of goods, people, capital, ideas, and culture facilitates the ongoing
processes of making sense of place, and thus identity as a small islands’ residents.
Through a comparative case study of local artists in Newfoundland Canada and
Tasmania Australia, Brinklow (2015) highlights how various factors, including
‘openness’ and ‘high connectivity to the mainland’ influenced a sense of place
of small islands and this contributes to sustain cultural distinctiveness of island
identity. Conversely, a case study from Waiheke small island in New Zealand by
Omondiagbe et al., (2021) suggests that openness in terms of connectivity and
social-political influences from the ‘mainland’, is an ‘exogenous infiltration’ that
threatens the unique identity of small island communities. Thus, connectivity
and transportation as nodes of infrastructure for mobility play a critical role in
the (un)making of island identity. What is the case for Indonesia?
SITUATING SANGIHE ISLANDS: STUDY SITE AND METHODS

Situated at the northern tip of Sulawesi and administered under North Sulawesi
Province, Sangihe islands are an archipelagic frontier in northeast Indonesia
(Figure 1). The islands consisted of 105 islands, of which 26 are inhibited; the
main island is Sangir Besar (Figure 2), where most population lived, and the
local government offices located.
Figure 1. Site location: The Archipelagic district of Sangihe (in Green),
North Sulawesi Province, Indonesia.
Source: Map prepared by authors.
The district is known for its tourism and marine conservation areas and
is one of the Indonesia Navy’ main posts covering the North-eastern sea of
Indonesia (Antaranews, 2021), which ocean-bordering with the Philippines, the
Papua New Guinea (PNG), and few SIDS in West Pacific ocean. The closest urban
area to Sangihe district is Manado, which is the capital city of North Sulawesi
Province.
In terms of political landscape, Sangihe’s dominant formal political power
reflected the national politics, whose District Chief is the member of Indonesian
Democratic Party of Struggle (PDI-P), that is one of the key ruling parties in
Indonesia (i.e., the President Joko Widodo’s political party). Of 25 seats in the
Local Representatives (DPRD), PDI-P and its current coalition, Golkar, both
secured 5 seats respectively, followed by their coalition-party Nasdem (4 seats);
while limited seats filled in from other parties Demokrat, Hanura, and Perindo
(Antaranews, 2019) and all of these parties are either opportunistic-based
coalition to the ruling coalition or simply ineffective opposition to them.
Figure 2. Fieldwork sites in four hamlets in the archipelagic district
of Sangihe (in Red)
Source: Map prepared by authors.

As the majority population uphold Christianity, churches play key roles in
social organisation and governance of everyday practices, such as social events
(particularly funeral and wedding), distribution of wealth, humanitarian relief,
and disaster risk management. Civil society organisations are vibrant, from
culture-based to social and environmental-underpinned movements; and few
are led by women. While they have been there for long, civil society organisations
start to gain national and international media traction after sequences of
community protests over the mine concession of 42.000 Ha (or over half of the
island) that was given to a multinational-backed mining company (Kompas.
com, 2022; Saturi, 2022). Thus, Sangihe local politics is a proximate reflection
of what happens at the national level.
The Tahuna Port serves as Sangihe’s primary connection to the “outer” world.
Although Tahuna Airport is available, it is less accessible and used compared to
the port. In 2016, the Tahuna Port was expanded as part of the national flagship
programme on logistics and transport connectivity, known as Tol Laut (Sea Toll).
The National Master Plan on Ports (Ministry of Transportation, 2017), which
was developed by the National Planning Agency, the Ministry of Transport, and
the Coordinating Ministry of the Economy, categorised Tahuna Port as a “local
supporting port” (Pelabuhan Pengumpan Lokal). This classification means that
the port supports two main ports (Pelabuhan Utama) in the North-eastern
region: the International Port of Bitung for fisheries and logistics; and the Port
of Manado for transport and logistics (Ministry of Transportation, 2022).
To explore the ways in which (im)mobilities shape identities, this study
uses a mix of qualitative methods. These methods include 42 individual in-
depth interviews (including boat/ride-along and walk-in port interviews), 32
household surveys, 6 focus group discussions (FGD) and community mapping
exercise, and life stories.10 Autoethnography was applied through the use of
researchers’ travel diary, and visual/audio-recordings to gain insights and
nuances. The study involved 65 participants, including 25 male and female
participants who took parts in multiple data collection methods, spread in
Tahuna and Beeng Darat island (Figure 2). Insights from in-depth interviews
with local government officials, including the District Planning Agency, the
Transport Agency, the Trade and SMEs agency; and Commercial passenger’ ship
10 Prior to this fieldwork, the first author lived in small island of Beeng Darat in Sangihe for 14 months
in 2012-2013 teaching at a primary school. The third author is a Sangir, who was born and raised in
Sangihe, and currently teaching at a primary school there.
crew, and Technical Unit of Port Management (Syah Bandar/ UPT Pelabuhan II
Tahuna) were used to triangulate data from interviews, surveys, and FGDs with
communities.
TYPOLOGIES OF TRAVELLING AND COMMUTING IN THE

SMALL ISLANDS
This section discusses how the local communities on Sangihe Islands perceived
physical mobility to other islands within the district, to the main island of
Sulawesi, and to the neighbouring archipelagic district of Talaud. We identified
four common patterns of mobility among Sangir communities. First, travel
route within islands, from one island to another, as well as from the island to
Sangir Besar in the mainland of Sangihe (as shown in Figure 3). Second, travel
route from Tahuna to Manado, the mainland of Sulawesi. Third, travel route to
and from the outermost islands of Talaud, Marore, Matutuang or Kawio (the
last three islands mentioned are administratively under Sangihe district, but on
“the outer” ring). Lastly, while we observed transnational sea travel to and from
the southern the Philippines by both by Indonesian and Filipino citizens, this
study focused on mobility within the national border. For participants in this
research, traveling to the Philippines was less significant compared to inter-
islands within the Indonesian border in terms of relevance and frequency.11
11 Sea travel to and from the southern of Philippines were reported by communities in Tahuna and in
outer islands, such as Marore, Matutuang, Kawio. Filipinos and Indonesians who travel there were
mostly medium-scale fishers and crews. To access basic services, all participants of this research,
citizens of Indonesia, reported to travel to Manado, as discussed in this section.

Figure 3. Typologies of Travelling and Commuting of Sangir Communities
within the Indonesia Border
The first pattern of travel route, characterised as a “commuting,” involves

traveling within “inner” ring of Sangihe archipelago and is typically done with
no special preparation. This type of travel allows for a one-day round trip and
requires only a small amount of cash (or none), water, snacks (or none), without
specific preparation (such as clothing) needed (denoted by A in Figure 3).
The second pattern of travel involves traveling from Sangihe to the mainland
of Sulawesi, such as from Tahuna to Manado; or from smaller islands to Tahuna
and then to Manado. This is considered as “big travel” and can take anywhere
from one to four days depending on the origin island (denoted by B in Figure
3). For those living in Marore, Matutuang, Kawio and other outermost islands,
travelling to Manado is perceived as a major journey that requires weeks to
months of planning, and up to a week of travel time, with Tahuna serving as a
transit hub (combining Pattern C and Pattern B in Figure 3).
The third pattern (denoted by C in Figure 3) describes travel routes from
Tahuna or other parts of Sangihe to the outermost islands of Talaud. This type of
travel requires specific preparation and anticipation of the uncertainties before
and on board, such as food for the journey, appropriate clothing, and sufficient
cash along with good health condition. Although the cargo-passenger ship,
Tol Laut- Sabuk Nusantara 95 (Figure 4), is available and follows on regular
schedule, the factual schedule is erratic due to unpredictable weather, wind,
and tidal waves conditions. As a result, people must call the Port unit or ask
their relatives to check whether the ship has docked in the Port of Tahuna. The
information of ship availability is also broadcast through the state-led local
Radio of the Republic Indonesia (RRI Tahuna) and is spread by words of mouth
in public spaces, such as wet markets, churches, mosques, offices, hospitals,
and schools. Sabuk Nusantara usually dock for 2 to 3 days before departure
following a certain route that eventually reaches Talaud after three days of
travel by sea with few stops.
Figure 4. Sabuk Nusantara 95
Photo credit: Christian Manangsang (2022).

Note: Transporting Basic Goods and Passengers from Tahuna to Smaller
Frontier Islands within the District and to Talaud District
Among the three patterns of travel routes within nation-state border,
patterns A and B are the most common and familiar trips for most community
members in Sangihe. When probed about their travel experience, the majority
of participants have experienced traveling from Tahuna to Manado (denoted by

B in Figure 3).12 When people from smaller islands want to travel to Manado,
they usually come one or two days before the trip and stay with their relatives
in Tahuna, or they ask their relatives to buy the ticket one day before the travel.
This practice is common, especially for those coming from the outermost islands,
where traveling from their islands to Sangir Besar takes 4 to 8 hours by sea,
and in some parts is completed by land transport. This travel can be extremely
exhausting, so they rest for a day or two before departing for Manado, which
takes around 7 to 8 hours, known as “the big trip”. While the pandemic has not
brought significance differences in traveling within the islands, it has affected
their travel to Manado.
A recent study was conducted to understand personal preferences for
choosing a fleet provider to travel from/to Sangihe - Manado (denoted by B
in Figure 3). According to Letunaung, Timboeleng and Lefrandt (2021), 370
respondents who are frequent passengers to and/from Sangihe-Manado
consider three aspects when selecting providers, namely waiting-time, ticket
price, and safety and cleanliness inside the fleet. Yet, less is known on about who
actually travels? and why they choose to travel, especially during the pandemic.
SEA TRAVEL AND (IM)MOBILITIES

When discussing (im)mobility during the pandemic, it is crucial to explore the
key drivers of travel to understand who travels, and who does not, and why. By
understanding the layers of (in)ability to travel, we can unravel the complexity
of vulnerability and resilience of the communities residing in small islands.
Two factors were found to accentuate the perception of “urgency”: rational
economic calculation and the perception of “distance.” These factors were
found to constitute a level of perceived risk that informed speculative decisions
to go or stay put. While people are aware of the risk of being exposed to the
coronavirus during the trip and in the destination place, some still decided to
travel for several reasons. First, individual health issues are one of the driving
factors that induced human mobility, creating a sense of urgency to travel
despite the perceived high-risk situation (both in terms of health problem itself
and the risk of coronavirus exposure) and stricter human mobility regulations.
12 To gauge communities understanding of ‘travelling’, the author promoted with, “what was your
experience of travelling in the past two years?”.
Such situation was explained blatantly by a female participant (aged in her
mid-60s), a neighbourhood shop owner in Tahuna. She and her daughter went
to Manado during the peak of the pandemic due to her daughter’s pregnancy.
She explained,
“My daughter and I went to Manado during the peak of the pandemic as she
needed help with her pregnancy. The unborn baby was found to be tangled with
the placenta, and the doctor here (in Tahuna) told us to seek for better medical
services in Manado (- in the main island of Sulawesi). We did not wish to go as
it was in the middle of the (pandemic) situation, but we had to. We lived there
(in Manado) for 3 months in a rent-one room flat. I brought a small-stove and
coconut oil to cook ourselves so we can save some money.”
Figure 5. KM Merit Teratai in Port of Manado
Photo credit: Naimah Lutfi Abdullah Talib (2022).

Note: It took 8-10 hours from Manado to Tahuna, depending on weather
and the speed of the passenger’ fleet. Wooden stair is used to get into the
ship.
The Sangihe Statistics Agency noted a range of 25 to 74% decline in
the number of people travelling to and from Manado during the peak of the
pandemic from March 2020 to August 2021, especially prior to the introduction
of vaccinations (BPS Sangihe, 2020). However, the number gradually increased
throughout 2021, returning to a normal rate, after the gradual introduction
of the first COVID-19 vaccination. These data were confirmed during our
fieldwork and were especially apparent among older adults who only travelled
for emergency health or economic reasons.

The peak of the pandemic was understood by individuals as a temporal
window when the number of affected people increased (as reported in the
mass media). This was then followed by a stricter travel regulation, such as
requirement of rapid antigen test, mask, and proof of vaccination, as well as the
presence of health officials in the ports. A former village chief (Kapita Laung) of
Beeng Darat Island who is also a lifelong traditional fisherfolk (aged in his mid-
70) mentioned that he and his wife went to Manado to treat his coronary artery
disease during this period. He considered the hospital treatment expensive and
even more with additional rapid test that costs IDR 165.000 per person (AUD
17). He added that people in Beeng Darat, a smaller island, south-eastern of
Sangir Besar, could barely keep money in cash. To him, it became even hard
when passengers were required to wear mask all the time on the fleet, as it felt
suffocating.
For most male across age groups, economic reasons due to financial
hardship provide adequate push factor for traveling during the pandemic.
People of productive age, particularly male breadwinner, travel to the main
island of Sulawesi or to the northern-most islands of Sangihe as medium-
scale fishing fleet crews (known as the ABK for Pajeko ‘kapal Tuna’ or ‘kapal
Cakalang’) or engaged in construction work for infrastructure projects, such
as ports or tidal wave breakers. The uncertainty of situation and financial
hardship, a combination of fluctuation of fish’ prices, weak purchasing power
of the communities, and unpredictable weathers, have pushed them to travel
to places they perceived as ‘far away’ from home. However, for a few, financial
hardship has become a factor that prevents them from travelling during the
pandemic. People can only travel when they have a minimum level of cash; even
for people who pushed to travel for work.
On a different occasion, when someone within the household provides
adequate earnings (either spouse, kids, or grandkids), they decide to stay put
and fish. For ones that are able to stay put, they optimise the use of their existing
productive capital, such as fish with their adult male member of household that
stay put, using traditional boats, both motored-katinting or paddled-dayung.
The ship’s fare, extra cost for rapid antigen, and extra time-use from undergoing
health check in the ports, have kept them from traveling. Thus, having some
earnings has lessened the ‘urgency’ to travel for work during the pandemic. In
a way, at the least, they still have option to stay put. Most people decided to
postpone travel the reason is deemed as ‘not urgent’,
Approximately 70% of local community participants (42 persons), who
travelled in the past two years, said that while it was tiring and worrying, they
somehow still enjoyed the ship ride for its ocean environment; and the fleet
design facilitates such activities. Most participants who travelled said that the
upper deck and the side deck were nice spots where they could feel the ocean
breeze.
The idea that open space with good air circulation is an ideal place for
preventing the spread of the corona virus in fact resonates with the design of the
fleet (Figure 6). These alternative spaces, open decks on the upper, front, and on
the sides of the fleets, have provided more microplaces for people to enjoy the
sea trip, particularly during the pandemic. As explained by male participant in
his late 50s,
“When the weather is good (teduh – described as ‘not windy, not raining, and
soft waves’), I liked to stare at the sky, the stars. This is why it is fun to travel by
sea… (he took a while to continue, slightly smiled)... You do not have to wear mask
outside, so I could breathe easily... I could stretch my legs and walk on the upper
deck.”
Figure 6. The Upper Deck of the Fleet

Hand sanitizer is another material that is consistently mentioned by
participants of all ages, gender, and occupations as “the extra preparation” that
passengers had to have before sea travel, both to Manado and to frontier smaller
islands. When using toilets, before eating, passengers explained that they will
spray/ apply hand sanitizer to their palms. They also tended to prepare food
from home, which otherwise, they purchased on the fleet just before departing
(Figure 7).
Figure 7. KM. Saint Marry (No. 376) Bed Used as “Temporary Shop”

Note: Type of bed on the KM. Saint Marry (No. 376) that was used as
“temporary shop” selling snacks, packed-meal, bottled water, cigarettes,
black coffees, and candies.
While some of them started to apply hand sanitizer out of habits, other
shared that they only sprayed/applied it on specific activities that are perceived
to bring risk of exposure to corona, such as before eating or sleeping, after
went from the toilet. Hand sanitizer and face mask are mentioned by most
participants as the two-shared standards of ‘self-protection’ to passengers from
the “corona virus” within participants across age, gender, and occupations.
Applying/spraying hand sanitizer, limiting interaction with other passengers
and putting on masks have become new set of practices and rituals of sea travel
in times of pandemic, signifying individual measure to manage public health
risks.
On the commercial night passengers’ fleet (150-200 GT cargo and
passengers’ fleet), such as Saint Merry, KM Barcelona, Merit Teratai, people
of all ages and gender, tried to follow the rules within the fleet. However, the
idea of using a face mask as protection was understood differently. While some
people rigorously used face mask all night long on the fleet for fear of getting
infected by the SARS-CoV-2 virus, the others (both younger people and older
adults) only covered their mouths or took off their mask entirely when they
were sleeping.
The fleet as a place during the trip has also reshaped passengers’
perception on “public” and “private” space. The perception of public and private
space was blurred due to the design of the indoor cabin on the fleet. The idea
of public space where passengers must follow the rule, and the idea of private
space where they can exercise a degree of autonomy and freedom were blurred
due to the setting of the interior of the ship. Passengers define their individual
bed in the cabin as “private space” (Figure 8), while the washroom/toilet and
entertainment area is defined as “public space”(see Figure 9). Since there is no
clear separation among these spaces, the boundary between private and public
was undermined.
Figure 8. (Blurred) Private and Public Space

Note: Inside the boat from Manado to Tahuna (Round trip, three
times a week).

Figure 9. Public Space

Note: The toilet with a separate hand wash basin area.
Inside the cabin, the arrangement of the bunkbed made it difficult for
passengers to maintain 2-meters physical distance, particularly in the economy
class, where some bunkbeds are fixated to each other. This caused conflicting
emotions—i.e., tired of being on ship for 8 hours, fear of getting exposed by
the virus, and pasrah (at a desperate level of accepting the situation) of having
to travel and follow the rules. While there is no officials-on-the ship that
monitoring the situation, the captain of the ship/ferry through speakerphone
aired regularly to all cabins and outdoor part of the fleet, always reminded
passengers to put their face mask-on when on board, and to follow the COVID-19
control and prevention/health protocol.”
The different rituals between pre- and during the pandemic on fleet were
observed. One participant noted that prior to the pandemic, the leader of the
common prayer usually broadcasted generic good weathers wishes, e.g., “the
fleet, crews, and passengers were protected and blessed to reach the destination,
granted with good weather along the trip.” The additional prayer’s message
during the pandemic—i.e., to keep all passengers, captain, and crew healthy—
indicates an acknowledgement of such health risk to the people on the fleet.
Temporal dimension is also shaped the way people perceive travelling,
including the risk to travel during the pandemic. Sangihe has on-and-off face
mask’ requirement and rapid test regulation for travelling during the pandemic.
During the peak time, marked by peak number of people affected, the local
authority implemented a set of protocol, such as requirements to have a rapid
antigen test, face mask, and a copy proof of vaccination through the state
organised mobile app, i.e., Peduli Lindungi. By the end of 2020 and then mid-
2021, participants shared that they only went traveling for urgent reasons such
as health problems and work, mostly to Manado or frontier smaller islands
in Northern Sangir Besar. For those who must travel for work, and mostly are
male participants along the aged 30s to 60s, compromising risk is inevitable.
One of them considered uncomfortable to be snuck by the stick on the nose/
rapid (translated into Manado language: mau dicucu kita pe idung, mau apa jo
koa) during the pandemic. Yet, he had to do it due to his construction work in
Kawaluso, and then in Matutuang islands.
Interaction among passengers were also influenced by perception of risk
of exposure to the virus. Passengers were aware that virus could be transmitted
through conversation with other people who were not wearing face mask
properly and within a short physical distance. According to a female respondent
on her late forties, owner of a small restaurant in Tahuna, she used to enjoy
talking with my neighbour bed (ranjang sebelah), but now, she could only say
‘hello’ with no further conversation.
The notion of interaction in a closed- or semi opened-space that constrained
passengers to exercise “keep 2-m distance” has influenced their decision to
limit interactions with other passengers. However, few participants, mostly
male, through interviews said that they do not have issue nor feeling afraid of
getting exposed by the virus and still talk to other passengers in the upper deck
(open space). They did not feel afraid of being infected by COVID-19, went on
the upper deck to smoke, and talked to people as usual.
(Un)familiarity as An Embodied Experience

Subjectivity of perception of distance is apparent in this case, which is
influenced by physical miles-distance and familiarity of travelling experience.
While Manado-Tahuna is approximately 250 km and takes 8-10 hours by sea, it
is considered as “less far” compared to travel to Melonguane (the capital city of
Talaud), which is approximately 160 km from Tahuna. This is partially explained
by the time for travel—i.e., travelling to Talaud takes 10 hours to 3 days due to
its routes that requires few stops in several islands, and unpredictable weather
conditions. It is also explained by the physical experiences during the trip, which
were shaped by the unfamiliarity towards the exterior and interior of the ship.
According to a male participant from Melonguane/Talaud, it is the arrangement

of the bed, the situation inside the fleet (Kapal Perintis), and even its shape and
size that made the travel feel even longer. In fact, Talaud is geographically closer
to Tahuna, but communities perceived travel to Talaud as ‘big trip’ and felt ‘far
away’ compared to Manado.13
In present times, air travel remains the most uncommon way of traveling
for small island communities in Sangihe and it shapes the conception of
physical experiences of air traveling. Out of 65 participants, and more than 100
hours of casual chats with people from various ages, gender, and occupations,
only three who had used airplane in the past five years to go to Manado. They
were all civil servants (local government officials and teacher). According to
female respondent (mid-20s) who works as an airport staff, mostly government
officials who use the aeroplane (to travel). Sometimes, there are guests and
businesspeople from Manado or Jakarta or other places who take aeroplane.
The frequency of aeroplane service and its tight regulation become key
reasons of not having air travel as an option in the first place. There is only
one aeroplane provider flying Tahuna-Manado (Wings Air), twice a week, every
Wednesday and Saturday—compared to sea travel which is available twice a
day (morning and evening trip). The rigidity of aeroplane schedule was deemed
as “less compatible” with the Sangihe communities’ situation and preference.
Their erratic schedules, on top of weather uncertainty, were hindering them
to engage with a “time-lock” schedule. This particularly apparent when they
travel from smaller or outer islands. Meanwhile, they feel to have options when
traveling by sea.
From cultural perspective, travel by air was also problematic. Air travel
has a fix regulation of a list of things that passengers are allowed to bring and
its quantity. In comparison, compared to air travel, sea travel is more flexible
for individuals due to baggage allowance that enables them to bring bulky stuff,
while sizeable baggage in air travel is strictly prohibited.
Limited information to access such services and ‘formality’ treatment at
the airport also intensifies unfamiliarity experience with this way of travel.
A teacher on his end of 40s considered travel by plane was (an) interesting
experience. It is an experience that comes once in lifetime. However, he might
13 This sensation was also felt by the first author who went inside the Sabuk Nusantara fleet and
attempted to be ‘present’. First author inhaled, smelled, walked on, sit on, touched, and try to ‘feel’ the
interior boats. This results in the feeling of travel to faraway places, the infamiliarity of being in such
a huge fleet, the design of Bunkbeds, toilets and lack of entertainment space made the fleet feels like
semi-military boat.
not know how to start—i.e., where to buy the ticket, how to get in the plane. He
had to ask the former village chief who once went to Jakarta for training and
had this experience (air-travel). Exclusivity makes airport feels different with
seaport, despite both has a clear entrance and waiting room. He then explained
that all family members could only accompany until the front waiting room and
were not allowed to see him off to board, which was different with sea travel,
where people could stay and wave hands as the fleet departs.
In addition, travel by air is expensive, and financial constraint contributes
to limit the options. With a rate of IDR 700.000 – 980.000 per trip per
individual, traveling by air is not an option for most community members.
They prefer sea travel as they could take the whole family with that amount
of money (e.g., aeroplane fare) to from Tahuna to Manado or Melonguane (the
capital city of Talaud). In this case, it is a combination of lack of cultural/ social
appropriateness, limited access to information, and financial constraint that
shaped and heightened the unfamiliarity of travel by air.
Similar to perception of ‘distance,’ which was constructed as a combination
of physical miles-away from “home” and subjectivity of (un)familiar embodied
experience of traveling; level of ‘urgency’ accentuated and/or constrained by
rational economic-reasons and perception of distance. Level of urgency has
also shaped the way small island communities perceived risk to travel during
the pandemic; and often, in a circumstance considered as ‘urgent,’ risk was
compromised.
A combination of factors—i.e., level of urgency, perception of distance and
(un)familiarity—have informed the speculative decision to travel or stay put
until ‘situation gets better.’ People who travelled did not feel 100% safe from
being exposed by the SARS-CoV-2 virus despite practicing health protocol as
religious as possible, such as wearing mask, washing hand/ applying hand
sanitiser, and limiting interactions with other passengers in the indoor cabin.
Yet, by doing these rituals, they felt ‘more secure’ and ‘protected’ by applying
hand sanitizer often, before eating and after using the toilet. They felt less
guilty if they get exposed, as they have done what they could, while realising
should they be infected by the virus somewhere, it should be beyond their
control. Perception of being able to manage risk better is shown by maintaining
health protocol, which was an important aspect of people who travelled, and
contributed to the speculative decision to travel, even though not being sure
what will happen during and from the trip.

BECOMING CITIZENS THROUGH INFRASTRUCTURE FOR
MOBILITY
This study contributes to the idea of human mobility as a package of ‘beyond
movement’ experience. Perceptions of risks, needs, and potential gains were
reflecting the speculative decisions to travel during the pandemic. The COVID-19
pandemic can be seen as temporal conjecture in modern-history of human
lives, which provides opportunity to rethink the way in which small islands
communities travel.
Through the lens of new mobilities, Sheller and Urry (2006) provide
a relational approach to understand why and how people travel. From our
fieldwork, we can draw insights that participants who travel during the peak
of the pandemic was pushed from economic and health reasons. The ideas of
‘emergency’ and ‘urgency,’ emerged from the need to earn some cash and to treat
illness, forced people to travel, despite the long distance (feeling of tiredness),
costs (monetary and time), and risk of exposure to COVID-19. On the other
side, the provision of public transport, both operated by the state, such as Tol
Laut, or facilitated by the state, such as commercial cargo/passengers’ ferries,
has facilitated the urge to travel. Pandemic, then a public health issue, now has
become part of social reality, has become another crucial element to factor in to
have a safe and convenient sea trip, including to have rapid antigen test, wear
face mask, and apply hand sanitisers. In parallel, the pandemic has exacerbated
(im)mobility of some most vulnerable groups in small islands community, such
as older adults, sick people, and pregnant women with health conditions.
The corona virus itself, as tiny material organisms, has changed the way
human see travelling and perceive its urgency. While a set of COVID-19 spread
prevention protocol has been put in place and communicated trough several
channels, lack of support and access to (free) rapid antigen test and the rigidity
of the ship-design arrangement, have made it hard for people who really need to
go but unable to because of financial reasons and hesitant from the anticipation
of very uncomfortable set of design arrangement on the ship.14
14 Interior design of the ship as shown in Figure 8 limits personal space and challenges the idea of
“comfortable travel” i.e., parallels bunkbed that are close to each other and relatively small size. If
passengers bring many bags or luggage, they often put it on their bed, and reduce the space for bodies
to rest.
Travelling as a form of human mobility that can facilitate or hinder the
objectives of reaching wellbeing, such as social, economic, and health related
outcomes. Participants who travelled during the pandemic in the past two years
were on the move to maintain social relationships, to work, and to obtain better
health services. This insight resonates with other studies that demonstrate how
the pandemic heightened existing vulnerability by hindering ability or access
to other services (Witoelar and Utomo, 2022). Our study shows how existing
hardships (financial and health) interact with variables emerged due to the
pandemic, such as additional travel requirement, which results in additional
layers of barrier to travel and exacerbates vulnerability.
Maintaining social relationship and identity as a ‘mobile’ small island
community, is one of the key drivers to travel, while knowing the consequences
(the hassle, extra tiredness, fear, or anxiety from exposure to the virus) of
travel by sea from Tahuna to Manado. In several cases, participants shared that
their relatives have an understanding on their absence on wedding and burial
ceremonies, two major social events among North Sulawesi communities. As
for casual vacation, 63 out of 65 participants said they waited until the situation
gets better or did not visit at all until the time of writing. Only two older adults’
participants (above 65 years old), who perceived it was still too risky, due to
their age and health conditions, to travel for casual social events during the
pandemic.
At the same time, living in a small-frontier island but staying mobile is
deemed to be the ideal situation, particularly for the younger generations. Being
able to study in Manado and able to go back home (Sangihe) during holiday
makes them proud as a Sangir, and contributes to wellbeing. As RM, on her early
20s, a university student, explained,
“It’s desirable to be able to study here (in Manado) and going back to Sangihe
during university break. I could meet my family and friends at home. My friends
and lecturers here are also interested with our communities, cultures, history, and
places. It would be great to be able to take them to Sangihe, to Beeng.”
The infrastructure of citizenship making could also be explained through
the provision of private commercial passenger fleets with Tahuna-Manado
route, where citizenship is frequently mediated through the materiality of
infrastructure (Lemanski, 2020). This chapter has extended the discussion

on citizenship making through the provision of fleets as the material of sea
transport that enables connectivity between frontier small island with the
mainland of Sulawesi, which offers perceived better economic, educational, and
social opportunities. From the passenger’s perspective, it not only enables them
to serve their travelling purposes, which contributes to wellbeing, but also as
a form of recognition of citizenship. Based on household survey conducted in
four hamlets to 32 participants, the provision of regular, affordable, and reliable
logistic and passenger fleets, for instance Sabuk Nusantara 95 (under the Tol
Laut programme), made them feel more recognised as the citizen of Indonesia,
despite some islands in Sangihe is closer to the Philippines than to Manado,
let alone the capital city of Jakarta. The idea that the central government’
programme is reaching the outermost islands have consoled, or other way,
performed a ‘socially acceptable’ political performativity to the frontier small
island communities. This act of “serving” done by the state is important in
maintaining a sense of citizenship among Sangihe
Understanding what citizenship means for the people can provide insights
on how and why action to pursue and maintain sense of citizenship need to be
taken (Kabeer 2006). This question around citizenship making is relevant in the
case of Sangihe communities as they are geographically located in the frontier
and interacting directly with the state apparatus, who protect the sea border,
in everyday basis. In fact, state apparatuses are often part of the community
themselves, or to some degree, have assimilated into the community (as one
said, “they now can speak Sangir” or “So Sangir Opo itu”).
Based on in depth interviews and hundred hours of informal conversation,
the provision of consistent supply of daily necessities (sembako, such as rice, oil,
sugar, flour) is seen as a form of recognition that the state takes care of its citizens
in the frontier islands, despite its effectiveness to reduce the daily necessity
price that remains in question. Knowing that being in remote islands means
dealing with the constant uncertainties from weather and tidal waves, the effort
to offer ‘order’ and ‘a degree to certainty’ through state-led service provision
is appreciated by the communities and the local governments. These efforts
to provide a constant supply, particularly of necessities, in which construction
materials are included, are deemed as an act of the central government supports
to the local government’s development plan.
According to the Secretary of Local Planning Agency, they felt recognised
and serviced by the State due to the consistent provision of daily necessities
and materials. The stable frequency and the quantity of supply needs (muatan)
are crucial for supporting the regional medium term development plan. He
explained,
“We felt recognised and cared by the consistent provision of daily necessities
and materials for infrastructure development (such as cements, iron, etc) through
Tol Laut, including Sabuk Nusantara (for passenger transportation and logistics
to Talaud, the northest district in Indonesia). (This) provision of stable, in terms
of frequency and quantity supply of needs (muatan), especially materials that we
cannot produce ourselves in Sanger, is crucial for supporting the regional medium
term development plan.”
Physical infrastructure as an enabling, and signifier of, citizenship making
is a very useful concept to make sense of the way small islands communities’
perception and acceptance of government’ programme of Tol Laut (logistics
and transports from Tahuna to smaller norther islands within Sangihe district),
despite no proof of that the relatively stable price of daily-necessities in some
regions in Eastern Indonesia could not be attributed to Tol Laut (House of
Representative, 2022; ERIA 2022).15 For the people in Sangihe, the provision
of Tol Laut aside from commercial ships as an option for travelling to the
northernmost islands in Sangihe (with very affordable fare IDR 16.000 per trip
or AUD 2), has created a sense of being taken care of and feeling of being seen
and heard. Thus, provision of physical infrastructures (the fleets, ports, subsidy
for the trip fare) was seen as a form of recognition to small island communities
in the frontiers as distinct Sangir communities and Indonesian citizens as well.
This finding confirms similar studies on networked infrastructures (Larkin,
2013; Anand, Gupta and Appel, 2018; Sultana, 2020), where provision of services
by the state was seen as a form of recognition to communities as citizens.
Improvement of the programme was also mentioned several times,
reflecting local communities’ hopes and aspirations. The Head of Local
Transport Agency also added that Sabuk Nusantara, while has provided a
15 However, without attempting to support the programme, the situation is far more complex that
this chapter could possibly addressed. This need a deep understand why Tol Laut alone could not
be expected to reduce price of basic necessities as per the Presidential Regulation No. 27 of 2021,
particularly in the frontier, remote and small regions (daerah 3TP). To understand why, despite Tol Laut
is in place, the price of basic necessities could not be reduced, we need to understand a holistic supply-
chain mechanism; global commodities supply and prices; local commodities and its comparative
advantage; and where Tol Laut sits within these overlapped factors.

significant contribution to small island communities, needs to be more frequent
to cater the demand of transport not only to the main island of Sulawesi, but
also to Talaud. Many people in Sangihe have family in Talaud, the outermost-
Northern district in Indonesia. When they hear family members have passed
away, which is important to their culture to attend the ceremony (duka), they
cannot always be there due to the inconsistency of the frequency and timing
of Sabuk Nusantara’s services. However, they also understand that weather is
another crucial factor in affecting the service punctuality.
Transport agency also explained, which confirmed the findings from the
local communities, that sea travel is not only the most affordable way of traveling
to the main island (to Manado) and to Talaud (the neighbouring district), but it
is also perceived to be the most convenient and familiar for most people. This
is in fact the Maritime Culture (Budaya Bahari) of Sangir people that affects the
preferences of traveling by sea, from air or land, aside of economic calculation.
This finding was resonating with some participants who shared their story,
including 65 years old fisherfolk in Tahuna, who got carsick when he travelled
by car, and he was too afraid to travel by plane. Besides, he considered travelling
by air extremely expensive (e.g., IDR 700,000-1Million per person). Therefore,
sea travel is considered the most convenient way to travel and most affordable
too.
This is where the practice of travelling can contribute to the ongoing process
of place-making, which in the small islands’ context, is the making of islandness.
Places are connected through human and non-human materials (through waste
encounter and waste management), and islands, as place, are continuously
in the making (Arnall and Kothari, 2020). Communities encountered with
physical needs and feelings to travel and how they encountered with the ship
and regulation to sea travel during the pandemic, have shown a collective
and individual agency to manage and deal with the situation, which in turn
contributes to the place/ island-making as their ‘home’ and part of the ‘bigger
home’ (e.g., Indonesia).
The making of islandness can further contribute to the ongoing processes
of identity making (Pugh, 2013; Brinklow, 2015; Arnall and Kothari, 2020).
Rather different from the argument that “external infiltration”, through
decision-making from the outside, that has undermined identity of small island
communities (Omondiagbe et al., 2021), our study found that the identity
of Sangir communities as the residents of a distinct small island was not
undermined by their identity as part of the archipelagic nation of Indonesia.
These multiple identities—i.e., being Sangir as a distinct community and as part
of Indonesian citizens—co-exist and are performed in different ways, which are
often juxtaposed in subtle ways. This was illustrated by the speech of the church
leader on the fleet—i.e., while prayers to have a safe sail in the ocean are deemed
to be ‘very Sangir’ guided by Church leader,16 the provision of time and space to
inform passengers to follow Protokol Kesehatan based on the Government of
Indonesia’s regulation is a form of acknowledgement towards them as part of
the country. The ship on the sea, as an infrastructure for mobility, has been a
place that facilitates the process of maintaining of multiple identities of Sangir
communities.
CONCLUDING REMARKS
This chapter has given a broad overview of the nature of (im)mobilities in Sangihe
islands and how COVID-19 pandemic has reshaped the way people think of and
do travel and how it contributes to small frontier islands people’s perceptions of
themselves as “distinct small island communities” and “as citizens of Indonesia”.
In doing so, we have laid out three findings. First, we described the typologies
of travel and commute of small islands communities. Second, we discussed
how the sea travel was perceived and experienced by the communities, and the
circumstances under which (im)mobilities occurred. Third, how infrastructure
of mobilities in turn contributed to the infrastructure of citizenship making,
e.g., maintaining sense of citizenship and distinct identities of the communities
in the frontier islands. These elaborations have demonstrated that mobility is
indeed could be better understood as processes shaped by social, economic,
and political systems and circumstances (Sheller & Urry 2006). In broadening
our understanding of ‘circumstances’ in which mobility is shaped and operates,
we should consider constants uncertainties as part of the picture.
16 The first author always got goosebumps when the prayer was held and aired throughout the ship,
“Bapak di surga, kami mohonkan perjalanan kami ke Manado hari ini diberi kemudahan dan kelancaran.
Semoga Kaptain dan Awak Kapal Merit Teratai diberi kemudahan dan kesehatan dalam menjalankan
tugas, mengantar kami dengan selamat sampai ke Manado. Semoga angin dan cuaca memberikan
ketenangan dalam perjalanan kami. Semoga seluruh penumpang, kaptain dan awak kapal diberikan
Kesehatan, dan kami bisa berkumpul dengan keluarga. Amin.” [based on audio recording, first author’s
travel diary].

Dealing with Constant Climatic and Social Uncertainties
The pandemic becomes an additional factor and in parallel has heightened the
visibility of the constant uncertainties in sea travel for small islands communities
due to uncertain climatic condition, which is inherent part of being and living in
small islands in open sea. Pandemic, as in material viruses of SARS-CoV-2 and
its mutants, and imaginaries of fear and worry constitutes perception of risk
to travel during the pandemic. The ‘pandemic’ consistently described as “when
the number of cases is high as reported by the media” or “when it is trending in
the news.,” has been becoming another layer of perceived risk, aside from the
weather.
Weather, which consists of wind and tidal waves, has been a crucial aspect
that is factored in household and individual level decision making processes to
travel or stay put for generations. Communities describe it as bad weather with
strong wind-blowing/Angin Barat or good weather/Teduh to travel. During bad
weather, they would rather postpone traveling, unless it is extremely urgent,
such as seeking medical help or working purposes. Weather has been part of
Sangir people’s lives, it is on everyday conversation and folklore, as well as news
and regulations.
While the availability of private commercial ships is celebrated among
the communities, and the subsidized-Tol Laut ship helps bring necessities
and passenger to Sangihe-outer islands and Talaud (cargo/passenger fleets),
there is still question around who are unable to travel, not by choice, but due
to a set of constraints, such as health, economy/financial, and social factors.
Question on the access and the right to affordable transport becomes central to
understand the complexity of provision of transport, especially in the context of
small islands in the frontier islands.
These forms of provision—i.e., the state being present in providing the
services and/or managing the private operators to step in—must deal with
uncertainty itself, uncertainty that is derived from the climatic situation
(weather, tidal waves, wind), and social aspects—such as the pandemic and its
implication to norms and the way people perceived travel—as well as financial
uncertainties induced by pandemic and its spill over to people purchasing power
and employment opportunities. Cresswell (2010) argued that (im)mobility
should be taken seriously as it is reflected, influenced by political structure and
social relations, through mobility as movement, representation, and practice.
This chapter has demonstrated under what conditions such mobility is enabled,
and for some others, constrained or disabled.
Furthermore, there might be other invisible (vulnerable) groups or
individuals, who urgently need(ed) to travel but was not able to due to absent
of such infrastructures for mobilities, such as inability to access vaccinations
or Rapid Antigen Test; lack of support for people in need of extra care during
travelling; or economic and social constraints. This inability that creates
immobility (not by choice) becomes a social justice and policy issue, in terms of
questions on the right to mobility and the right to have access to basic services
in which transport facilitates.
The role of transports—particularly sea transport in the context of small
islands facilitated by the state—in facilitating people to access social, health,
and economic opportunities and services is pivotal. Through the use of
relational approach, that is adopting new mobilities paradigm (Sheller & Urry
2006), our study indicates that sea transports, both formal-informal or private-
public led, are parts of the enabling ecosystem for small islands communities to
access broader range of services and opportunities, which in turn contributes
to identity making as a small island communities and citizenship making as part
of Indonesia.
Finally, echoing other scholars who advocate that small islands are
significant ontological categories of place, this chapter has demonstrated
small islands as a geographical space, imaginaries, and meanings-making are
indeed significant in shaping social realities, in particularly in the context of
vast archipelagic nation-state, such as Indonesia. In this ongoing and dynamic
processes of the making of islandness as a spatial-unit and imaginaries,
communities must deal with constants uncertainties, such as, to mention a
few, weather, tidal waves, and public health concerns, slowdown economic
situations, which may complicates the idea of (im)mobility and aspiration of
such mobility, connectivity, and citizenships, both present and in the future.
ACKNOWLEDGMENT
We would like to thank Melbourne Research Scholarship for the provision of full
doctoral scholarship for the first author, and to Melbourne Sustainable Society
Institute for the Research Seed Fund 2021, which has enabled this fieldwork to
happen. The authors also would like to thank village chiefs and communities

in Tidore, Apeng Sambeka, and Beeng Darat for their time and kindness to this
‘student from Java who speaks poor Manadonese’; Colleagues from Politeknik
Nusa Utara Sangihe (Dr. Walter Balansa, Christian Manansang, Joddy Kakambong,
and Meita Toppol) for our intriguing discussion and assistance in the field;
Ria and Gabriel Dareho for showing me around and making good memories
during my short visit in Tahuna; Dr. Dedi S. Adhuri and Prof. Frans Gruber
Iyong for recommendation pre-fieldwork readings on Sangihe communities’
social, culture and history; Prof. Brian King for stimulating discussion during
our mentoring session and your encouraging feedback to our initial draft (and
Science Leadership Collaborative programme who facilitated the first author
encounters with Brian); and to Sakaeng Solata community coffeeshop (Stenly
Takarandehang and Stenly Pontolawokang) as well as Radio Republik Indonesia
Tahuna Pro 2 (Julfan) for their welcoming and kind gestures, and opportunity to
share this outsider view of the mobility and connectivity of small-frontier islands
communities (via RRI Tahuna radio broadcast). Lastly, first author would like to
thank all first author’ former students, who came from Manado, Bitung, and
different parts of Sangihe to say hello, and blessed me to see you grow.
REFERENCES
Adams, J. & A. Cuecuecha (2010). The Economic Impact of International
Remittances on Poverty and Household Consumption and Investment
in Indonesia. Rochester, NY. Available at: https://papers.ssrn.com/
abstract=1683904. Accessed on 13 January 2023.
Adey, P. (2006). ‘If Mobility is Everything Then it is Nothing: Towards a Relational
Politics of (Im)mobilities.’ Mobilities [Preprint]. Available at: https://doi.
org/10.1080/17450100500489080.
Adey, P. et al. 2021. ‘Pandemic (Im)mobilities.’ Mobilities, 16(1): 1–19. Available
at: https://doi.org/10.1080/17450101.2021.1872871.
Arnall, A. & U. Kothari (2020). ‘Becoming an island: Making connections and
places through waste mobilities.’ Transactions of the Institute of British
Geographers, 45(4): 891.
Barnett, J. & E. Waters (2016). ‘Rethinking the Vulnerability of Small Island
States: Climate Change and Development in the Pacific Islands.’ In J. Grugel
& D. Hammett (eds). The Palgrave Handbook of International Development.
London: Palgrave Macmillan UK, 731–748. Available at: https://doi.
org/10.1057/978-1-137-42724-3_40.
Bissell, D. (2010). ‘Passenger Mobilities: Affective Atmospheres and the Sociality
of Public Transport.’ Environment and Planning D: Society and Space, 28(2):
270–289. Available at: https://doi.org/10.1068/d3909.
BPS Sangihe. (2020). Badan Pusat Statistik Kabupaten Kepulauan Sangihe.
Available at: https://sangihekab.bps.go.id/pressrelease/2020/07/02/
289/perkembangan-transportasi-laut-dan-udara-sulawesi-utara-
mei-2020.html. Accessed on 6 January 2023.
Cresswell, T. (2021). ‘Valuing mobility in a post COVID-19 world.’ Mobilities,
16(1): 51–65. doi:10.1080/17450101.2020.1863550.
Domínguez-Mujica, J., Parreño-Castellano, J.M. & Moreno-Medina, C. (2022).
‘Pandemic Mobility/(Im)mobility in the Canary Islands: irregular migrants
becoming hotel guests.’ Belgeo, 3. Available at: https://doi.org/10.4000/
belgeo.56022.
ERIA (Economic Research Institute for ASEAN and East Asia), 2022, ERIA
project Report No. 24 on Maritime Highway and Indonesia Development:
https://www.eria.org/uploads/media/Research-Project-Report/2021-
24-Maritime-Highway-and-Eastern-Indonesia-Development/Maritime-
Highway-and-Eastern-Indonesia-Development.pdf
Gugganic, M. (2021). Island Imaginaries: Introduction to a Special Section.’
Science as Culture, 30(3): 321–341. Available at: https://doi.org/10.1080
/09505431.2021.1939294.
Hampton, M.P. and Jeyacheya, J. (2020). ‘Tourism-Dependent Small Islands,
Inclusive Growth, and the Blue Economy.’ One Earth, 2(1): 8–10. Available
at: https://doi.org/10.1016/j.oneear.2019.12.017.
Hardjono, J. (1986). ‘Transmigration: Looking to The Future.’ Bulletin of
Indonesian Economic Studies, 22(2): 28–53. Available at: https://doi.org/1
0.1080/00074918612331334814.
KKP (2019). KKP mereviu Kawasan Konservasi Perairan Sangihe / Ministry
of Marine Affairs and Fisheries reviewed the Sangihe-Tatoareng Marine
Protected Areas’ evaluation report. Available at: https://kkp.go.id/djprl/
artikel/13367-kkp-evaluasi-penetapan-kawasan-konservasi-perairan-
daerah-kkpd-tatoareng-kepulauan-sangihe-sulawesi-utara. Accessed on
19 March 2023.

KKP (2023). KKP Targetkan Penambahan 200 Ribu Hektare Kawasan Konservasi
di Tahun 2023, Ministry of Marine Affairs and Fisheries Directorate
General of Sea Space Management. Available at: https://kkp.go.id/kkhl/
artikel/49892-kkp-targetkan-penambahan-200-ribu-hektare-kawasan-
konservasi-di-tahun-2023 Accessed on 28 March 2023.
Kompas.com (2022). ‘Blokade Jalan Tolak Alat Berat Perusahaan Tambang
di Sangihe, Warga dan Polisi Saling Dorong Halaman all.’ Kompas.com,
16 June. Available at: https://regional.kompas.com/read/2022/06/16/
085213378/blokade-jalan-tolak-alat-berat-perusahaan-tambang-di-
sangihe-warga-dan . Accessed on 15 January 2023.
Kwan, M.-P. & T. Schwanen (2016). ‘Geographies of Mobility.’ Annals of the
American Association of Geographers [Preprint]. Available at: https://www.
tandfonline.com/doi/abs/10.1080/24694452.2015.1123067 Accessed
on 6 May 2022.
Letunaung, J., J.A. Timboeleng, & L.I.R Lefrandt (2021). ‘Analisis Tingkat
Pelayanan Transportasi Laut Dengan Pengguna Jasa Pada Pelabuhan
Manado (Studi Kasus: Manado - Tahuna).’ Jurnal Ilmiah Media Engineering,
11(1). Available at: https://ejournal.unsrat.ac.id/index.php/jime/article/
view/36596. Accessed on 9 May 2022.
Ministry of Transportation (2017). Master Plan of National’s Ports System.
Rencana Induk Pelabuhan Nasional. Legal Repository System of the
Ministry of Transportation of the Republic of Indonesia. JDIH | Kementerian
Perhubungan. Available at: https://jdih.dephub.go.id/produk_hukum/
view/UzFBZ05ETXlJRlJCU0ZWT0lESXdNVGM9. Accessed on 2 January
2023.
Ministry of Transportation (2022). Sistem Informasi Pelabuhan, Port Information
System. Available at: https://simpel.dephub.go.id/index.php/outerapp.
Accessed on 2 January 2023.
Murphy, M.M. et al. (2020). ‘COVID-19 containment in the Caribbean: The
experience of small island developing states.’ Research in Globalization, (2),
100019. Available at: https://doi.org/10.1016/j.resglo.2020.100019.
Nadjib, M. (2015). ‘Penangkapan Ikan Ilegal di Laut Kawasan Perbatasan
Sangihe: Dari Londe ke Pumpboat’, Jurnal Ekonomi dan Pembangunan,
23(1): 25–38. Available at: https://doi.org/10.14203/JEP.23.1.2015.25-38.
Nyamnjoh, F.B.( 2007). ‘The Disillusioned African.’
Omondiagbe, H. et al. (2021). ‘Out of character: Reiterating an island’s
imaginaries in the face of a changing identity.’ Journal of Marine and Island
Cultures, 10(1). Available at: https://doi.org/10.21463/jmic.2021.10.1.06.
Pugh, J. (2013). ‘Island Movements: Thinking with the Archipelago - EBSCO’,
Island Studies Journal, 8(1): 9–24.
Salvacion, A.R. (2022). ‘COVID-19 Susceptibility Mapping: A case study for
Marinduque Island, Philippines.’ Spatial Information Research, 30(5): 563–
570. Available at: https://doi.org/10.1007/s41324-022-00444-7.
Saturi, S. (2022). ‘Kala Warga Sangihe Tuntut Cabut Izin Tambang Emas dan
Desak Bebaskan Robison.’ Mongabay.co.id, 29 November. Available at:
https://www.mongabay.co.id/2022/11/29/kala-warga-sangihe-tuntut-
cabut-izin-tambang-emas-dan-desak-bebaskan-robison/. Accessed on 15
January 2023.
Sheller, M. & J. Urry. (2006). ‘The New Mobilities Paradigm’, Environment and
Planning A: Economy and Space, 38(2): 207–226. Available at: https://doi.
org/10.1068/a37268.
Sheller, M. & J. Urry. (2016). ‘Mobilizing the new mobilities paradigm’, Applied
Mobilities, 1(1): 10–25. Available at: https://doi.org/10.1080/23800127.2
016.1151216.
Tirtosudarmo, R. (2009). Mobility and Human Development in Indonesia.
Available at: https://mpra.ub.uni-muenchen.de/19201/. Accessed on 13
January 2023.
Urry, J. (2016). Mobilities: New Perspectives on Transport and Society. London:
Routledge.

CHAPTER 3
Does Economic Diversity Help

Economic Resilience of the
Agglomeration Areas in Java during
the COVID-19 Pandemic?
Aloysius Gunadi Brata, Threesya Siburian, Dinar Fandan Sari,
AM. Rini Setyastuti, and Andreas Sukamto
INTRODUCTION
The current COVID-19 pandemic, as argued by Coulson et al. (2021), is a spatial
phenomenon and suggests the crucial resilience of economies to shocks,
indicating an ability of a system to anticipate, absorb, accommodate, or recover
from the effects of a shock event (Martin, 2012; Martin & Sunley, 2014; Martin et
al., 2016; Mayor & Ramos, 2020). Ascani et al. (2020) also show that the spread
of the disease is also driven by the local specialisation in economic activities
which are spatially concentrated. According to Ascani et al. (2021), people’s
mobility within and across local labour markets is the mechanism of spatial
diffusion of the pandemic, which then can strongly affect local labour resistance.
They show that local market areas (LMAs) with industrial districts were more
vulnerable to COVID-19, indicating the density of systemic interactions within
LMAs depicted by firms operating within an industrial district. It supports the
finding by Gereffi (2020) in which COVID-19 is more likely to affect sectors with
physical proximity and workforce groups. Hence, COVID-19 can affect the entire
economy; nevertheless, the effects are not always negative and may vary across
sectors or industries.
45
The above studies indicate that spatial units which have concentrated
economic activities should be differently evaluated in understanding the
implication of the pandemic as well as the resilience of these units during the
pandemic. The agglomeration area is an example of interesting spatial units
during COVID-19 pandemic since they tended to have a severe infection, including
in developing countries. In Indonesia, a seroprevalence survey in November-
December 2021 found that the inhabitants of the agglomeration areas have a
higher antibody to the virus (CNNIndonesia, 21/3/2022) confirming that these
areas, especially in Java, were seriously affected by the pandemic. Besides,
agglomeration areas also become an important target of government policies in
dealing with the spread of the virus which has been also applied in Indonesia.
Handayani et al. (2022) show that it is important to evaluate the spatial
distribution of COVID-19 at the scale of urban agglomerations in explaining key
aspects of transmission and policy.
Analyzing the 2020 data, Handayani et al. (2022) find that there were
various interventions in the agglomeration areas in Indonesia, however only
interventions that were consistently implemented that have resulted in great
impact in stemming the transmission of COVID-19. Similarly, Aritenang (2022)
finds a higher number of average daily infections across regions in Java are
observed among large and core-metropolitan cities as they have a higher
economic connectivity and mobility of commuters within metropolitan districts.
One example of the Government of Indonesia (GoI) policies to control
the transmission of COVID-19 which is designed for the agglomeration
areas is the Regulation of the Minister of Transportation No. 25/2020 about
the transportation regulation during the holiday of Eid al-Fitr 1441H. This
regulation points out that the prohibition also applies to the agglomerations
which have been stated as large-scale restriction areas, but it does not explicitly
mention the name of the agglomeration areas. However, in the following years,
the GoI clearly stated that people can move, without strict permits, from one
location to another location if the moves are in the same agglomeration area
during the holiday of Eid al-Fitr 1442H (Kompas, 16/04/2021), even no need to
show the result of COVID-19 test to move during the Eid al-Fitr 14423H holiday
(Kompas, 04/04/2022). Eight agglomeration areas included in this study are
Medan metropolitan area (Medan-Binjai-Serdang-Karo/Mebidangro) in North
Sumatera, Makassar metropolitan area (Makassar-Maros-Sungguminasa-
Takalar/Mamminasata) in South Sulawesi, and the rest are those located in Java.

The latter cover Jakarta metropolitan area (Jakarta-Bogor-Depok-Tangerang-
Bekasi/Jabodetabek) including some cities/districts of Banten and West Java,
Bandung metropolitan area (Bandung Raya) in West Java, Semarang metropolitan
area (Kendal-Demak-Ungaran-Semarang-Purwodadi/Kedungsepur) in Central
Java, Solo Raya, Jogja Raya, and Surabaya metropolitan area (Gersik-Bangkalan-
Mojokerto-Surabaya-Sidoarjo-Lamongan/Gerbangkertosusila) in East Java.
We focus on Java since most agglomeration areas in Indonesia are in
Java. According to Mulyaningsih et al. (2019), Java contributes at least four
agglomeration areas, namely Greater Jakarta, Greater Bandung, Greater
Semarang, and Greater Surabaya. The agglomeration areas in Java also play
an important role in manufacturing in which manufacturing firms tend to
locate their plants in more populous and densely populated areas to gain
localisation economies as well as urbanisation economies (Kuncoro 2003). It is
not surprising that during the pandemic, Java also had a high infection rate of
COVID-19. Therefore, it is important to study how the economy of these areas
responds to the devastating effect of the current pandemic. The contribution
of Java to the national economy is also large. Based on the 2019 BPS data, 59%
of the Indonesian economy is concentrated in Java, while in 2018 and 2020 is
58.48% and 58.75%, respectively (Resosudarmo & Irhamni, 2021).
Although studies on the regional or local economic resilience during
COVID-19 are growing, however they rarely refer to Indonesia, moreover to
agglomeration areas in Java. Among others are studies by Purwandari et al.
(2022), Mujiyanto et al. (2023), Desdiani et al. (2022), and Maftukhah & Pramono
(2022). Purwandari et al. (2022) proposed a new method in measuring city
economic resilience, but they do not cover the period of COVID-19. Mujiyanto
et al. (2023) described fiscal policies, monetary policies as well as policies to
restrict people mobility to maintain economic resilience at the national level.
Meanwhile, Desdiani et al. (2022) focused on local budget resilience in which
they found that provinces heavily dependent on intergovernmental transfers
struggling with countercyclical mitigating fiscal policies, while those with high
local own-source revenue suffered from reduced tax revenues and constrained
fiscal space due to the crisis and low economic activity. It was only Maftukhah
& Pramono (2022) who investigated the regional economic resilience by using
data at the provincial level with focus on the effect of structural change both
before and after COVID-19. They found that prior to the pandemic, an increase
of sectoral productivity could encourage regional resilience. A similar result
Does Economic Diversity Help Economic Resilience of the Agglomeration 47

Areas in Java during the COVID-19 Pandemic?
was also found during the COVID-19 shock, suggesting the significant role of
both flexibility and responsiveness of industry and its employees. The finding of
Maftukhah & Pramono (2022) reflects the benefits of economic diversity, as an
indicator of economic structure (Tan et al. 2020), amidst the pandemic. Recent
studies also showed that economic diversity is one of the possible explanations
for economic resilience during the COVID-19 crisis (Gong et al. 2020; Tan et al.
2020, Hu et al. 2022).
Unlike the previous studies mentioned above, this chapter focuses on
regional economic resilience and its relationship with economic diversity,
with specific emphasis on agglomeration areas in Java. Due to the limited
period of data during the pandemic, this study aims to measure the short-
term economic resilience based on Gross Domestic Regional Product (GDRP)
in six agglomeration areas in Java during the current pandemic covering 43
districts and cities. Resistance is one of dimensions of resilience represents the
sensitivity of a regional economy to disturbances and disruptions (Martin, 2012;
Martin et al., 2016). Specifically, we also examine whether economic diversity
in the pre-COVID-19 period has a strong correlation with the level of economic
resilience. It is hypothesized that economically diversified regions have greater
economic resistance than regions with a less diversified structure during the
crisis caused by COVID-19. Therefore, this study contributes not only to the
literature of regional economic resilience during a shock, namely COVID-19,
but also helps provide empirical evidence on the role of economic diversity on
regional economic resilience from agglomeration areas in a developing country.
The study also has policy relevance to anticipate other shocks as pandemics
may lead to relatively long-lasting transformations of cities and regions (Florida
et al., 2021).
LITERATURE REVIEW ON REGIONAL ECONOMIC

RESILIENCE
According to Martin (2012), resilience has four dimensions. First, resistance that
represents the vulnerability or sensitivity of a regional economy to disturbances
and disruptions, such as recessions. Second, it refers to the speed and extent
of recovery from such a disruption. The degree of resistance to the shock in
the first place determines the speed and extent of recovery. Third, the extent
to which the regional economy undergoes structural re-orientation and what
implications such re-orientation has for the region’s output, jobs, and incomes.

Four, the degree of renewal or resumption of the growth path that characterised
the regional economy prior to the shock.
Furthermore, Martin et al. (2016) emphasizes that resilience is a
multifaceted process. Figure 1 shows four sequential steps of resilience: risk,
resistance, reorientation, and recoverability. These sequential aspects of the
resilience process are determined by many factors. Evaluating many studies on
the determinants of regional economic resilience, a study by Di Caro & Fratesi
(2018) finds some important conclusions. First, the determinants of economic
resilience show consistent patterns across time and space, with territorial
capital, including human capital and agglomeration economies, being notably
prominent in many countries. Second, factors that contribute to explaining
the economic performance of a location in normal times are also useful for
understanding the patterns that are observed during and after recessionary
events. Third, policymakers can play an active role in maintaining a resilient
economy by managing resources and efforts in the right policy areas in
anticipating a crisis.
Figure 1. Dimensions of Regional Resilience to Economic Shocks
Source: Martin et al. (2016)

There are many factors that may influence regional economic resilience;
however, plenty of studies indicate that the structure of the regional economy
prior to the economic shock is the most important factor that determines
resilience (Duranton & Puga, 2000; Martin et al., 2016; Pudelko et al., 2018; Di
Caro & Fratesi, 2018; Gong et al., 2020; Tan et al., 2020; Hu et al., 2021). The
economic structure refers to the economic profile of a region that is regionally
fundamental and has historical roots, which can be represented by its regional

economic diversity. Economic diversity refers to the variety of economic
activities in a region and the extent to which regional activity is uniformly
distributed among these activities. An economy with a diverse structure can
adjust and adapt, as different sectors respond differently to recessions, and can
spread risk to other distributed activities (Xiao et al., 2018).
A diversified economic structure provides greater regional resistance to
several arguments. This structure can absorb shocks to effectively eliminate the
impact of a shock (Ringwood et al., 2019; Cainelli et al., 2019). Xiao & Drucker
(2013) also argue that economic diversity is hypothesized to improve regional
economic resilience for three reasons. First, the impacts of an adverse shock
in a specific industry become less extensive since the portion of the affected
industry in the local economy is relatively small and risk is spread more widely,
meaning that the respective industry is shock resistant (Xiao & Drucker, 2013).
If many industries are shock resistant, the regional economy becomes more
resilient. Second, it facilitates skill matching between employers and employees,
it augments production efficiency, reduces job search costs, and alleviates
frictional unemployment. Third, regions with diverse economic activity tend
to have more frequent and beneficial cross-industrial transfers of ideas and
knowledge. Therefore, the economic structure plays a vital role in shaping the
sensitivity and resilience of a regional economy to shocks (Martin et al., 2016;
Tan et al., 2020). This means “not putting all eggs in the same basket” is an
important principle for a region to have more ability of shock absorption.
Tan et al. (2020) noted that there is a continues debate whether
diversification or specialisation is good or not for regional resilience. For
instance, specialisation offers both advantages, such as the reduction of urban
crowding and development of localisation economies that result from having
closely related producers located near each other; and disadvantages such as a
potential for less innovation and greater exposure to risks with the fluctuations
of specific sectors or technologies (Puga & Diego, 2000).
Empirical studies in different countries found that there is a relationship
between economic diversity and economic resilience. Di Caro (2015 ; 2017)
found the positive impact of diversification on economic resilience across the
Italian regions. Pudelko et al. (2018) investigated regional economic resilience
in Germany after the great recession of 2008/2009, in particular, the economic
effects of regional agglomeration (arising from specialisation), related variety
and unrelated variety and the explicit division of short-term resilience into

sensitivity and recovery. Related variety indicates the presence of diverse but
interconnected industries and economic activities with a specific geographic
area that encourages knowledge spillovers, innovation, and the exchange of
ideas and technologies between industries.
Meanwhile, unrelated variety refers to a situation where a geographic
area has a diverse range of industries or economic activities but has minimal
or no significant connection with one another. In addition, by explicitly dividing
short-term resilience into sensitivity and recovery components, regional
policymakers can develop targeted strategies in dealing with weaknesses and
capacity of the regions which are important for regional economic resilience.
Pudelko et al. (2018) also found that diversification policy should not (only) be
directed at unrelated diversity because unrelatedness tends to weaken a region’s
ability to recover which indicates a deficiency in the synergistic reallocation of
production factors between regional industries. In contrast, Pavel et al. (2020)
did not find any evidence that economic diversity is a resilience driver for urban
communities in the North-West region of Romania. Moreover, the cities with
diversified economies were affected the most by the crisis; in these cities, the
share of services in the total economy increased after the crisis from 2008
to 2011, while the share of the industry decreased. They also found that the
economic crisis hit urban communities earlier than the rural communities,
but the rural communities started to recover earlier and more than urban
communities.
In the context of the current COVID-19 pandemic, the recent discussion
focuses on the resistance that is the second step of resilience in Figure 1. In this
case, the level of regional economic diversity refers to the situation prior the
COVID-19 pandemic. If a region has a high economic diversity prior COVID-19,
it is likely that this area will be more resilient during the pandemic (Gong
et al. 2020; Tan et al. 2020, Hu et al. 2022). Gong et al. (2020) analysed the
specific characteristics of the COVID-19 crisis, as well as their effect on regional
recovery and resilience potential in China. They emphasized on the nature of
shocks due to the pandemic which is one of the key factors that have an impact
on the resilience of an area. They conclude that the complex combination of
characteristics of the COVID-19 crisis, institutional experience in dealing with
previous pandemic and epidemic crises, government support schemes, as well
as regional industrial structures, has the potential to affect the recovery rate
and resilience of Chinese society.

Meanwhile, Tan et al. (2020) focus on the industrial structure or agency
factors that determine economic resilience in Chinese cities which rely heavily
on natural resources (Resource-Based Cities/RBCs). Tan et al. (2020) found
that economic resilience is generally determined by regional competitive
effects in various regions because of some unique competitive advantage of the
region. Additionally, industrial structure, in general, has a positive effect, while
specialisation tends to weaken economic resilience, especially in mining cities.
The tertiary industry has a positive effect, while the primary and secondary
sectors have a negative effect on the industrial structure. The regional
competitiveness of the primary, secondary, and tertiary sectors in RBCs have
a positive effect on resilience, and the secondary sector provides the highest
contribution.
METHODOLOGY
Index of Resistance
We use a resistance index that is adapted from Martin et al. (2016) to measure
the rate of the resilience of the local economy (annotated as Resist) (see also
Sensier et al. 2016; Östh et al. 2018). This index has also been used in Hu et al.
(2021). The formula of this index is as follows:
( )
( )
where Resisti is the resistance index of region i which is a part of an agglomeration

area, ∆EiContraction is the contraction economy of region i, |(∆EiContraction)expected| is the
contraction economy of the respective agglomeration area Z (as the reference
economy) and |(∆EiContraction)expected| is the contraction economy of the respective
agglomeration area Z (as the reference economy) in absolute terms. For instance,
to calculate Resist of District A that is in the agglomeration area Z, we need data
of its GDRP growth, and the growth of the whole GDRP of the agglomeration
area Z. A positive value of the Resist index indicates that the provinces or
regions are more resilient to the shocks caused by COVID-19, compared to their
reference economy. On the other hand, if the index value is negative, it means
that the provinces or regions are less resistant to the recent shock. For example,
the Resist of District A in agglomeration Z is 0.5 meaning that the economy of

District A is 50% more resistant than the economy of the agglomeration areas,
and a value of -0.5 indicates that the economic resistance of the district is only
half of the resistance of the reference province (see Martin et al., 2016).
Index of Economic Diversity

Economic diversity can be measured by using various indices, such as the Ogive
Index, the entropy index, the Herfindahl-Hirschman Index, and Hackman Index.
The selection of the used index is also affected by the easiness to calculate and
its acceptance in empirical studies. The Herfindahl-Hirschman Index is easy
to calculate and to interpretate; and it does not require additional data. For
this study, we also use the Herfindahl-Hirschman Index to measure the local
economic diversity (Martin et al., 2016; Tan et al., 2020) based on value added
data, i.e., GDRP. As this index uses GDRP data, it does not reflect the different
productivity between capital and labour. The formula of this economy diversity
index is as follows:
where HHI is the Herfindahl-Hirschman index of region i that indicates the level
of economic diversity; si is the share of sector i in the GDRP of the respective
region. By contraction, the value of HHI is restricted between 0 and 1. If the
HHI is 0, the economic diversity of the respective region is perfect. On the other
hand, if the index value is 1, it means that the economy of the respective region
completely depends on one economic sector. In other words, the lower the HHI,
the more diversified the regional economy will be. It then can be expected that
regions with low HHI will be able to resist the shock caused by the COVID-19
pandemic.
We also calculate two variations of the standard HHI that can be used as
robustness check. For each alternative indices, the formula is based on specific
reason. First, there are other versions of HHI in which the number of economic
sectors is also included in the computation. It is argued that by its nature, the
number of sectors (n in equation 2) affects the possible range of HHI (Cracau
& Lima, 2016). This approach is particularly important when the underlying n
changes. In our study, there is no difference in the number of economic sectors
across regions since it is based on the BPS classification. However, a region
possibly has no production in agriculture if it is a city. We then also apply two

normalised versions of HHI, as shown in Cracau & Lima (2016). The first is an
index that is normalised by using the common procedure, denoted as HHIN and
an index which is proposed by Baumann in 2009, annotated as HHIB. According
to Cracau and Lima, the Bauman version fits better with the original HHI.
With n represents the number of economic sectors, the formulas of these two
alternative indices are as follows.
√ √
√
CROSS-SECTIONAL REGRESSION MODEL

To examine the relationship between economic resistance and economic
diversity, we firstly correlate the resistance index and the share of all economic
sectors and the economic diversity indexes. This step provides a picture of
what sectors potentially support the economic resistance or the opposite. For
instance, it is argued that the agricultural sector helps the regions during the
COVID-19 crisis. Meanwhile, the manufacturing sector contributes to reducing
the economic resistance of a region.
We then regress the economic diversity indexes on the resistance index of
the local economy to examine their relationship by using data at the district/
city level. It is expected that the higher the economic diversity index prior to
the shock, the weaker the economic resistance of a region. We add three control
variables to the model (Table 1). These variables are GDRP per capita in 2019
(annotated as YC), the number of COVID-19 confirmed cases until December
2020 (annotated as CC), and a dummy variable of cities (annotated as DC). Data
on the population is also taken from the BPS which is used to calculate GDRP
per capita. Since the diversity index and GDRP per capita are based on the year
prior to the pandemic (or 2019), we can also minimize the potential problem of
endogeneity that can affect the estimation results, especially from the possible
impact of pandemic on the local economy through the variable of confirmed
cases of COVID-19. The cross-sectional regression we use is as follows:

Table 1. Description of Variable
Variable Definition Source
RESIST Resistance index at the district/city Calculated using equation (1),
level (2020) based on the BPS data
HHI Herfindahl-Hirschman Index at the Calculated using equation (2-5),
district/city level (2019) based on the BPS data
logYC Log of Gross Domestic Regional GDRP divided by population,
Product (GDRP) per capita (constant based on the BPS data
price) at the district/city level (2019)
logCC Log of Number of COVID-19 confirmed
cases at the district/city level (per
December 2020)
DC Dummy variable of cities (1 for cities, 0
for districts)
The GDRP per capita reflects the economic level of local economic
development. Regions with a higher level of economic development tend to show
a more rapid spread of disease (Gong et al., 2020; Hu et al., 2022). Therefore, the
economy of these regions will be more vulnerable, or their economic resistance
is weak. Containing the spread of the virus is also more difficult in regions with
high population density (Gong et al., 2020; Chen et al., 2021). A city typically
has a high population density implying that cities become more vulnerable to
the negative implication of the current pandemic. It should be mentioned that,
except for the COVID-19 confirmed cases, the left-hand side variables in the
equation refer to the year 2020, while economic diversity indexes and GDRP
per capita represent the situation before the pandemic. Besides checking the
correlations among the control variables, we conduct the Oster test (Oster,
2019) by implementing psacalc routine in STATA to ensure the robustness of
the regression results as we have a small number of observations that limit the
number of control variables.
USE OF 2019–2020 GRDP DATA

Our main data is the real Gross Regional Domestic Product (GRDP) in 2019 and
2020 consisting of 17 subsectors of the 43 regions in six agglomeration areas
in Java provided by Badan Pusat Statistik (BPS 2021). Data in 2019 represents
economies before the pandemic, while data in 2020 indicates the immediate
impact of the pandemic.

The COVID-19 pandemic has highly affected six agglomeration areas in
Java and contributed to 520.036 COVID-19 confirmed cases in Indonesia (Table
2). The most significant contributor is Gerbangkertosusila (Gersik-Bangkalan-
Mojokerto-Surabaya-Sidoarjo- Lamongan) in East Java. However, the highest
average cases at the district level is found in Jabodetabek. Table 2 also shows
the cities with the highest confirmed cases in many agglomeration areas, such
as Jakarta Timur in the Jabodetabek and Surabaya in the Gerbangkertosusila.
In contrast, regencies that record the lowest number of confirmed cases is
Kepulauan Seribu in Jabodetabek, while Salatiga is the only city that has the
lowest confirmed-cases of COVID-19. This is reasonable since Salatiga is the
smallest city not only in Kedungsepur (Kendal-Demak-Ungaran-Semarang-
Purwodadi) area but also in the agglomeration areas in Java in terms of
population size. This data suggests that the economic impact of COVID-19 is
probably not homogenous across regions within an agglomeration, especially
between cities and regencies as also observed by Handayani et al. (2022) and
Aritenang (2022). It should be noted that cities reflect urban, while regencies
reflect rural characteristics.
Table 2. COVID-19 Confirmed Cases, December 2020
Total Average Region with Lowest Cases Region with Highest Cases
Area
Cases Cases (number of cases) (number of cases)
Jabodetabek 193.883 13.849 (R) Kepulauan Seribu (171) (C) Jakarta Timur (38.951)
Bandung Raya 15.583 3.117 (R) Sumedang (979) (C) Bandung (6.374)
Kedungsepur 22.756 3.793 (C) Salatiga (908) (C) Semarang (11.080)
Solo Raya 15.445 2.206 (R) Wonogiri (1.199) (R) Sragen (2.727)
Jogja Raya 12.351 2.470 (R) Gunung Kidul (872) (R) Sleman (4.997)
Gerbangkertasusila 260.018 5.779 (R) Bangkalan (1.043) (C) Surabaya (18.205)
Notes: R refers to regency, C refers to city. Jabodetabek consists of (R) Kepulauan Seribu, (C)
Jakarta Selatan, (C) Jakarta Timur, (C) Jakarta Pusat, (C) Jakarta Barat, (C) Jakarta Utara, (C) Bogor,
(R) Bogor, (C) Depok, (C) Tangerang, (R) Tangerang, (C) Bekasi, (R) Bekasi, and (C) Tangerang
Selatan. Bandung Raya consists of (R) Bandung, (C) Bandung, (C) Cimahi, (R) Sumedang, and (R)
Bandung Barat. Kedungsepur consists of (R) Kendal, (R) Demak, (C) Salatiga, (R) Semarang, (C)
Semarang, and (R) Grobongan. Solo Raya consists of (R) Sukoharjo, (C) Surakarta, (R) Boyolali,
(R) Klaten, (R) Karanganyar, (R) Wonogiri, and (R) Sragen. Jogja Raya consists of (R) Kulon Progo,
(R) Bantul, (R) Gunung Kidul, (R) Sleman, and (C) Yogyakarta. Gerbangkertosusila consists of (R)
Bangkalan, (R) Mojokerto, (C) Surabaya, (R) Sidoarjo, (R) Gresik, and (R) Lamongan.

In 2020 the economy of Java contracted by 2.51%. Gerbangkertosusila
experienced the most severe economic contraction. Its contraction is more
than 150% higher than the average contraction across the agglomeration areas,
and the one with the lowest contraction was Solo Raya in Central Java (Figure
2). The contraction of the Gerbangkertosusila economy is more than 2.6 times
Solo Raya’s economic contraction. This also indicates differences in the level of
economic resilience across regions in the six agglomeration areas.
Figure 2. Economic Growth at the Agglomeration Level, 2020 (%)
Source: own calculations based on BPS (2021)

Less than half of the regions in the agglomeration areas in Java are not
resistant to the crisis during the COVID-19 pandemic (Figure 3). Looking at the
agglomeration level, Solo Raya and Bandung Raya seem to have less resistance
than other areas. In these two areas, the number of resistant regions is smaller
than the number of non-resistant regions. However, the region with the lowest
resistance index is the city of Tangerang in Jabodetabek, while the most resistant
region is Demak in Kedungsepur.

Figure 3. Distribution of Regions Based on the Resistance Category at the
Agglomeration Level
Source: Author’s own calculations based on BPS (2021)

Table 3 provides correlations between the resistance index, as measured
by the conventional HHI, at the local level and the share of sectors of the
economy for each agglomeration area. In the Jabodetabek, there are three
sectors that have a positive correlation with the resistance index. The fact that
the sector of information and communication, and the financial and insurance
activities sector have a strong positive correlation with the resistance index also
confirms the role of these sectors in some regions in this area to contribute to
the regional as well as national economy. During the pandemic, the information
and communication sector played a significant role in supporting, for instance,
the policy of working and studying from home to manage the spread of viruses.
Table 3. Correlation Between the Resistance Index and the Structure
of Economy
JBDBK BR KS SR JR GKS R C All
Sector
1 -0.28 0.48 0.63 0.44 0.41 0.05 0.33 -0.16 0.18
2 -0.27 -0.51 0.22 0.72* 0.01 -0.61 -0.51* -0.19 -0.27*
3 -0.35 -0.79 0.13 -0.18 -0.18 0.6 0.11 -0.63* -0.17
4 0.19 -0.08 -0.13 -0.52 0.18 -0.19 -0.20 0.23 0.08
5 -0.15 -0.01 -0.24 -0.38 0.56 -0.29 0.30 -0.18 0.00
6 0.3 0.36 -0.7 -0.28 -0.68 -0.54 -0.02 -0.04 -0.06
7 0.36 0.31 0.22 -0.19 -0.21 -0.46 0.31 0.13 0.15
8 -0.26 0.22 -0.18 0.35 -0.65 -0.31 0.05 -0.39 -0.26*

9 0.38 0.33 -0.2 -0.3 0.16 -0.35 0.06 0.13 0.08
10 0.50* 0.22 -0.83* -0.15 0.35 0.33 0.28 0.43* 0.28*
11 0.51* 0.66 -0.57 -0.4 -0.09 -0.38 0.05 0.52* 0.35*
12 0.16 0.4 -0.43 -0.39 -0.12 0.02 -0.13 0.15 0.02
13 0.31 -0.38 -0.39 -0.39 -0.37 -0.41 -0.35* 0.27 0.12
14 0.44 0.92* 0.04 -0.08 0.62 -0.21 0.23 0.29 0.25*
15 0.35 0.85* 0.52 0.90* -0.18 -0.52 0.15 0.25 0.18
16 0.36 0.93* -0.05 0.76* 0.16 0.03 0.03 0.35 0.19
17 0.53* 0.31 0.6 0.63 0.19 0.16 0.25 0.45* 0.31*
Notes: Agglomeration areas: JBDBK (Jabodetabek), BR (Bandung Raya), KS (Kedungsepur), SR

(Solo Raya), (JR (Jogja Raya), GKS (Gerbangkertasusila), R (Regencies), C (Cities), All (All Regions).
Sector: 1) Agriculture, Forestry and Fishing, 2) Mining and Quarrying, 3) Manufacturing, 4)
Electricity and Gas, 5) Water supply, Sewerage, Waste Management and Remediation Activities,
6) Construction, 7) Wholesale and Retail Trade; Repair of Motor Vehicles and Motorcycles, 8)
Transportation and Storage, 9) Accommodation and Food Service Activities, 10) Information and
Communication, 11) Financial and Insurance Activities, 12) Real Estate Activities, 13) Business
Activities, 14) Public Administration and Defense; Compulsory Social Security, 15) Education, 16)
Human Health and Social Work Activities, 17) Other Services Activities. * p < 0.10
Meanwhile, the resistance index in Bandung Raya has a strong positive
correlation with the services sectors, such as human health and social work
activities. These correlations are also statistically significant for Solo Raya.
Another interesting finding for Solo Raya is that the sectors of mining and
quarrying have a positive and statistically significant correlation with the
resistance index. In contrast to Jabotabek, the relationship between the
information and communication sector and the resistance index in Kedungsepur
is negative. This negative relationship indicates that Kedungsepur does not
receive the resistance effect of the information and communication sector. For
the last two agglomerations, Jogya Raya and Gerbangkertosusila, the correlation
presented in Table 3 does not provide any strong correlation between the
component of the local economy and the resistance index.
We also estimated the correlation between those two variables for the
groups of regencies as well as for the cities. The results show that the sectors of
mining and quarrying, and business activities have a negative and statistically
significant correlation with the resistance index for the regencies. This indicates
that primary sectors do not always provide advantages for regencies to have
better resistance to the shock. During the pandemic, many projects on the
infrastructures have been cancelled or delayed, which then affected the mining
and quarrying sector.

Meanwhile, the sector that has a negative correlation with the index of
economic resistance for the cities is manufacturing. This finding is consistent
with the important contribution of the manufacturing sector to the cities’
economy. Therefore, when the pandemic hit the cities, manufacturing is one of
the sectors that suffered negative impact of the pandemic. The manufacturing
sector is also affected by the government policy to handle the spread of
viruses. There are three sectors that have a positive and statistically significant
correlation with the resistance index for the cities. Similar to Jabodetabek, these
sectors are information and communication, financial and insurance activities,
and other services activities.
Overall, sectors that support economic resistance during the pandemic
are information and communication, financial and insurance activities, public
administration and defense and other services activities. Meanwhile, the
sectors of mining and quarrying, and transportation and storage have a negative
correlation with the resistance index.
RESULTS OF REGRESSION
To examine the influence of the diversity index on the resistance index, we
estimate equation 5. To minimise the issue of endogeneity with our main variable
of interest, we put our main variable of interest in its lag variable. We only used
a short period of data, then we could not identify whether the variables contain
trend. We checked the correlations among the control variables and found that
they are relatively small (less than 0.50). We do this falsification test to argue
that our main variable of interest is relatively independent. For the Oster test,
the R-squared value increases by 30%, which resulted in 0.115, 0.115, and 0.124
for regression (1), (2), and (3), respectively. The results of this test presented in
the last row of Table 4 show that all delta values are greater than one, meaning
that the present of unobservable bias can be ignored.
As expected, all coefficients of the diversity indexes show a negative sign
and are statistically significant, except for the estimation that uses a Baumann
version of the Herfindahl-Hirschman index (HHIB). However, the confidence
interval is 90%, meaning that these relationships are considered as low
significance. Since the higher value of the diversity indexes means low economic
diversity, these results indicate that regions with low economic diversity are
more vulnerable to shock. In addition, regarding the result for column 4, we can
interpret regions with higher economic diversity than the average diversity in

their respected agglomeration area, tend to have a higher economic resistance
during the pandemic. Therefore, the findings for the local economies in the
agglomeration areas in Java are in line with Gong et al. (2020), Tan et al. (2020),
Hu et al. (2022) who confirmed that having a diversified economic structure
provides the ability for a region to resist when a shock distracts the economy of
this region.
Table 4. Regression Results
(1) (2) (3)
HHI(t-1) -1.26*
(-2.20)
HHIN(t-1) -1.19*
(-2.20)
HHIB(t-1) -1.06
(-1.63)
logYC(t-1) -0.17** -0.17** -0.17*
(-2.59) (-2.59) (-2.52)
logCC 0.12 0.12 0.13
(1.21) (1.21) (1.34)
DC -0.19 -0.19 -0.20
(-1.48) (-1.48) (-1.34)
_cons 1.04* 0.97 1.04*
(2.11) (1.95) (2.04)
No. of regions 43 43 43
R2 0.088 0.088 0.095
delta (Oster test) 1.390 2.443 2.665
Notes: Economic Diversity Index: 1) Herfindahl Hirschman index (Common), 2) Herfindahl

Hirschman index (Normalized), 3) Herfindahl Hirschman index (Baumann). In the ( ) is t-statistics,
* p<0.10, ** p< 0.05, p< *** 0.010, Robust Standard Errors (clustered by agglomeration areas). For
the Oster test, we increased the value of R-squared by 30%.
The variable of GDRP per capita (logYC) in 2019 also has a negative and
statistically significant coefficient in all estimations. It means that there is
evidence that regions with a higher development level before the pandemic
experience serious negative implications of the pandemic. The results are
consistent with the findings of other studies by Gong et al. (2020), and Hu et
al. (2022), confirming that the more developed local economy became more
vulnerable during the pandemic, or their economic resistance is weak. The
other two control variables show the expected sign of the coefficient; however,
they are all statistically insignificant.

Based on the regression results, we provide a contribution to the economic
resistance discussion, especially in the current pandemic, that economic
diversity prior the shock contributes to the ability of a local economy to be
more resistant during the COVID-19 pandemic. This findings suggest that
it is important to improve economic diversity as a preparedness strategy to
anticipate other economic shocks in the future. This strategy is also important
for local economies which have already developed since economic shocks in the
future, as the current situation tends to directly hit them.
Table 5. Regression Results (Western Java Vs. Eastern Java)
Western area of Java Eastern area of Java
(1) (2) (3) (4) (5) (6)
HHI(t-1) -1.71** 1.05
(-12.86) (1.03)
HHIN(t-1) -1.61** 0.99
(-12.86) (1.03)
HHIB(t-1) -1.58** 0.59
(-42.64) (0.88)
_cons 0.32 0.22 0.38 -0.12 -0.06 -0.07
(3.76) (2.85) (5.03) (-0.69) (-0.52) (-0.49)
No. of regions 19 19 19 24 24 24
R 2
0.104 0.104 0.129 0.032 0.032 0.024
* p<0.10, ** p< 0.05, p< *** 0.010, Robust Standard Errors (clustered by agglomeration areas).
We also detect heterogeneity by splitting the observations based on their
area in Java and based on their administrative status. We classified the regions
in Jabodetabek and Bandung Raya into the Western area of Java (19 regions),
while other regions are grouped as the Eastern area of Java (24 regions).
Based on administrative status, we have 17 cities and 26 regencies. Since the
number of observations in each group is relatively small, the regressions only
included our variable of interest, namely economic diversity index. The results
in Table 5 show that all economic diversity indexes have a significant negative
relationship with resistance index only in the Western part of Java. Their
significance level is higher than the main results in Table 4. The same significant
negative correlations are also found for regencies in Table 6, except for column

6, which uses a Baumann version of the Herfindahl-Hirschman index, with a
confidence interval of 90%. Based on the results in Tables 5 and 6 we can argue
that economic diversity has a stronger influence in the Western area of Java and
in the regencies.
Table 6. Regression Results (Cities Vs. Regencies)
Cities Regencies
(1) (2) (3) (4) (5) (6)
HHI(t-1) -7.41 -1.07*
(-1.21) (-2.11)
HHIN(t-1) -6.97 -1.00*
(-1.21) (-2.11)
HHIB(t-1) -4.98 -0.77
(-1.35) (-1.53)
_cons 1.11 0.67 0.93 0.28** 0.22** 0.27*
(1.22) (1.23) (1.37) (2.82) (2.85) (2.09)
No. of regions 17 17 17 26 26 26
R 2
0.174 0.174 0.202 0.103 0.103 0.084
* p<0.10, ** p< 0.05, p< *** 0.010, Robust Standard Errors (clustered by agglomeration areas).
CONCLUSION
This chapter measures the short-term economic resilience, or resistance, of
the regions in the agglomeration areas in Java, Indonesia during the current
pandemic. The COVID-19 pandemic has highly affected six agglomeration areas
on the island. Particularly, the study examines whether the level of economic
resilience has a strong correlation with the economic diversity in the pre-
COVID-19 period. It hypothesized that economically diversified regions have
greater economic resistance than regions with a less diversified structure
during the crisis caused by COVID-19. The estimations are robust as there is no
multicollinearity, the presence of unobservable bias can also be ignored, and
the heterogeneity has already been checked.
The findings show that different economic sectors have different
relationships with economic resistance. Some sectors support economic
resistance, such as information and communication, and financial and insurance

activities, while the mining and quarrying sector tends to have a negative
correlation with the resistance index. This adds to finding the influence of
economic diversity on regional resistance as found in similar studies such as Di
Caro (2015, 2017), Gong et al. (2020), Tan et al. (2020), and Hu et al. (2022). The
study also finds that the more developed local economies, the more vulnerable
they were during the pandemic. This study then contributes to the importance
of economic diversity to resist such economic shocks as the current pandemic.
Improving economic diversity is also crucial for local economies that have
already achieved higher development levels.
REFERENCES
Aritenang, Adiwan F. (2022). ‘The impact of urban characteristics on the spread
of Covid-19 in 2020: The case of Java Island cities, Indonesia.’ Regional
Statistics 03:3-17. Available at: (https://www.ceeol.com/search/article-
detail?id=1065922
Ascani, Andrea, Alessandra Faggian, and Sandro Montresor (2020). ‘The
Geography of Covid‐19 and the Structure of Local Economies: The Case of
Italy’. Journal of Regional Science 61 (2): 407–41. Available at: https://doi.
org/10.1111/jors.12510.
Ascani, Andrea, Alessandra Faggian, Sandro Montresor, & Alessandro Palma
(2021). ‘Mobility in times of pandemics: Evidence on the spread of COVID19
in Italy’s labour market areas’. Structural Change and Economic Dynamics
58: 444-454. Available at: https://doi.org/10.1016/j.strueco.2021.06.016.
Badan Pusat Statistik (2021). Tinjauan Regional Berdasarkan PDRB Kabupaten/
Kota 2016-2020, Buku 2 Pulau Jawa-Bali. Available at: ttps://www.bps.
go.id/publication/2021/09/06/3f6357ab7dc0edeb0fb459aa/tinjauan-
regional-berdasarkan-pdrb-kabupaten-kota-2016-2020--buku-2-pulau-
jawa-bali.html.
Cainelli, Giulio, Roberto Ganau, & Marco Modica (2019). ‘Does related variety
affect regional resilience? New evidence from Italy.’ Ann Reg Sci 62: 657–
680. Available at: https://doi.org/10.1007/s00168-019-00911-4
CNNIndonesia. 2022. ‘UI: Warga Wilayah Aglomerasi Punya Antibodi Lebih
Tinggi Lawan Covid. March 21’. https://www.cnnindonesia.com/
nasional/20220321102625-20-774088/ui-warga-wilayah-aglomerasi-
punya-antibodi-lebih-tinggi-lawan-covid.

Coulson, Edward., Steven Brakman, Alessandra Faggian, & Yasusada Murata
(2021). ‘COVID‐19 and regional economies: An introduction to the special
issue.’ Journal of Regional Science 61 (4): 691–695. doi:10.1111/jors.12553.
Cracau, Daniel, and José E. Durán Lima (2016). ‘On the Normalized Herfindahl-
Hirschman Index: A Technical Note.’ International Journal on Food System
Dynamics 7(4): 382–386. doi:10.18461/ijfsd.v7i4.748.
Desdiani, Nauli Aisyiyah, Syahda Sabrina, Meila Husna, Amalia Cesarina
Budiman, Fachry Abdul Razak Afifi, & Alin Halimatussadiah (2022).
‘Local Budget Resilience in Times of COVID-19 Crisis: Evidence from
Indonesia.’ Economies 10(5): 108. Available at: https://doi.org/10.3390/
economies10050108
Di Caro, Paolo. 2015. ‘Recessions, recoveries, and regional resilience: Evidence
on Italy.’ Cambridge Journal of Regions, Economy, and Society 8(2): 273–291.
doi: 10.1093/cjres/rsu029.
Di Caro, Paolo. 2017. ‘Testing and explaining economic resilience with an
application to Italian regions.’ Papers in Regional Science 96: 93–113. doi:
10.1111/pirs.12168.
Di Caro, Paolo, & Ugo Fratesi (2018). ‘Regional determinants of economic
resilience.’ The Annals of Regional Science 60:235-240. Available at:
https://doi.org/10.1007/s00168-017-0858-x
Duranton, Gilles., & Diego Puga. 2000. ‘Diversity and Specialization in Cities:
Why, Where and When Does it Matter?’ Urban Studies 37: 533–555. doi:
10.1080%2F0042098002104.
Florida, Richard, Andrés Rodríguez-Pose, & Michael Storper (2021). ‘Cities in a
post-COVID world.’ Urban Studies 0(0). doi: 10.1177/00420980211018072.
Gereffi, Gary (2020). ‘What does the COVID-19 pandemic teach us about global
value chains? The case of medical supplies.’ J Int Bus Policy 3: 287–301.
Available at: https://doi.org/10.1057/s42214-020-00062-w
Gong, Huewin, Robert Hassink, Juntao Tan, & Huang Dacang (2020).’Regional
Resilience in Times of a Pandemic Crisis: The Case of COVID-19 in China.’
Tijdschrift Voor Economische En Sociale Geografie 111(3): 497–512. doi:
10.1111/tesg.12447.

Handayani, Wiwandari, Tia Dianing Insani, Micah Fisher, Tae-Hyoung Tommy
Gim, Santi Mardhotillah, & Urban El-fatih Adam (2022). ‘Effects of
COVID-19 restriction measures in Indonesia: A comparative spatial and
policy analysis of selected urban agglomerations.’ International Journal of
Disaster Risk Reduction 76: 103015. Available at: https://doi.org/10.1016/j.
ijdrr.2022.103015.
Hu, Xiaohui., Liangan Li, & Ke Dong (2022). ‘What matters for regional economic
resilience amid COVID-19? Evidence from cities in Northeast China.’ Cities
120:103440. doi: 10.1016/j.cities.2021.103440.
Kompas (2021). ‘8 Wilayah Aglomerasi yang Bisa Mudik Lokal pada 6-17
Mei.’ April 16. Available at: https://www.kompas.com/tren/read/2021/
04/16/090400065/8-wilayah-aglomerasi-yang-bisa-mudik-lokal-pada-6-
17-mei?page=all.
Kompas (2022). ‘Mudik di Wilayah Aglomerasi Tak Perlu Tunjukkan Hasil
Tes Covid-19.’ April 4. Available at: https://nasional.kompas.com/
read/2022/04/04/13251021/mudik-di-wilayah-aglomerasi-tak-perlu-
tunjukkan-hasil-tes-covid-19?page=all.
Kuncoro, Mudrajad (2003). ‘Why Manufacturing Industry Persisted to Cluster
Spatially in Java?’ Gadjah Mada International Journal of Business 5(2): 221-
247. doi: 10.22146/gamaijb.5410
Maftukhah, Siti & Retno Widodo Dwi Pramono (2022). ‘How Different
Geographical Areas React to Covid19 Shock Regional Resilience and
Structural Transformation.’ The Journal of Indonesia Sustainable
Development Planning 3(3): 256-277. Available at: http://journal.
pusbindiklatren.bappenas.go.id/lib/jisdep/article/download/343/144.
Martin, Ron (2012). ‘Regional economic resilience, hysteresis, and recessionary
shocks.’ Journal of Economic Geography 12(1): 1–32. doi: 10.1093/jeg/
lbr019.
Martin, Ron, & Peter Sunley (2015). ‘On the notion of regional economic
resilience: Conceptualization and explanation.’ Journal of Economic
Geography 15: 1–42. doi: 10.1093/jeg/lbu015.
Martin, Ron, Peter Sunley, Ben Gardiner, & Peter Tyler (2016). ‘How Regions
React to Recessions: Resilience and the Role of Economic Structure.’
Regional Studies 50(4): 561–585. doi:10.1080/00343404.2015.1136410.

Mayor, Matias, & Raul Ramos (2020). ‘Regions and economic resilience: New
perspectives.’ Sustainability 12(11): 10–13. doi: 10.3390/su12114693.
Mujiyanto, Mujiyanto, Zaenal Abidin Sahabuddin, & Guntur Eko Saputro (2023).
‘Government’s Strategy in Maintaining Economic Resilience during the
Covid-19 Pandemic.’ Journal of Social Work and Science Education 4(2):
468-479. Available at: https://ejournal.karinosseff.org/index.php/jswse/
article/view/396
Mulyaningsih, Tri, Riyana Miranti, Anne Daly, & Chris Manning (2023).
‘Regional Skill Differentials: A Study of The Indonesian Labor Market.’ The
Singapore Economic Review 68(03): 703-728. https://doi.org/10.1142/
S0217590819500371
Oster, Emily. 2019. Unobservable Selection and Coefficient Stability: Theory
and Evidence.’ Journal of Business and Economic Statistics 37 (2): 187–204.
Available at: https://doi.org/10.1080/07350015.2016.1227711.
Östh, John, Aura Reggiani, & Peter Nijkamp (2018). ‘Resilience and accessibility
of Swedish and Dutch municipalities.’ Transportation 45: 1–23. doi:
10.1007/s11116-017-9854-.
Pavel, Alexandru, Bogdan Andrei Moldovan, Karima Kourtit, & Peter Nijkamp
(2020). ‘Urban or Rural: Does It Make A Difference for Economic Resilience?
A Modelling Study on Economic and Cultural Geography in Romania.’
Sustainability 12(9): 3776. doi: 10.3390/su12093776.
Pudelko, Franziska, Christian Hundt, & Linus Holtermann (2018). ‘Gauging two
sides of regional economic resilience in Western Germany—Why sensitivity
and recovery should not be lumped together.’ Working Papers on Innovation
and Space 1(17): 141–189. doi: 10.1007/s10037-018-0124-4.
Purwandari, Titi, Sukono, Yuyun Hidayat, & Wan Muhamad Amir W. Ahmad
(2022). ‘Developing New Method in Measuring City Economic Resilience
by Imposing Disturbances Factors and Unwanted Condition.’ Computation
10(8): 135. Available at: https://doi.org/10.3390/computation10080135
Resosudarmo, Budy P., & Milda Irhamni (2021). ‘Consequences of the COVID-19
Pandemic on Human Capital Development’. In B. Lewis and F. Witoelar
(eds.). Economic Dimensions of COVID-19 in Indonesia: Responding to the
Crisis. Singapore: ISEAS, pp. 170-189.

Ringwood, Lauryn, Philip Watson, and Paul Lewin (2019). ‘A quantitative
method for measuring regional economic resilience to the great recession.’
Growth and Change 50(1): 381-402. doi: 10.1111/grow.12265.
Sensier, Marianne, Gillian Bristow, & Adrian Healy (2016). ‘Measuring
Regional Economic Resilience across Europe: Operationalizing
a complex concept.’ Spatial Economic Analysis 11(2): 1–24. doi:
10.1080/17421772.2016.1129435.
Tan, Junato, Xiaohui Hu, Robert Hassink, & Jianwei Ni (2020). ‘Industrial
structure, or agency: What affects regional economic resilience? Evidence
from resource-based cities in China.’ Cities 106 (September). doi: 10.1016/j.
cities.2020.102906.
Xiao, Jing, Ron Boschma, & Martin Andersson (2018). ‘Industrial diversification
in Europe: The differentiated role of relatedness.’ Economic Geography 94
(5): 514–549. doi:10.1080/00130095.2018.1444989.
Xiao, Yu, & Joshua Drucker (2013). Does Economic Diversity Enhance Regional
Disaster Resilience? Journal of the American Planning Association 79 (2):
148– 160. doi:10.1080/01944363.2013.882125.

CHAPTER 4
The Productivity of Urban Farming

in the City of Yogyakarta
Catur Sugiyanto, Dion Saputra Arbi, and Ahmad Syahrul Fauzi
INTRODUCTION
Urban Farming is an industrial activity that produces, processes, and markets
food to meet the needs of people who live in urban areas. It is done by utilizing
land and waters that they own through the application of intensive methods
and utilizing natural resources and urban waste to produce agricultural and
livestock products (Smith et al., 1996). Urban farming has been implemented in
many cities. It contributes to food supply, a better environment, and increasing
income. When the food supply is interrupted, for example due to the COVID-19
pandemic, there is a claim that urban farming can serve as a buffer for households
to provide food (Jakarta Post, 2020).
Many studies support the above claim that urban farming has many
benefits, such as increasing food security and creating significant employment
opportunities (Agbonlahor et al., 2007). Besides helping reduce income spent
on food (Mougeot, 2005) and malnutrition (Mougeot, 2005, Zezza et Tasciotti
2010; Orsini et al., 2013), it provides high value horticultural products (Korth et
al 2014; Grafius et al., 2020). In addition to the economic impact, urban farming
can also be beneficial to the urban environment, disaster preparedness, and
sustainability (Yuan et al., 2023).
69
Likewise, in Indonesia urban farming serves as a means of empowering
local community (Surya et al., 2020), increasing activities for the women, adding
food sources (Anggrayni, Andrias, & Adriani, 2015; Suryandari & Abdullah,
2012), developing healthy urban greenery (Santoso, Aulia, & Widya, 2015;
Taufani, 2017), increasing urban resilience (Atmaja, Yanagihara, & Fukushi,
2020), and reducing malnutrition (Siswati et al 2022). The size of urban
farming in several major cities in the country has been expanding in the last few
years. It has become one of the most dynamic household business activities in
those cities. Most urban farming was not motivated only by profit but by other
factors such as healthy food production, environmental impact, relaxation, and
as a means of socializing through shared goals exhibited by the community
farmers (Glavan et al., 2018). Urban farming in Indonesia encompasses a range
of practices, including rooftop gardens, balcony gardens, vertical gardens,
community gardens, hydroponics, aquaponics, and rooftop beekeeping. These
diverse approaches allow for flexible and creative use of limited urban spaces,
(Suparwoko & Taufani, 2017).
Urban farming has gained traction as a community-driven movement in
Indonesia. Many individuals, households, and communities are actively involved
in setting up and maintaining urban farms, often collaborating with local
governments, NGOs, and educational institutions. The Indonesian government
has recognised the significance of urban farming and has implemented
several initiatives to support and encourage its development. These include
programmes for community-based urban farming, provision of urban farming
training and education, and the establishment of urban farming demonstration
sites (Chandra & Diehl, 2019; Nasrudin & Dwiyantama 2022).
Urban farming in Indonesia, however, faces several challenges, including
limited access to land, limitations of innovation and technology, inadequate
infrastructure, and supporting market that has not been maximized (Chandra
& Diehl, 2019; Sutrisna, 2020; Pribadi et al. 2017). Furthermore, there is a
limited amount of research conducted on the productivity performance of
urban farming and the factors influencing it in Indonesia. This knowledge gap
hinders the government’s ability to prioritise and provide adequate support for
the development of urban farming initiatives. Therefore, this chapter tries to
shed some light on this discussion, by examining the productivity estimate of

urban farming in the city of Yogyakarta. We apply total factor productivity (TFP)
measurement. The data used is from our own survey of 250 urban farmers,
which was conducted in all 14 sub-districts in the city of Yogyakarta during the
pandemic 2021.
The decision to choose Yogyakarta as our case study is appropriate. In
line with the practice of urban farming in many cities worldwide (World Bank,
2013), Yogyakarta City is one of the cities where many households currently
practice urban farming. The interest to practice urban farming has been
increasing as the city government support the activities since 2018. The city
underlines the support programme by raising the tag line “eat what you grow
and grow what you eat”. It turns out that the people, especially the women
groups, responded positively to the programme. The phenomenon of farming in
the city of Yogyakarta has increased during the pandemic (Jogjadaily, 2020). Not
only that urban farming contributes to the food production in the city, but it also
strengthens the communication and collaboration amongst the neighbours as
they share or sell their product to neighbours (Jogjadaily, 2020). Urban farming
in Yogyakarta has become one of the most dynamic urban farming activities in
Indonesia.
The main finding of this chapter highlights variations in productivity levels
among different types of urban farming in the City of Yogyakarta. Fish production
emerges as the most productive, closely followed by urban farming focused
on flowers. These two types of urban farming demonstrate high productivity
compared to other types of urban farming in Yogyakarta, which exhibit lower
productivity levels.
PREVIOUS STUDIES ON AGRICULTURAL PRODUCTIVITY

Total factor productivity (TFP) is a broad agricultural productivity metric that
considers the contribution of all inputs to production, calculated as the ratio of
total output to agricultural inputs (Comes et al., 2019). Total output consists
of plant and animal products, while total inputs consist of land, labour, capital,
and resources. TFP growth occurs when total agricultural output grows faster
than inputs. Global TFP growth since the 1990s has been largely driven by
innovations that improve the efficiency of using labour, land, capital, and other
inputs.
The Productivity of Urban Farming in the City of Yogyakarta 71

Comes et al. (2019) find a contrast in the relationship between TFP and
partial factor productivity that focuses on individual factors of production, such
as land area. The future of agriculture still depends heavily on agricultural land
use models and socioecological trade-offs along the traditional growth path
(partial factor productivity) of intensification or extensification. Intensification
that is conducted by increasing productivity can encourage new land and
resources to produce. TFP values should be the main factor in the broader
societal values of ecosystem services, biodiversity, food security, and social
equity (Comes et al., 2019).
Recent research by Fang et al. (2021) shows that agricultural insurance
deductions, agricultural economic status, and industrialisation have a positive
effect on China’s agricultural green TFP. This requires a willingness to apply
policies, such as taxes, subsidies, and regulations, to pursue the best combination
of technological gains in agriculture and broader societal goals. During the crisis
due to the pandemic, the impact of COVID-19 affects agricultural production
in China. Restrictions applied during the pandemic has blocked the outlet of
agricultural products, hindered the necessary production inputs, destroyed
the production cycle, and ultimately caused the damage of production capacity
(Pu & Zhong, 2020). So, in the face of the serious impact of COVID-19, special
strategies are needed to protect domestic agricultural production, especially in
developing countries.
TFP has been used as an indicator to measure Indonesian agricultural
productivity. Fuglie (2004, 2010) show that based on the Tornqvist-Thiel index,
Indonesian TFP accelerated in the 1970s and 1980s but stagnated in the 1990s.
Also confirmed by Warr (2008) that growth of total factor productivity has
contributed to 41% of output growth in Indonesian agriculture since 1975.
Indonesia has made efforts to intensify agriculture to increase productivity,
especially for food crops by expanding crop area and absorb more labour
(Resosudarmo, 2012). On average, the Indonesian agricultural output increased
by 3.6% per year between 1961 and 2006 (Resosudarmo and Yamazaki, 2010).
The Indonesian TFP growth accelerated again in the 2000s stemming
from diversification into non-food commodities such as tropical perennials,
horticulture, livestock, and aquaculture. Agricultural extensification is an
important source of agricultural growth in Indonesia, where forest areas
were previously converted to agricultural land. Another source of increasing

productivity is literacy and improving the quality of human resources. The
enactment of government policies that support agricultural extensification in
Indonesia makes the measurement of productivity an important, interesting,
and challenging study. An increase in TFP indicates an increase in the quality of
resources to produce more output from each unit of input.
Some important notes in estimating agricultural productivity, from previous
research are worth to note. First, the use of TFP will minimise bias that may
arise from changes in the relative prices of inputs and outputs resulting from
the aggregation weights (Fuglie, 2004). Second, the use of the Tornqvist-Thiel
index can minimise the aggregation bias that may arise from revenue growth
and other factors that change over time but require annual data for quantities
and prices (Fuglie, 2010). Third, the measurement of inputs—i.e., the area of
land and the agricultural media used—requires a fairly complicated technique.
Therefore, estimation of agricultural productivity is always interesting and a
challenging study.
Inline to the agriculture TFP growth, the Food and Agriculture Organization
(FAO 2018) predicts that urban agriculture will produce more than 15% of the
world’s food. Urban agriculture is thought to reduce greenhouse emissions, be
climate resilient, address food security issues for growing urban populations,
and provide chemical-free food without the risk of pests and diseases. Although
more study is needed, there are opportunities to improve crop yields and quality
by combining advances in environmental control, phenomics and automation
with breeding efforts to adapt traits for architecture, development, and quality
(taste and nutrition) that allow more diverse crops to be grown on farms with
controlled environments (O’Sullivan et al., 2019). Urban agriculture is unique
in utilising energy, water, and nutrients from urban waste, but innovation is
needed to utilise these resources safely and economically.
Another research by McDougal et al. (2018) demonstrated the contribution
of urban food production to food security by analysing inputs and yields over a
1-year period in 13 small-scale organic farms and gardens in Sydney, Australia.
The result indicated an average yield of 5.94 kg per square meter, which is about
double the yield of commercial vegetable farms in Australia. Although relatively
small, the contribution of urban agriculture is still considered potential for
further development.

Nichols, et al. (2020) reviewed the potential of urban and peri-urban
agriculture and its contribution to sustainable food production in the city of
Brighton and Hove, UK. Urban and peri-urban landholdings tend to be small
and labour-intensive, characterised by high crop diversity, including perennials
and annuals. Results show that these types of farming systems can be highly
productive and have environmental and social advantages over industrial
agriculture as crops are typically produced with few synthetic inputs and are
destined for local consumption. Overall, food products grown on small-scale
land in and near cities make a significant contribution to food security while
possibly benefiting some ecosystem services, such as supporting healthy
soils. Such development, raise the interest to understand the urban farming
characteristics and its performance.
This chapter will employ the technique adopted by the FAO to estimate
urban farming productivity. The FAO technique is known for its simplicity,
practicality, and relative reliability. TFP has become an important concept in
agricultural economics and policy over the past three decades (Kryszak et al.,
2023) making it suitable for implementation in the context of urban farming in
Yogyakarta.
THE USE OF FAO TECHNIQUE

To measure productivity, the FAO provides a simple explanation of productivity
as the relationship between the production of commodities and services
produced against the use of input goods and services used (FAO, 2018). Total
factor productivity (TFP) compares total outputs relative to the total inputs
used in production of the output. TFP index is defined as the ratio between
an Output Index (i.e., the change in production volumes over a considered
period) and an Input Index (the corresponding change in inputs/factors used to
produce them). Output and input indices are calculated as weighted averages of
changes in produced quantities and in input quantities respectively, where the
weights are represented by the production value of the various products and the
expenditure for each of the four considered production factors (intermediate
inputs, land, labour, and capital).
Depending on the type of average applied and the chosen reference period
for the weights, the TFP indicator assumes different analytical forms. Laspeyres
indices are defined as arithmetic means with weighting factors referring to the
time 0 (base year):

[ ]
⁄
( )
[ ]
⁄
( )
while Paasche indices are harmonic means with weighting factors referring to
the time t (current year):
[ ]
{ ⁄
( )}
[ ]
⁄
( )
{ }
where qjt and ikt are respectively the quantity of product (j) and factor k at
time (t), while wjt and xkt are the weights of product j and factor k within the
agricultural sector, and L and P are Laspeyres and Paasche, respectively. Finally,
the geometrical average of the Laspeyres and the Paasche index gives the
Fischer index, which benefits from the most suitable statistical properties. In
formula, the TFP Fisher index is computed as follows:
√
In 1992, Diewert presented a comprehensive argument advocating for the

use of the Fisher Index to calculate Total Factor Productivity (TFP), which is
also found in the work of Coelli et al. (1997). This study adopts the Fisher Index
as the method for measuring TFP or Multi-Factor Productivity (MFP), and it is
applied as follows:
TFP =
If the TFP has a value of more than one (TFP> 1), then the productivity is
high and vice versa (Zen and Budiasih, 2018). Our intension is to apply method
outlined above. However, as we only have one period observation data, we could
not construct index of output, Fisher, Paasche or Laspeyres. Therefore, we end
up calculating productivity as ratio of the total value of output and total input
used. Results from this method will be our main results for this chapter.

Please also note that, in addition, we estimate total factor productivity (or
multifactor productivity) as an average residual of the production function of
Cobb-Douglas type, or also known as the growth accounting method. However,
the result from this growth accounting method is not robust, and so we keep the
result from this method in the appendix.
DATA COLLECTION
In 2019, the number of urban farmers in the city if Yogyakarta was 3,056
(Yogyakarta City Agriculture and Food Security Office, 2021), and most of
them actively join the farmer group. There are 230 urban farmer groups. The
distribution of the farmers across the sub-region is displayed in Figure 1.
Figure 1. Urban Farmers in the City of Yogyakarta
Source: Yogyakarta City Agriculture and Food Security Office, 2021

Data collection was carried out by interviewing farmers (mostly the
mother) in June 2021. Interview was carried out by students from the Faculty of
Economics and Business UGM. We set the number of farmers chosen 460, based
on the Slovin formula. Quota to each sub districts is set proportionally to the
number of farmers. Then, we distribute the sample evenly among the farmers
group in each sub-district, and finally, conveniently choose farmers from the
farmer groups. Because of the pandemic, some time the sub-district was lock
downed so it was difficult to do interview. We managed to visit 376 farmers.
Figure 2 shows the distribution of the sample after deleting the incomplete
response.

The sub-district with the largest samples of urban farmers is Kotagede with
a total of 42 urban farmers. Kotagede is located at the suburb of Yogyakarta City,
there is still a lot of agricultural land in the area. Then, the sub-district with the
least number of urban farmers sampled is Pakualaman with only a total of 6
urban farmers. Pakualaman is located in the central area of Yogyakarta City and
has the narrowest area when compared to other sub-districts, agriculture in
this sub-district is limited due to dense settlements and limited land.
Figure 2. The Distribution of the Selected Farmers
Source: Survey (2021)

We collect data relevant to the urban farming, such as inputs, production,
and the distribution (types of use) of the products. We noted there are many
ways of planting including hydroponics, vertical farming, planting above the
fishponds using buckets/drums and others. There is no standard of the size
of equipment, growing media, and measuring the product used by farmers.
Therefore, we work together with the local extensionists in converting the sizes
to obtain a uniform measurement. Table 1 provides list of variables collected
that are relevant to calculating the productivity.

Table 1. Unit of Measurement
No. Variable Unit No. Variable Unit
1 Land Area m 2
5 Flower Bundled
produced
2 Input Urban Farming IDR 6. Livestock Tail (the
Spending grown livestock itself)
3 Urban Farming Income IDR 7 Fish harvested Fish
4 Vegetable produced Bundled 8 Harvest Time Month
Figure 3 shows the urban farming land area in each sub-district. The total
land area owned by all sample urban farmers in the City of Yogyakarta vary
across the sub-districts. In the central City of Yogyakarta, the urban areas such
as Ngampilan, Gondokusuman, Danurejan, Gondomanan, and Pakualaman,
tend to have less and limited land areas. For sub-districts Mantrijeron, Jetis,
and Tegalrejo, have area close to rural. So, in general, farmers in a more urban
area have limited land/space for farming, while those living in the suburb have
more space.
Figure 3. Land Area Each District by Sample (M2)

Urban farming uses various types of planting media such as hydroponics,
polybags, pot, including vertical plants. Hydroponic agriculture uses water
utilisation methods without using soil media as the main place and emphasizing
the fulfilment of nutrients for agriculture that is carried out (Figure 4). Various
planting media used in hydroponic agriculture are husk charcoal, sponge, sand,
gravel, water, and others. Hydroponic system provides more yields, easy to
harvest, reduces environmental pollution, and among others. The hydroponic
planting process is carried out with limited land media and is effective in
increasing agricultural production.
Figure 4. Hydroponic Planting at KWT Makmur, Mantrijeron, Yogyakarta
Source: Survey documentation (2021)

In addition to planting using hydroponics, farmers also use polybags and
potted plants. When farming with polybag, the size of land used is calculated
based on the size of the polybag. Farming using polybags is usually done in
nurseries and vegetable crops as well as family medicinal plants (Tanaman
Obat Keluarga/TOGA) (Figure 5). The size of these polybags varies, both small
with a width of 10-14 cm and large (15 cm and above) depending on farmer
preference and the type of plant being planted. For example, the polybag with
a size of 20 cm x 20 cm can be interpreted as the land used is 400 cm2 or 0.04
m2. Then we calculate how many polybags used by the farmer. We do similarly
for pot and vertical farming. The size of land for pot media is measured based
on the diameter of the pot base. The smallest with a diameter of 10 cm. Farmers
use pots to grow fruits, family medicinal plants, flowers, and others

Figure 5. Polybag and Potted Plants with Vertical Media
Source: Survey Documentation (2021)

Vertical farming or viticulture is agricultural cultivation is carried out
vertically using paralon (PVC pipe) or bottles in stages in a narrow space (Figure
6). Farmers grow vegetables for household consumption such as celery, lettuce,
spinach, mustard greens, and others. Another vertical farming is planting on the
wall. It is a plastic sheet with many pockets hung on the wall. Farmers put some
media in the pockets to grow vegetables.
Figure 6. Potted Plants with Vertical and Wall Plant
Source: Survey Documentation (2021)

The yield of each urban farming sector (vegetable, flower, livestock, and
fish) can be seen in Table 2 and Table 3. Table 2 shows the yields of vegetables
and flowers. Mustard has the highest yield, 1.289 bundles per harvest. Mustard
can grow in polybags, hydroponics, and vertical plants, so it is widely planted
by urban farmers. Other vegetables produced are chili, kale, pulses, tomatoes,
eggplant, and swamp cabbage. Table 2 also presents the yield for flowers. Orchid
is the most preferred plant to grow, with the total production of 1.214 orchid
plants. It also can be planted on vertical media. Other flowers are caladium,
adenium, rose, aglaonema, rose moss, and jasmine.

Table 2. Vegetable and Flower Yield
Urban Farming Sector Harvest Time (months) Total Yield (Bundled)
Vegetables
Chili 3 707
Mustard 1 1,289
Eggplant 3 152
Swamp Cabbage 1 88
Tomato 3 403
Pulses 2 423
Kale 1 488
Other Vegetables – 44
Flowers
Orchid 4 1,214
Caladium 3 704
Aglaonema 3 146
Rose 4 251
Adenium 3 595
Jasmine 3 27
Rose moss 3 56
Other Flowers – 639
Urban farmers also produce livestock (chicken, birds, duck, and rabbit)
and fish, as reported in Table 3. Chicken is the highest, with a total production
of 846 chickens. Chickens have the fastest yield and require a smaller space
compared to other livestock. The four main fish produced are catfish, tilapia,
gourami, decorative fish, and butterfish. Catfish have good resistance when
compared to other types of fish and can survive in small ponds, so catfish are
widely farmed. The production is 35.239 catfish per harvest.
Table 3. Livestock and Fish Yield
Urban Farming Sector Harvest Time/monthly Production
Livestock
Chickens 3 846 chickens
Birds 12 36 birds
Ducks 2 15 ducks
Rabbits 12 14 rabbits

Fish
Catfish 3 35,239 fish
Tilapia 6 7,839 fish
Gourami 6 229 fish
Decorative Fish 4 13 fish
Butterfish 4 3,925 fish
As shown in Table 4, the net income of Urban Farming per month. On
average, urban farmers in Yogyakarta City can earn net income IDR 223,148.75
in a month. The Regional Minimum Wage (UMR) for the City in 2021 is IDR
2,069,530 per month. Therefore, urban farming activities in the city of
Yogyakarta cannot be a secondary income for urban farmers. Besides, it could
also have other non-economic benefits considering that the motivation of
farming is mostly for own consumption (81,9%), while the rest are hobby/
recreation (71%), for market (34,5%), and for social activities (19,1%).
Table 4. Income Urban Farming per Month in Yogyakarta (in IDR)
Urban Total Number Net Income
Total Input Net Income
Farming Revenues of Each Farmer
Cost (IDR)
Sector (IDR) Farmers (IDR)
Vegetables 9,836,000.00 4,874,046.00 4,961,954.00 65 76,337.75
Flowers 7,497,000.00 2,464,288.00 5,032,712.00 33 152,506.42
Livestock 20,730,000.00 846,439.00 19,883,561.00 28 710,127.18
Fish 10,510,000.00 1,782,495.00 8,727,505.00 47 185,691.60
Total 48,572,000.00 9,967,267.00 38,604,733.00 173 223,148.75

Table 5 shows farmers expenditure on input (IDR)/month. Most of the
inputs such as seeds, fertilizers, media, feed, and others, have been provided by
the Yogyakarta City Agriculture Office. Therefore, the costs incurred by urban
farmers are quite low for all types of urban farming. The largest expenditure
is spent on vegetables, which is IDR 74,985.00 each farmer. For flowers, input
expenditure of IDR 74,675.00 each farmer, then fish amounting to IDR 37,925.00
each farmer, while the lowest expenditure is for livestock IDR 30,230.00 each
farmer.

Table 5. Average Monthly Cost for Each Urban Farming Sector (IDR)
in Yogyakarta City
Urban Farming Number of Input Cost Each Farmer
Input Cost (IDR)
Sector Farmers (Average) IDR
Vegetables 4,874,046 65 74,985
Flowers 2,464,288 33 74,675
Livestock 846,439 28 30,230
Fish 1,782,495 47 37,925
Total 9,967,267 173 57,614
TOTAL FACTOR PRODUCTIVITY ESTIMATES

As mentioned earlier, we calculated productivity by determining the ratio of
the total value of output to the total input used. We categorised urban farming
activities in Yogyakarta into four types: (1) livestock farming, (2) fish farming,
(3) vegetable farming, and (4) flower plant farming. The productivities for each
of these farming activities are presented in Table 6.
Table 6. Total Factor Productivity (TFP) Estimate of the Urban Farming in
Yogyakarta City
Productivity Score
Livestock Productivity 0.77
Fish Productivity 2.47
Vegetable Productivity 0.57
Productivity of Flower Plants 1.19
Among the commodities produced by urban farmers, the productivity
levels are as follows: livestock (0.77), fish (2.47), vegetables (0.57), and flower
plants (1.19). The highest productivity is observed in fish farming, particularly
catfish. Fish farming requires minimal inputs, such as feed and space, while
yielding a significant output. Many farmers utilize buckets provided by the city
office for catfish farming, which can serve as a valuable source of protein and
income for families.

Following fish, flower farming demonstrates a productivity level of 1.19.
Harvesting flowers usually involves taking parts of the tree or the flower itself,
eliminating the need for seeds or new plants. Additionally, flower gardening
may be more familiar to farmers compared to vegetable cultivation, resulting in
higher productivity in flower farming.
On the other hand, livestock farming exhibits a lower productivity level of
0.77. The lengthy production process, with birds and rabbits taking 12 months
to harvest, contributes to this lower productivity. Urban farming activities for
livestock necessitate a significant allocation of inputs, such as feed, and require
adequate land area. Among the commodities, vegetable production shows the
lowest productivity (0.57) in urban farming in Yogyakarta City. This may be
attributed to farmers’ relative unfamiliarity with vegetable farming. Vegetables
are typically cultivated in higher-altitude areas or greenhouses with controlled
temperatures rather than in the open air of the city. Consequently, farming
vegetables in the city requires higher input expenditures, including fertilizer
and high-quality growing media.
It can be concluded that the urban farming activities in Yogyakarta, fish
and flower farming demonstrate relatively higher efficiency. On the other hand,
livestock and vegetable farming in Yogyakarta have not reached optimal levels
of efficiency.
PRODUCTIVITY LEVELS OF DIFFERENT COMMODITIES

The productivity levels observed among different commodities in urban farming
activities in Yogyakarta provide some important policy implications. First, given
the high productivity of fish farming, particularly catfish, there is a need to
encourage and support fish farming initiatives. This can be done by providing
training, resources, and access to technology for efficient fish farming practices.
Additionally, the provision of necessary infrastructure, such as buckets or tanks,
can further enhance fish farming productivity.
Second, flower farming has shown relatively high productivity in urban
areas. To capitalize on this potential, policymakers can facilitate the availability
of land or spaces for flower farming. Providing technical assistance and
knowledge-sharing platforms can also help farmers improve their flower
farming techniques and marketing strategies.

Third, livestock farming has lower productivity levels, largely due to
the lengthy production process and significant input requirements. Policy
interventions should focus on streamlining the production process, improving
access to feed and resources, and providing training on efficient livestock
farming practices. Additionally, exploring alternative livestock options that
require fewer resources and have shorter production cycles could be considered.
Fourth, vegetable farming has the lowest productivity among the studied
commodities, which may be attributed to farmers’ limited familiarity with
vegetable farming techniques. Policymakers should prioritize providing
training programs and knowledge transfer platforms to educate farmers on best
practices for vegetable cultivation in urban environments. This could include
techniques for maximizing space utilization, optimizing resource allocation, and
adopting innovative farming methods such as vertical farming or hydroponics.
Finally, continuous research and development efforts should be undertaken
to explore innovative farming methods, improve productivity, and address
specific challenges faced by urban farmers in Yogyakarta. Collaboration with
agricultural institutions, universities, and relevant stakeholders can help in
developing tailored solutions and technologies that enhance overall urban
farming efficiency.
CONCLUSION
This chapter presents an estimation of the productivity of urban farming in
Yogyakarta City. The data used for this analysis were collected through a survey
conducted among 250 urban farming households across all sub-districts in the
city. Although the sample size was lower than the initial target of 460 households
and 376 farmers were visited due to incomplete responses—i.e., partially
impacted by the 2021 pandemic—we provide descriptive characteristics of
urban farming and report estimated productivity levels for future discussions.
This information can be valuable for the Office of Agriculture and Food Security
of the City.
The productivity levels in the urban farming sector in Yogyakarta City vary
across different types of farming. The productivity values obtained are 2.47 for
fish production, 1.19 for flowers, 0.77 for livestock, and 0.57 for vegetables. Fish
farming and flower gardening can be considered efficient and productive, while
livestock and vegetable farming exhibit lower efficiency and productivity.

To better support the urban farming programme, it is crucial to provide
additional encouragement and facilities particularly for those engaged in
fish and flower farming. Simultaneously, prioritising training and knowledge
transfer to improve productivity is of utmost importance for those involved in
livestock and vegetable farming.
Further surveys are necessary to gain a better understanding of the factors
contributing to these productivity differences. Analysing these factors will
enable the development of appropriate policies to support urban farming in
Yogyakarta, as well as in Indonesia as a whole. It is worth noting that measuring
productivity in farming, as often encountered in research on agricultural
and traditional sectors, is not a simple task. Farming practices are often not
standardised in terms of inputs and output measurement. Therefore, caution
must be exercised when surveying farmers. Further surveys are still required
to calculate productivity accurately and assess its growth using more precise
indices.
REFERENCES
Agbonlahor, M.U, Momoh S., & Dipeolu A.O. (2007). ‘Urban Vegetable Crop
Production and Production Efficiency.’ International Journal of Vegetable
Science, 13 (2): 63-72. doi: 10.1300/J512v13n02_06
Buhaug, Halvard & Henrik Urdal (2013). ‘An urbanization bomb? Population
growth and social disorder in cities.’ Global Environmental Change 23 (1):
1-10. Available at: https://doi.org/10.1016/j.gloenvcha.2012.10.016.
Chandra, A. J & J.A.Diehl (2019). ‘Urban agriculture, food security, and
development policies in Jakarta: A case study of farming communities at
Kalideres – Cengkareng district, West Jakarta.’ Land Use Policy 89. Available
at: https://doi.org/10.1016/j.landusepol.2019.104211.
Coelli, T., S. Rahman, & C. Thirtle (2003). ‘A stochastic frontier approach to total
factor productivity measurement in Bangladesh crop agriculture, 1961-
92.’ Journal of International Development, 15(3): 321–333. Available at:
https://doi.org/10.1002/jid.975
Coelli, T.J., D.S. Prasada Rao, & G.E. Battese (1997). An Introduction to Efficiency
and Productivity Analysis. Boston: Kluwer Academic Publishers.

Davis, A., R.E. Major, & C.E. Taylor (2016). ‘Do trees flower longer in the city?
A comparison of flowering eucalyptus trees in streets, remnants and
continuous forest and their association with nectarivorous birds.’ Urban
Ecosyst 19: 735–747. Available at: https://doi.org/10.1007/s11252-015-
0515-z
Dorr, et. al. (2023). ‘Food production and resource use of urban farms and
gardens: a five-country study.’ Agron. Sustain. Dev. 43, 18. Available at:
https://doi.org/10.1007/s13593-022-00859-4
FAO. (2018). Statistics, R. (n.d.). Guidelines for the measurement of productivity
and efficiency in agriculture.
Fuglie, K. O. (2004). ‘Productivity growth in Indonesian agriculture, 1961-2000.’
Bulletin of Indonesian Economic Studies, 40(2): 209–225. Available at:
https://doi.org/10.1080/0007491042000205286
Fuglie, K. O. (2010). ‘Sources of growth in Indonesian agriculture.’ Journal of
Productivity Analysis, 33(3): 225–240. Available at: https://doi.org/10.
1007/s11123-009-0150-x
Glavan, et al. (2018). ‘The economic performance of urban gardening in three
European cities – examples from Ljubljana, Milan and London.’ Urban
Forestry & Urban Greening 36. Available at: https://doi.org/10.1016/j.
ufug.2018.10.009.
Grafius, D.R., J.L. Edmondson, B.A.Norton et al. (2020). ‘Estimating food
production in an urban landscape’. Sci Rep 10: 5141. Available at: https://
doi.org/10.1038/s41598-020-62126-4
Kamel, H. (2015). Estimate of the elasticities of substitution of the CES and
translog production functions in Tunisia Estimate of the elasticities of
substitution of the CES and translog production functions in Tunisia Kamel
Helali * and Maha Kalai. March. Available at: https://doi.org/10.1504/
IJEBR.2015.068544
Korth, Marcel, Ruth Stewart, Laurenz Langer, Nolizwe Madinga, Natalie Rebelo
Da Silva, Hazel Zaranyika, Carina van Rooyen, & Thea de Wet (2014) ‘What
are the impacts of urban agriculture programs on food security in low and
middle-income countries: A systematic review.’ Environ. Evid. 3.

Kryszak, Łukasz., Katarzyna Świerczyńska, & Jakub Staniszewski (2023).
‘Measuring total factor productivity in agriculture: A bibliometric review.’
International Journal of Emerging Markets 18, no. 1: 148-172.
McDermott, et. al. (2010). ‘Sustaining intensification of smallholder livestock
systems in the tropics.’ Livestock Science 130 (Issues 1–3): 95-109. Available
at: https://doi.org/10.1016/j.livsci.2010.02.014.
Mougeot, L. (2000). ‘Urban Agriculture - Definition, Presence, potential and
Risks and Policy Challenges, Growing Food.’ Urban Agriculture on Policy
Agenda. La Ha Vana, Cuba, International Workshop on Growing cities.
Nasrudin, Nasrudin & Yusuf Donner Dwiyantama (2022). ‘Urban F Urban
Farming: Empowerment t arming: Empowerment to Increase Economic,
Economic, Education, and Nutritional Benefit for the Sub-Urban
Community.’ ASEAN Journal of Community Engagement, 6(2): 294-306.
Available at: https://doi.org/10.7454/ajce.v6i2.1159
Nicholls, Elizabeth., et al. (2020). ‘The contribution of small‑scale food
production in urban areas to the sustainable development goals: a review
and case study.’ Sustainability Science, 15:1585–1599. doi: https://doi.
org/10.1007/s11625-020-00792-z.
Orsini, F., R. Kahane, R. Nono-Womdim, & G. Gianquinto (2013) ‘Urban agriculture
in the developing world: a review.’ Agron. Sustain. Dev. 33: 695–720.
Pribadi, et al. (2017) Multifunctional adaption of farmers as response to
urban growth in the Jabodetabek Metropolitan Area, Indonesia. Journal
of Rural Studies 55, 100-111. Available at: https://doi.org/10.1016/j.
jrurstud.2017.08.001
Resosudarmo, B.P. & S. Yamazaki (2010). “Training and Visit (T&V) Extension
vs. Farmer Field School: The Indonesian Experience”. In V. Beckmann,
N.H. Dung, X. Shi, M. Spoor, & J. Wesseler (eds.). Economic Transition and
Natural Resource Management in East and Southeast Asia. Aachen: Shaker
Verlag, 260-298.
Resosudarmo, B.P. 2012. ‘Implementing a National Environmental Policy:
Understanding the ‘Success’ of the 1989-1999 Integrated Pest Management
Programme in Indonesia.’ Singapore Journal of Tropical Geography, 33(3):
365-380.

Siswati, T., S. Iskandar, N. Pramestuti, J. Raharjo, A.K. Rubaya, & B.S. Wiratama.
(2022). ‘Drivers of Stunting Reduction in Yogyakarta, Indonesia: A Case
Study.’ Int. J. Environ. Res. Public Health, 19: 16497. Available at: https://
doi.org/10.3390/ijerph192416497
Smit, J., A. Ratta, & J. Nasr (1996). “Urban agriculture: Food, jobs, and sustainable
cities. United Nations Development Programme.” Publication series for
Habitat II, Vol. 1. New York: UNDP.
Smith, J., J. Nasr, & A. Ratta (2001). ‘Benefits of Urban Agriculture, Urban
Agriculture Food, Jobs and Sustainable Cities.’ The Urban Agriculture
Network, Inc
Taylor, J. R. (2020). ‘Modelling the Potential Productivity of Urban Agriculture
and Its Impacts on Soil Quality Through Experimental Research on Scale-
Appropriate Systems.’ Frontiers in Sustainable Food Systems, 4 (July): 1–18.
Available at: https://doi.org/10.3389/fsufs.2020.00089
Sullivan, C. A. O., G.D. Bonnett, C.L. Lyre, Z. Hochman, & A.P. Wasson (2019).
‘Strategies to improve the productivity, product diversity and profitability
of urban agriculture.’ Agricultural Systems, 174 (December): 133–144.
Available at: https://doi.org/10.1016/j.agsy.2019.05.007
Suparwoko & Betri Taufani (2017). ‘Urban Farming Construction Model on the
Vertical Building Envelope to Support the Green Buildings Development
in Sleman, Indonesia.’ Procedia Engineering 171: 258-264. Available at:
https://doi.org/10.1016/j.proeng.2017.01.333.
Sutrisna, Nana (2020). ‘Urban Agricultural Development for Food Security
at The Time Of COVID-19 Pandemies in Indonesia.’ Sumatra Journal of
Disaster, Geography and Geography Education 4, no. 2 (December): 165-
172.
Yuan, et al. (2023). ‘Corrigendum to A review on urban agriculture: Technology,
socio-economy, and policy.’ Heliyon, 9. Available at: https://doi.org/10.
1016/j.heliyon.2023.e13165
World Bank (2013). “Urban Agriculture, Finding from Four City Case Studies.”
Urban Development Series Paper, Urban Development & Resilience Unit

Zezza, A. & L. Tasciotti (2010). ‘Urban agriculture, poverty, and food security:
Empirical evidence from a sample of developing countries.’ Food Policy
35: 265–273. Available at: https://jogjadaily.com/2020/11/pertanian-
perkotaan-solusi-ketahanan-pangan-saat-pandemi/
The Jakarta Post (2020). Available at: https://www.thejakartapost.com/
news/2020/10/27/urban-farming-a-solution-to-food-security-issues-
during-pandemic.html
APPENDIX
Appendix 1. Estimates of Urban Farming Production Function Using the
Cobb-Douglas Type
Vegetable Flower Livestock Fish
Ln_size land 0,194** 0,0279 Ln size land (M)2 0,0186 0,00577
(M)2
0,0774 0,0927 0,0279 0,075
Ln_hours 0,183 0,552* Ln_hours allocated 0,0431 0,00291
allocated farming
farming
0,265 0,313 0,0889 0,251
Ln_seeds (Rp) 0,0715 0,0713** Ln_feed (Rp) 0,175*** 0,252***
0,0753 0,0333 0,01 0,0227
Ln_fertilizer 0,0949 0,0527 Ln_water (Rp) 0,0258 0,119
(Rp)
0,0764 0,0387 0,0288 0,114
Ln_water (Rp) 0,0235 0,0543 Ln_cage-space 0,0316 0,0575
(Rp)
0,0362 0,0438 0,0202 0,0432
Constant 0,822** 0,436 Ln_medicine (Rp) 0,0583** 0,0504
0,316 0,386 0,0282 0,0432
Constant 0,1 0,0655
R-squared 0,215 0,215 0,109 0,294
R-squared 0,831 0,687

Note: All price variables are in real terms. Although the estimates are not statistically impressive,
for the vegetable production, land is the key input that is statistically significant; hours of work
spent in the farm and seeds are significant in flower production; feed and medicine are key to
livestock production; and feed is also significant in fish production functions.

CHAPTER 5
Mental Health Consequences of

COVID-19 Pandemic on Teachers
and Caregivers of Early Childhood
Education
Ni Wayan Suriastini, Fita Herawati, Ika Yulia Wijayanti,
Indrawan Firdauzi, Setyo Pujiastuti, and Sukamtiningsih
INTRODUCTION
In the early time of COVID-19 pandemic, people’s mental health got worsened,
and the number even doubled in some countries (OECD, 2021). According to
World Health Organisation (WHO), the prevalence of anxiety and depression
increased by 25% globally (WHO, 2022a). Several research has examined
mental health impact on various group of people; nonetheless, there is limited
number which focus on teachers and parents of Early Childhood Education
(ECE).
Early stage of COVID-19 pandemic has brought unexpected change to the
role of teachers and parents due to school closure and online learning policy.
Online learning is an unprecedented case, especially for ECE (Alan, 2021).
It is likely that the transition to online learning will give negative impact on
young children attending ECE since early years education is aimed at providing
real-life sensory experiences, hands-on activities, and meaningful face-to-face
interactions, which are hard to achieve virtually (Spiteri, 2021).
91
Moreover, teachers and parents have very little guidance on online learning
that is appropriate for young children (Ford et al., 2021). In addition to other
challenges during the COVID-19 pandemic—i.e., such as economic insecurity
and social restriction—, teaching and caregiving responsibilities potentially
heightened the risk of mental health problem. Mental health issue among
caregivers and teacher is an important concern since they are the key human
resources to determine the quality of learning. Quality of child-teacher and child-
parent relationship are among potential resource factors that enable children
in ECE to adapt positively with current and future adversity (Miller-Lewis et
al., 2013). Because of that, caregivers and teachers should be considered as
important dependent variables in terms of how they affect children resilience
(Siu & Chan, 2020).
Previous research revealed that some countries made adjustment to
online learning system, including curriculum adaptation (Aletheiani, 2021;
Lafave et al., 2021; Spiteri, 2021). However, a curriculum that does not take
into consideration a pandemic scale creates unnecessary pressure and anxiety
not just on young children but also on the adults who care for and teach them
(Spiteri, 2021).
The Government of Indonesia has implemented certain adjustments in the
education system in response to the COVID-19 pandemic (Aletheiani, 2021).
In light of the learning from home policy that was enforced on 17 March 2020
(Suriastini et al., 2020), the Ministry of Education and Culture introduced an
early education curriculum consisting of 12 learning modules for ECE. These
modules were designed to facilitate a simplified learning process for ECE
teachers. However, there are concerns that the adaptation to the new curriculum
may also pose a burden on teachers and impact their mental well-being. Yet,
there is a lack of evidence regarding whether this burden leads to mental health
issues among teachers.
This chapter, hence, aims to identify mental health condition of ECE
teachers and caregivers during the COVID-19 pandemic in Special Region of
Yogyakarta Province (Daerah Istimewa Yogyakarta), Indonesia. Identifying the
problem is important to decide appropriate strategies to solve the problem and
ensure the success of the learning process in early childhood education and
school readiness for the next educational level. This study used cross-sectional
data taken from the Study of Early Childhood Education Curriculum Adaptation

during the COVID-19 Pandemic in Special Region of Yogyakarta Province
conducted from January 2021 to February 2021 to achieve the objective of the
study. The main finding is that higher education attainment is the significant
protective factor affecting anxiety and depression symptoms during online
learning. These findings highlighted that low educated teacher and caregivers
were in the higher risk of mental health problems, so that they have to be the
target of technical support as well as mental health services to prevent further
negative impact on children.
LITERATURE REVIEW ON ECE-RELATED MENTAL HEALTH

ISSUES
A growing body of research has focused on identifying the mental health issues
that have arisen during the COVID-19 pandemic. The OECD has highlighted a
significant shift in the prevalence of mental health conditions, with the pandemic
leading to a pronounced deterioration in this area (OECD, 2021). Similarly, the
World Health Organization (WHO) has reported that prior to the pandemic,
a substantial number of individuals experienced major depressive disorder
and anxiety disorders. Adjusted estimates after accounting for the impact of
COVID-19 show a significant increase in these conditions, representing a 28%
rise for major depressive disorders and an increase of 26% for anxiety disorders
in just one year (WHO, 2022b).
Within the realm of Early Childhood Education, mental health has become
a prominent research topic during the COVID-19 pandemic, with a particular
focus on children, teachers, and families (Su et al., 2022). This interest has been
driven by the disproportionate impact of social restrictions on teachers and
caregivers (Peris & Ehrenreich-May, 2021a; WHO, 2022b). Teachers have had
to enhance their IT skills during the transition from offline to online teaching,
while parents have taken on the role of tutors during remote schooling (Su et
al., 2022). These challenges—coupled with factors such as economic hardship,
illness, restless children, intense isolation, and poor social support—have
contributed to an increase in mental health problems among caregivers (Alan,
2021; Peris & Ehrenreich-May, 2021a). Exposure to a wide range of disasters
has been found to have a negative impact on the mental health of families,
leading to prolonged periods of heightened anxiety and depression symptoms
(Russell et al., 2020).
Mental Health Consequences of COVID-19 Pandemic on Teachers 93

and Caregivers of Early Childhood Education
Studies have shown that teachers have experienced adverse psychological
symptoms, including stress, anxiety, and depression, during the pandemic
(Ozamiz-Etxebarria et al., 2021; Santamaría et al., 2021). Similarly, many
caregivers have faced mental health problems such as stress, anxiety, depression,
and post-traumatic stress disorder (Calvano et al., 2022; Khoury et al., 2021).
A study conducted by Kishida et al. (2021) in Japan revealed that full school
closure was associated with significantly higher scores for both child and parent
mental health problems compared to full school reopening.
The demands of parenting during the COVID-19 crisis have necessitated
the active planning of new caregiving, work, and education routines, potentially
compromising parents’ ability to sustain their own emotional well-being
and self-care (Russell et al., 2020). Given the educational responsibilities of
caregivers, their mental health conditions are of vital importance as young
children often rely on their assistance for accessing and completing online
learning activities (Lau et al., 2021).
Supporting parents is crucial in order to provide quality home-based
learning for young children during the COVID-19 pandemic (Zhang et al.,
2021). Conversely, parental stress during this period has been shown to have
a negative effect on children’s engagement in home learning activities and may
even increase the risk of adverse childhood experiences (Calvano et al., 2022;
Sonnenschein & Stites, 2021). From teacher side, mental wellbeing is closely
related to their ability to support their students. As pointed out by Jennings
(2015), depression was negatively associated emotional support, classroom
organisation, and instructional support. In contrast, teachers’ coping abilities
were related to their positive reactions to children’s negative emotions (Buettner
et al., 2016).
CROSS-SECTIONAL DATA OF EARLY CHILDHOOD

EDUCATION
This study used cross-sectional data from the Study of Early Childhood
Education (ECE) Curriculum Adaptation during the COVID-19 Pandemic in
Special Region of Yogyakarta Province (Daerah Istimewa Yogyakarta) conducted
from January 2021 to February 2021. SurveyMETER collected the data with
support from Knowledge Sector Initiative (KSI) core funding. Henceforth,
Yogyakarta province refers to Special Region of Yogyakarta to make it simple.

Total population of ECE in Yogyakarta province is 5,300 including Kindergarten
(Taman Kanak-kanak/TK), Playgroup (Kelompok Bermain/KB) and other
informal ECE or Special Programme School (Satuan PAUD Sejenis/ SPS). The
sample target of this survey is 500 ECE, 1,000 teachers (two teachers per ECE),
and 500 caregivers. Sample was selected by proportional random sampling for
each type of ECE in five districts (kabupaten/kota) in Yogyakarta province. We
conducted simple random sampling to select school samples and individual
samples in each school sample.
In fact, many playgroups and SPS have only one teacher during the COVID-19
Pandemic. After scrutinising process, we obtained a total of 751 teachers and
500 parents/caregiver samples with completed responses. It is important to
note that the final samples might not representative toward population. Indeed,
it is difficult to obtain representative samples in rapid surveys undertaken
during the COVID-19 crisis (OECD, 2021). Moreover, we used phone survey
in this study due to social restriction, so that time constrain in each interview
really mattered.
We are fully aware that this survey has limited variables that potentially
affect mental health conditions since its focus was observing adaptation process
toward 12 learning modules emergency curriculum implementation. As a
consequence, we might not be able to construct a good model in predicting the
determinants of mental health problems. However, we tried to include variables
used by previous researchers in analysing mental health problems during the
pandemics.
As shown in Appendix 1, the prevalence of anxiety symptoms amongst
teachers was 17.18%, while prevalence of depressive symptoms was higher
at 21.17%. Almost half of respondents (49.27%) were kindergarten teachers.
Respondents were dominated (64.05%) by urban schoolteachers. Almost all
teachers (99.07%) were female. The average age of teachers was 42.96 years old.
Education attainment among teacher respondents was quite good with most of
them (66.84%) completing higher education (universities or college), while the
rest of them completed Senior High School with 0.67% only completing Junior
High School or less.
Such high proportion of teachers who had side job (40.75 percent) raise
the curiosity that income from teaching might not be sufficient to meet their
needs and increase vulnerability of mental health problem caused by economic

hardship during COVID-19 pandemic crisis. Unfortunately, most of teachers did
not use 12 learning modules to provide lesson plan which was importantly used
by students and parents in online learning setting. There was only 31.56% of
them used it.
There were also such high proportion (42.32%) of teachers faced difficulties
to access internet, in contrast only few of them (5.86%) faced difficulties with
regard to Information, Communication, and Technology (ICT). Around half
of teacher respondents (52.73%) received technological support which was
useful in the online teaching. The prevalence of anxiety and depression among
caregivers were higher than teacher. It accounted for 20.40% for anxiety and
24.20% for depression (see Appendix 2). Caregiver respondents were also
dominated by kindergarten caregivers (47.20%) and located in urban school
(57.80%). Similar with the teachers, almost all caregiver respondents were
female (96.40%).
The average age of caregivers was 33.63 years old. Around half of them
completed Senior High School (53.60%), while the remaining caregivers who
completed Junior High School or less and Higher Education share relatively
the same proportion at around 23%. Less than half of caregiver respondents
were working during the survey (43.40%). In online learning setting, ideally
caregivers assist their children to study according to the lesson plan provided by
teachers. Nonetheless, not all of caregiver respondents utilised the lesson plan.
Those who have utilised it accounted for 61.80%. With regard to difficulties
in online learning, around 32.80% of them revealed that they faced difficulties
with internet access, while 4% of caregivers stated that they have difficulties
with information and technology.
EMPERICAL STRATEGY
The main model used in this study is a logistic regression model, which examines
the relationship between variables associated with mental health issues. The
model includes measures of mental health conditions on the one hand and
various potential factors on the other. It is important to note that this study
focuses on associations rather than establishing causality.
Some studies use prevalence symptoms of anxiety and depression to
address mental health problems during COVID-19 pandemic (Alegría et al.,
2022; OECD, 2021; Russell et al., 2020; Vahratian et al., 2021). According to WHO,

anxiety disorders are characterised by excessive fear and worry and related
behavioural disturbances. Symptoms are severe enough to result in significant
distress or significant impairment (WHO, 2022b). Meanwhile, depression or
depressive disorder involves a depressed mood or loss of pleasure or interest in
activities for long periods of time. People who have lived through abuse, severe
losses or other stressful events are more likely to develop depression. Women
are more likely to have depression than men (WHO, 2023).
Regardless many limitations, this survey used validated instruments, i.e.,
General Anxiety Disorder (GAD-7) for anxiety and Patient Health Questionnaire-9
(PHQ-9) for depression which were difficult to conduct by other surveys in
the early of pandemic (OECD, 2021). These two measures of mental health
conditions had been used in many other studies (Solomou & Constantinidou,
2020; Vahratian et al., 2021).
The GAD-7 is commonly used as a measure of general anxiety symptoms.
It measures the symptoms that can be identified through 7 signs: (1) feeling
nervous, anxious, or on edge; (2) being able to stop or control worrying; (3)
worrying too much about different things; (4) trouble relaxing; (5) being
restless; (6) becoming easily annoyed or irritable; and (7) feeling afraid as if
something awful might happen (Johnson et al., 2019). Each item is scored on a
four-point Likert scale (0–3) with total scores ranging from 0 to 21 with higher
scores reflecting greater anxiety severity.
The total score is categorised into four levels of anxiety severity, which are:
(1) Minimal (0-4); (2) Mild (5-9); (4) Moderate (10-14); and (4) Severe (15-21)
(Spitzer et al., 2006). In this study, we re-categorised these four levels into binary
categories, i.e., no anxiety which consist of minimal level of anxiety and anxiety
consisting of mild to severe anxiety. The Patient Health Questionnaire-9 (PHQ-
9) is the most commonly used depression screening tool in primary and general
settings, with cut-off value ≥10 used as a standard to identify major depression
(Negeri et al., 2021). It consists of nine items measuring depressive symptoms
corresponding to the diagnostic criteria for major depressive disorder. Each
item is scored on a four-point Likert scale (0–3) with scores ranging from 0
to 27, with higher scores reflecting greater depression severity. According to
Kroenke et al. (2001), there are five categories for the total score, which are: (1)
Minimal (0-4); (2) Mild (5-9); (4) Moderate (10-14); and (4) Moderately severe
(15-190; and (5) Severe (20-27).

Even though we used validated instrument to measure mental health
conditions in this study, some limitations have to be noted in interpreting the
finding. As pointed by Vahratian et al. (2021), since this survey data were based
on self-report and were not confirmed by health professional, the question about
mental health condition might be predictive but do not necessarily reflect a
clinical diagnosis. According to WHO (2022a), mental health conditions are led by
interaction of individuals vulnerabilities and stressors. Therefore, as covariates,
we used some individual characteristics to indicate individual vulnerabilities.
As for potential stressors, and we included school characteristics and some
variables related online learning and emergency curriculum adaptation.
Type of ECE and location of school were selected as school characteristics.
Our respondents were grouped into three types of ECE in Indonesia, i.e.,
kindergarten (TK), playgroup (KB), and special programme school (SPS).
Kindergarten is ECE programmes for children aged 4-6 years old, while
playgroup for children aged 2-4 years old. In addition, SPS is other type of ECE
which is available for children aged 0-6 years old and integrated with other
community service. We did not employ heterogeneity analysis because our
sample is quite small to be broken down.
According to previous studies, demographic characteristics are identified
as determinant of teachers and parents of mental health (Deeb et al., 2022;
Kishida et al., 2021; OECD, 2021; Solomou & Constantinidou, 2020). In this
study, we used sex, age, education, and working condition as individual
characteristics. Working status was applied for caregivers, while for teachers
we categorised whether they have side job or not. Those working conditions are
used to describe economic security of respondents since according to previous
studies, it is very important factors affect mental health during COVID-19
pandemics (Alegría et al., 2022; Deeb et al., 2022; OECD, 2021).
Technology is fundamental component in online learning, so that teachers
must have the necessary technological skill to deliver learning material (Alan,
2021). To accommodate this matter, we include some difficulties such as
difficulties in internet access and difficulties in ICT. We also add ICT support
received by teachers since other study pointed out that teachers need training
in technical and technological issues (Dayal & Tiko, 2020). Finally, curriculum
adaptation is captured by lesson plan based on 12 learning modules for teachers
and lesson plan utilisation for caregivers.

Determinants of ECE Teacher’s and Caregiver’s Mental Health during
Covid-19 Pandemic
The results of logistic regression analysis are presented in Appendix 3. The
type of school has a significant impact on the mental health of teachers and
caregivers. In comparison to kindergarten teachers, SPS teachers had a 0.56
times lower probability of experiencing anxiety symptoms. Additionally,
playgroup caregivers were 0.61 times less likely to experience depression
symptoms compared to kindergarten teachers. These findings can be attributed
to the lower burden faced by SPS and playgroup teachers and caregivers, as
their roles are less focused on ensuring school readiness for the next formal
educational level. Moreover, SPS is considered non-formal ECE, allowing for
more flexible curricula and targets.
The location of the school did not show a significant association on teacher’s
mental health. However, caregivers in urban areas are associated with a higher
probability of experiencing anxiety and depressive symptoms. Urban caregivers
had a 1.56 times higher likelihood of anxiety symptoms and a 1.76 times higher
likelihood of depression symptoms compared to their counterparts in non-
urban areas.
Higher education attainment consistently had a negative association with
the mental health problems of both teachers and caregivers. For instance,
caregivers who completed Senior High school had a 0.51 times lower probability
of experiencing anxiety and a 0.56 times lower probability of experiencing
depression compared to those who completed Junior High school or less. Similar
results were found for other mental health indicators among teachers and
caregivers. A study by Vahratian et al. (2021) supported these findings, showing
that the prevalence of anxiety and depressive symptoms increased the most
among individuals with less than a high school education. Quality education is
considered one of the protective factors for mental health (WHO, 2022b).
Online learning is closely related to ICT utilisation. Higher education
attainment enhances the ability to use the internet, programs, and devices
required for online learning. Additionally, more educated teachers may
experience less depression and stress due to their increased capability in
dealing with challenging child behaviour (Friedman-Krauss et al., 2014).

These findings align with prior research on mental health outcomes
among parents during the COVID-19 pandemic, which indicated that higher
parental education levels were associated with significantly lower odds of
anxiety, depression, and PTSD (Deeb et al., 2022). Alongside family income and
teacher support, parental education is an important protective factor against
parental stress during the pandemic (Zhang et al., 2021). Individuals with lower
education levels have experienced higher rates of mental distress during the
COVID-19 crisis (OECD, 2021).
Sex did not have a significant association with mental health indicators,
although the estimation of anxiety symptoms in the teacher sample was
omitted. As reflected in the descriptive statistics, the majority of teacher and
caregiver respondents were female, potentially influenced by gender roles. The
low percentage of working caregivers also did not significantly associate with
mental health, despite numerous studies indicating that economic hardship
is an important determinant of mental health problems during the COVID-19
pandemic. In Asia, including Indonesia, women’s labour force participation is
low as they are often expected to prioritise family and childcare responsibilities
due to traditional gender roles (Schaner & Das, 2016).
Teacher’s anxiety symptoms were not associated with having a side job,
but those with a side job experienced depressive symptoms 1.43 times more
frequently than those without a side job. Since a side job serves as a proxy for
insufficient income from teaching salaries, this finding suggests that economic
insecurity increases the risk of mental health problems during the COVID-19
pandemic crisis, which is consistent with other studies (Alegría et al., 2022).
Given the low proportion of teachers utilising the 12 learning modules to
create lesson plans, it did not have a significant association with their mental
health condition. Almost all teachers (96.27% of our samples) prepared lesson
plans, but not all of them used the 12 learning modules emergency curriculum
as a reference. Our observation indicates that the majority of teachers prefer
to use the Basic Competence and Core Competence of the 2013 Curriculum
(Kompetensi Dasar dan Kompetensi Inti), accounting for 68.97% of teachers in
our sample. Other commonly used references were the School-based curriculum
(Kurikulum Tingkat Satuan Pendidikan/KTSP) and the Standard Level of Child
Development Achievement (Standar Tingkat Pencapaian Perkembangan Anak/
STPPA), with 60.19% and 55.93% utilisation, respectively. The preference
for these references reflects the familiarity of teachers with regular curricula
or guidelines used in normal circumstances. Although the new emergency
curriculum offers a simpler teaching standard, teachers require more time and
technical assistance to learn and become familiar with it. Some respondents
noted a lack of socialisation regarding the 12 learning modules, with only
41.15% of teachers receiving training and information on these modules
according to our data.
In our study context, introducing new modules during the pandemic crisis
was less effective to reduce the teaching burden and the risk of mental health
problems. Limited socialisation and teacher difficulties in understanding the
new modules would reduce their preference to utilise it. Since teachers have
been familiar with existing curriculum or guidelines, it will be more effective
to provide guidelines and material supplements regarding how to simplify the
learning target or how to deliver learning material in a practical way under
online learning scheme.
However, parents who utilise the lesson plan shared by teacher were less
likely to experience both anxiety and depressive symptoms than those who did
not utilise it. The probability to experience anxiety and depression is reduced
by 0.5 times respectively. It indicated the importance of lesson plan as the
guidance of caregivers in assisting their children to study at home. During the
crisis, lesson plan can facilitate family-school cooperation since the families
have neither the tools nor enough time to educate their children at home (Otero-
Mayer et al., 2021). Therefore, guideline about how to adapt and prepare lesson
plan based on familiar curriculum in emergency situation is urgently needed by
teachers to reduce their burden. It should be considered by education authority
in anticipating future similar crisis.
While no significant association with caregivers, some difficulties regarding
online learning tend to heighten mental health problems among teachers.
Difficulties in the internet access increased the probability to experience anxiety
symptoms by 1.5 times, while no significant effect on depressive symptoms.
Furthermore, difficulties in ICT increased the risk of teachers to experience
anxiety symptoms by 2.49 times, while increased the depressive symptoms by
2 times. This finding is relevant with study by Jeon et al. (2018), which found
that teacher’s competence has negative association with depressive symptoms.

Technical support like training or webinar about online learning technology
associated with lower probability of experiencing depressive symptoms among
teachers. Those who received the support were 0.51 less likely to experience
depressive symptoms than their counterparts. Some technical supports in
curriculum adaptation during emergency situation are crucial because teachers
need to improve their technological competencies, increase the number of
interactive resources, establish a user-friendly online learning platform, and
serve families while struggling to maintain their psychological well-being (Alan,
2021). They need to be well-equipped with diverse training on content, formats,
and delivery methods for teaching mentoring (Ford et al., 2021).
Strategies to Reduce the Burden on Kindergarten Teachers and

Caregivers
Policy measures should be taken to address the mental health of teachers and
caregivers in Early Childhood Education (ECE). Specifically, attention should
be given to those working in kindergarten settings, as they are more likely
to experience anxiety and depression symptoms compared to SPS (Special
Programme School) teachers and playgroup caregivers. Strategies should be
developed to reduce the burden on kindergarten teachers and caregivers and
provide necessary support to promote their mental well-being.
Urban areas tend to be associated with higher rates of anxiety and depressive
symptoms among caregivers. Policymakers should consider implementing
targeted interventions and support systems to address the mental health needs
of caregivers in urban settings, potentially including access to mental health
services and resources. Policies should prioritise promoting higher education
attainment for both teachers and caregivers. Higher education levels are
consistently associated with lower probabilities of anxiety and depression
symptoms. Initiatives should be developed to improve access to quality
education of teachers and caregivers. Some of alternative programs for teachers
are professional development and training to enhance the mental well-being
and cope with the work stressor. In addition, policymakers should promote
initiatives that support and enhance parental education levels, including access
to educational resources, guidance, and support systems.
Policies should address economic insecurity and insufficient income for
teachers. Side jobs serving as proxies for insufficient teaching salaries are
associated with increased depressive symptoms. Measures should be taken to
improve the economic well-being of teachers, including fair compensation and
financial support.
Guidelines and support for lesson plan adaptation are essential during
emergency situations. Teachers prefer familiar curriculum references rather
than new modules, suggesting the need for guidance on simplifying learning
targets and delivering practical online learning materials. Education authorities
should provide clear guidelines and materials to assist teachers in adapting
lesson plans to emergency situations, reducing their burden and promoting
effective home-based learning. Online learning and ICT utilisation should be
supported through technical assistance and training for teachers. Policymakers
should ensure that teachers receive the necessary technical support, training,
and teaching resources to effectively adapt to online learning environments.
Technical support and training for teachers in online learning technologies
are crucial. Teachers who receive technical support are less likely to experience
depressive symptoms. Policymakers should prioritise providing teachers with
diverse training opportunities to improve their technological competencies and
ensure they have the necessary resources to deliver effective online instruction
while maintaining their well-being.
CONCLUSION
This study aimed to investigate the mental health consequences of the COVID-19
pandemic and the utilisation of emergency curriculum on teachers and caregivers
in Early Childhood Education (ECE). Among the various determinants of mental
health, we focused on several key findings.
One significant finding was that higher education attainment acted as a
protective factor against anxiety and depression symptoms during online
learning. These findings underscore the increased risk of mental health
problems among teachers and caregivers with lower levels of education. As a
result, targeted technical support and mental health services should be provided
to the latter group to prevent further negative impacts on children.
The introduction of new modules as an emergency curriculum during
the pandemic crisis proved to be less effective. The lack of socialisation and
difficulties that teachers faced in understanding the new modules reduced
their preference to utilise them. In such emergency situations, it becomes more
important to provide guidelines on how to simplify the existing curriculum and

communicate it in an easy and practical manner to caregivers. Additionally,
teachers require technical support to find suitable methods and overcome other
challenges posed by the online learning setting.
Policy measures are necessary to address the mental health of ECE teachers
and caregivers. Specifically, strategies should focus on reducing the burden on
kindergarten teachers and caregivers and providing support for their mental
well-being. Interventions and support systems should be implemented to
address the mental health needs of caregivers in urban areas. Promoting higher
education attainment for teachers and caregivers is crucial for protecting
against anxiety and depression symptoms. Policymakers should ensure teachers
receive technical support, training, and resources to adapt to online learning
environments effectively. Emphasizing parental education as a protective factor
for parental stress and mental health is important. Policies should address
economic insecurity and insufficient income for teachers. Clear guidelines and
support are needed for lesson plan adaptation during emergency situations.
Providing technical support and training in online learning technologies is
essential for reducing depressive symptoms among teachers.
Despite the limitations inherent in the data and sampling procedures
used in this study, the findings align with previous research and offer valuable
insights for health and education policymakers. These findings underscore
the significance of identifying risk factors for adverse outcomes among
vulnerable groups and emphasize the need for policies that provide support
to these groups. However, it is important to acknowledge the need for further
research in this area to ensure the reliability and robustness of our conclusions.
Continued investigation and exploration of these topics will contribute to a
more comprehensive understanding of the mental health consequences of the
COVID-19 pandemic on teachers and caregivers in the field of Early Childhood
Education.
REFERENCES
Alan, Ü. (2021). ‘Distance Education During the COVID-19 Pandemic in Turkey:
Identifying the Needs of Early Childhood Educators.’ Early Childhood
Education Journal, 49(5): 987–994. Available at: https://doi.org/10.1007/
s10643-021-01197-y
Alegría, M., M. Cruz-Gonzalez, I.S. O’Malley, K. Alvarez, G. L. Stein, L. Fuentes, K.
Eddington, C. Poindexter, S.L. Markle, A.N. Thorndike, L. Zhang, & P.E. Shrout
(2022). ‘Role of social determinants in anxiety and depression symptoms
during COVID-19: A longitudinal study of adults in North Carolina and
Massachusetts.’ Behaviour Research and Therapy, 154. Available at: https://
doi.org/10.1016/j.brat.2022.104102
Aletheiani, D. R. (2021). ‘Curricular responsiveness to the Covid-19 crisis: The
case of Indonesia.’ Prospects, 51(1–3): 463–472. Available at: https://doi.
org/10.1007/s11125-020-09537-3.
Buettner, C. K., L. Jeon, E. Hur, & R.E. Garcia (2016). ‘Teachers’ Social–
Emotional Capacity: Factors Associated With Teachers’ Responsiveness
and Professional Commitment.‘ Early Education and Development, 27(7):
1018–1039. Available at: https://doi.org/10.1080/10409289.2016.1168
227
Calvano, C., L. Engelke, J. Di Bella, J. Kindermann, B.Renneberg, & S.M.Winter
(2022). ‘Families in the COVID-19 pandemic: parental stress, parent mental
health and the occurrence of adverse childhood experiences—results
of a representative survey in Germany.’ European Child and Adolescent
Psychiatry, 31(7): 1–13. Available at: https://doi.org/10.1007/s00787-
021-01739-0
Dayal, H. C., & L. Tiko (2020). ‘When are we going to have the real school? A
case study of early childhood education and care teachers’ experiences
surrounding education during the COVID-19 pandemic.’ Australasian
Journal of Early Childhood, 45(4): 336–347. Available at: https://doi.
org/10.1177/1836939120966085
Deeb, S., D. Madden, T. Ghebretinsae, J. Lin, U. Ozbek, V. Mayer, & N. Vangeepuram
(2022). ‘Child Disruptions, Remote Learning, and Parent Mental Health
during the COVID-19 Pandemic.’ International Journal of Environmental
Research and Public Health, 19(11). Available at: https://doi.org/10.3390/
ijerph19116442

Ford, T. G., .K.A. Kwon, & J.D. Tsotsoros (2021). ‘Early childhood distance learning
in the U.S. during the COVID pandemic: Challenges and opportunities.’
Children and Youth Services Review, 131(November). Available at: https://
doi.org/10.1016/j.childyouth.2021.106297
Friedman-Krauss, A. H., C.C. Raver, P.A. Morris, & S.M. Jones (2014). ‘The Role
of Classroom-Level Child Behaviour Problems in Predicting Preschool
Teacher Stress and Classroom Emotional Climate.’ Early Education and
Development, 25(4): 530–552. Available at: https://doi.org/10.1080/104
09289.2013.817030
Jennings, P. A. (2015). ‘Early Childhood Teachers’ Well-Being, Mindfulness, and
Self-Compassion in Relation to Classroom Quality and Attitudes Towards
Challenging Students. ‘Mindfulness, 6(4): 732–743. Available at: https://
doi.org/10.1007/s12671-014-0312-4
Jeon, L., C.K. Buettner, & A.A. Grant (2018). ‘Early Childhood Teachers’
Psychological Well-Being: Exploring Potential Predictors of Depression,
Stress, and Emotional Exhaustion.’ Early Education and Development,
29(1): 53–69. Available at: https://doi.org/10.1080/10409289.2017.134
1806.
Johnson, S. U., P.G. Ulvenes, T. Øktedalen, & A. Hoffart (2019). ‘Psychometric
properties of the GAD-7 in a heterogeneous psychiatric sample.’ Frontiers
in Psychology, 10 (July): 1–8. Available at: https://doi.org/10.3389/
fpsyg.2019.01713.
Khoury, J. E., H. Kaur, & A. Gonzalez (2021). ‘Parental Mental Health and
Hostility Are Associated With Longitudinal Increases in Child Internalizing
and Externalizing Problems During COVID-19.’ Frontiers in Psychology, 12
(July): 1–9. Available at: https://doi.org/10.3389/fpsyg.2021.706168
Kishida, K., M. Tsuda, P. Waite, C. Creswell, & S. Ishikawa (2021). ‘Relationships
between local school closures due to the COVID-19 and mental health
problems of children, adolescents, and parents in Japan.’ Psychiatry Research,
306. Available at: https://doi.org/10.1016/j.psychres.2021.114276
Kroenke, K., Spitzer, R. L., & Williams, J. B. W. (2001). ‘The PHQ-9: Validity
of a brief depression severity measure.’ Journal of General Internal
Medicine, 16(9): 606–613. Available at: https://doi.org/10.1046/j.1525-
1497.2001.016009606.x
Lafave, L., A.D. Webster, & C. McConnell (2021). ‘Impact of COVID-19 on Early
Childhood Educator’s Perspectives and Practices in Nutrition and Physical
Activity: A Qualitative Study.’ Early Childhood Education Journal, 49(5):
935–945. Available at: https://doi.org/10.1007/s10643-021-01195-0
Lau, E. Y. H., J. Li., & K. Lee (2021). ‘Online Learning and Parent Satisfaction during
COVID-19: Child Competence in Independent Learning as a Moderator.’
Early Education and Development, 32(6): 830–842.
Miller-Lewis, L. R., A.K. Searle, M.G. Sawyer, P.A. Baghurst, & D. Hedley (2013).
‘Resource factors for mental health resilience in early childhood: An
analysis with multiple methodologies.’ Child and Adolescent Psychiatry and
Mental Health, 7 (1): 1–23. Available at: https://doi.org/10.1186/1753-
2000-7-6.
Negeri, Z. F., B. Levis, Y. Sun, C. He, A. Krishnan, Y. Wu, P.M. Bhandari, D. Neupane,
E. Brehaut, A. Benedetti, & B.D. Thombs (2021). ‘Accuracy of the Patient
Health Questionnaire-9 for screening to detect major depression: updated
systematic review and individual participant data meta-analysis.’ The BMJ,
vol.375. BMJ Publishing Group. Available at: https://doi.org/10.1136/bmj.
n2183
OECD (2021). Tackling the mental health impact of the COVID-19 crisis: An
integrated, whole-of-society response. Available at: https://read.oecd-
ilibrary.org/view/?ref=1094_1094455-bukuf1f0cm&title=Tackling-the-
mental-health-impact-of-the-COVID-19-crisis-An-integrated-whole-of-
society-response
Otero-Mayer, A.,A. González-Benito, B. Gutiérrez-de-Rozas, & C. Vélaz-de-
Medrano (2021). ‘Family-School Cooperation: An Online Survey of Parents
and Teachers of Young Children in Spain.’ Early Childhood Education
Journal, 49(5): 977–985. Available at: https://doi.org/10.1007/s10643-
021-01202-4
Ozamiz-Etxebarria, N., N. Idoiaga Mondragon, J. Bueno-Notivol, M. Pérez-
Moreno, M., & J. Santabárbara (2021). ‘Prevalence of Anxiety, Depression,
and Stress among Teachers during the COVID-19 Pandemic: A Rapid
Systematic Review with Meta-Analysis.’ Brain Sciences, 11(9). Available at:
https://doi.org/10.3390/brainsci11091172

Peris, T. S., & J. Ehrenreich-May (2021a). ‘The Parents Are Not Alright: A Call
for Caregiver Mental Health Screening During the COVID-19 Pandemic.’
Journal of the American Academy of Child & Adolescent Psychiatry, 60(6):
675–677. Available at: https://doi.org/10.31234/osf.io/xzf2c
Peris, T. S., & J. Ehrenreich-May (2021b). ‘The Parents Are Not Alright : A Call
for Caregiver Mental Health Screening During the COVID-19 Pandemic.’
Journal of the American Academy of Child & Adolescent Psychiatry, 60(6):
675–677.
Russell, B. S., M.Hutchison, R. Tambling, A.J. Tomkunas, & A.L. Horton (2020).
‘Initial Challenges of Caregiving During COVID-19: Caregiver Burden,
Mental Health, and the Parent–Child Relationship.’ Child Psychiatry
and Human Development, 51(5): 671–682. Available at: https://doi.
org/10.1007/s10578-020-01037-x
Santamaría, M. D., N.I. Mondragon, N.B. Santxo, & N. Ozamiz-Etxebarria (2021).
‘Teacher stress, anxiety and depression at the beginning of the academic
year during the COVID-19 pandemic.’ Global Mental Health, 8. Available at:
https://doi.org/10.1017/gmh.2021.14
Schaner, S., & S. Das (2016). “Female Labor Force Participation in Asia: Indonesia
Country Study.” ADB Economics Working Paper Series. 474. Available at:
https://www.adb.org/sites/default/files/publication/180251/ewp-474.
pdf
Siu, A. F. Y. & A.H. C. Chan (2020). ‘Preschool teachers’ resilience and their
readiness to develop resilience to young children in a Chinese context.’
Asia-Pacific Journal of Research in Early Childhood Education, 14(1): 49–69.
Available at: https://doi.org/10.17206/apjrece.2020.14.1.49
Solomou, I., & F. Constantinidou (2020). ‘Prevalence and predictors of anxiety
and depression symptoms during the COVID-19 pandemic and compliance
with precautionary measures: Age and sex matter.’ International Journal
of Environmental Research and Public Health, 17(14): 1–19. Available at:
https://doi.org/10.3390/ijerph17144924
Sonnenschein, S., & M.L.Stites (2021). ‘The Effects of COVID-19 on Young
Children’s and Their Parents’ Activities at Home.’ Early Education and
Development, 32(6): 789–793. Available at: https://doi.org/10.1080/104
09289.2021.1953311
Spiteri, J. (2021). Quality early childhood education for all and the Covid-19 crisis:
A viewpoint. Prospects, 51(1–3): 143–148. https://doi.org/10.1007/
s11125-020-09528-4
Spitzer, R. L., Kroenke, K., Williams, J. B. W., & Löwe, B. (2006). ‘A brief measure
for assessing generalized anxiety disorder: The GAD-7.’ Archives of Internal
Medicine 166, no.10: 1092–1097. Available at: https://doi.org/10.1001/
archinte.166.10.1092
Su, J., Ng, D. T. K., Yang, W., & Li, H. (2022). Global Trends in the Research on
Early Childhood Education during the COVID-19 Pandemic: A Bibliometric
Analysis. Education Sciences, 12(5). Available at: https://doi.org/10.3390/
educsci12050331
Suriastini, W., Setyo, P., Herawati, F., Naryanta, Moddilani, G., & Pellini, A. (2020).
COVID-19 and the education response in Indonesia: exploring the learning
crisis. EdTech Hub.
Vahratian, A., Blumberg, S. J., Terlizzi, E. P., & Schiller, J. S. (2021). ‘Symptoms
of Anxiety or Depressive Disorder and Use of Mental Health Care Among
Adults During the COVID-19 Pandemic-United States.’ MMWR Morb
Mortal Wkly Rep, 70: 490–494. Available at: https://doi.org/http://dx.doi.
org/10.15585/mmwr.mm7013e2
WHO (2022a). World mental health report: Transforming mental health for all.
Available at: https://www.who.int/publications/i/item/9789240049338
WHO (2022b). Mental disorders. Available at: https://www.who.int/news-
room/fact-sheets/detail/mental-disorders
WHO (2023). Depressive disorder (depression). Available at: https://www.who.
int/news-room/fact-sheets/detail/depression
Zhang, C., W. Qiu, H. Li, J. Li, L.Zhang X. Li, & J. Li (2021). ‘Parental Stress and
Home Activities for Young Children during the Stay-at-home Quarantine
Time in China.’ Early Education and Development, 32(6): 843–862. Available
at: https://doi.org/10.1080/10409289.2021.1949561

APPENDICES
Appendix 1. Characteristics of Teachers
Teachers (N=751)
Characteristics
N % M SD Min Max
Mental Health Condition
Anxiety Symptoms 129 17,18 .171771 .3774326 0 1
Depressive Symptoms 159 21,17 .2117177 .408798 0 1
Type of School
Kindergarten 370 49,27
Playgroup 204 27,16 1.743009 .8143716 1 3
SPS 177 23,57
Location
Urban 481 64,05
.6404794 .4801797 0 1
Rural 270 35,95
Sex
Female 744 99,07
.9906791 .0961579 0 1
Male 7 0,93
Age
Age (year) 751 42,96 8,96
Education
Junior High School and less 5 0,67
Senior High School 244 32,49 2661784 .4872962 1 3
Higher education 502 66,84
Teachers have side job
Yes 306 40.75 .4074567 .4916886 0 1
Lesson plan based on 12
modules
Yes 237 31.56 .3155792 .4650559 0 1
Difficulties in internet access
Yes 318 42.34 .4234354 .4944324 0 1
Difficulties in ICT
Yes 44 5.86 .0585885 .2350095 0 1
Technical support for
teachers
Yes 396 52.73 .5272969 .499587 0 1
Appendix 2. Characteristics of Caregivers
Caregivers (N=500)
Characteristics
N % M SD Min Max
Mental Health Condition
Anxiety Symptoms 102 20.40 .204 .4033726 0 1
Depressive Symptoms 121 24.20 .242 .4287232 0 1
Type of School
Kindergarten 236 47.20
Playgroup 146 29.20 1764 .8084623 1 3
SPS 118 23.60
Location
Urban 289 57.80
.578 .4943731 0 1
Rural 211 42.20
Sex
Female 482 96.40 .1864767 0 1
.964
Male 18 3.60
Age
Age (year) 457 33,63 5,55 22 59
Education
Junior High School and less 115 23.00
Senior High School 268 53.60 2.004 .6818459 1 3
Higher education 117 23.40
Caregiver’s working status
Working 217 43.40 .434 .4961212 0 1
Lesson plan utilization by
caregivers
Yes 309 61.80 .618 .4863631 0 1
Yes 164 32.80 .328 .469955 0 1
Difficulties in ICT
Yes 20 4.00 .04 .1961554 0 1

Appendix 3. Determinants of Teacher’s and Caregiver’s Mental Health
during COVID-19 Pandemic
Mental Health Problems
Teacher’s (N=751) Caregiver’s (N=500)
Covariates Anxiety Depressive Anxiety Symptoms Depressive
Symptoms Symptoms Symptoms
OR Std. Err OR Std. Err OR Std. Err OR Std. Err
Type of School (Kindergarten as reference)
Playgroup 1.195 0.325 0.676 0.173 0.763 0.220 0.606* 0.164
SPS 0.786 0.242 0.557** 0.158 1.131 0.322 0.848 0.229
Location (Rural as reference)
Urban 0.818 0.173 0.884 0.175 1.561* 0.382 1.759** 0.408
Sex (male as reference)
Female 1.422 1.566 3.238 2.569 2.044 1.384
Age
Age (year) 1.011 0.012 0.997 0.0108 1.034 0.0213 0.991 0.0196
Education (Junior High School and less as reference)
Senior High
0.139** 0.132 0.0689** 0.0796 0.507** 0.138 0.556** 0.143
School
Higher
0.105** 0.101 0.0526** 0.0613 0.472** 0.170 0.410** 0.143
education
Teachers have side job
Yes 0.987 0.209 1.429* 0.278
Caregivers working status
Working 1.232 0.298 1.051 0.240
Provide lesson plan based on 12 modules
Yes 1.149 0.247 1.147 0.231
Lesson plan utilization by parents
Yes 0.504*** 0.119 0.497*** 0.110
Yes 1.498* 0.314 1.112 0.218 1.500 0.380 1.265 0.305
Difficulties in information and technology
Yes 2.486*** 0.865 2.014** 0.697 0.876 0.529 0.981 0.555
Receives technical Support regarding online learning technology
Yes 0.861 0.179 0.511*** 0.0996
Constant 0.977 1154 5.259 9.166 0.0418*** 0.0479 0.413 0.423
*** p<0.01, ** p<0.05, * p<0.1
CHAPTER 6
Marine Fisheries Sustainable

Livelihoods in Pulau Tiga,
Natuna Islands
Teguh Asandy and Wilmar Salim
INTRODUCTION
Indonesia is an archipelago with two-thirds of its area being marine waters.
With the vast extent of the seas, Indonesia had a high fisheries potential of 10
million tons in 2015 (Suman, Irianto, Satria, & Amri, 2016). Approximately 12%
of the total national fish production comes from Wilayah Pengelolan Perikanan/
Fish Management Area (WPP) 711 covering Karimata strait, Natuna sea, and
South China Sea (Suman, Irianto, Satria, & Amri, 2016), where Natuna Islands
is located. With this geographical condition, fishing is the primary source of
livelihood for the people of Natuna Islands.
The main commodities of Natuna fishermen are reef and demersal
fish, which are not only consumed locally but also exported nationally and
internationally (Wardono B et al, 2020). In 2015, the total fish catch of
Indonesian fishermen in the Natuna Sea reached nearly 49 thousand tons, and it
continued to experience a growth of 2.87%. Unfortunately, this growth is faster
than the regeneration of fish in the sea. As a result, the amount of fish stocks in
the Natuna Sea continues to decline (Ministry of Marine Affairs and Fisheries
Indonesia, 2016). If this trend continues, there will eventually be a point where
there is not enough fish to sustain the livelihoods of fishermen in Natuna.
113
To improve the condition of fisheries in Natuna, the government of Indonesia
had implemented the illegal ship sinking policy during the administration of the
former Minister of Marine and Fishery, Susi Pudjiastuti, in 2016, yet such attempt
has not been able to permanently improve fishermen’s welfare (Resosudarmo
and Kosadi, 2018). It indicates that increasing fishermen’s welfare is more
complex than just sinking illegal ship to preserve fish. For this reason, an
analytical tool that can describe contributing factors more comprehensively is
needed. This research initiates the use of the sustainable livelihoods concept
proposed by the World Commission on Environment and Development (WCED)
(1987) to analyse the fisheries sector in the context of developing coastal rural
areas in Natuna Islands.
Figure 1. Maps of Indonesia Fisheries Management Area Including WPP
711 Natuna Sea (Top Yellow); Natuna Islands (Left Bottom); Pulau Tiga
District (Right Bottom)
Source: Ministry of Marine Affairs and Fisheries Indonesia (2014); Natuna

Public Works and Spatial Planning Office (2020); Own Analysis (2020)
The sustainable livelihood approach can provide an overview of the
complexity of the problems of coastal communities that depend solely on the
fisheries. Sustainable livelihood itself means that there is enough income, in the
form of food and money, to fulfil the basic needs to sustain life (WCED, 1987).
This means that to achieve a sustainable livelihood for Natuna fishermen, at least
enough production is needed for consumption and regeneration. Sustainability
refers to maintaining or increasing the productivity of resources over the long
term (WCED, 1987).
This study can be useful not only for the current conditions but also for the
future so that people can be adaptive to the problems they face especially for
policy making in Natuna Islands and programmes related to fishermen and the
marine fisheries. The purpose of this study was to determine the condition of
Sustainable Livelihoods (SL) in the fisheries sector in the context of developing
coastal rural areas in Natuna Islands. The discussion will focus on the sustainable
livelihood framework, namely local capital assets, vulnerabilities, local policies,
and institutions. The location of this study is in Sabang Mawang Village, Sabang
Mawang Barat Village, and Tanjung Batang Village in Pulau Tiga District, Natuna
Islands, where an Integrated Marine and Fisheries Center/Sentra Kelautan dan
Perikanan Terpadu (SKPT) is also located (see Figure 1). The fieldwork in Pulau
Tiga District was conducted between July and August 2020.
SUSTAINABLE LIVELIHOODS FRAMEWORK

A household can secure a sustainable livelihood in many ways through
stable work with adequate wages or a variety of activities (WCED, 1987). On
a primary level, sustainability means there is an ability to recognise current
and past conditions and make connections between them; on a secondary
level, sustainability means going beyond what is ‘primary’ by thinking about
the future—and plans one might have, thus “higher consciousness enables
its possessors to transcend the limits of the currently remembered primary
consciousness” (Edelman, 2003). Carney (1998) defines sustainable livelihood
as the ability to cope with and recover from stresses and shocks and maintain
or enhance its capabilities and assets both now and in the future, while not
undermining its natural resource base.
According to Serrat (2017), parameters that can be used to analyse
sustainable livelihoods are local capital assets or livelihood assets, vulnerability
contexts, structure and process trough policies and institutions, and sustainable
Marine Fisheries Sustainable Livelihoods in Pulau Tiga, Natuna Islands 115

livelihood strategies and outcomes. In further discussion, strategy and output
components become part of the results of the analysis of local capital assets,
vulnerabilities, and institutions. This analysis will only use three parameters
by using the Sustainable Livelihood Framework (SLF) set by Department of
International Development (DFID) as stated by UNDP in 2017.
The SLF’s role is to provide a comprehensive analytical overview
emphasizing strong linkages where community capacity and vulnerability are
discussed, and coping strategies are developed (UNDP, 2017). The SLF adopts
a systemic approach to understanding livelihoods and provides guidance
in conceptualizing: (1) The goods and assets that people need; (2) How the
community earns income; (3) The context in which the needed assistance is
provided; (4) Other factors that provide resilience in times of crisis (UNDP,
2017). The three parameters mention above are related to each other as shown
in Figure 2.
Figure 2. Sustainable Livelihoods Framework
Source: DFID (1999), in UNDP (2017)

Figure 2 shows that SLF can be used to elaborate the component of
each sustainability parameters. As shown in Figure 2, we can look at the
vulnerabilities that occur, from shocks, trends, and seasonality. The impact of
vulnerability will create difficulties for the affected community. The extent to
which a community can reduce vulnerability is determined by its livelihood
assets. Livelihood assets consist of human capital, natural capital, financial
capital, social capital, and physical capital. To support livelihood assets in the
form of these five capitals, there are structures and processes in the form of
policies from institutions that can assist in increasing the community’s capital
in facing vulnerability. Institutions and livelihood assets influence each other
and provide access to inputs in achieving the expected livelihood outcome.
Livelihood Assets
Livelihood assets consist of five elements, which might have three potential
relationships among them (Rahman et al., 2018), i.e., sequentially related (one
asset ensures the availability of others, substitutability, and combinability
(combination of assets sustain livelihood activities. They may be described as
follows:
Natural Capital
Natural capital means a stock of natural resources from which further resources
and services can be developed that may be useful for livelihoods such as land
and soil, food production, marine and forest resources, air quality, erosion
protection, waste disposal, storm protection, water supply, carbon storage, and
sequestration (UNDP, 2017). Environmental sustainability can be determined
by analysing the condition of soil and land quality, water, fish availability, forests,
and biodiversity (Allison & Ellis, 2001).
Social Capital
Analysis in social capital is conducted to understand the ability of a community
to strengthen community organisational skills, build existing institutions,
increase awareness of the structure and function of social organizations, build
trust, provide leadership training, encourage the inclusion of marginalized
groups, and support networks (Allison & Horemans, 2006).
Human Capital
Analysing the condition of human capital can be done through analysing
training, education, awareness-raising, food improvement, security and
access to a better diet, and increased access to health and education services
(Allison & Horemans, 2006). Direct support can focus on asset accumulation
like development of infrastructure for health, education, or training, capacity

building of staff involved in the health, education, or training sector, and
development of relevant experience and skills among the most vulnerable
population groups (UNDP, 2017).
Physical Capital
Physical capital is the basic infrastructure and facilities built to support the
community’s livelihood process by improving the physical environment so that
people can be more productive. Infrastructure is generally a public facility that
is for free as it is funded by taxpayers except for certain infrastructures such as
housing, electricity, toll roads, and drinking water. Physical capital includes basic
infrastructure such as roads and transportation, markets, irrigation buildings,
housing, etc. (DFID, 2001).
Financial Capital
Financial/economic capital includes reserves or inventory, savings, deposits,
or movable goods that are easily cashable to achieve people’s livelihood goals
(DFID, 2001). It can be in the form of basic resources such as cash, credit or
debt, savings, and other economic resources or assets including infrastructure
equipment, basic production, and technology that are essential for pursuing
any livelihood strategy (Scoones, 1998).
Vulnerability
Vulnerability is characterized as insecurity in the well-being of individuals,
households, and communities in the face of changes in their external environment.
The vulnerability has two aspects—i.e., external shocks caused by seasonality
and critical trends; and internal aspect caused by the lack of capabilities
to overcome thread. Shock can be caused by conflict, disease, flood, storm,
drought, and pest. Seasonality will affect prices and employment opportunities.
Meanwhile, critical trends determine demographic, environmental, economic,
governance, and technological trends (Serrat, 2017).
Institutional Structure and Processes

Policy structures are public and private sector organisations that establish and
implement policies and legislation, provide services, and buy trade and perform
all manner of other functions that affect livelihoods (Serrat, 2017). Processes
include laws, regulations, policies, operational arrangements, agreements,
societal norms, and practices that, in turn, determine how policy structures
operate. Policy-making structures cannot be effective if there are no proper
institutions and processes to implement policies.
In supporting the analysis of local policies and institutions, stakeholder
analysis will be used to define aspects of social and natural phenomena that are
influenced by factors, which are among others: decisions or actions; identities
of individuals, groups, and organizations that are affected by or may influence
parts of the phenomenon (this may include non-human and non-living entities
and future generations); and prioritizing these individuals and groups to be
involved in the decision-making process (Reed et al, 2009). To understand
how stakeholders can be engaged, for example, for instrumental purposes
classification to determine key players, context setters, subjects, and crowds by
understanding interest and influence of each stakeholder (Eden & Ackermann,
1998; Lopez, 2001).
DATA GATHERING AND METHOD OF ANALYSIS

This is qualitative research using data gathered from interview with key
informants/ resource persons who understand the problems studied and
questionnaire with structured questions. Observation is also done to validate
data gathered (Sugiyono, 2011). Interview with key informants is done by several
steps. First, eight respondents from seven local government were interviewed,
followed by interviews with ten sources representing the community at the level
of sub-district, village, and local fisheries. Documentation and observations
related to the physical condition are then gathered to validate information
provided by the sources.
Questionnaires to local fishermen are used to support the triangulation
process by gathering information to find out the phenomenon of fishermen’s
sustainable livelihoods in more depth and as a form of final validation in the
data collection process. There are total of 56 fishermen responded to the
questionnaire about their assets: 12 respondents from Sabang Mawang Village,
17 from Sabang Mawang Barat Village, and 27 from Tanjung Batang Village.
Secondary data are used to support data that have been obtained verbally and
have been documented during the research survey process. This secondary
data include annual reports, maps, rules, and other written data that can assist
in improving the quality of analysis in the research.

The analytical method used descriptive qualitative analysis assisted by
a sustainable livelihood framework, which consists of three components. For
target one, after the variables of local capital assets have been analysed in-depth,
an overall assessment or scoring is carried out representing each of the existing
local capital assets, namely natural, social, human, physical, and financial capital
with definition of each score presented in Table 1. To analyse vulnerability
qualitative descriptive analysis is obtained from sources related to the things
that have an impact on the livelihoods of local fishermen using. To analyse local
policies and institutions, the actors involved will be identified according to the
rules and their impact on local fishermen, then the interests and influence of
each actor will be determined into four categories: high influence-high interest
(the key players); high influence-low interest; low influence-high interest; and
low influence-low interest (the crowd).
Table 1. Scoring Assessment of Capital Assets
Score Range
No. Variable Sub-Variable
0 (none) 1 (low) 2 (Medium) 3 (high)
1. Natural Land ownership Not having Land owned by Land owned Having
assets land one landlord who by several individual
ownership own majority of landlords, and ownership of
formally village land each doesn’t land
own majority of
village land
Food land use/ No land used Less than 1/3 Less than ½ Each farmer
Food production for plantation of the total of the total owns 2 ha of
existing land existing land paddy fields
for plantations for plantation (Law No. 56
compared to compared to 1960)
the number of the number of
farmers farmers
Availability Growth rate Growth rate of Growth rate of Growth rate of
of marine of fish is fish is negative fish is positive fish is positive,
resources negative and and fishermen but fishermen and fishermen
fishermen fishing outside fishing outside can catch
fishing on EEZ the catchment the catchment according to
border (200 area area the fishing
miles) catchment area
Water Water sources There are natural Derived water Derived from
availability are obtained water sources but from natural natural sources
from outside insufficient for sources of the of the region
the region daily needs region and and sufficient
insufficient during the dry
during the dry season
season
Score Range
2. Social Network access Did not Received Obtained Obtaining
assets information receive assistance from assistance assistance
(vertical) assistance less than 1/3 of less than ½ more than ½
the respondent the number the number
fishermen of respondent of fishermen
fishermen respondents
Effectiveness There are no There are groups There are There are
of institutional groups in each but they are not groups and they groups where
participation village active are active most members
(formal group) are local
fishermen
Trust No reciprocity No reciprocity There is no There is
and and only reciprocity, reciprocity and
cooperation cooperation but there is exchange in
without cooperation cooperation
transaction cost with reduced
reduction transaction
costs
3. Human Skills and No training The number of Number of Number of
assets abilities fishermen who fishermen who fishermen who
participated participated participated in
in training is in training training more
less than 1/3 of less than ½ than more than
the number of the number of ½ the number of
respondents respondents respondents
Knowledge No technology Use of technology Use of Use of
(Technology) service used services, less than technology technology
1/3 of total fisher service service
respondents technology, technology,
less than ½ more than ½ of
of total fisher total fishermen
respondents respondents
Access to health No health There is only a There are health There are
facilities clinic or auxiliary centres and puskesmas
health centre clinics, and and clinics,
insufficient and sufficient
capacity capacity
according to SNI according to SNI
Access to Only one Only two school All school levels All school levels
education school level levels are are available but are available
available in available in one do not fulfil the and meet the
one district district (primary, capacity based capacity based
(primary, junior secondary, on SNI on SNI
junior senior secondary)
secondary,
senior
secondary)

Score Range
4. Physical Road Absence of Some access Most access Access roads
assets rural road roads have been roads have been are available in
access built (Up to 50% built (More than all parts of the
of roads have 50% of roads village
been built) have been built)
Transportation No access to Availability Availability Availability
land and sea of Medium of Medium of public
transportation Transportation Transportation transportation
Facilities Facilities (STM) serving the
at an affordable district
Transportation cost
(IMT) at an
unaffordable cost
Safe and No marine Has marine Has marine Has marine
sustainable support support facilities, support facilities support
housing facilities, but no utilities and utilities, but facilities,
unsustainable and unsustain- has unsustain- utilities, and
utilities, and able settlements able settlements sustainable
settlements settlements
Electricity Not yet served Not yet served in Already served in Already served
by electricity all areas and the all areas, but the in all areas and
quality is not yet quality is not yet the quality is
good good good
Clean water Obtained from Obtained with There is access There is a clean
outside the labour (human/ to clean water water network
village or sub- animal) within the village to residential
district area (wells, ponds) areas
Telecommuni- No telecom- There is a Internet access There is internet
cation network munication telecommunica- is available but access in all
signal tion signal uneven areas of the
but limited to village
telephone and
SMS
5. Financial Loan / debt No loan Loan with interest Loans with Independent
assets and collateral low interest of loan or
security (bank) (micro-credit) independent of
or interest- loan
free by mutual
agreement
Saving No saving There is a saving There is a saving There is a saving
habits habit (less than habit (less than habit (more than
1/3 of the total ½ of the total ½ of the total
local fishermen local fishermen local fishermen
respon-dents) respon-dents) respon-dents)
Asset No assets Less than 1/3 of Less than ½ More than ½
the total fisher of the total of the total
respondents have fishermen fishermen
assets respondents have respondents
assets have assets
Source: Own Analysis, 2020
LOCAL CAPITAL (LIVELIHOOD) ASSETS
Local capital (livelihood) assets consist of five types of capital, namely natural,
social, human, physical, and financial capital.
Natural Capital
On the land ownership, local communities in Pulau Tiga District still have access
to their land, thus making local communities, especially local fishermen, less
vulnerable do to having alternatives in times of uncertainties in fisheries for
example due to changing season.
Sabang Mawang and Sabang Mawang Barat Village have exceeded the
farmers’ land needs so that food land sustainability can be achieved. As for
Tanjung Batang Village, the existing plantation land area is still insufficient for
the existing farmers and less than 1/3 of the total land area, which is 16.1%, so
land use conditions are still relatively low.
Analysis on the availability of marine resources still shows that the
potential growth of fish yields decreased by 5.4% which indicates that there is a
decrease in fish stock especially in Natuna Sea. This condition is also supported
by the behaviour of fishermen who take fish outside the catch zone (between
30 to 100 miles) due to decreasing fish quantity at low miles (up to 12 miles),
which can me risky and dangerous for the safety of local fishermen.
Water availability is sufficient in all three villages. Clean water can be
obtained from springs in the forest and are distributed to houses in all the three
villages. In time of drought, water from reservoirs can supply community needs,
including fishermen’s activities while at sea, for three months and are free of
charge.
Table 2. Natural Capital Scoring
Village Value
No. Natural Capital Sabang Sabang Tanjung Score Range
Mawang Mawang Barat Batang
1. Land Ownership 4,5 4,5 3 0-4,5
2. Landuse for Food 4,5 4,5 1,5 0-4,5
3. Availability of Marine Resources 1,5 1,5 1,5 0-4,5
4. Availability of Water 4,5 4,5 4,5 0-4,5
Total 15 1 10,5 0-18

Social Capital
All fishermen in Sabang Mawang Village are getting trained by the village
government to optimised catches. Meanwhile, such training is not carried in
Sabang Mawang Barat Village. This indicates a low scores in vertical network
access in both village.
In terms of the effectiveness of institutional participation, there are two
cooperatives in Pulau Tiga District, namely Koperasi Tiga Saudara located in
Tanjung Batang Village and the Koperasi Nelayan Serumpun Tiga in Sabang
Mawang Village. Tiga Bersaudara Cooperative is active and has received some
administrative assistance. Meanwhile Serumpun Tiga Fisherman’s Cooperative
has not received any assistance mainly because the cooperative is still relatively
new. There are no cooperatives at all in Sabang Mawang Barat Village indicating
lower institutional participation.
Table 3. Social Capital Scoring
Village Value
No. Social Capital Sabang Sabang Tanjung Score Range
1. Information Network Access (Vertical) 6 2 2 0-6
2. Effectiveness of Institutional
2 0 4 0-6
Participation (Formal Groups)
3. Trust 4 6 4 0-6
Total 12 8 10 0-18

In Sabang Mawang Barat Village the majority of fishermen work with large
collectors, such as CV. Mina Sejahtera as tekong or pandhega or ship captains
and crew on Bagan Apung. This partnership indicates a higher trust and
reputation of fishermen in Sabang Mawang Barat Village who can be trusted in
managing the ship’s owned by larger collectors. Meanwhile, in Sabang Mawang
and Tanjung Batang village, local fishermen sell their fish to local collectors
without any reciprocal relationship between them. However, on order to lure
fishermen sell their fish, collectors often providing reduced transaction costs in
various ways, such as reducing the price of fuel, ice, or other means.
Human Capital
In the context of skills, most local fishermen have not received training in the
last five years except for some fishermen in Tanjung Batang Village. Due to
unavailability of intervention to elevate fishermen skills, in general the analysis
found that fishing ability among fishermen is still low.
Furthermore, referring to the knowledge, the ownership of technology in
local fishermen is good, as farmers in the three villages already use fish finder
technology or radar, GPS, and radio. In the process of introducing technology,
people get knowledge transferred through other local fishermen who have
used technology first. One of the reasons for the transfer of knowledge is that
fishermen’s education is not optimal for obtaining fishery knowledge and is
only based on experience.
Then on the aspect of health access, the fulfilment of the needs for existing
facilities/infrastructure in Pulau Tiga District is sufficient and according to the
calculation of the standard needs for health facilities/infrastructure. In addition,
the presence of official vehicles and ambulances can assist local communities in
facilitating public health needs that cannot be facilitated by the existing sub-
district health facilities/infrastructure. This assessment is seen based on SNI
03-1733-2004 concerning Procedures for planning the housing environment
in urban areas.
Lastly, on access to education, educational facilities, or facilities in Pulau
Tiga District is still sufficient to support the needs of local fishermen’s families
in their sustainable livelihoods, it can be seen from the capacity of each facility
still exceeds the number of residents in Pulau Tiga District. This assessment is
also seen based on SNI 03-1733-2004 concerning Procedures for planning the
housing environment in urban areas.
Table 4. Human Capital Scoring
Village Value
No. Human Capital Sabang Sabang Tanjung Score Range
1. Skills and Abilities 0 0 1,5 0-4,5
2. Knowledge (Technology) 4,5 4,5 4,5 0-4,5
3. Health Access 4,5 4,5 4,5 0-4,5
4. Education Access 4,5 4,5 4,5 0-4,5
Total 13,5 13,5 15 0-18

Physical Capital
Regarding the existing road in Sabang Mawang, West Sabang Mawang, and
Tanjung Batang, in general, they already have road access, although unpaved in
some segments, which makes it paralysed during rainy season. One example is
the Sabang Mawang Barat and Tanjung Batang border roads which are still in
the form of soil and has an impact on community productivity in carrying out
activities in rainy weather.
Figure 3. Inter- Island Transportation in Pulau Tiga District, Natuna Islands
Source: Documentation, 2020

Regarding transportation access, people use motorbikes as local
transportation. As for moving between islands, which are limited by the sea,
people use pompong or small boats which have a general schedule a day, start at
07 AM to 09 AM and 12 PM to 14 PM for IDR10 thousand one way from Pulau
Tiga to the main island or vice versa. This long-distance condition is the reason
for the Damri subsidy in 2020 which serves the Ranai route to the Lampa Strait
once a day to make it easier for the public to use alternative modes to go to the
Lampa Strait to travel by ship (Bukit Raya and Sabuk Nusantara) or for those
who need to go to the district centre, i.e., Ranai. Another public transportation
that can be used from Selat Lampa is a shared vehicle (functioning as a taxi)
that can carry several passengers to the city centre. The last alternative is to use
private vehicles such as cars and motorcycles.
In general, the houses on Pulau Tiga are made of wood and are located
along the coastline the coast of the island. This condition is relatively common
in all villages in Pulau Tiga Subdistrict, which have the characteristics of the
majority being fishermen. In terms of supporting facilities related to fisheries,
each house has these facilities, such as docks and boat moorings. As for utilities,
almost every basic housing need has been facilitated, starting from clean water,
electricity, telephone, and gas which are part of the physical capital analysis.
So, if categorized, the condition of the three villages is in moderate condition,
namely having marine support facilities and utilities, but not yet sustainable in
the aspect of local fishermen’s houses.
Figure 4. Housing and Settlement of Coastal Communities, Pulau Tiga
District, Natuna Islands
Source: Documentation, 2020

In terms of energy infrastructure, local people use electrical power as an
energy source in their daily activities. Based on the two weeks survey, access
to electricity on Pulau Tiga was found to be well served so that local people
can work 24 hours a day. As for the water availability which is also part of the
analysis of access to clean water on physical capital, in general, the availability
of clean water is sufficient, even in times of drought there are already dams to
store water reserves. In the existing condition, there are two water reservoirs in
Sabang Mawang Village and Tanjung Batang Village, respectively. They function
as water reserves in times of a clean water crisis. The condition of water needs

is also sufficient in each village with the existence of a clean water network
that serves all houses in the three villages so it can be concluded that the water
network has served up to community settlements without any obstacles in the
process of draining and taking water.
Finally, the trend of access to information in the villages of Sabang Mawang,
West Sabang Mawang, and Tanjung Batang is relatively good with indications
of good telecommunication signals, namely 4G. This indicates that information
will be relatively easy to obtain, and many fishing communities also already
have mobile phones.
Table 5. Physical Capital Scoring
Village Value
No. Physical Capital Sabang Sabang Tanjung Score Range
1. Roads 3 2 2 0-3
2. Transportation 3 3 3 0-3
3. Access to safe and sustainable
2 2 2 0-3
housing
4. Electricity 3 3 3 0-3
5. Clean Water 3 3 3 0-3
6. Telecommunication Network 3 3 3 0-3
Total 17 16 16 0-18
Financial Capital
In Sabang Mawang Village as many as 12 fishermen (70.6%) still depend on the
fish collectors (pengumpul) to support their financial conditions. Fishermen in
Tanjung Batang Village also depend on fish collectors or banks, but not majority
(40.7%). Meanwhile, fishermen in Sabang Mawang Barat Village are mostly
financially independent and no longer dependent on loans from fish collector,
banks, or other financial sources (58.3%).
Furthermore, local fishermen from those three villages still do not prefer
saving regularly. Less than 1/3 of the total respondents from each village
expressed the low preference for saving, which results in mixing of money for
daily needs and money that should be saved for emergencies.
Finally, most local fishermen from Sabang Mawang Barat Village and
Tanjung Batang Village have individual assets such as gardens, fish cages,
remittance businesses, food stalls, and so on. Meanwhile, in Sabang Mawang
Village, less than 1/3 of respondents from local fishermen have those assets,
which indicate the low assets owned by the fishermen. However, they consider
assets other than their profession as fishermen do not significantly guarantee
the same income as income from selling fish, although those assets can be an
alternative financial capital apart from income from selling fish products.
Table 6. Financial Capital Scoring
Village Value
No. Financial Capital Sabang Sabang Tanjung Score Range
1. Loans 2 6 4 0-6
2. Savings 2 2 2 0-6
3. Assets 2 6 6 0-6
Total 6 14 12 0-18

The five local capitals—i.e., natural, social, human, physical, and financial
capitals—discussed above are summarised in Table 7 below, showing that all
the three villages in Pulau Tiga are just sustainable, with Sabang Mawang Barat
Village has the highest score.
Table 7. Final Scoring of Local Capital Assets
Village Value Score
Local Capital
Sabang Mawang Sabang Mawang Barat Tanjung Batang Range
Natural 15 15 10,5 0-18

Social 12 8 10 0-18
Human 13,5 13,5 15 0-18
Physical 17 17 16 0-18
Financial 6 14 12 0-18
Telecommunication Network 63,5 67,5 63,5 0-90
Total 17 16 16 0-18
Note: 1-30 = Not sustainable; 41-60= Fairly sustainable; 61-90 = Sustainable


FACTORS AFFECTING THE LOCAL ECONOMIC
DEVELOPMENT
Vulnerability
In the context of the vulnerable shocks that occur to fishermen, there are
natural hazards that always threaten them, namely the wind season. The
monsoon phenomenon in Natuna greatly affects fishermen, especially during
the northern season (November to January) and the southern season (May to
July). In the northern season, fishermen cannot go to sea at all because the waves
are relatively high. However, during the northern season, there is a ‘shady time’
every month, namely at high tide which occurs around the 6th, 7th, 8th, 20th, 21st,
and 22nd day of lunar months. This condition can then affect the financial capital
of local fishermen because they cannot go to the sea and catch fish.
Due to the occurrence of the wind season annually, it has influenced price
uncertainty and available job opportunities. The price of fish increases during
the season when fish yields are not produced, such as during the north and
south seasons, but fishermen cannot take advantage of this condition because
they consider the safety risks that may occur, and on the contrary, fish prices fall
during the east and west seasons.
This uncertainty in the working hours of fishermen then causes job losses
that can occur spontaneously (especially due to the weather). Fishermen who
have good financial capabilities and have alternative incomes such as having
clove gardens, cages, and stalls, can adapt to these uncertain conditions.
However, fishermen who do not have alternative income will tend to work as
clove collecting laborers in the northern season every two years and work as
construction workers which are usually held by the local village government,
which is highly dependent on the availability of construction work carried out.
In the context of trends, two points can affect the sustainable livelihoods
of fishermen in Natuna, especially in Pulau Tiga District, namely the trend of
preference in continuing the fishing profession and the trend of fishery products
that occur. In this case, preferences of fishing profession are influenced by
several factors—such as the uncertainty of income and the level of education
that is getting better. Of course, when discussing income uncertainty, this is also
greatly influenced by limited infrastructures, such as ships that still have low
Gross Tonnage (GT) and poor navigation technology as well as the lack of good
coordination in predicting dynamic weather between fishermen, government,
and the related institutions. In addition, better education also reduces the level
of preference for becoming a fisherman.
Furthermore, the male population in Sabang Mawang Village and Tanjung
Batang Village experienced growth, which indicates that there is the next
generation of fishermen. Meanwhile, the male population in Sabang Mawang
Barat Village has decreased, which may potentially change the dynamics of the
fishing profession in the future.
Finally, based on the trend of potential fishery products, there is a decrease
in the growth rate of potential fishery products per year in general by 5.4%
in the year 2017. This is also supported by competition in catching fish in the
Exclusive Economic Zone which is also the catch area for local fishermen outside
Natuna and foreign fishing boats.
Institutional Structure
Using the stakeholders’ interest and influence matrix, the key actors in local
policies and institutions are the Ministry of Maritime Affairs and Fisheries/
Kementerian Kelautan dan Perikanan (KKP), the three village governments
(Sabang Mawang, Sabang Mawang Barat, and Tanjung Batang), fish
entrepreneurs or bulk collectors, and local collectors. As is known, Indonesia
Marine and Fisheries Ministry/Kementerian Kelautan dan Perikanan (KKP) has
provided a lot of assistance to Natuna fishermen directly and established the
Integrated Marine and Fisheries Center/Sentra Kelautan dan Perikanan Terpadu
(SKPT) and Indonesian Fishery Company/Perum Perikanan Indonesia (Perindo)
facilities to make Natuna a fishing centre.
Meanwhile, villages government act as the main facilitators of local
communities aimed at improving welfare which will later affect the economic
development of the village. Fish entrepreneurs (Pengusaha Ikan) have a role
in maintaining the stabilization of fish prices when SKPT and Perindo have not
consistently taken that role as a whole and bulk collectors are highly valued by
local fishermen. The role of local collectors (Pengumpul Lokal) is to ensure that
fish sales can still be carried out outside Natuna to keep the livelihoods of local
fishermen running. Pulau Tiga District Government has influence on society
but it is low in importance, since it only deals with administrative matters and
does not have its own territory.

Figure 5. Matrix of Marine and Fishery Stakeholders in Pulau Tiga
District
Source: Analysis, 2020

In the context of fisheries, the local institutions that play key roles are the
Department of Marine Affairs and Fisheries (Branch of the Natuna Service) of
Riau Islands Province (Dinas Kelautan dan Perikanan/DKP Provinsi Kepulauan
Riau), Natuna District Fisheries Service (Dinas Perikanan Natuna), and SKPT/
Perindo. The Department of Marine Affairs and Fisheries (Branch of the Natuna
Service) of the Riau Islands Province (DKP Provinsi Kepulauan Riau) has an
interest in exercising maritime authority up to 12 miles. However, until now,
it is still in the stage of completing the development of facilities/infrastructure
to be able to facilitate the needs of the service to accommodate the needs of
local fishermen. Meanwhile, the Natuna Fishery Department (Dinas Perikanan
Natuna/Disperikan Natuna) has a history of local fishermen because they once
had the authority to manage marine affairs.
However, after the elimination of the maritime authority of local
government, Dinas Perikanan Natuna acted as a facilitator for the central and
provincial governments in providing input on marine and fishery development
in Natuna District. Then SKPT/Perindo—which is an extension of the central
government—has not been able to optimise its role as a fishery centre. With the
dynamics of fishermen in Natuna, the influence of SKPT/Perindo has not been
maximised. Hence, a strengthening strategy is needed to increase mutual trust
between SKPT/Perindo and local fishermen.
Lastly, Cooperatives (Koperasi) and Fuel Oil Providers (Penyedia Bahan
Bakar Minyak) belong to the crowd. The cooperative itself has so far only
focused on internal affairs and has not had an impact on village areas and local
fishermen. Then the fuel provider also has no influence and interest in fisheries
because fuel is a basic need that will always be needed by everyone. Therefore,
it is impossible for some groups to monopolise fuel as a means of control,
including fishermen. This is because the disconnection of fuel access from one
party will easily be substituted by other parties, so that in the end there is no
single fuel provider that has an influence on fishermen through the fuel access
provided.
CONCLUSION
The purpose of this study was to determine the condition of Sustainable
Livelihoods (SL) in the fisheries sector in the context of developing coastal rural
areas in Natuna District. Through the sustainable livelihood framework, we can
see that there are capitals—either those with lower scores or the ones that are
higher—that still need some improvements. The Sabang Mawang Village which
is not yet sustainable in financial capital will require greater encouragement
from others or external parties. They also need a sustainable increase in social
capital. The Tanjung Batang village also needs an increase in natural capital,
social capital, and financial capital to be sustainable in the context of fisheries.
As for the Sabang Mawang Barat Village, it is necessary to increase the social
capital which is lower than in the other two villages.
The vulnerability that occurs is also a condition that cannot be easily
eliminated as it always happens due to nature that is inevitable. This condition
of vulnerability will then directly have an impact on local fishermen who will
later depend on the local capital in each village. In this case, it can be concluded
that Sabang Mawang Village and Tanjung Batang Village need more support
because they are easy to be exposed to vulnerability. In a situation of special
vulnerability, the trend also affects the village of Sabang Mawang Barat regarding
the trend of decreasing the number of male residents which can later become a
threat to the sustainability of the livelihoods of local fishermen.

When discussing local institutions, two types of capital are always a
problem, namely social and financial capital. In general, local fishermen
cannot work in groups—i.e., either in group of fishermen or in formal groups,
which causes local fishermen to sell their fish freely because the decision to
sell is up to everyone. As a result, fishermen tend to look at market conditions
before deciding to sell fish. Hence, the arising problem is lack of commitment,
especially for SKPTs, to work professionally and based on formal agreements.
Furthermore, this condition will lead to financial uncertainty for local fishermen,
which leads to dependence on local collectors. Local collectors do not have a
bad impact on local fishermen. However, due to social closeness as neighbours
or relatives, fishermen give a bad stigma to local collectors because their close
relationship is used to cover the financial shortcomings experienced by local
fishermen in vulnerable situations.
In order to improve the livelihoods of fishermen in Natuna Regency, the
results of this study suggest the following recommendations:
1. Fishermen still depend on collectors in selling fish because SKPT does
not yet have the scheme expected by local fishermen, such as the lack
of flexibility in working hours, namely operating hours—while local
fishermen are very dynamic in landing fish so they can return to land at
night, SKPT cannot accommodate fish sales. This opportunity is therefore
taken by collectors who have flexibility in working hours. In addition, there
is a finding that the payment scheme by SKPT is not paid immediately
when the fish lands, so this creates uncertainty for local fishermen to go to
sea even though local fishermen have a schedule to go to sea at least once
a week. This financial scheme is quite burdensome for local fishermen,
which even becomes an opportunity for collectors as they can directly pay
money when local fishermen land from the sea.
2. Local fishermen should be encouraged to form an association or
organisation such as cooperative. Because one of the advantages of forming
a cooperative is getting fisheries assistance from the government, which
is by regulation must go through cooperatives. Many local fishermen in
Pulau Tiga Subdistrict do not get assistance because no cooperative exists
there. It should be emphasized that the formation of cooperatives should
be carried out with training that emphasizes learning about the financial
management as local fishermen still do not manage their money well
independently. In addition, cooperatives must be able to increase their
capacity in terms of ensuring the quality of fish according to the standards,
especially if they want to cooperate with export oriented SKPT. So, the role
of SKPT is crucial in synergising local fishermen’s cooperative programmes
for the benefit of the community and the existence of SKPT.
3. There needs to be a profitable scheme by SKPT and the government (central
and regional) so that local fishermen do not worry about the future of their
families. The schemes that can be carried out can be indirectly, such as
schools for fishermen’s children, health insurance, incentives for fishing
materials (fuel, ice, fishing line, bait, etc.), and other programmes that can
ensure the sustainability of local fishermen’s lives.
4. Increasingly open competition between local fishermen and fishermen
outside Natuna (national or international) can result in the exclusion
of local fishermen because local fishermen in general still catch fish
traditionally, both from fishing gear and small boat sizes, namely 2 to 5 GT.
Therefore, there needs to be strict restrictions both in terms of security
through technology and ships to protect Indonesia’s maritime sovereignty
from foreign ships so that local fishermen can go to the sea safely. In
addition, it is necessary to periodically check for unsustainable fishing
gear that has the potential to be used by ships outside Natuna to maintain
the sustainability of marine ecosystems.
ACKNOWLEDGMENT
The study was funded by the PPMI research scheme of the School of Architecture,
Planning and Policy Development of Institut Teknologi Bandung (ITB) year
2020.
REFERENCES
Allison, E. H., & F. Ellis (2001). The livelihoods Approach and Management of
Small-Scale Fisheries. Norwich: Elsevier.
Allison, E. H., & B. Horemans (2006). ‘Putting the principles of the Sustainable
Livelihoods Approach into fisheries development policy and practice.’
Marine Policy (30): 757–766.
Carney. (1998). Sustainable Rural Livelihoods. What Contribution Can We Make?
Nottingham: Russell Press Ltd.

DFID (1999). Sustainable Livelihoods Guidance Sheets. London: DFID.
DFID (2001). Sustainable Livelihoods Guidance Sheets. Eldis Document Store.
Edelman (2003). ‘Naturalizing Consciousness: A Theoretical Framework.’
Proceedings of the National Academy of Sciences of the United States of
America.’ 5520-5524.
Eden, C., & F. Ackermann (1998). Making Strategy: the Journey of Strategic
Management. London: Sage Publications.
Kementerian Kelautan dan Perikanan Indonesia (2016). ‘Kelautan dan
Perikanan dalam Angka Kabupaten Natuna’. Jakarta: Pusat Data, Statistik,
dan Informasi, Jakarta: KKP.
Kementerian Kelautan dan Perikanan Indonesia (2019). Laporan Tahunan
Kementerian Kelautan dan Perikanan. Jakarta: KKP.
Kementerian Kelautan dan Perikanan Indonesia. (2015). Keputusan Menteri
Perikanan dan Kelautan (KKP) Nomor 48 Tahun 2015. Jakarta: KKP.
Lopez, D. (2001). ‘Stakeholder management for conservation projects: a case
study of Ream National Park, Cambodia.’ Environmental Management 28:
47-60.
Rahman, H.M.T., B.E. Robinson, J.D.Ford, et al. (2018). ‘How Do Capital Asset
Interactions Affect Livelihood Sensitivity to Climatic Stresses? Insights
From the Northeastern Floodplains of Bangladesh.’ Ecological Economics,
(150): 165-176.
Reed, M., A.Graves, N. Dandy, H. Posthumus, K. Hubacek, J. Morris, et al. (2009).
Who’s in and why? A Typology of Stakeholder Analysis Methods for Natural.
Aberdeen: Elsevier.
Republik Indonesia. (2016). Instruksi Presiden Nomor 7 Tahun 2016 tentang
Percepatan Pengembangan Industri Perikanan Nasional. Jakarta: Republik
Indonesia.
Resosudarmo, Budy & Ellisa Kosadi (2018). ‘Illegal Fishing War: An
Environmental Policy during the Jokowi Era.’ Southeast Asian Economies,
(35): 369-385. 10.1355/ae35-3d.
Scoones (1998). Sustainable Rural Livelihoods: A Framework for Analysis.
Brighton: Institute of Development Studies, University of Sussex.
Serrat, O. (2017). The Sustainable Livelihoods Approach. In Knowledge Solution.
Singapore: Springer.
Sugiyono (2011). Metode Penelitian Kuantitatif, Kualitatif, dan R&D. Bandung:
Alfabeta.
Suman, A., H.E. Irianto, F. Satria, & K.Amri (2016). Potensi Dan Tingkat
Pemanfaatan Sumber Daya Ikan Di Wilayah Pengelolaan Perikanan Negara
Republik Indonesia (WPP NRI) Tahun 2015 Serta Opsi Pengelolaannya.
Indonesian Fisheries Policy Journal KKP.
UNDP (2017). Guidance Note: Application of The Sustainable Livelihoods
Framework in Development Projects. Panama City: Regional Centre for
Latin America and The Carribean.
Wardono, B., R. Muhartono, T. Hikmah Apriliani, & A. Zulham (2020). ‘Traditional
Fisheries Supply Chain Management in The Border Area: The Natuna
Islands, Indonesia’. E3S Web of Conferences 147. Available at: https://doi.
org/10.1051/e3sconf/202014702009.
WCED (1987). Global Policies for Sustainable Agriculture, a Report of the Advisory
Panel on Food Security, Agriculture, Forestry and Environment to the World
Commision on Environment and Development. London and New Jersey: Zed
Books Ltd.

CHAPTER 7
Designing an Appropriate
Renewable Energy System for
Small-Isolated Island Communities
Alan Dwi Wibowo and Andante Hadi Pandyaswargo
INTRODUCTION
Small island communities are known to be the most vulnerable because they
are prone to weather and climate changes (IPCC, 2015). Their geographical
situation also poses challenges to reliable energy access. In developing countries,
communities living on small islands have specific challenges in which their
energy needs differ from those living on the mainland (Hong and Abe, 2012;
Pandyaswargo et al., 2014). Separated by water to the main island means a
higher challenge to transportation access, weather-dependency of both sea and
land-based income, and market access limitation that force the communities
to be independent with their limited resources. The exorbitant cost and
technical challenges for grid extension have made off-grid renewable energy
(RE) the most adopted solution to supply energy in remote island communities
(Renewable Energy Agency, 2019). The declining prices for RE technologies and
the challenge of transporting fossil fuels from the mainland have increased the
adoption of RE in such communities (IRENA, 2020).
Electricity is an instrument to improve the quality of human life (Mandelli
et al., 2016) and plays an important role in solving poverty (Kanagawa & Nakata,
2008). In a previous study, it was revealed that access to electricity in a village
could significantly increase the number of new establishments of small and
139
medium enterprises in that village (Wirawan & Gultom, 2021). However, many
villages face sustainability challenges in implementing and maintaining off-
grid renewable energy systems. Some of the challenges are the unavailability
of funding for the initial and operational costs (Ahlborg & Hammar, 2014;
Bertheau et al., 2020a; Chmiel & Bhattacharyya, 2015), technical operational
challenges (Julius et al., 2014), limited power generation resources, lack of
village community know-how, poor communication between stakeholders
(Bertheau et al., 2020a), and cultural incompatibility (Urmee & Anisuzzaman,
2016).
Among the challenges to introducing off-grid RE systems, the financial
capacity of the island is one of the key challenges because it determines whether
investment will take place despite the availability of external support. External
support from the government, international organisations, or the private sector
has its own hurdles, such as time limit (IESR, 2019) and topical support without
continuous assistance. The high investment costs and low-profit opportunities
make the private sector reluctant to make an investment (Come Zebra et al.,
2021). Only after an initial investment is made do opportunities to improve the
small island’s capacity open. When stable financial performance is achieved,
external investors may have better confidence in scaling up RE systems in small
islands. Selecting the right RE system design and configuration determines
its efficiency and financial sustainability. Therefore, it is an essential part of
developing electricity systems in small islands.
Studies on the techno-economic assessment for RE designs in remote
areas are widely available (Chambon et al., 2020a; Ghasemi et al., 2013;
Shahzad et al., 2017a; Veilleux et al., 2020a; Vendoti et al., 2021a). However, the
characterisation of power generation systems in small islands is still limited. The
main characteristics in developing integration of RE power generation systems
for small islands basically covers the flexibility of power generation systems,
the profile of electricity load and local demand, the structure of transmission
and distribution network, implementation strategy and generation expand
plan, operational aspects, and the last but not least the influence of governance
of the electricity sector (International Renewal Energy Agency, 2018).
The structure of transmission and distribution network is one of the
power generation systems characteristics that will have an impact on power
reliability, operational efficiency, and investment costs. There are two common
types of transmission RE systems: distributed and centralised power systems.
Distributed RE power systems takes the form of power systems at distributed
locations near load centers, whereas centralised power generation systems are
generally located at the point of greatest resource availability. Both distributed
and centralized power generation have advantages and disadvantages. From a
technical standpoint, distributed power systems have the potential to deliver
energy at periods of high demand and extremely near to the location of the need,
hence reducing transmission loss. However, the intermittency of RE output hard
to managed smoothly. On the other hand, centralised power generation systems
adhere to the existing electrical power management model and may be situated
in areas where the resource is most accessible. However, new transmission
lines may be required to deliver power from these power generation units to
the load centers, and this model may necessitate major financial investments
(Woods Institute for the Environment, 2010).
Particularly, selecting between a centralised RE system or a distributed
RE system as a distribution network is rather tricky. The selection of this type
of power generation system is critical in developing the RE generation system,
which is often associated with the accessibility of electricity to the entire society
as well as the potential expenses including investment costs and operating costs.
The goal of this chapter is to present some insights about the RE power
generation system for small islands community in a developing country by
providing a technical and financial assessment on the basis of the constructed
scenarios. Our study tasks are as follows in order to reach our research objectives:
(1) Conducting field surveys to collect primary data through interviews and
questionnaires; (2) Calculating the energy demand in the islands; and (3)
Simulating the RE power generation on the basis of the constructed scenarios.
This chapter focuses on the design and implementation of RE power generation
on a small island that has tourism potential. Pulau Burung islands have been
selected as a case study location. Why this location? Based on our discussions
with the local government, we believe that this island has significant tourism
potential. Actually, this island is one of the hidden tourism spots with the
mangrove park in Kalimantan Selatan.
Finally, this chapter shows the comparison between the two systems in
terms of their technical and economic performances to suggest which one is
more appropriate for Pulau Burung Island. In this case, centralised RE power
systems is more suitable for Pulau Burung island. However, sustaining such a
system presents several difficulties.
Designing an Appropriate Renewable Energy System for Small-Isolated 141

Island Communities
STUDY AREA: Pulau Burung Island
Pulau Burung island is located at Tanah Bumbu District, Kalimantan Selatan
Province, Indonesia (see Figure 1). The island can only be reached by boat.
The travel time from the mainland (Batulicin city, capital of the Tanah Bumbu
district) to the Pulau Burung island is about 20 minutes with a transportation
fee of IDR 15,000 – IDR 25.000 per person. The number of inhabitants recorded
in the local government is about 80 households, unfortunately the actual only
found about 50 households who are still presently living on that island.
Figure 1. Location of Pulau Burung Island
Pulau Burung island is one of the tourist communities without access to

the power grid (off-grid island) in Kalimantan Selatan, which offers tourism
potential in the shape of a mangrove forest region. The Pulau Burung island as
an exotic of mangrove tourism in Kalimantan Selatan has become viral in local
communities. Several culinary facilities, meeting venues and public facilities are
well available on this island (see Figure 2).
Figure 2. Tourism Facilities in Pulau Burung Island
In general, the residents of the Pulau Burung island work as fishermen.

The catch is distributed and sold directly to Kalimantan mainland. In addition,
some residents work as farmers for fruits such as durian, longan, mango, and
mangrove breeding (see Figure 3). Our observation of the daily activities of the
Pulau Burung island residents shows that there is no reliable source of biomass
or biomass residue that can be used for energy generation feedstock. Likewise,
this island also does not have rivers or waterfall; therefore, there is no potential
for hydropower.

Island Communities
Figure 3. Mangrove breeding in Pulau Burung Island
PREVIOUS STUDIES ON ELECTRICITY SYSTEM

Studies about electricity system in small off-grid islands have been done in
various countries (Kalamaras et al., 2019; Lozano et al., 2019; Veilleux et al.,
2020b). However, there are still 770 millions of people living without access
to electricity (IEA, 2021). A comprehensive study about the challenges of
electrification in small islands developing states in the pacific (Dornan, 2014)
highlighted the low percentage of electrification rates in small islands and
revealed the following issues: (1) Traditional grid-extensions are not appropriate
for island electrification; (2) More funding is required to establish off-grid,
isolated power generation systems in these islands; and (3) Institutional and
regulatory reforms are needed to accommodate a sustainable operation and
maintenance of such systems.
Although electrification progresses in the past decade have been significant
and positive (World Bank, 2023), the fact that islands are still struggling to get
access to clean and reliable energy may indicate that the existing solutions are
not enough to “leave no one behind” as pledged in the Sustainable Development
Goals (SDGs). Another possibility is that the currently offered solutions are
not to be taken as a one-fits-all solution. In this section, we will review the
remaining challenges that may need to be addressed to establish a sustainable
electrification solution for small islands. The challenges are categorised into the
techno-economic issues and social issues.
Techno-Economic Issues
There are a number of techno-economic studies for small island electrification
that have been conducted in the recent years (Table 1). A study about an off-grid
island in Greece simulated the feasibility of several combined heat and power
(CHP) standalone systems for the island. The study found that although the
initial cost of the proposed hybrid system was high, the extension of electricity
grid is more expensive at a prohibited level because of the topography of Greek
islands. Besides, the study also found that the utilisation of diesel engines is
unsustainable and environmentally unfriendly for the long term because of the
necessity to transport gasoline on regular basis (Kalamaras et al., 2019).
On the other hand, another study conducted in the Philippines (Lozano et
al., 2019), which compared a diesel engine system with a photovoltaics (PV)
+ diesel hybrid system and a PV system only, showed that the hybrid system
was more cost-effective. However, the study indicated that these results may
be specific for the islands especially influenced by the location of the island,
which is still in a reasonable proximity to the main island. Hence, the cost of
transporting gasoline might not be as expensive as the Greek island in the study
(Kalamaras et al., 2019). The importance of local characteristics in determining
the type of energy systems in off-grid island is also supported by (Come Zebra
et al., 2021) in their comprehensive study. Both (Come Zebra et al., 2021)
and (Lozano et al., 2019) also emphasize the importance of determining the
appropriate load profile such as size and capacity of the system to ensure the
efficiency of the system.
The support for a hybrid energy system is also shown by another Philippine
study from different locations (Pascasio et al., 2021). The study confirmed after
examining 147 off-grid islands of different sizes in the Philippines that a hybrid
energy system is significantly less-costly than a diesel engine-based system
because the former can allow increase in demand while keeping costs low.
The study argued that a hybrid energy system is robust against uncertainties
in component costs and increasing demand. Supports for a hybrid system by
various Philippine studies may indicate that the topography of the archipelagic
islands in the Phillipines may be suitable for this particular model. Authors of
this study even urge the Philippine government to quickly implement the hybrid
system to provide access to clean and affordable energy in the off-grid islands.
(Pascasio et al., 2021) also highlights the importance of considering both the
solar and wind energy for the hybrid systems.

Island Communities
Despite of the strong urge and backed with robust evaluations from various
studies mentioned above, there are still 11 million people in the Philippines
having no access to electricity today (Maritime Fairtrade, 2023). It may indicate
that more complicated economic problems remain that might have been
challenging the governments to take actions from the recommendations. For
example, a study by (Bertheau et al., 2020) revealed that the common practice
of operating renewable energy system a cooperative management in the
Philippine islands are facing various challenges such as the slow and ineffective
policy implementation in the country. This may include the financial supports
mentioned in the policies. Also, there are lack of comprehensive energy sector
planning at different levels, which makes planning for funding schemes more
complicated. Furthermore, contrary to the previous studies, the projected
revenues from renewable energy systems in the islands are unattractive
and have prohibited the private sectors to join venture with island energy
cooperatives (Bertheau et al., 2020).
A study in a Thailand island revealed another important financial challenge
(Veilleux et al., 2020a). The study argued, that in order to reach high renewable
energy fractions (60% and above), batteries with larger capacities are required.
Therefore, the price of batteries (in this case Li-ion battery) will determine the
project price. Currently, the price of battery is very high and is the leading cause
of why rural electrification today with renewable energy is still less financially
sustainable (Veilleux et al., 2020). As the price of battery directly increases the
price of the initial capital costs, only when battery prices become affordable,
initiating renewable energy-intense off-grid electrification systems can be done
in affordable manners.
A study with a different view were demonstrated by Roy (2023). Similar to
the previously reviewed case studies, this particular one from remote islands
in India, tested various scenarios of renewable technology combinations.
However, the results do not show any significant difference in terms of technical
and financial performances. Differences between the scenarios results were
found on the social performance indicators instead. The social indicators used
in this study were Human Development Index (HDI), job creation index, and
local transport employment (Roy, 2023).
Table 1. Summary of Techno-Economic Issues Indicated
by Previous Studies
Authors, Year Location Main Findings Methods
(Kalamaras et Greece 1. The extension of electricity grid is prohibitively HOMER
al., 2019) expensive
2. Diesel engines are unsustainable
(Lozano et al., Philippines 1. A hybrid renewable energy system is more HOMER
2019) cost-effective
2. Determining an appropriate load profile is key
for the system’s efficiency
(Bertheau et Philippines 1. Institutional reforms to manage the energy HOMER
al., 2020a) system are necessary
2. The private sectors are hesitant to join venture
with energy cooperatives due to the low
projected revenues
(Veilleux et al., Thailand 1. To reach a higher percentage of renewable HOMER
2020a) energy, larger battery capacity is required
2. The high price of battery today is the leading
cause of why renewable energy for small island
electrification is less financially feasible.
(Pascasio et al., Philippines 1. A hybrid system is less costly than a diesel HOMER
2021) engine system
2. A hybrid system is more accommodating for
demand fluctuations
(Roy, 2023) India 1. The techno-economic performance of various HOMER,
renewable energy combination tested were the Regression
same. machine
2. The difference lays on the social indicators learning
affected by various energy source modelling
combinations.
(Come Zebra et Multiple 1. Grid connecting is economically viable Multiple
al., 2021b) countries compared to an off-grid system, however, for methods
remote and sparsely populated areas, the (A review
off-grid solution may be more cost-effective study)
compared to a grid-connected option
2. When choosing an optimal layout, the future
developments of the main grid in the region
should be considered.
3. A hybrid off-grid system is more technically
and financially viable than single source.

Island Communities
Social Issues
A comprehensive study about off-grid renewable energy (Come Zebra et al.,
2021) indicated that while studies conducting techno-economic evaluations in
this area have been extensively performed, very limited studies have been done
discussing the social aspects. Table 2 summarises some of the rare studies that
have discussed the social aspects of off-grid electrification.
The study by Dornan (2014) from the Pacific islands mentioned the
necessity to reform regulations and institutions. In similar tones, the study
by Bertheau et al. (2020) required more effective policy implementation and
institutional reforms. However, how exactly these recommendations translate
to local situations may require a deeper understanding of the context in each
study location. The lack of tools and indicators to assess social performance
might be the cause of the limited study in this area.
The study by Roy (2023) had provided a way to make the approach of social
assessment more tangible by indicating that HDI, job creation index, and local
transport employment are the aspects that could be influenced by the choice
of renewable energy for off-grid electrification system. However, Hirmer &
Guthrie (2016) have provided a more comprehensive list of factors that would
determine the sustainability of such systems. The study provided 46 factors
categorised into 6 values: emotional, epistemic, function, indigenous, intrinsic,
and social significance values. These values were developed in the experiments
conducted in seven villages across rural Uganda (Hirmer & Cruickshank, 2014).
However, (Pandyaswargo et al., 2014) revealed that complicated indicators
presented to subject of social analysis in communities with low literacy levels
are not effective. Therefore, we built a simplified version based on the Hirmer
& Guthrie ( 2016) in our previous study (Pandyaswargo et al., 2022) with an
Indonesian agricultural community as our study focus. In the method that we
built, survey questionnaires containing simple socio-economic questions were
used to collect data from the community. We then find relationships between
attributes of the people in the community to allow identification of social
challenges. Some of the results from this study are: (1) It is necessary to create
a people-centred public–private partnership business model to be paired with
the appropriate system design to address common sustainability issues of off-
grid RE systems; and (2) Since the study focused on an agricultural village,
future studies in other remote areas with the challenge of access to the grid,
such as small islands are necessary.
Table 2. Summary of Social Issues Indicated by the Previous Studies
Authors, Year Location Main Findings
(Bertheau et Philippines 1. Slow and ineffective policy implementation in the
al., 2020a) country hinders island electrification.
2. Poor planning at various levels hinders island
electrification.
(Dornan, 2014) Small Island 1. Institutional and regulatory reforms are needed to
Developing accommodate a sustainable operation and maintenance
States (SIDS) of islands energy systems.
of the Pacific
(Hirmer & Uganda 1. A methodology for social assessment is produced in this
Guthrie, 2016) study: The Hirmer’s wheel.
2. Emotional, epistemic, function, indigenous, intrinsic,
and social significance values can be used for social
assessment.
(Hirmer & Uganda 1. Values of the communities should be considered in rural
Cruickshank, electrification because it may determine the acceptance
2014) of the introduced system.
(Pandyaswargo Indonesia 1. A methodology for social assessment is produced in this
et al., 2022) study: A combination of Survey, Multiple Correspondence
Analysis (MCA) and Nonparametric tests.
2. It necessary to create a people-centered public–
private partnership business model to be paired with
the appropriate system design to address common
sustainability issues of off-grid RE systems.
(Roy, 2023) India 1. Differences between different renewable energy
combinations affects the social performance indicators.
2. The social indicators used in this study were: Human
Development Index (HDI), job creation index, and local
transport employment
THE USE OF HYBRID OPTIMISATION MODEL FOR

ELECTRIC RENEWABLE (HOMER)
In order to show how to construct an appropriate RE power system, this
chapter analyses the off-grid tourism small island of Pulau Burung (see
Figure 4) in South Kalimantan Province, Indonesia. Field surveys as the first
step was successfully performed. Interviews with the head of the village,
and questionnaires distributed to the village using door-to-door interviews
conducted by experienced facilitators are all used to collect the data. Random
sampling was performed in this study for approximately 82% of the population
(41 of 50 actual households). The second phase is calculating the energy
demand in the islands. The third step is simulating the RE power generation.

Island Communities
In this step the Hybrid Optimization Model for Electric Renewable (HOMER)
was used to assess the technology and economic aspect. The use of HOMER
to assess the techno-economic of off-grid RE power generation systems has
been widely used by many studies in the developing countries (Chambon et al.,
2020b; Shahzad et al., 2017b; Veilleux et al., 2020c; Vendoti et al., 2021b).
The HOMER performs simulation, optimisation, and sensitivity analysis.
Since we want to know whether the proposed energy designs in this study
are financially feasible and able to suffice the load demand under our inputted
circumstances (see Table 1), this study employed the simulation and sensitivity
analysis functions using HOMER. First, it perform an hourly time series
simulation of its operation over a year. The hourly calculation considers the
available renewable power, compares it to the electric load demand, and then
decide what to do with the surplus of power during the time of excess, or how
to best generate additional power in times of deficit. After completing the hourly
calculation for the first year, the model determines whether the proposed system
design can satisfy the constraints inputted (e.g., battery, PV, and wind turbine
lifetime, or population served). In the sensitivity analysis, HOMER finds a system
configuration that is optimal among several particular set of input assumption
(Table 4 & Table 5). The variable that is inputted is called the sensitivity
variable. Example of these variables are fuel price, wind speed, interest rate,
or the estimated life span of a PV panel. Further details such as the mechanism
of HOMER in calculating the total net present cost (NPC) are described in the
official explanatory document of HOMER (Lambert et al., 2006).
Figure 4. Helicopter View of Pulau Burung island
Source: Google Map captured
DATA COLLECTION
Data were taken through field surveys, in-depth interviews with the head of
the village, and questionnaires to the village residents collected through door-
to-door interviews. Forty-one respondents were successfully interviewed (the
respondents were estimated to be approximately 82% of actual households
that live on the island).
To get a balanced perspective of both gender groupings, respondents
are made up of a mix of adult males and females. The proportion of male and
female respondents for is 46% and 54%, respectively. The average length of
respondents living on that island is around 34 years. Respondents’ levels of
education are elementary school (46%), junior high school (20%), senior high
school (27%), and no school (7%).
The data gathered were utilised to calculate the demand for electricity’s
load pattern. Figure 5 shows the Pulau Burung island electricity daily load
pattern. The trend reveals that demand rises in the morning and evening, with
the evening experiencing a greater peak. The morning peak mainly caused by the
lighting, phone charging, water pumping, rice cooking, and speaker. While the
evening peak is mainly caused by lighting, phone charging, radio and television,
rice cooking, water pumping, speaker, and computer.
Figure 5. Electricity Daily Demand Load Pattern in Pulau Burung Island

Island Communities
Data from the field survey were input into HOMER software to assess the
techno-economic and sensitivity of the microgrid renewable energy design.
Furthermore, the software calculated the net present cost (NPC) and levelised
cost of electricity (COE). Based on the field survey data, the peak daily load of
electricity in Pulau Burung island is approximately 2.79 kW. We set the daily load
configuration in the HOMER software to 11.26 kWh per day to accommodate
the peak loads.
Solar radiation data are taken from The National Aeronautics and Space
Administration (NASA) Prediction of Worldwide Energy Resource (POWER)
database (Figure 6) and the National Renewable Energy Laboratory (NREL)
(National Renewable Energy Laboratory, 2022). The wind speed data are also
taken from NASA POWER (available in HOMER software) (Figure 7).
Figure 6. Monthly Average Solar Irradiance over 22 years in Pulau
Burung island
Source: NASA,2022
Figure 7. Monthly Average Wind Speed at 50 m above the Surface of the

Earth over 30 years in Pulau Burung Island
Source: NASA,2022
There are two renewable energy system configurations in this study to be
compared. Configuration 1 is a centralised photovoltaic system hybrid with a
diesel generator as a backup (Figure 8), and Configuration 2 is a distributed
photovoltaic system combined with a wind turbine system (Figure 9). In
this study, we compare which configuration has better techno-economic
performance. We developed the proposed design configurations by considering
the field survey to the site and direct interviews with the head of Pulau Burung
island village. We also employ the primary parameters (Table 3) that we
gathered from the observation survey. These parameters are inputted into the
HOMER software for analysis.
Figure 8. System Configuration 1 (Centralised)
Figure 9. System Configuration 2 (Distributed)

Island Communities
Table 3. System Configuration Parameters
Configuration 1 Configuration 2
No Parameters
Value Unit Value Unit
1 Average daily load 11.26 kWh/day 11.26 kWh/day
2 Generator cost 500 USD/kW 150 USD/kW
3 Fuel cost 0.7 USD/litre 0.7 USD/litre
4 PV capital cost 2448 USD/kW 783 USD/kW
5 Wind turbine capital – USD/kW 783 USD/kW
cost
6 PV & wind turbine 20 Years 20 Years
lifetime
7 Battery type GS SLR – 12 V, 83.4 –
1000 2 V Ah, 1 kWh
2kWH Lead Acid
8 Battery cost 523.81 USD/ 229 USD/
battery battery
9 Battery lifetime 20 Years 3 Years
10 Discount rate 7.5 % 7.5 %
11 Village location -3.25 – -3.25 –
latitude
12 Village location 116.25 – 116.25 –
longitude
TECHNO-ECONOMIC ASSESSMENT
This study assessed two kinds of configuration microgrid hybrid-renewable
energy systems. The first configuration (Figure 8) provided two centralised solar
power systems and a diesel engine system, and the second configuration (Figure
9) consists of distributed solar and wind power systems. Both configurations
are designed to fulfil the daily electricity load on Pulau Burung Island. The
HOMER software simulated an optimised systems of both configurations. In
addition, the software also processes the sensitivity parameters we inputted
into the software, such as gasoline price, discount rate fluctuation, and different
wind speeds, to run a sensitivity analysis.
At the time of visit, we found that Pulau Burung already has a hybrid
energy system consisting of a centralised PV energy system and a diesel engine
system. However, due to poor gasoline supply (caused by increased price from
transportation from the main island), the diesel generator engine has not been
operating since the engine was installed. Based on the interview with the
head of the village, the initial investment of the centralised PV energy system
amounted to USD 69,000 for 5100 Wp. Meanwhile, the distributed centralised
PV energy system, which could be installed in each house with a capacity of
80 Wp, requires approximately 64 units that cost USD 550 per unit. Therefore,
the estimated total installation cost for a distributed PV energy system in Pulau
Burung island is approximately USD 35,200.
The HOMER software simulation results show that under the present
discount rate (9%), the cost of power generation with a centralised PV system
is lower than that with a distributed PV system. Specifically, the NPC of the
centralised PV system is USD 8,118, and the COE is USD 0.0948/kWh; meanwhile,
the NPC for the distributed PV system is USD 18,118, and the COE is USD 0.2115/
kWh. The net present cost (NPC) of RE systems is the present value of all the costs
of developing and operating the RE power systems over the project lifetime,
devided by the present value of all the revenues that it earns over the project
lifetime. HOMER calculates the net present cost of each RE system as a whole
system. Based on the calculation, the overall cost of developing a centralised
PV system on Pulau Burung Island is lower than developing a distributed PV
system. Furthermore, HOMER defines the levelised cost of energy (COE) as the
average cost per kWh of useful electrical energy produced by the system. In our
case, the cost of energy production for a centralised PV system is cheaper than
a distributed PV system.
This finding is in line with the study by Aghamolaei et al. (2020), where
NPCs for centralised RE systems are more cost-effective than distributed RE
systems. The finding of that study shows that installing centralised PV could
save almost 35% of the cost. In this study, installing centralised PV will result
in potential savings of around 55% compared to distributed PV. However, our
finding shows that COE are consistent with the study Mudgal et al. (2019),
where the COE value is aligned with the NPC value. Hence, the centralised RE
system is more viable than the distributed RE system for Pulau Burung island.
SENSITIVITY ANALYSIS
Using HOMER software, we performed sensitivity analysis to both
configurations—the centralised RE system and distributed system—as as
shown in Figure 8 and Figure 9, respectively. The HOMER sensitivity performs

Island Communities
multiple optimisations under a range of input assumptions to gauge the effects
of uncertainty or changes in the model inputs. In this case, we inputted the
sensitivity variation of diesel fuel price ranging from USD 0.7/L to USD 1.4/L
and discount rates ranging from 7.5% to 9% to the centralised RE system (Table
4). To the distributed RE system, we added an additional variation of wind speed
(Table 5). We determine the ranges of sensitivity variable based on the local and
national values that we observed in the case study location.
The sensitivity analysis shows that when the discount rate is lowered to
7.5%, the NPC of the centralized PV increases to USD 6951.82, and the COE
increases to USD 0.106762. This condition is stable both when the diesel price
is normal and when the gasoline price doubles (Appendix 1). In the distributed
system, the wind speed was found to be too weak for a wind turbine to generate
any electricity (Appendix 2). The system is also sensitive to diesel prices and
discount rates. In these circumstances, the centralised system is still better than
the decentralised system even when the generators do not function due to a
lack of gasoline supply.
Simulation of sensitivity analysis on centralised RE is carried out using
two variables such as price change—i.e., change in diesel price and change in
discount rate. On the other hand, sensitivity analysis on distributed RE systems
is done using a variable considered less stable, i.e., wind scale. The result shows
that centralised RE system model for Pulau Burung island is not sensitive to
oil prices and discount rates; whereas the simulation of sensitivity analysis on
distributed RE using the wind scale variable found that the wind speed is not
applicable in this design.
THE IMPORTANCE OF CONSIDERING THE SOCIAL

FACTORS IN THE COMMUNITY
While the techno-analysis of this study showed that the centralised RE system
is already optimal compared to the distributed RE system, discussions with
the head of the village revealed that the lack of knowledge on how to manage
the centralised PV system leads to several problems. First, in the event of a
breakdown, they always had to call experts from outside the islands to come and
fix the problem. Such a situation results in exorbitant expense in maintaining
the system. We also observed on site that some damaged batteries have not
been replaced because they have depleted their battery stock on the island. It is
costly to purchase battery replacements from the main island. Furthermore, as
the population is quite sparsely distributed, there are some transmission and
distribution efficiency issues resulting in insufficient supply to some households
with more electric appliances.
Our findings supports the previous studies’ emphasizes on the importance
of considering the social factors in the community (Hirmer & Guthrie, 2016;
Pandyaswargo et al., 2022). The identified problem with battery price is also in
line with the previous study in Thailand (Veilleux et al., 2020). Furthermore, our
concern about the challenge of sustaining something that must be continuously
provided from the external is also in line with the Greek study (Kalamaras et al.,
2019).
Similar to our previous study in an Indonesian agricultural community
(Pandyaswargo et al., 2022), these types of complex social and technical
issues should be resolved by various stakeholders of the islands. Perhaps a
collaboration between the government, the private sector, and the community
to build a social business directly connected to the RE system would help. As
long as this collaboration is not well-formed, the RE system on small islands
remains complicated to maintain. In the end, if the people of the isolated islands
only expect external charity without solving the problem holistically, there will
always be sustainability problems entailing any type of energy solutions that
would be adopted.
CONCLUSION
From the analysis, we identified that the centralised system is better in terms
of techno-economic performance, as reflected in the NPC and COE. This finding
similar with the previous literature that the NPC for the centralised RE system
is more viable than the distributed RE system. Moreover, our sensitivity analysis
revealed that the price of electricity produced by the centralised RE system is
more stable compared to the distributed system that fluctuates with changes in
gasoline price, discount rate, and wind speed.
However, there are a number of challenges in maintaining a centralised
RE system: (1) a capacity gap in the community to implement and maintain the
system; (2) a lack of resources for battery replacement and gasoline supply; and
(3) the sparsely distributed population may lower transmission and distribution
efficiency.

Island Communities
While our findings are confirmed by previous studies as listed in Table 1,
we believe that each off-grid community has unique socio-economic as well as
RE potential, and geographical characteristics. These factors affect the social
challenges, financial and technical feasibility of a standalone renewable energy
system in each location. Perhaps, such high need of personalisation is what
hinders archipelagic countries such as Indonesia and the Philippines to achieve
their 100% electrification goals. Therefore, while findings in this chapter may
guide policy makers and practitioners on where and what to look for, it is not
intended to be technically applied as it is in different islands.
Furthermore, if the centralised RE system is taken as the energy access
solution, building the residents’ capacity to maintain the system is necessary.
This study suggests a collaboration of various stakeholders of small islands to
address the complex social and technical issues through investment in a social
business directly connected to the RE system. We recommend the assessment
tool we developed in our previous study (Pandyaswargo et al., 2022) to identify
the socio-techno-economic aspects for off-grid electrification system planning.
REFERENCES
Aghamolaei, R., M.H. Shamsi, & J. O’Donnell (2020). ‘Feasibility analysis of
community-based PV systems for residential districts: A comparison of
on-site centralized and distributed PV installations.’ Renew Energy 157:
793–808.
Ahlborg, H. & L. Hammar (2014). ‘Drivers and barriers to rural electrification
in Tanzania and Mozambique - grid-extension, off-grid, and renewable
energy technologies.’ Renew Energy 61: 117–124.
Bertheau, P., J. Dionisio, C. Jütte, & C. Aquino (2020). ‘Challenges for
implementing renewable energy in a cooperative-driven off-grid system in
the Philippines.’ Environ Innov Soc Transit 35: 333–345.
Chambon, C.L., T. Karia, P. Sandwell, & J.P.Hallett (2020). ‘Techno-economic
assessment of biomass gasification-based mini-grids for productive energy
applications: The case of rural India.’ Renew Energy 154: 432–444.
Chmiel, Z. & S.C. Bhattacharyya (2015). ‘Analysis of off-grid electricity system at
isle of eigg (Scotland): Lessons for developing countries. Renew Energy 81,
578–588.
Come Zebra, E.I., H.J.van der Windt, G. Nhumaio, & A.P.C. Faaij (2021). A review of
hybrid renewable energy systems in mini-grids for off-grid electrification
in developing countries. Renewable and Sustainable Energy Reviews 144.
Dornan, M., 2014. ‘Access to electricity in Small Island Developing States of the
Pacific: Issues and challenges.’ Renewable and Sustainable Energy Reviews
31: 726–735.
Ghasemi, A., A. Asrari, M. Zarif, & S. Abdelwahed (2013). ‘Techno-economic
analysis of stand-alone hybrid photovoltaic–diesel–battery systems for
rural electrification in eastern part of Iran—A step toward sustainable
rural development.’ Renewable and Sustainable Energy Reviews 28: 456–
462.
Hirmer, S. & H. Cruickshank (2014). ‘The user-value of rural electrification:
An analysis and adoption of existing models and theories.’ Renewable and
Sustainable Energy Reviews.
Hirmer, S. & P. Guthrie (2016). ‘Identifying the needs of communities in rural
Uganda: A method for determining the ‘User-Perceived Value’ of rural
electrification initiatives.’ Renewable and Sustainable Energy Reviews 66:
476–486.
Hong, G.W. & N. Abe (2012). ‘Sustainability assessment of renewable energy
projects for off-grid rural electrification: The Pangan-an Island case in the
Philippines.’ Renewable and Sustainable Energy Reviews.
IEA (2021). Access to electricity – SDG7: Data and Projections – Analysis - IEA
[WWW Document]. Available at: URL https://www.iea.org/reports/sdg7-
data-and-projections/access-to-electricity
IESR (2019). “Akses Energi yang Berkelanjutan untuk Masyarakat Desa” [WWW
Document]. Seri Diskusi Pojok Energi. Available at: URL http://iesr.or.id/
wp-content/uploads/2019/05/Proceeding-PE-11.pdf
International Renewal Energy Agency (2018). Transforming Small-Island
Power Systems Technical Planning Studies for the Integration of Variable
Renewables.
IPCC (2015). “Climate Change 2014, Small Islands, Climate Change 2014: Impacts,
Adaptation and Vulnerability.” Part B: Regional Aspects: Working Group
II Contribution to the Fifth Assessment Report of the Intergovernmental
Panel on Climate Change.

Island Communities
IRENA (2020). “Renewable Power Generation Costs in 2020.” International
Renewable Energy Agency.
Julius, E., A.O. Olufemi, & D.J. Chuks (2014). ‘Challenges of sustaining off-
grid power generation in Nigeria rural communities.’ African Journal of
Engineering Research 2: 51–57.
Kalamaras, E., M. Belekoukia, Z. Lin, B. Xu, H. Wang, & J. Xuan (2019). ‘Techno-
economic Assessment of a Hybrid Off-grid DC System for Combined Heat
and Power Generation in Remote Islands.’ Energy Procedia 158: 6315–
6320.
Kanagawa, M. & T. Nakata (2008). ‘Assessment of access to electricity and the
socio-economic impacts in rural areas of developing countries.’ Energy
Policy 36: 2016–2029.
Lambert, T., P. Gilman, & P. Lilienthal (2006). ‘Micropower System Modelling
with HOMER.’ In Felix A. Farret B.E. & M. Godoy Simões (eds).
Integration of Alternative Sources of Energy. Available at: https://doi.
org/10.1002/0471755621.ch15
Lozano, L., E.M. Querikiol, M.L.S. Abundo, & L.M. Bellotindos (2019). ‘Techno-
economic analysis of a cost-effective power generation system for off-grid
island communities: A case study of Gilutongan Island, Cordova, Cebu,
Philippines.’ Renew Energy 140: 905–911.
Mandelli, S., J. Barbieri, R. Mereu, & E. Colombo (2016). ‘Off-grid systems for
rural electrification in developing countries: Definitions, classification,
and a comprehensive literature review.’ Renewable and Sustainable Energy
Reviews 58: 1621–1646.
Maritime Fairtrade (2023). “Filipino Women Speak Out On Equitable Energy
Access.” [WWW Document]. URL https://maritimefairtrade.org/filipino-
women-speak-out-on-equitable-energy-access/
Mudgal, V., K.S. Reddy, & T.K. Mallick (2019). ‘Techno-economic analysis of
standalone solar photovoltaic-wind-biogas hybrid renewable energy system
for community energy requirement.’ Future Cities and Environment 5.
National Renewable Energy Laboratory/NREL (2022). RE Data Explorer. [Data
Set Title: Himawari Global Horizontal Irradiance]
Pandyaswargo, A.H., N. Abe, & G.W.C. Hong (2014). “Participatory workshop on
bottom-up study contributing to the realization of Sustainable Development
Goals: Pangan-an Island case study.” [WWW Document]. Technical Report
of International Development Engineering. URL http://www.ide.titech.
ac.jp/en/tride
Pandyaswargo, A.H., A.D. Wibowo, A.D., & H. Onoda (2022). ‘Socio-techno-
economic assessment to design an appropriate renewable energy system
for remote agricultural communities in developing countries.’ Sustain Prod
Consum.
Pascasio, J.D.A., E.A. Esparcia, M.T. Castro, & J.D. Ocon (2021). ‘Comparative
assessment of solar photovoltaic-wind hybrid energy systems: A case for
Philippine off-grid islands.’ Renew Energy 179: 1589–1607.
Renewable Energy Agency, I. (2019). “Off-grid renewable energy solutions to
expand electricity access: An opportunity not to be missed.”
Roy, D. (2023). “Modelling an off-grid hybrid renewable energy system to deliver
electricity to a remote Indian island.” Energy Convers Manag 281: 116839.
Shahzad, M.K., A. Zahid, T. Rashid, M.A. Rehan, M. Ali, M., & M. Ahmad (2017).
‘Techno-economic feasibility analysis of a solar-biomass off grid system for
the electrification of remote rural areas in Pakistan using HOMER software.’
Renew Energy 106: 264–273.
Urmee, Tania & Md. Anisuzzaman (2016). ‘Social, cultural, and political
dimensions of off-grid renewable energy programs in developing countries.’
Renew Energy 93: 159–167.
Veilleux, G., T. Potisat, D. Pezim, C. Ribback, J. Ling, A. Krysztofiński, A. Ahmed,
J. Papenheim, A.M. Pineda, S. Sembian, & S. Chucherd (2020). ‘Techno-
economic analysis of microgrid projects for rural electrification: A
systematic approach to the redesign of Koh Jik off-grid case study.’ Energy
for Sustainable Development 54: 1–13.
Vendoti, S., M. Muralidhar, & R. Kiranmayi (2021). ‘Techno-economic analysis
of off-grid solar/wind/biogas/biomass/fuel cell/battery system for
electrification in a cluster of villages by HOMER software.’ Environ Dev
Sustain 23: 351–372.

Island Communities
Wirawan, H. & Y.M.L. Gultom (2021). ‘The effects of renewable energy-based
village grid electrification on poverty reduction in remote areas: The case
of Indonesia.’ Energy for Sustainable Development 62: 186–194.
Woods Institute for the Environment (2010). “Distributed vs. Centralized Power
Generation.” Large-Scale Solar Technology and Policy Forum, 8-9 April.
World Bank (2023). Access to electricity (% of population) [WWW Document].
World Bank Global Electrification Database from “Tracking SDG 7: The
Energy Progress Report.” URL https://data.worldbank.org/indicator/
EG.ELC.ACCS.ZS
APPENDICES
Appendix 1. Sensitivity Analysis Results for the Centralised RE System
Appendix 2. Sensitivity Analysis Results for the Distributed Hybrid RE/

Wind System
CHAPTER 8
The Shift of Ecosystem Services

Use in Nusa Lembongan and Nusa
Ceningan due to the COVID-19
Pandemic
Alan Dwi Wibowo and Andante Hadi Pandyaswargo
INTRODUCTION
With strategic value determined by their position, condition, and geographical
potential, small islands have their own challenges in the development process,
However, small islands are often limited by the availability of potential land
resources (mineral/mining potential, potential for agricultural farming, land
availability, etc.), transportation accessibility, and the existence of infrastructure
that poses more challenges for local communities to improve or obtain a decent
standard of living. The potential for disasters, on the other hand, also poses
threats such as abrasion, rising sea levels as a result of global warming, and
tsunamis. Apart from the limiting factors that give rise to various “problems,”
small islands as an ecosystem still offers various forms of services that provide
various direct and indirect benefits.
Ecosystem services in the context of coastal areas and small islands
often dominate the use of cultural services as indicated by the Indonesian
government’s agenda which directs their development as competitive tourist
destinations (Bappenas, 2015). Maritime tourism—i.e., the one that exists in
coastal and marine areas—is one of the oldest forms of tourism and the largest
163
segment of the tourism industry. In 2005, the marine tourism market was
estimated to represent more than 10% of total tourism expenditure worldwide
(Diakomihalis, 2007), being the most visited area by tourists. Hence, marine
tourism is often a major economic driver and job generator. In general, tourism
plays an important role in regional economic growth, generating a source of
income and employment for small islanders (Kurniawan et al., 2016). While
there is no doubt that tourism plays a key role economically, its continued
growth with a high concentration of tourists and its existence often in small
tourist destinations area will present major externalities for ecosystems and
their functions (Drius et al., 2019).
Despite their value, coastal and marine ecosystems are among the most
threatened and degraded ecosystems in the world (Barbier, 2017; Martínez et
al. 2007). Therefore, it becomes important to focus on policy and management
interventions from the global to the local level (Patterson and Glavovic, 2013).
The phenomenon of the COVID-19 pandemic that shook the entire world at
the beginning of 2020, forced a massive transformation of lifestyles, and had
a significant impact on various sectors of the world economy. Tourism has
received the greatest impact in this situation (Škare et al., 2021) due to global
restrictions on human mobility that have reduced the number of international
tourist arrivals, as well as the implementation of the lockdown and fears of
COVID-19 spread that reduced domestic tourist visits (Kartiko, 2020).
Local communities on small islands in Indonesia are one of the groups that
are vulnerable to COVID-19 (Hakim, 2020), apart from being threatened with
the risk of spreading, the pandemic also creates socio-economic problems. Nusa
Lembongan and Nusa Ceningan in Bali, as a small island and tourist destination,
face this kind of problem. Most of the people lost their main jobs and were
slowly forced to transform and look for other alternative sources of livelihoods.
This phenomenon causes a shift in the use of ecosystem services that were
previously dominated by cultural services to provisioning services, as indicated
by the revival of seaweed farming.
This chapter aims to understand the process of how the use of ecosystem
services and the livelihoods of the local communities have shifted. The research
was conducted quantitatively and qualitatively using data that were obtained
through observation, questionnaire, and in-depth interview. This aims to enrich
information to comprehensively examine the dynamics of processes that cause
the shift in the use of ecosystem services, an issue which is rarely discussed.
USE OF CLUSTER RANDOM SAMPLING METHOD
This study was carried out in Nusa Lembongan and Nusa Ceningan, small
islands that are part of the Nusa Penida Marine Protected Area in the Province
of Bali. The survey and data collection were conducted on 21 – 28 October 2021.
A quantitative approach is used to see the shift in the use of ecosystem services
and changes in the livelihoods of the local communities of Nusa Lembongan
and Nusa Ceningan. This study used cluster random sampling method, and the
respondents are people who work as seaweed farmers when the survey is being
conducted. The data collection process was carried out alone for 7 days, asking
the respondent’s occupation before farming seaweed, the amount of assets
owned, and the estimated amount of income.
The number of samples in each village was determined using the Slovin
Formula—one of the equations for determining the number of representative
samples—with a 10% margin of error. The sampling population was total
number of people who work as farmers (1.845 people) in general in 2020, based
on the village’s profile of Jungutbatu and Lembongan (Profil Desa Jungutbatu,
2020; Profil Desa Lembongan, 2020).
A qualitative approach is also used to enrich information that allows
researchers to comprehensively examine the dynamics of the processes that
cause the shifting use of ecosystem services due to COVID-19 pandemic.
Information obtained through in-depth interviews with the village
administration, institutional representatives of the Marine Protected Area
(MPA) of Nusa Penida, the initiator of seaweed farming and local communities
who changed professions from tourism sector to seaweed farmers.
BACKGROUND OF STUDY SITE

Tourism in Nusa Lembongan and Nusa Ceningan and the Disruption
of COVID-19
Nusa Lembongan and Nusa Ceningan are small islands located in the southeast
of Bali, and has been known as one of the popular tourist destinations. Nusa
Lembongan and Nusa Ceningan islands have an area of 6.15 km², and are
divided into two villages, namely Lembongan and Jungutbatu. Most of the
land use is designated for residential areas to support tourism activities, and a
small portion of the remaining land is vegetation which is agricultural land and
The Shift of Ecosystem Services Use in Nusa Lembongan and Nusa Ceningan 165
due to the COVID-19 Pandemic
mangrove forest. Nusa Lembongan and Nusa Ceningan as a tourist destination
has begun to be visited in the 2000s, and grew more rapidly in the early 2010s,
as indicated by an increase in the number of visits and tourism service providers
(Rahmawati, Trianasari and Martin, 2019).
The high dependence on the tourism sector cannot be denied, shown by
how it is able to shift the existence of seaweed farming which has been the main
livelihood for decades. Mathematically, tourism provides far more benefits at
a lower cost, compared to seaweed farming which shows the opposite. This
shift was strongly influenced by market demand and exceptionally low prices
in 2015. Therefore, this shift—which in this study will refer to as the first
shift—can be said to have occurred without a hitch. At the beginning of 2020,
the world was shocked by the presence of COVID-19, which shook various
dimensions of human life. Locally, this was also a turning point that caused
a major blow to tourism conditions in Nusa Lembongan and Nusa Ceningan.
Many businesses and tourism service providers could no longer operate in the
absence of tourists. This uncertain condition forced a shift—which in this study
will refer to as the second shift—of livelihoods back towards seaweed farming
that previously showed signs of disappearing at times of flourishing tourism
(Andréfouët et al., 2021).
The Bali Provincial Government, through the Department of Fisheries and
Marine Affairs, has made a management and zoning plan that regulates the
utilisation and activities that take place in the Nusa Penida Marine Protected
Area, with the aim of regulating the use of natural resources, managing
ecosystems and their services, and maintaining local wisdom so as to ensure the
availability, sustainability, and diversity of coastal ecosystems (KKP Nusa Penida,
2017). Zoning divides the permitted activities in the Nusa Lembongan and Nusa
Ceningan and Nusa Ceningan coastal area, with the main zones being based on
sustainable fisheries zones (traditional fishing, seaweed farming, and special
tourism) and utilisation zones (ports and marine tourism in general). Although
zoning arrangements have been in place, the reality is that the existing activities
still overlap, especially in the location of seaweed farming and tourism, which
became one of the reasons that eventually caused the seaweed to “disappear”
for a short period of time, and will be discussed in the next section.
Figure 1. Zoning Sheet of Nusa Lembongan and Nusa Ceningan island
(Source: Management and Zoning Plan For A Marine Protected Area
(Kawasan Konservasi Perairan) Nusa Penida)
Seaweed Farming as the ‘Lifesaver’ of Nusa Lembongan and Nusa

Ceningan
In 2019, Indonesian seaweed production reached 9,746,946 tons. Of the total
exported seaweed raw materials, 82% were sent to China (BPS 2020). The
potentials of Indonesian seaweed farming have contributed to the increased
income of coastal communities in the areas of Sumatra, Java, Bali, Nusa Tenggara,
Sulawesi, and Maluku.
As an archipelagic country, Indonesia has a remarkably high diversity
of seaweed species and high production potential. Yet, in reality its seaweed
industrialisation has not been maximally developed, which makes it still lag far
behind compared to Japan, Korea, Taiwan and China (Kadi, 2004).
Nonetheless, the Central Government of Indonesia has created a national
seaweed industry development roadmap for 2018-2021 through presidential
decree. The final expected target in 2021 is to become the market leader in
the semi refined carrageenan (SRC) and gelatine industry, achieving a 50:50
export composition for raw materials and finished goods, and reach 5% of
recommended mix of seaweed in fish and livestock feed (Perpres Nomor 33
Tahun 2019).
Nusa Lembongan and Nusa Ceningan are famous for seaweed farming,
which is thought to have existed since the middle of 1980s, and became the
main livelihood for the local population at that time. Various types of seaweed
have been tried to be cultivated on the coast of Nusa Lembongan and Nusa
Ceningan, and finally Euchema spinosum and Euchema cottonii are the two
seaweed species that are able to adapt and be cultivated. Seaweed farming has
been the longest-lasting livelihood for decades, and has been a lifesaver for local
communities long before tourism became their main source of income. There was
a period of time where seaweed farming had experienced a significant decline
in production and the area of planted land, and seemed to have disappeared.
These symptoms occurred in 2015 due to the rapid development of the tourism
sector (Andréfouët et al., 2021), and at the same time, production costs are not
comparable to the selling price of dried seaweed, which then became unstable
and tended to be low, prompting the shift of livelihoods to jobs engaged in
tourism services.
The Regional Government of Klungkung Regency and stakeholders from
the communities in 2017 had initiated a movement to restore seaweed farming
aimed to revive the image and identity of Nusa Lembongan and Nusa Ceningan as
a leading producer of seaweed in Bali. Unfortunately, these efforts do not attract
the public because of seaweed farming, considering the cost of production and
the instability of prices and market demand, cannot compete with tourism that
provides a more instant source of income. However, the COVID-19 pandemic
and unexpectedly increasing market demand for seaweed at the same time, had
stimulated a second shift towards intensive seaweed farming which began to
occur in mid-2020, in line with the period of lockdown in Indonesia.
IDENTIFICATION OF COASTAL ECOSYSTEM SERVICES

(CES) IN NUSA LEMBONGAN AND NUSA CENINGAN
Ecosystem services (ES) are benefits that humans obtained directly and
indirectly from ecosystem functions (Costanza et al. 1997; Millennium
Ecosystem Assessment, 2005); Costanza et al., 2014). In general, the
classification of ecosystem services in coastal and marine areas is divided
into provisioning services (source of food/fishery, raw materials, medicines
and genetic resources), regulating services (flood/storm/erosion and climate
regulation), cultural services (tourist and recreation, education, knowledge,
history, culture and traditions), and supporting services (primary production,
nutrient cycling, species/ecosystem protection) (Friess et al., 2020). These four
classes are often regrouped into two domains, namely intermediate and final
ecosystem services.
Intermediate ecosystem services are described as ecosystem services with
little or no direct benefit to human, consisting of the biophysical structures and
processes that maintain ecosystems, and how they interact and contribute to
the provision of final ecosystem services from time to time. In contrast, final
ecosystem services are benefits that are received explicitly (or directly) and
are both biophysically and monetarily accountable (Fisher and Turner 2008;
Barbier et al., 2011; Lamothe and Sutherland, 2018). Dependence on the use of
ecosystem services, pressures received by ecosystems from the accumulation
of human activities that give greater burdens, and human failure to take into
account the various services provided by ecosystems is a major factor causing
the degradation and loss of these ecosystem services (Waite et al. 2014; Christie
et al., 2015).
Both terrestrial and aquatic ecosystems are threatened by increased
number of human populations resulting in cumulative impacts that degrade
ecosystem functions and ecosystem services, and the importance of them is
only realised when they are already lost (Carrasco De La Cruz, 2021; Heckwolf
et al., 2021). Assessment of ecosystem services helps bridge the conceptual gap
between the ecological and social sciences, linking the state of ecosystems to
human well-being and activities (Böhnke-Henrichs et al., 2013), and to evaluate
progress towards sustainable development (Griggs et al., 2013; Liu et al., 2015).
Five ecosystem types were identified in the observations of Nusa Lembongan
and Nusa Ceningan, with each potential final ecosystem service provided for
each ecosystem type, as presented in Table 1 below.
Based on visual observation and identification through satellite imagery,
five types of ecosystems were found in the coastal area of Nusa Lembongan and
Nusa Ceningan. Although not all benefits can be obtained and directly measured
in monetary value, each identified ecosystem type in the Nusa Lembongan and
Nusa Ceningan area has provided various benefits that support the lives of
local communities, which can be enjoyed directly in the form of final ecosystem
services. In terms of provisioning services (products that are directly obtained
from ecosystems), coastal and marine areas provide various sources of food,
especially fisheries and habitats for various types of wild and cultivated plant
and animal species that maintain biodiversity. The habitat here also includes a
role as a location for seaweed farming carried out by local communities in the
continental shelf area.
Table. 1 Coastal Ecosystems and Its Services in Nusa Lembongan and
Nusa Ceningan
Coastal Type of Ecosystem
Ecosystem Coralligenous
Services Sandy shores Sea grass Mangroves
habitats
1. Provisioning Potential of Raw material Raw material Raw material
services renewal energy Food Food Food production
production production production
2. Regulation Coastal Coastal Coastal Coastal protection
services protection protection protection Erosion control
Erosion control Erosion control Erosion control Storm protection
Water protection
Biological control
Climate, air
regulation
3. Cultural Recreation and Recreation and Recreation and
services tourism tourism tourism
Symbolic, Symbolic and Symbolic,
–
aesthetic, aesthetic values aesthetic,
tradition values Cognitive effect tradition values
Cognitive effect Cognitive effect
4. Supporting Habitat Habitat Habitat
services Biomass Biomass Biomass
– production production production
photosynthesis photosynthesis photosynthesis
Nutrient cycle Nutrient cycle Nutrient cycle
Description: The benefits are No immediate Immediate Opportunities that
hypothetical benefits were benefits were can be developed
referring to found found
other studies.
Source: author’s data analysis - modified after (Liquete et al., 2013) and (Drius et al., 2019)
The underwater ecosystem that is still preserved in the waters of Nusa
Lembongan and Nusa Ceningan (and Nusa Penida broadly) makes it a favourite
place for diving, and also plays a role in maintaining fish catches, especially in
the current pandemic conditions, apart from farming seaweed, there are also
many local communities who turn to fishermen. An interesting finding at this
location is that the habitat of seagrass overlaps with the seaweed farming area.
Seagrass grows at the bottom of the water, while seaweed lives on the surface.
Competition for sunlight causes seagrass to slowly decrease, and its function in
reducing the size of sea waves is being replaced by seaweed. Therefore, there is
no significant ecological damage in coastal areas from the loss of seagrass.
The Shift in the Use of Ecosystem Services

As a tourist destination and local communities whose livelihoods depend
on tourism, cultural services are the most dependable or utilised form of
coastal ecosystem services in Nusa Lembongan and Nusa Ceningan. All local
communities work in tourism-related fields, and on each side of the island are
found various tourism-related activities. Traces of other economic activities
that can still be seen are fisheries and a bit of land agriculture, whereas traces
of seaweed farming were not even visible in 2017, when the pandemic had not
yet occurred. Ecosystems are seen as having more monetary value as objects
that can be enjoyed its scenery, rather than as an area that needs to produce raw
materials productively.
Based on the types of ecosystem services that have been identified and the
reality of the COVID-19 pandemic that is still being felt in the mid of 2022, there
is a pattern that shows a shift towards consumption or utilisation of coastal
ecosystem services. The shift happened, in particular, from utilizing cultural
services to provisioning services. This shift also includes changes in the work
of local communities, followed by changes in the landscape in coastal areas
through periodic observations of Maxar satellite imagery (Google Earth).
Observations of the seaweed farming landscape through remote sensing and
household survey in Nusa Lembongan and Nusa Ceningan have previously been
carried out by Andréfouët et al. (2021), which tries to see the phenomenon of
seaweed collapse. This research can be said as a follow-up to previous research,
which emphasizes the reverse phenomenon, namely the revival of seaweed
farming. Satellite imagery captures the landscape condition of Seaweed Point
which is one of the largest seaweed farming sites located in Lembongan Village
based on three different times.
In 2013 (Figure 2) tourism showed an intensive development,in line with
seaweed farming that was still surviving. According to Andréfouët et al. (2021),
symptoms of seaweed loss started in 2015 and peaked in 2017 (Figure 3). The
physical appearance of seaweed farming is no longer found, indicating that it
has been completely displaced by the presence of tourism. On the outskirts of
Seaweed Point, many restaurants and inns are found, which makes Seaweed
Point a backdrop for photographic scenes. Some respondents admitted that
for tourists, seaweed is an attraction to enjoy in Nusa Lembongan and Nusa
Ceningan, but on the other hand for tourism organizers, which are local
communities themselves, their presence is considered a visual distraction that
eliminates the aesthetic value of the scenery. Conditions in 2021 (Figure 4),
along with the ongoing disruption of COVID-19 and the absence of tourism,
the coastal waters were again converted to seaweed farming with the land use
being much wider and denser than in 2013.
Figure 2. Satellite imagery of Seaweed Point in 2013. Seaweed farming
still exists, along with increasing tourism visits
Source: Maxar Satellite Imagery (Google Earth)
Figure 3. Satellite imagery of Seaweed Point in 2017. Traces of seaweed

farming not found, showing symptoms of “losing”.
Figure 4. Satellite imagery of Seaweed Point in 2021. Seaweed farming is
growing rapidly along with the absence of tourism visits

Although the rise of seaweed farming was affected by the COVID-19
pandemic, the initial steps to re-establish seaweed farming even started
before COVID-19 occurred. In pre-pandemic conditions, it was found that
three respondents owned land and farmed seaweed. The three respondents
coincidentally are pioneers or early figures in seaweed farming. One of the three
is a figure who received a request from the Klungkung Regency Government
to re-initiate seaweed farming. The intention to re-initiate seaweed farming
already exists in most local communities as Andréfouët (2021) also found in a
small number of respondents in the previous research. The inhibiting factors
are based on various considerations. First, there are conditions of demand and
the market price is not friendly for farmers, which is around IDR 2000 – IDR
5.000 per semi-dried (40%) kilogram of seaweed. Second, starting seaweed
farming in a new area (the traces of previous seaweed farming have been lost)
require extra labor to clear the planting area, especially from dead coral and
seagrass debris, and install a large number of stakes reaching ±100 cm deep.
Third, seaweed farming not only requires production factors but also dedication
and strong physical conditions. Fourth, there is a sense of shame when doing
seaweed farming, because they are underestimated and tourism sector jobs are
seen as a better option.
The current condition is different, when seaweed farming is the only option
that most respondents consider the only answer, including fishing for some
people who own boats. The selling price of seaweed is currently improving,
around IDR20,000 per semi-dried (40%) kilogram for Euchema cottonii. Apart
from vulnerable groups of people, who previously worked in the tourism
sector, groups of people who are more economically stable (have a permanent
profession – teacher; civil servant) are also involved in this seaweed farming
movement as a second source of livelihood. The five respondents who worked
as teachers and civil servants gave the same statement, that the reason for their
participation in seaweed farming was to consider a good selling price so as to
increase their income.
Figure 5. Pie chart showing the respondent’s previous occupation before
doing seaweed farming
Source: author’s data analysis
Figure 6. The number of respondents who have done and have done
seaweed farming before Covid-19

Figure 7. Bar chart showing the number of “ris” that respondents have
based on 4 classes
Most respondents are reluctant to mention the amount of monthly income
they earned from seaweed farming, and others found it difficult to project the
amount of income they earned due to uncertain harvests within a certain period
of time. However, all local communities are willing to answer regarding the
amount of “ris” they have. “Ris” is the term used by local communities for the
rope used to tie seaweed seed, with a length of 2.5-3 meters. Each respondent
started seaweed farming individually and at different times. This causes the
distribution of land used by each person is relatively different. People who start
seaweed farming early tend to have a wider planting area, so they have higher
“ris” and higher yields. The three initiators of seaweed farming are the owners
of the largest land with the number of “ris” that they each own more than 1000
units.
Using questionnaire data obtained from 180 people, a simple calculation
model is used to project the value of the seaweed commodity. Each individual,
based on the number of “ris” owned and assuming that every 1 “ris” can produce
1kg of semi-dried seaweed, then for every one “ris” an income of IDR 20.000,00
will be obtained. The amount of “ris” will be linear with the amount of income.
Hence, with another assumption that all “farmers” in Nusa Lembongan and
Nusa Ceningan are “seaweed farmers”, the projected income for the seaweed
subsector is IDR 10.762.500.000, which is five times greater than the national
village funds allocation. This shows that the economic value of seaweed has the
potential to become a leading commodity and the main livelihood that supports
local communities, apart from tourism.
Table. 2 The Economic Value of the Nusa Lembongan and Ceningan
Seaweed Commodities
Average Income of All National Village
Village Samples Farmers
Income Farmers Funds Allocation
Jungutbatu 85 IDR 4.329.412 557 IDR 2.411.482.353 IDR 1.153.966.000
Lembongan 95 IDR 7.178.947 1288 IDR 9.246.484.211 IDR 979.592.000
Total 180 IDR 5.833.333 1845 IDR 10.762.500.000 IDR 2.133.558.000
The national seaweed industry development roadmap targets that in the
first year of realisation, in 2018, it is to accelerate the increase in processed
seaweed production capacity, and the operation of new factories near the
farming agricultural sites. Seaweed farming in Nusa Lembongan and Nusa
Ceningan was active again in mid-2020. However, the capacity of local human
resources is still limited to the production of raw materials and has not been able
to carry out further processing. This condition is out of the roadmap that has
been planned, as a result of the inability to maintain seaweed farming and the
unstoppable shift to the tourism sector, yet no one is to blame for this problem.
It seems difficult to maintain seaweed farming in the same area where
tourism activities take place, because local communities will be faced with the
choice of which sector can provide more income with less effort, while seaweed
farming is unlikely to be a favorite choice when the pandemic ends. But with the
experience of the tourism crisis during the pandemic, it is the time for the local
communities to consider revitalising seaweed farming to the maximum.
Figure 8. Seaweed Farming Activities in Nusa Lembongan and Nusa
Ceningan
In the view of local tourism entrepreneurs, the coastal ecosystem of Nusa

Lembongan and Nusa Ceningan—which was previously a recreational object
with an appreciation of the aesthetic value of its scenery—should be made
free from all forms of visual disturbances, including the existence of seaweed
farming. But now, the coastal ecosystem is seen as a production factor with
productivity needs to be maximised. In this regard, shifting the use of ecosystem
services, livelihoods, followed by changes in its landscapes that occur flexibly
and rapidly over short periods of time, remains possible since the ecosystems
still have the capacity to support these efforts, and traces of previous seaweed
farming experience are still available.
Small islands facing similar problems do not necessarily have the
opportunity and capacity to recover quickly as happened in Nusa Lembongan
and Nusa Ceningan. Signs of recovery have been seen and Bali tourism is slowly
starting to revive. Slowly Nusa Lembongan and Nusa Ceningan will return to
being busy with tourism. Is there a chance for a third shift to emerge which
will lead to the reduction and disappearance of seaweed farming? There is a
possibility for that. However, learning from the experiences that have occurred,
local communities may have had their own wisdom in taking the next step.
The COVID-19 pandemic provides lessons about the importance of giving
more recognition to coastal ecosystem management, and the need to learn to
maximise the potential and capacity of ecosystem services.
CONCLUSION
The coastal ecosystem of Nusa Lembongan and Nusa Ceningan shows a capacity
that is still able to provide services that can be utilised by local communities,
in the midst of uncertainty due to COVID-19. The shift in the use of ecosystem
services—i.e., from the previous seaweed farming in the mid-1980s toward
the dominance of cultural services for tourism a decade ago, and later back to
provisioning services—is indicated by the revival of seaweed farming which
is the main source of livelihood during the absence of tourists in the midst of
pandemic in 2020. The process that occurred in a fast time and without many
major obstacles that hindered was possible because the ecosystem still has the
capacity to support these changes. The revival of seaweed cultivation supports
the central government’s mission to increase the production of seaweed raw
materials and processed goods, to meet domestic needs and export targets.
Nevertheless, Nusa Lembongan and Nusa Ceningan needs to increase the
capacity of local human resources in further processing the seaweed, in order to
sustain its economic potential, which otherwise would be overshadowed by the
income generated from tourism industry. Looking at the uncertain future and
learning from the past experiences, the basis for developing small islands needs
to have other alternative sources of income and not solely rely on tourism,
which is easily interrupted as has been the case during the pandemic time. Thus,
planning for small islands development in the future needs to be designed not
only based on one leading sector, in this context it is tourism which has proven
to be vulnerable, as the main focus. Other alternative economic sectors that are
more resilience and not susceptible to sudden disruptions, such as seaweed
farming, could be the answer to solve the problem.
ACKNOWLEDGMENT
The study was funded by the PPMI research scheme of the School of Architecture,
Planning and Policy Development of Institut Teknologi Bandung (ITB) year
2021.
REFERENCES
Andréfouët, S., I.M.Dewantama, & E.E. Ampou (2020). ‘Seaweed farming
collapse and fast changing socio-ecosystems exacerbated by tourism and
natural hazards in Indonesia: A view from space and from the households
of Nusa Lembongan island.’ Ocean and Coastal Management, 207 (June).
doi: 10.1016/j.ocecoaman.2021.105586.
Badan Pusat Statistik (2020). Ekspor Rumput Laut dan Ganggang Lainnya
menurut Negara Tujuan Utama, 2012-2020. Available at: https://www.bps.
go.id/statictable/2019/02/25/2025/ekspor-rumput-laut-dan-ganggang-
lainnya-menurut-negara-tujuan-utama-2012-2020.html (Accessed: 15
June 2022).
Bappenas(2015). ‘Strategi Kebijakan Perencanaan Pembangunan Pariwisata di
Pulau-Pulau Kecil Sebagai Pusat Perumbuhan.’ Jakarta.
Barbier, E. B. et al. (2011). ‘The value of estuarine and coastal ecosystem services.’
Ecological Monographs, 81(2): 169–193. doi: 10.1890/10-1510.1.
Barbier, E. B. (2017). ‘Marine ecosystem services.’ Current Biology, 27(11),
R507–R510.
Böhnke-Henrichs, A. et al. (2013). ‘Typology and indicators of ecosystem services
for marine spatial planning and management.’ Journal of Environmental
Management, (130): 135–145.
Breslow, S. J. et al. (2016). ‘Conceptualizing and operationalizing human
wellbeing for ecosystem assessment and management.’ Environmental
Science & Policy, (66): 250–259.
Carrasco De La Cruz, P. M. (2021). ‘The Knowledge Status of Coastal and Marine
Ecosystem Services - Challenges, Limitations and Lessons Learned from
the Application of the Ecosystem Services Approach in Management.’
Frontiers in Marine Science, 8 (May). doi: 10.3389/fmars.2021.684770.
Christie, M. et al. (2015). ‘Valuing marine and coastal ecosystem service
benefits: Case study of St Vincent and the Grenadines’ proposed marine
protected areas.’ Ecosystem Services, (11): 115–127. doi: 10.1016/j.
ecoser.2014.10.002.
Costanza, R. et al. (1997). ‘The value of the world’s ecosystem services and
natural capital’, Nature, 387(6630): 253–260. doi: 10.1038/387253a0.
Costanza, R. et al. (2014). ‘Changes in the global value of ecosystem services.’
Global environmental change, (26): 152–158.
Diakomihalis, M. N. (2007). ‘The impact of maritime tourism on the Greek
economy via the tourism satellite account.’ Tourism and Hospitality
Planning & Development, 4(3): 231–244.
Drakou, E. G. et al. (2017). ‘Marine and coastal ecosystem services on the science
– policy – practice nexus : challenges and opportunities from 11 European
case studies.’ International Journal of Biodiversity Science, Ecosystem Services
& Management, 13(3): 51–67. doi: 10.1080/21513732.2017.1417330.
Drius, M. et al. (2019). ‘Tackling challenges for Mediterranean sustainable
coastal tourism: An ecosystem service perspective.’ Science of the Total
Environment, (652): 1302–1317. doi: 10.1016/j.scitotenv.2018.10.121.
Fisher, B. and Turner, R. K. (2008). ‘Ecosystem services: Classification for
valuation.’ Biological Conservation, 141(5): 1167–1169. doi: 10.1016/j.
biocon.2008.02.019.
Friess, D. A. et al. (2020). ‘Indicators of scientific value: An under-recognised
ecosystem service of coastal and marine habitats.’ Ecological Indicators,
113 (September 2019), 106255. doi: 10.1016/j.ecolind.2020.106255.
Griggs, D. et al. (2013). ‘Sustainable development goals for people and planet.’
Nature, 495 (7441): 305–307.
Hakim, L. (2020). ‘COVID-19, tourism, and small islands in Indonesia: Protecting
fragile communities in the global Coronavirus pandemic.’ Journal of Marine
and Island Cultures, v9n1: 130–141.
Heckwolf, M. J. et al. (2021). ‘From Ecosystems to Socio-Economic Benefits:
A Systematic Review of Coastal Ecosystem Services in The Baltic Sea.’
Science of the Total Environment, (755): 142565. doi: 10.1016/j.
scitotenv.2020.142565.
Kadi, A. (2004). ‘Potensi Rumput Laut Di beberapa Perairan Pantai Indonesia.’
Oseana, XXIX(4), 25–36.
Kartiko, N. D. (2020). ‘Insentif pajak dalam merespons dampak pandemi
COVID-19 pada sektor pariwisata.’ Jurnal Pajak Dan Keuangan Negara
(PKN), 2(1): 124–137.
KKP Nusa Penida .(2017). ‘Rencana Pengelolaan dan Zonasi Kawasan Konservasi
Perairan (KKP) Nusa Penida.’ 175.
Kurniawan, F. et al. (2016). ‘Vulnerability assessment of small islands to
tourism: The case of the Marine Tourism Park of the Gili Matra Islands,
Indonesia.’ Global Ecology and Conservation, (6): 308–326. doi: 10.1016/j.
gecco.2016.04.001.
Lamothe, K. A. & I.J. Sutherland (2018). ‘Intermediate ecosystem services :
the origin and meanings behind an unsettled concept unsettled concept.’
International Journal of Biodiversity Science, Ecosystem Services &
Management, 14(1): 179–187. doi: 10.1080/21513732.2018.1524399.
Liquete, C. et al. (2013). ‘Current Status and Future Prospects for the Assessment
of Marine and Coastal Ecosystem Services: A Systematic Review.’ PLoS ONE,
8(7). doi: 10.1371/journal.pone.0067737.
Liu, J. et al. (2015). ‘Systems integration for global sustainability.’ Science,
347(6225), 1258832.
Martínez, M. L. et al. (2007). ‘The coasts of our world: Ecological, economic and
social importance.’ Ecological economics, 63(2–3): 254–272.
Millenium_Ecosystem_Assessment. (2015). ‘Ecosystems and human well-being -
Synthesis: A Report of the Millennium Ecosystem Assessment.’
Millennium_Ecosystem_Assessment. (2005). Ecosystems and human well-
being: wetlands and water. World Resources Institute.
Patterson, M. & B. Glavovic (2013). ‘From frontier economics to an ecological
economics of the oceans and coasts.’ Sustainability science, 8(1): 11–24.
Perpres_Nomor_33_Tahun_2019. (2019). ‘Peta Panduan Pengembangan Industri
Rumput Laut Nasional 2018-2021.’
Quijas, S. and P. Balvanera. (2013). ‘Biodiversity and Ecosystem Services.’
Encyclopedia of Biodiversity: Second Edition, 341–356. doi: 10.1016/B978-
0-12-384719-5.00349-X.
Rahmawati, P. I., N. Trianasari, & A.A.N.Y. Martin (2019). ‘The Economic Impact
of Mount Agung Eruption on Bali Tourism.’ Teams (69): 98–107. doi:
10.2991/teams-18.2019.18.
Škare, M., D.R.Soriano, & M. Porada-Rochoń. (2021) ‘Impact of COVID-19 on the
travel and tourism industry.’ Technological Forecasting and Social Change
163: 120469.
Waite, R. et al. (2014). ‘Coastal capital: ecosystem valuation for decision making
in the Caribbean.’
CHAPTER 9
Network of Relationships between

Actors in Developing Bonjeruk
National Ecotourism Village:
A Case Study
Mega Nisfa Makhroja and Lalu Agung Haris Atmaja
INTRODUCTION
Ecotourism is a type of tourism development that is responsible for the
preservation of unspoiled areas and for developing them into something that
provides economic benefits while maintaining the cultural and natural integrity
of the local community (Fennell, 2003). Due to its commitment to sustainable
development, ecotourism has grown to be a significant issue in recent years
(Gal Oberhofer Dimitrijević, 2022). When it comes to the economy, ecotourism
contributes to the growth of the nation’s economy while also promoting the
pace of inclusive development (including both local populations and nature)
(Vaugeois, 2015). Regarding the social side, the ecotourism industry offers
tourists a high-quality experience, particularly in terms of the distribution of
messages about conservation and education. The ecotourism industry then has
the ability to lessen the adverse effects of environmental harm brought about
by excessive visitor traffic. Developing countries, such as Indonesia, developed
ecotourism since it experienced the overtourism11 (Ecotourism, 2019).
1 Overtourism is defined as a serious issue that is hurting the entire tourism sector is overtourism.
Even though it only pertains to a small portion of travel, it has been increasingly discussed in the
media and in periodicals as an issue that originates from the fundamental idea of travel. In addition
to being wholly unsustainable, overtourism is an issue that needs to be addressed. Its policy solutions
are frequently just as clumsy and poorly thought out as the policies that led to them in the first place.
The small-scale sustainable tourism sector as well as the attitudes of passengers may be harmed by
overtourism (Ecotourism, 2019)
183
In general, ecotourism develops its implementation in accordance with its
positive value for sustainable development. In this regard, the use of ecotourism
is not without criticism, though. Businesspeople in the ecotourism industry
have not properly grasped the importance of ecotourism (Damanik, 2014).
Environmental preservation, educational value, and community empowerment
are crucial components of ecotourism that are not well implemented. Despite
their claims that ecotourism is conducted for green investments, businesspeople
frequently concentrate on mass tourism, which is related to the unsustainable
expansion of ecotourism infrastructure that is growing to accommodate the
growing demands of tourists while disregarding the principles of environmental
preservation (Hogh, 2003). For example, the construction of a green hotel area
(Travelling, 2015) that should support environmentally friendly development
has the opposite effect because the construction site does not consider land
feasibility studies, which has an impact on the ecotourism area itself (Hogh,
2003). As a result, tourism management needs to resort the necessary
infrastructure development. More improvement of road infrastructure to
tourism, lighting, and communication regions is required for ecotourism than
the establishment of luxury hotels (Rukendi, 2015).
Indonesia, as one of the countries with the largest biodiversity in the world,
has been facing challenges for ecotourism development. Most of the regions
have a minimum initiative to develop their ecotourism potential. Meanwhile, the
tourism sector contributes around 17% of total exports of goods and services and
contributes to Gross Domestic Product (GDP) of 3.8%, amounting to USD10.05
billion/year (Parekraf, 2014). To improve local community participation and
engagement about tourism issue, the government has initiated tourism village
awards, which is expected to become a chance to open attractive ecotourism
business prospects by building and developing villages (Aminoto, 2019)
One of the tourism villages that is currently being developed is the Bonjeruk
Tourism Village, which is located in Jonggat District, Central Lombok Regency.
Bonjeruk Tourism Village has many advantages, which do not only consist of
natural and historical environment but also the richness of traditional culinary
and food processing innovations that are creative and out of the box. Bonjeruk
Tourism Village is also one of the tourism villages that still upholds the customs
of the Sasak Tribe.
In addition, the Bonjeruk Tourism Village was once the centre of the Dutch
East Indies government during the colonial period. Due to its long history,
some of the buildings in this tourism village are still in European style, which
stand out among traditional buildings in rural areas. Beautiful natural scenery,
interesting culture, delicious food, and friendly people have attracted many
visitors to come and learn about Bonjeruk Tourism Village (Ustman, 2021).
The uniqueness and attractiveness of the tourism village of Bonjeruk make it
interesting to visit and study.
Due to its high potential as tourism village, Bonjeruk Tourism Village
has been facing challenges of its local management. Several actors who are
involved comprise businesspeople, NGOs, society/community members, and
government. However, those stakeholders have their own interests, which
becomes obstacle to develop Bonjeruk Village. This problem resulted in inequity
of Bonjeruk development since out of the existing 22 hamlets, only 12 hamlets
have been developed (Usman, 2021).
The main concern Bonjeruk Ecotourism development is about collaboration
among stakeholders to support the process. However, it seems hard since each
stakeholder has different interests. The stakeholders of Bonjeruk Ecotourism
include among others: (1) businesspeople who are concerned about historical
tourism; (2) nongovernmental organisations (NGOs) which concern about
environment and modern attraction; and (3) travel agents and community
in Bonjeruk Ecotourism Village who were impacted during the pandemic. In
this chapter, we will analyse how the interaction between the government,
entrepreneurs, and nongovernmental organisation in the development of the
Bonjeruk Tourism Village. This is important to explain about the slow process of
development in the said village. The result of the study presented in this chapter
is expected bring about positive mind about the importance of collaboration
in the future, which in the long run the development of Bonjeruk Ecotourism
will result in positive impact to the local economy, thus creating welfare for the
surrounding community accordingly.
PREVIOUS STUDIES ON TOURISM INDUSTRY

The tourism industry plays a key role in national economic growth because
tourism ranks 2nd in Indonesia’s foreign exchange (BI, 2018). According
to Balaam and Dilman (2011), tourism and migration are connected with
mobilisation from one country to another country, or it can be termed as
human mobilisation. Fossati & Panella (2000) define a framework for tourism
and sustainable development based on three elements, namely: (1) the idea of
movement (trip, change of one’s normal residence, or one’s normal habitat);

Island Communities
(2) the idea of a person’s behaviour as a tourist (the economic-social role
temporarily assumed); and (3) the environment, or tourist space (the territory’s
tourist vocation).
In the case of tourism, the movement of people is followed by global
connections that bring prosperity to the community and have implications not
only for the state, but also for local and international NGOs. Several international
NGOs are interested to support community development project in ecotourism
area in Indonesia, such as WWF, and Greenpeace. Hence, economic activity in
tourism industry is much more influenced by local and international factors. It
has become the main source of income for the government, especially in tourist
receiving countries (Dain Bolwell, 2008). The growing strength in the tourism
sector is a major supporter of the economy. Job creation, especially for other
economic sectors in the tourism industry, is an opportunity to increase the
welfare of local communities.
Governments play significant role to maintain tourism destinations due to
its policy that will have significant impact on tourism infrastructure aspect. In
addition to having an impact on the economy, the tourism industry is an agent
of social, cultural, and environmental change (Mowforth, 2015). There has been
a change in the culture and economy of the community in managing tourism in
their country.
Milne (2001) reviews the complex relationship between the tourism
industry and economic development. The economic industry plays a role
on an international and national scale in which tourism industry is involved.
Over times, the development of the tourism industry has become a form of
global-local nature and processed development. Over the past 3 decades the
relationship between tourism and the local economy has evolved—i.e., from a
top-down effort or completing a business concept to support process of tourism
development. However, complexity manifests in several dimensions. In this
case, shifts in industrial structure and regional development have an effect on
the tourism industry (Milne, 2001).
According to Hansen (2007), the creation of certification in ecotourism
is a tool to increase the legitimacy and visibility of the ecotourism industry in
developing countries. However, the existence of such certification has an impact
on the obligation to achieve international standards that are still difficult to
fulfil by developing countries.
The implementation of the Sustainable Tourism Eco Certification
Programme (STEP), in developing an ecotourism accreditation system, offers
certification options based on self-reports and work within the limitations of the
ecotourism market and a step in the right direction for the future of ecotourism
certification. Thus, this certification programme is a challenge for developing
countries to collaborate with north-south networks in developing ecotourism,
which on the other hand is a form of intervention in tourism management in
developing countries. Global demands are a challenge that must be faced, but
the most important thing is how significant domestic policy regulations are
created to achieve the goals of long-term environmental development.
Duffy (2013) describes the relationship of ecotourism in the dynamics of
politics between international capital, business actors, policies, and corruption.
Emphasized about the state being a shadow player from the political elite
who take advantage of the ecotourism industry (Duffy, 2013). There has
been a development of the pattern of community relations into the form of a
network which is represented in the form of ecotourism development and on
the other hand into the drug trade. The case study that conducted by Duffy in
Belize is a relevant example that illustrates north-south relations in a global
network, which shows how banking, medicine and environmental protections
are regulated. Policies on tourism and ecotourism do not work effectively due
to the pressures from the interests of government regulations, particularly
when domestic corruption strengthens and controls all government policies.
Moreover, the production of liquor, drugs, and money laundering cases still
linger.
Using the Marxist/structuralist perspective, Bianchi (2002) explains how
the relationship between core and periphery in global tourism. He further
explained about how developing countries are involved in debt, with the aim
of developing the tourism sector as a form of its strength. He also stated that
the potential of Asian countries in the tourism development sector is quite
large. Tourism is more visible because of the power relations between capitalist
countries and their former colonies. Based on a strong analysis with neo-
classical dependency model approach, he shows a hierarchical relationship in
the management of the tourism industry from local, regional, national, global,
to transnational levels that involve many interests, which requires institutional
changes that shift the power structure (Bianchi, 2003).

Island Communities
Walpole (2000) sees that ecotourism has a significant impact on the
economic development of a country. He conducted research in eastern Indonesia
and saw the development of Ecotourism in Indonesia as a form of support for
labour distribution, local community development. However, there is inequality
between regional and national areas as well as unbalanced core-periphery
relations. To support job opportunities, one of the cooperation policies is
carried out.
Murianto (2019) analyses the development of Bonjeruk village as a nature
and culture-based tourism village in Central Lombok based on the management
function. In carrying out the management function, there are stages that
include: planning, organising, actuating, and controlling (Murianto, 2019).
First, in the planning stage, Bonjeruk Village as a new tourist destination must
prepare all needs such as public facilities, infrastructure, and human resources
(HR) in the tourism sector. Following the programme planning, obstacles that
arise from both the external and internal environment need to be identified.
Second, in the organising stage, a strong organisation needs to be developed
in order to contribute to the Bonjeruk Tourism Village. Third, in the actuating
stage, the organisation in Bonjeruk must be able to actively work and be run
according to the respective divisions of labour within the organisation. Fourth,
in the controlling stage, the Bonjeruk Tourism Village development programme
must be supervised and monitored on an ongoing basis so that resources and
funds can be used effectively and efficiently (Murianto, 2019). The four points
explain step by step the management of the Bonjeruk Tourism Village in order
to achieve what is expected.
While a system development was analysed through the principles of
Community Based Tourism (CBT) as described by Prasiasa (2010) and Subra,
et al. (2006), Stichting Nederlandse Vrijwilligers (SNV)2 established four basic
principles of Community Based on Tourism (CBT)—i.e., sustainable economy,
ecological sustainability, applicable institutions, and justice for the entire
community—in order to create a mechanism that can keep the economy healthy
and reliable to achieve community welfare (Murianto, 2019). Murianto also
identifies natural and cultural tourist attractions in Bonjeruk Village, ranging
from tourist areas to traditional music. He further examines the Bonjeruk
Tourism Village development model that uses a promotional strategy, namely
the promotion mix, target visitors, and various other indications along with
2 NGOs from Netherlands which is focused on financial support in village.
studies of social, cultural, and economic developments on the progress of
Bonjeruk Tourism Village.
In his article, Murianto also discusses specifically how the process of
the establishment of the Bonjeruk Tourism Village, the performance of the
organisation that plays a role in it and the potential of Bonjeruk village as a
successful tourism village. Permadi, et.al (2020) explained the efforts of the
Bonjeruk Tourism Village manager through the Tourism Awareness Group3
which is named Pokdarwis Bonjeruk Permai in supporting community-based
tourism village development activities. This community group was then split
into two because of the initiative of several members of the Pokdarwis Bonjeruk
Permai who left and formed a new tourism awareness group with support from
the descendants of Lalu Wirasaid bin Raden Nune Umas. In commemoration
of Lalu Wirasaid, this tour group was named Wirajaya Putra Jonggat (WPJ).
Each Pokdarwis still has problems in managing the Bonjeruk Tourism Village.
WPJ Pokdarwis only has 12 members remaining from the initial 25 members.
This reduction in the number of members was caused by the majority of the
members having other main jobs and thus overriding their duties as Pokdarwis
members. In addition, WPJ which focuses on historical tourism has not been
able to develop tourism management properly. Meanwhile, Pokdarwis Bonjeruk
Permai continues to voice promotions, which has resulted in the inability of
Pokdarwis Bonjeruk Permai to provide tourist spots or infrastructure to provide
comfort for tourists (Permadi, 2020)
Internal problems that obstruct the management of Bonjeruk Tourism
Village do not come only from Pokdarwis, but also from the community itself,
where there are still many local residents who still do not understand the
meaning of a tourism village. Hence, the community has not actively participated
in empowerment activities. In addition, local communities play an important
role in making decisions, influencing, and providing benefits to life and the
environment. Based on the complexity of the problems, Permadi et al. (2020)
carried out community service which was then realised through improving
the performance of the Pokdarwis organisation. They explain in detail the
methods of activities and the results of the training activities of the Pokdarwis
organisation in the Bonjeruk Tourism Village.
3 In Bahasa it is called Pokdarwis, which means Kelompok Sadar Wisata

Island Communities
CONCEPTUAL FRAMEWORKS
There are several concepts that we use in this analysis. First, we use Triple
Bottom Line Concept in Ecotourism Development. This concept is important to
understand how important ecotourism in development agenda. Second, we use
relationship network among actors during the development process and how
they are connected together.
Understanding Triple Bottom Line Concept in Ecotourism

Development
Along with the development of the tourism sector in Indonesia, many regions
are increasingly showing the tourism potential that exists in their area. With this
development, regional income is increasing. However, they should not only get
profits, but must also pay attention to the surrounding environments, namely
in terms of society and the natural environment. Several aspects that need to
be considered—i.e., profit, people, and planet (3P)—are elements in the Triple
Bottom Line (TBL) concept. This concept has been widely used in Indonesia
since 2007 along with the development of full cost accounting which is widely
used by companies in the public sector (Azis et al., 2010; Felicia, 2014).
The concept of the Triple Bottom Line was first introduced by Elkington
in 1994 in his book entitled Cannibals with Forks. He explains the Triple
Bottom Line that covers economic prosperity, environmental quality, and social
justice. The Triple Bottom Line concept implies that companies prioritise the
interests of stakeholders (all parties involved and affected by the activities of
a company) rather than the interests of shareholders. The interests of these
stakeholders include profit, people (society), and planet (natural environment)
(Felicia, 2014). Profit in this concept is the advantages obtained from Bonjeruk
ecotourism. This advantage is not only in the form of money but can also be
in the form of awards so that the local ecotourism becomes famous. People
are defined as a group of people who run a business or company, or people
who work in a company or business. In the concept of ecotourism, it means
people who are involved in the development of ecotourism, such as the Central
Lombok Tourism Ambassador in Bonjeruk Village who is assigned to promote
ecotourism through social media, those who are involved in organizing events,
and those who sell unique or distinctive items from the community. Planet
means environmental management. It means whether the ecotourism has a
good influence or not on the surrounding environment, such as producers who
make souvenirs or goods or regional specialties whether or not it will cause
waste that is safe or not for the natural environment.
Understanding Relationship Network Between Actors

Every development process must have actors involved in it. Actors are individuals
and groups who have influence and play a key role in the development issues
and problems in society (Magdalena et al., 2013; Yunindyawati et al., 2022).
In this context, community-based ecotourism development in Bonjeruk
Village. Yunindyawati et al. (2022) divides actors into three types: individuals,
organisations, and institutions. Individual actors are those who have an
influence and a significant role in the development of ecotourism in Bonjeruk
Village, including businesspeople, religious leaders, and community leaders.
Organisational actors are a group of individuals who have the same goals
in developing ecotourism in Bonjeruk Village including Kelompok Sadar Wisata
(Pokdarwis), Karang Taruna, Bonjeruk Organik (Bonjor) and Reduce, Reuse,
Recycle Waste (Tempat Pengolahan Sampah/TPS3R). Meanwhile, institutional
actors are formal and non-formal organisation that help realise ecotourism
development in the Bonjeruk Tourism Village, including village government,
educational institutions, and private companies. These three types of actors
interact with each other according to their interests and goals in developing
ecotourism in the Bonjeruk Tourism Village. The types of actors in the
development of the Bonjeruk Tourism Village can be seen in Table 1.
Development of ecotourism in Bonjeruk Village can be achieved if the
interaction between the actors involved in it goes well. This interaction
can be seen from how the network of relationships between existing actors.
Sociologically, the network of relations between actors is influenced by three
factors including power, interest, and sentiment (Yunindyawati et al., 2022).
First, power is closely related to power and the distribution of power owned by
each actor involved in the development of the Bonjeruk Tourism Village.

Island Communities
Table 1. Actors in the Development of Bonjeruk Tourism Village
No Actors Type
1 Individuals Society, entrepreneurs, religious leaders, community leaders
2 Organisation Pokdarwis Bonjeruk Permai, Karang Taruna, Bonjor, TPS3R
3 Institutions Local government, NGO, educational institutions, private
companies
Source: Author analysis based on interview results on 6 November 2021
Second, interest relates to the goals and targets to be achieved by each
actor. Third, sentiment relates to views and feelings towards each other’s
existence among actors. These three factors can be used as variables to measure
how close the network of relationships between actors in Bonjeruk Tourism
Village is. The more variables that are aligned between existing actors, the more
cooperative the interaction is and vice versa.
USE OF SECONDARY AND PRIMARY DATA

In this chapter, we use descriptive qualitative research methods to explain how
the interaction between the government, entrepreneurs, and non-governmental
organisation in local economic development in Bonjeruk Village. In order
to support this research, we use primary and secondary data. Primary data
were obtained through observations and interviews with several informants
in Bonjeruk Village, while secondary data were obtained through literature
studies of more than 15 articles from journal and books. The keywords that
we are looking for are ecotourism, interaction among actors, and tourism
development.
On the primary data, the sampling method is purposive sampling by choosing
the main actors involved in Bonjeruk Ecotourism Village. The respondents are
Local Government, NGOs, and business actors. During the field research we
interviewed 3 persons or each person for one category. Our observations aim
to see how far the preparations that have been made by the Bonjeruk Tourism
Village developer considering the potential they have is good enough to become
a tourism village. In addition, we also conducted interviews to find out the
extent of the interaction between the government, entrepreneurs, and non-
governmental organisation in local economic development in Bonjeruk Village.
The data that we collect are qualitative data and transcript of the interview.
BONJERUK TOURISM VILLAGE DEVELOPMENT
The development of the Bonjeruk Tourism Village is currently carried out
with community-based management (community-based tourism). Community
Based Tourism (CBT) is tourism that involves community participation
and aims to bring benefits to local communities by allowing tourists to visit
local communities and learn about their culture and recognize the natural
surroundings (Permadi and Dkk, 2020). The 4 principles of CBT include
sustainable economy, ecological sustainability, unified institutions, and fair
distribution of costs and benefits to all community members.
Based on information from the source, SL, who is one of the entrepreneurs
in Canteen 21, Dusun Batu Jering, entrepreneurs in Canteen 21 only need to pay
a fee to Pokdarwis Bonjeruk Permai 10% of the total net profit received. This is
very beneficial for the entrepreneur because the 10% contribution includes the
rent for the place and various facilities at Canteen 21 (SL, 2021). The fee from
the entrepreneur is then managed by Pokdarwis for various activities in the
development of the Bonjeruk Tourism Village. Community-based management
applied in Bonjeruk Tourism Village fulfils the Triple Bottom Line concept that
we use in this chapter—i.e., society (people), the environment (planet) and
economic profit (profit).
In addition, all interviewees agreed that the development of Bonjeruk as
a tourism village had a positive impact on local communities, especially in the
economic aspect. Interestingly, each resource person gave an opinion about
the positive impact of developing a tourism village in accordance with their
respective fields. According to the informant SL, the development of Bonjeruk as
a tourism village had a positive impact on the economy of the local community.
People who previously had difficulty finding a place to sell local products such
as fruits, vegetables and livestock have been helped by being able to sell their
products in tourism villages at even higher prices.
Activities that were previously part of the community’s routine have also
become economically valuable after the existence of a tourism village. For
example, the community will get contribution money from a walking package
that shows the process of roasting coffee in the village. Resource person (SL)
added that now the villagers no longer need to go to the market to sell their
products because the market has been moved to Bonjeruk Village. Another
resource person (LE), as Tourism Ambassador for Central Lombok Regency,

Island Communities
admitted that the development of a tourism village made the village where he
was born better known to the wider community and even to foreign countries
(LE, 2021).
Meanwhile, the Bonjeruk Village Head considered that the development
of a tourism village had a positive impact on the mental development of the
village community. The development of tourism villages spurs the community
to innovate and adapt to the tourism village environment. Although, currently
the majority of people only make tourism activities as a side job from their main
livelihood.
Interaction between government, businesspeople, and non-governmental
organisation in the development of Bonjeruk Tourism Village. Various parties
including community elements have their respective roles in the development
of the Bonjeruk Tourism Village. Based on interviews with resource persons, the
roles of each party in the development of the Bonjeruk Tourism Village include:
Government
The government, especially the village government, is the party most
responsible for preparing various infrastructure and other needs that support
the sustainability of the Bonjeruk Tourism Village. So far, according to source SL,
the Bonjeruk Village Government has built various supporting infrastructures
such as road access and electricity. The village government also often holds
various trainings such as sewing and other self-development trainings.
According to the village head, his party continues to support the development
of the Bonjeruk Tourism Village. One of the efforts made by the government
is through the Village-Owned Enterprises (BUMDes). Through BUMDes, the
village government offers capital loan assistance to Micro, Small, and Medium
Enterprise (MSME) owners in Bonjeruk Village (Rahman, 2021).
In 2019, the village government also budgeted a village fund of IDR 86
million to Pokdarwis Bonjeruk Permai to develop a culinary stall at the Bamboo
Market tourist attraction.4 This is done because the government wants to
embrace business actors, especially in the culinary field who do not yet have a
place for their business. The traveling merchants no longer need to be tired of
selling through the village because they are facilitated by the village government
and Pokdarwis selling places. Entrepreneurs also do not need to think about
4 Based on interviews that we conducted in 2021.
the price of renting a place because it is included in the 10% fee as previously
explained. However, from the statement of the village head, the government
is preparing rules, so that the village government can also get income from
tourism activities in Bonjeruk Village.
Various government institutions at the district, provincial and even national
levels and educational institutions also provide support in the development
of the Bonjeruk Tourism Village (Rahman, 2021). According to informant SL,
Kantin 21 tourism object received assistance in the form of proper tables and
chairs from the Manpower Office of Central Lombok Regency (Rahman, 2021).
In addition, the District Manpower Office and the Public Works Office also
provided bicycle assistance to support bicycling packages of 6 and 12 bicycles,
respectively. In addition, one of the MSMEs in Ombak Hamlet, Ombak Food,
received capital assistance of 100 million rupiah from the Ministry of Tourism
and Creative Economy (LE, 2021).
According to the Head of Bonjeruk Village, the coordination between
the village government and the Provincial Tourism Office is also quite good.
However, the village government considers that the Regency Tourism Office
has not paid enough attention to the locals. This was confirmed by resource
person who is Tourism Ambassadors of Central Lombok Regency. Currently as
the Tourism Ambassador of Central Lombok Regency in Bonjeruk Village, he
is still working alone and has not been sufficiently supported by the Regency
Tourism Office (LE, 2021).
Meanwhile, government-owned educational institutions also participated
in the development of the Bonjeruk Tourism Village. Lombok Tourism Polytechnic
(Politeknik Pariwisata/Poltekpar) in 2020 held the FAMOUS (Familiarisation
Trip of Sustainable) event which lasted for 3 days (Disbudpar, 2021). This event
aims to promote various tourism objects on the island of Lombok, especially
those in Central Lombok Regency. Poltekpar invites influencers and various
parties involved in promotional and tourism activities from all over Indonesia.
Entrepreneur
MSME actors have a particularly prominent position in the development
of the Bonjeruk Tourism Village. They become a support for the running of the
economy of the surrounding community. According to source SL, entrepreneurs
in Bonjeruk Village mostly use raw materials for their business from the
surrounding community. For example, in one of the culinary packages at Canteen
21 called begibung. Almost all of the raw materials in this group culinary

Island Communities
package, such as free-range chicken, vegetables and fruits, are purchased from
the local Bonjeruk community (SL, 2021). In addition, entrepreneurs also have
responsibilities that are no less important than just driving the local economy.
They are also responsible for maintaining and promoting traditional products
typical of Bonjeruk to tourists.
On the other hand, entrepreneurs must also continue to innovate so that
consumers can accept these traditional products. Innovating without losing
the characteristics of traditional products is a challenge that is not easy for
entrepreneurs (Rahman, 2021). According to LE resource person, one of the
MSMEs in Bonjeruk Village received assistance of IDR100 million from the
Ministry of Tourism and Creative Economy because of its innovative traditional
culinary products. UD. Ombak Bersatu or better known as Ombak Food produces
traditional foods that have been modified to attract consumers. Some of the
products from Ombak Food include Jaje Yeast, Moringa Leaf Sticks, Fish Thorn
Sticks, Dragon Fruit Sticks, Plecing Sticks, Rolled Ginger and many more (Olivia,
2021). UD Ombak Bersatu has also been registered on the website (https://
wisatatangguh.kemenparekraf.go.id/objek-detil.php?id=22065) as one of the
MSMEs included in the Geospatial Tourism and Creative Economy.
Not only MSMEs, Bonjeruk Tourism Village often gets assistance from
large companies as a form of Corporate Social Responsibility (CSR). According
to the village head, Bonjeruk is one of the assisted villages of PT. Indonesia
AirAsia, although recently the company’s support has declined in Bonjeruk.
According to a source from SL, Kantin 21 tourism object received donations for
the construction of various facilities including an open hall from an overseas
travel agency company. Bonjeruk Tourism Village is also supported by several
local and foreign travel agents including; Wise Steps Travel and Shoesstring
(Bonjeruk, 2020). In addition to providing support in the form of financial
donations, these travel agency companies also make Bonjeruk Tourism Village
one of the destinations in their tour packages. That way there is a mutually
beneficial interaction between the travel agency company and the Bonjeruk
Tourism Village.
Non-Governmental Organisation
Since the beginning of its development, Bonjeruk Tourism Village is the result of
collaboration between the government and NGOs. The village government and
Pokdarwis Bonjeruk Permai were among the first to realise the tourism potential
of their village. According to the Head of Bonjeruk Village, the success that has
so far been achieved by Bonjeruk cannot be separated from the hard work of
Pokdarwis and other NGOs in Bonjeruk Village (Rahman, 2021). The village
government claims to be only the provider of moral and material support as well
as a facilitator in uniting NGOs and the community to work together to build the
Bonjeruk Tourism Village. Meanwhile, the management of the tourism village
is entirely done by Pokdarwis Bonjeruk Permai and other NGOs. According to
source SL, currently there are at least three NGOs that are collaborating with
each other in the development of the Bonjeruk Tourism Village, namely, Waste
Processing Sites-Reduce, Reuse, Recycle (TPS3R), Organic Bonjeruk and Youth
Organisation. Each NGO is particularly important in accordance with their
respective functions and duties. For example, TPS3R which is engaged in waste
management is responsible for maintaining the cleanliness of the Bonjeruk
Tourism Village. A clean environment will provide comfort for tourists.
In addition to collaboration between NGOs, the development of the
Bonjeruk Tourism Village is also supported by non-governmental organisation
outside Bonjeruk. According to LE resource person, on 25-26 August 2021, the
Indonesian Ecotourism Network (INDOCON) sponsored by the International
Labour Organisation (ILO-UN) held training on the development of tourism
products to the surrounding community in Bonjeruk Village. The Mataram
Tourism College (Sekolah Tinggi Pariwisata/STP), which is under the auspices
of the Kertya Wisata Mataram Foundation, also provided equipment needed
for several events in Bonjeruk Village. Scientific journals are a means for STP
academics to share their research results on tourism, especially on the island of
Lombok. For example, Hospitality Scientific Journal volume 8 number 1 presents
an article entitled ‘Bonjeruk Village as a Nature and Culture-Based Tourism
Village in Central Lombok’, which discusses about development as well as input
and suggestions to the development of Bonjeruk Tourism Village. Various moral
and material supports from non-governmental organisation certainly encourage
the acceleration of the successful development of the Bonjeruk Tourism Village.
Network of Relationships between Actors
In the development of the Bonjeruk Tourism Village, each actor plays a role
to interact with each other and form a network of relationships that can be
analysed using the concept of a network of relationships between actors in a
sociological approach. This concept explains that the network of relationships
between actors can be influenced by three factors including power, interest, and
sentiment. We make these three factors as variables to analyse the interactions
between actors in the development of the Bonjeruk Tourism Village.

Island Communities
Regarding the network of relationships between actors in Bonjeruk
Tourism Village based on power among the three types of actors, there are 5
positive (+) power relationship networks, and 1 negative (-) power relationship
network as follows (Table 2):
1. Power relations between individual actors are going well (+):
The development of the Bonjeruk Tourism Village involves collaboration
between the surrounding community, the majority of whom work as
farmers as producers and the community who work as entrepreneurs and
MSMEs as distributors. Most of the ingredients of the products that are
managed and marketed in the Bonjeruk Tourism Village come from the
gardens and farms of the surrounding community (SL, 2021).
2. Power relations between individual and organisational actors are going
well (+):
Organisational actors get support from individual actors in managing the
Bonjeruk Tourism Village. Pokdarwis Bonjeruk Permai as the manager
of the tourism village provides business premises facilities with a profit-
sharing rental system to entrepreneurs. They only need to pay a rental fee
10% of the total net profit each month. Pokdarwis Bonjeruk Permai has
also organised various community development activities such as forming
an English language learning community, training, and so on.
3. Power relations between individual actors and institutions are running
well (+):
The government of Bonjeruk Village often organises trainings that can
improve the capacity of the community. Likewise, the community plays an
active role in Clean Friday mutual cooperation activities held by the village
government (Rahman, 2021).
4. Power relations between organisational actors and the organisation are
going well (+):
The types of organisational actors in the Bonjeruk Tourism Village
collaborate with each other according to their respective abilities and
fields (LE, 2021).
5. Power relations between organisational actors and institutional actors are
going well (+):
Organisational actor actively consults with the village government
(institutional actors) and assists the village government in carrying out the
work programmes related to the development of tourism villages. Village
government also often includes several types of organisational actors on
various occasions (Rahman, 2021).
6. Power relations between types of institutional actors are not going well (-):
The village government admits that it is still difficult to manage village-
owned enterprises (BUMDes), while it should be able to support the
development of tourism villages due to the slow resolution of the BUMDes
problem after the transition from the leadership of the Bonjeruk Village
Head (Rahman, 2021). In addition, communication between the Bonjeruk
Village Government and the Central Lombok Regency Government was not
running smoothly.
Table 2. Network of Relationship between Actors in Bonjeruk Tourism

Village based on Power
No. Relationship Type and Status Relations between Actors
1. Power (+) individuals-individuals
2. Power (+) individuals-organisation
3. Power (+) individuals-institutions
4. Power (+) organisation-organisation
5. Power (+) organisation-institutions
6. Power (-) institutions-institutions
Source: Author analysis based on interview results on 6 November 2022.
As shown in Table 3 regarding the network of relationships between actors
in Bonjeruk Tourism Village based on the interests between the three types of
actors, there are 5 positive (+) power relationship networks and 1 negative (-)
power relationship as follows:
1. The interest relationship between individual actors and individuals goes
well (+):
The Bonjeruk Tourism Village Community both agreed to develop the
village’s potentials through the development of a tourism village (SL,
2021).
2. The interest relationship between individual actors and the organisation is
going well (+):
The people who work as entrepreneurs and the Bonjeruk Permai Tourism
Awareness Group both want to increase the village’s economy through
the development of the Bonjeruk Tourism Village. The Bonjeruk Permai

Island Communities
Tourism Awareness Group as the manager is also willing to preserve the
culture desired by the community leaders of Bonjeruk Tourism Village
(Rahman, 2021).
3. Between types of individual and institutional actors is going well. The
community and village government want progress in various sectors
through the development of the Bonjeruk Tourism Village.
4. Interest relations between types of organisational actors and the
organisation is not going well. Until now, the Bonjeruk Tourism Village
still has two Pokdarwis, each of which manages several hamlets in the
Bonjeruk Tourism Village. Pokdarwis Bonjeruk Permai and Pokdarwis
Wirajaya Putra Jonggat have different visions in managing tourism in
Bonjeruk Village. This has hampered collaboration between the two types
of organisational actors(LE, 2021).
5. Interest relations between types of organisational actors and types of
institutional actors is going well. Pokdarwis and the Government of
the Bonjeruk Tourism Village have the same interest in developing the
Bonjeruk Tourism Village
6. Interest relations between types of institutional actors and types of
institutional actors is going well. Even though the power relationship
between types of institutional actors is not good in terms of interest, the
Central Lombok Regency Tourism Office wants regional tourism progress
in line with the Bonjeruk Tourism Village Government which wants the
development of the Bonjeruk Tourism Village.
Table 3. Network between Stakeholders of Bonjeruk Tourism Village
based on Interests
No. Relationship Type and Status Relations between Stakeholders
1. sentiments (+) individuals-individuals
2. sentiments (+) individual-organisations
3. sentiments (+) individual-institutions
4. sentiments (-) organisations-organisations
5. sentiments (+) organisations-institutions
6. sentiments (-) institutions-institutions
Source: Author analysis based on interview results on November 6, 2021
Challenges in the Development of Bonjeruk Tourism Village
Every business must have its own challenges, so must the development of the
Bonjeruk Tourism Village have. These challenges come from the community,
non-governmental organisation, and governments. According to the informant
SL, the people of Bonjeruk Village are religious and still hold tightly to their
regional cultural values. They initially rejected the tourism village development
plan because it was considered to be a threat to their religious and cultural values.
Pokdarwis Bonjeruk Permai as the initiator of the development of the Bonjeruk
Tourism Village took an approach by involving religious leaders and community
leaders in tourism village development activities. Pokdarwis chooses mentors
from among community leaders to oversee the activities of their organisation.
This method was also quite successful in attracting the sympathy of the village
community who gradually began to accept the development of the Bonjeruk
Tourism Village.
It does not stop there, it is exceedingly difficult to change the thinking of
the people who mostly work as farmers to switch to the tourism sector. The
community considers that the agricultural sector is more promising than
tourism so that their participation in the development of the Bonjeruk Tourism
Village is still incredibly low. Another challenge is that there are currently two
Pokdarwis in Bonjeruk Village.
In 2018, there was actually only one Pokdarwis, but several members
decided to separate because they had different views on the development of
the Bonjeruk Tourism Village. The village head of Bonjeruk admitted that he
was continuously trying to reunite Pokdarwis Bonjeruk Permai with Pokdarwis
Wirajaya Putra Jonggat. According to him, the split between Pokdarwis only
hinders development in the Bonjeruk Tourism Village. The challenge that is no
less influential in the development of the Bonjeruk Tourism Village is the lack of
coordination between the village government and the Tourism Office of Central
Lombok Regency. The village government complained that the district tourism
office was less responsive in assisting the development of the Bonjeruk Tourism
Village.

Island Communities
CONCLUSION
Bonjeruk Tourism Village is one of sample for Indonesian Tourism Village
which has various potentials from traditional to modern attraction. However,
during the process itself, the main challenge is from internal conflict among
actors. The research finds that interaction among actor has several forms
such as sentiment positive and negative. The interaction Subsequently, the
interaction between the government, entrepreneurs, and non-government
organisation in developing the local economy with the concept of ecotourism
in Bonjeruk Village is quite intense. Each party carries out their respective
duties and functions in the development of the Bonjeruk Tourism Village. The
village government is providing infrastructure and other supporting facilities.
Likewise, entrepreneurs, especially SMEs, are starting to adapt and continue
to innovate in their businesses. Large companies such as PT. Indonesia AirAsia
also fulfil their social and environmental obligations (CSR) by providing various
forms of support for the development of the Bonjeruk Tourism Village. Non-
governmental organisation are less active, various forms of training in the field
of tourism are provided to the local community and businesspeople as a form of
their participation in the development of the Bonjeruk Tourism Village.
In addition, stakeholders’ (government, business actors, and NGOs) lack
of awareness and different interest become challenge to improve Bonjeruk
development. Some actors have sentiment with other actors. Leadership
is particularly important to solve the problem. The community-based
development of the Bonjeruk Tourism Village is also quite sustainable based
on the Triple Bottom Line concept. Pokdarwis Bonjeruk Permai as the manager
of the Bonjeruk Tourism Village has been able to create a beneficial mechanism
for each member of the Bonjeruk Tourism Village through a contribution
programme and the use of local commodities from the community.
The management’s commitment to maintaining cultural and religious
values in the community is clearly applied, as they involve community leaders
in the development of the Bonjeruk Tourism Village. Every tourist who visits
the Bonjeruk Tourism Village will also be required to respect the customs and
traditions of the village community. Lastly, environmental sustainability is
important in the development of a sustainable tourism village. Bonjeruk Tourism
Village has a non-governmental organisation TPS3R which is engaged in waste
management. They are tasked with maintaining the cleanliness of the Bonjeruk
Tourism Village environment by managing waste from each tourist attraction
and village community. Based on the explanation above, the development of
the Bonjeruk Tourism Village is quite sustainable by meeting the criteria of the
Triple Bottom Line concept, namely profit, people and planet.
REFERENCES
Aminoto, J. (2019). “Ecotourism as a tourism development strategy from an
Islamic economic perspective.” S-1 Thesis. IAIN Purwokerto.
Balaam, D. N. (2011). Migration and Tourism : People on the Move. In David
N. Balaam & Bradford Dilman (eds). Introduction to International Political
Economy. Boston: Pearson.
BI (2018). Bank Indonesia. Retrieved from Bank Indonesia: https://www.
bi.go.id/id/publikasi/E-Magazine/Documents/GeraiInfo-73_Mendulang-
Devisa-Melalui-Pariwisata.pdf
Bianchi, R. V. (2003). Towards New Political Economy of Global Tourism. In R.
a. Sharpley (ed.). Tourism & Development: Concepts and Issues, 265-299.
Dain Bolwell, W. W. (2008). Reducing poverty through tourism. Geneva: ILO.
Damanik, J. (2014). Kebijakan Bisnis Ekowisata di Indonesia : Menyandingkan
Konsep dengan Praktek. In Pembangunan Pariwisata Berkelanjutan
: Peningkatan Kualitas Hidup Masyarakat melalui Ekowisata. Jakarta:
Kementerian Pariwisata dan Ekonomi Kreatif, 13-14.
Duffy, R. (2013). Shadow Players: Ecotourism Development, Corruption and State
Politics in Belize. Carpax Publishing, Taylor & Francais Group.
Ecotourism (2019). Ecotourism Is The Solution To Overtourism. Retrieved
from https://ecotourism.org/news/ecotourism-is-the-solution-to-
overtourism/
Gal Oberhofer Dimitrijević, D. S. (2022).”Can Ecotourism Offer Sustainable
Development:Why Costa Rica Succeeded.” IBIS - International Conference
on Business and Integral Security. Swiss.
Hogh, W. G. (2003). Endangered Visitors: A Phenomenological Study if Eco-
Resort Development. In Michael Lück & Torsten Kirstges (eds.). Global
Ecotourism Policies and Case Studies : Perspectives and Constraints. London:
Channel View Publication, 82-83.

Island Communities
ILO. (2022). The future of work in the tourism sector: Sustainable and safe
recovery and decent work in the context of the COVID-19 pandemic. Geneva:
ILO.
Milne, S. (2001).’ Tourism, Economic Development and The Global-Local Nexus :
Theory Embracing Complexity.’ Tourism Geographies 3(4): 369-393.
Mowforth, M.A. (2015). Tourism and Sustainability: Development, Globalisation
and New Tourism in the Third World. Abingdon: Routledge.
Murianto (2019). ‘Desa Bonjeruk Sebagai Desa Wisata Berbasis Alam dan
Budaya di Lombok Tengah.’ Hospitality, viii(1): 21.
Parekraf (2014). Siaran Pers Menteri Pariwisata dan Ekonomi Kreatif :
Masyarakat Ekonomi ASEAN Berpotensi dorong pertumbuhan wisatawan
ke Indonesia di atas 10%. Available at: http://www.parekraf.go.id/asp/
detil.asp?id=2555. Accessed on 14 October 2021.
Permadi (2020). Peningkatan Kinerja Organisasi Kelompok Sadar Wisata di
Desa Bonjeruk, Kecamatan Jonggat, Lombok Tengah. Jurnal PEPADU, 1(1).
Rukendi, C. (2015). Tourism Policy Research and Development. (M. N. Makhroja,
Interviewer)
Travelling, G. (2015). Green Hotel. Available at: http://greentraveling.weebly.
com/definition.html
UNWTO. (2023). UNWTO. (May).Available at: https://www.unwto.org/tourism-
in-2030-agenda
Usman. (2021). Bonjeruk Ecotourism Village. (L. A. Atmaja, Interviewer)
Ustman. (2021, 11 10). Chief of Pokdarwis Bonjeruk Permai. (L. Agung,
Interviewer)
Vaugeois, N. (2015). Tourism in Developing Countries: Refining a Useful Tool for
Economic Developmen. Available at: http://fama2.us.es:8080/turismo/
turismonet1/economia%20del%20turismo/economia%20del%20
turismo/tourism%20and%20developing%20countries.pdf. Accessed on
15 October 2020.
Wood, M. E. (2002). Ecotourism: principles, practices and policies for sustainability.
UNEP Publication
CHAPTER 10
Land Property Rights, Income,

and Food Security: Insights from
Ambon Island, Indonesia
Wardis Girsang, Ali Magana, Adelina Siregar,
L. Elia Madubun,, and L.O. Kakisina
INTRODUCTION
Rice food security is crucial for nearly half of the world’s 7.5 billion people. The
food crops area in Asia is estimated at a total of 115 million hectares (Weiss,
2009). This land can supply about 90% of rice in Asia and 45% of calories to 560
million poor people in the Pacific (Mohanty, 2013). As the world population is
projected to increase to around 9 billion in 2050, rice consumption will increase
by about 60%. Therefore, in the long-term rice yield growth in Asia should be
more than 3% per ha per year (Dowling and Greenfield, 1998). However, food
security is vulnerable because food crop production is projected to drop by 25%
by 2050 due to land conversion and climate change impacts (Rosegrant et al.,
2014). This was exacerbated by the COVID-19 pandemic since 2020 and global
economic recession, food insecurity, and low economic growth issues in 2023.
In Indonesia, rice land is about 7.5 million hectares to produce about 35
million tons of rice to support 280 million people. Rice consumption tends
to increase because of population growth. However, the area of fertile rice
land, particularly in Java that is home to 60% of the total rice production, has
declined significantly because of urban development and land conversion,
205
which was estimated at 80,000-100,000 hectares per year (Suswono, 2011).
As an alternative, central governments have accelerated to opening about
179,000 ha of new paddy fields per year, but 84% of this is in Eastern Indonesia
(Sudaryanto, 2003).
Different from Java and other main islands in Indonesia, rice extensification
in Eastern Indonesia is argued to be a rice bias policy to achieve food-self-
sufficiency and farmer’s prosperity. However, it has influenced the decision to
convert forest land to paddy fields. The rice bias policy in Maluku islands has
conversed about 100,000 hectares of forest and sago palm into irrigated rice
fields, plantations, and resettlements (Girsang, 2014b). This paddy field has
been argued to increase the usage of high chemical fertilizers that, in turn, has
caused toxicity not only to soil, crops, sea, and fish but also to food safety and
human health (Resosudarmo and Thorbecke, 1998; Resosudarmo, 2008).
A further issue in the small and remote islands of Maluku is the weak land
property right, particularly primary forest land. Supriatna et al. (2022) argued
that less secure land property rights would make it easy to attract private
companies to buy land and convert forest land for food crops and plantations,
particularly palm oil and cocoa plantations (Ducos, 2014). The combination
of rice bias policy implementation and plantation crop development has been
argued to accelerate deforestation in Maluku. Consequently, deforestation
has increased from 1.2% in the 1990s (Matinahoru, 2014) to 7.7% in the
conservation forests, 12.4% in the protected forests, and up to 41.8% in the
production forests (Kastanya, 2019).
Unlike other islands such as Java and Sumatra, Maluku regions are
considered small islands. In Indonesia, small islands are smaller than 2,000
km2 (Retraubun, 2013), short and small watersheds area, hilly topography, and
limited flat land area. Thus, the prime food crop is not rice but multiple tree
food crops such as sago palm, breadfruit, banana and tubers, spices, and edible
fruits. It occurs because small islands are also geographically isolated (remote),
peripheral, dependent, based on narrow production, and insufficiently
economic diversified (Legarda, 1984). Legarda (1984) further states that small
islands are characterised by diseconomies of scale (smallness), high transport
costs and delivery time, limited harbours, and airports for large carriers of
modern tourists and trade, high per capita costs, limited skilled workforce,
weak bargaining position towards investors, and vulnerable to fluctuation of
the global economy market.
The bigger islands in Maluku, such as Seram and Buru, also have small
island characteristics—i.e., hilly topography and short and direct watershed
from the forest to the coastal area. That means deforestation in the upstream
will cause environmental degradation in the downstream, particularly in
rice fields and food crops. Based on previous observation, illegal logging and
deforestation in the upstream area will cause floods during heavy rain. That
will destroy the irrigation water dam and divert the river flow direction. Then,
floods, soil erosion, river sedimentation, and soil, stone, and sand will come
into the paddy field through irrigation canals. In the dry season, the river dries
up, which results in dry irrigation canal dam. Therefore, most farmers cannot
plant rice because of the water shortage. Field observation showed that farmers
changed primary commodities from rice to horticulture crops. That situation
will cause food insecurity because of declining paddy production and rice food
unavailability.
In the case of Maluku, however, evidence for such argument is rare,
especially in island areas where local people do not grow paddy rice but sago
palm, breadfruits, and tubers. To fill out this gap, the main goal of this research
is to observe whether secure land property rights influence food security in the
case of Maluku. The main hypothesis is that secure land property rights are a
significant factor in determining household food security. The benefit of this
chapter is to promote a sustainable model of land property rights and food
security relationships in small islands amidst the global food and fuel crisis
threat.
Previous studies, such as by Chheng and Resosudarmo (2021), found that
secure land property rights and land conversion will strengthen food security.
The reason is that secure land property rights can be used as a bank loan
guarantee to get financial capital from Bank with the aim to upgrade technology
to improve soil fertility, buy high-yield seeds, and enhance rice production.
This will improve household income and food security. Furthermore, Kubitza
et al. (2018) stated that secure land property rights would reduce deforestation
and environmental degradation because farmers transform farm habits from
shifting cultivation to settled and good agricultural practices.
Land Property Rights, Income, and Food Security: Insights from Ambon Island, Indonesia 207
THE USE OF PLS-SEM TECHNIQUE
Previous research has examined the relationship between secure land property
rights, land size, and land use change (Supriatna, et al. 2022), as well as the
influence of strong land property rights on food insecurity (Chheng and
Resosudarmo, 2021). This chapter adopted the same idea by using the Partial
Least Square-Structural Equation Modelling (PLS-SEM). Richter et al. (2016)
and Hair et al. (2017) state that PLS-SEM has been used broadly by authors,
particularly in management, business, and marketing, but Chakim et al. (2019)
have applied PLS-SEM in social, and economic agriculture research. Research
variables have developed according to the Maluku islands’ local condition:
agriculture and fisheries are the primary livelihood. These variables are crop
biodiversity, crop and fishery productivity, household income from agriculture,
fisheries, and nonfarm economic activities. Figure 1 depicts the research model
comprising eight latent constructs (variables).
Figure 1 describes the association between land property rights,
productivity, and income to influence food security. The influence of land
property rights on food insecurity was adopted from the work of Chheng and
Resosudarmo (2021) and Supriatna et al. (2022). The role of agriculture and
fisheries incomes in supporting food security was adopted from the previous
study about the Dusun systems in small islands (Girsang, et al., 2019). The main
inputs to food security are land property rights, agriculture income, fishery
income, and nonfarm income. This argument is based on the previous study of
van Oostenbrugge et al. (2004) and Babatunde (2013), who found that fisheries
and nonfarm are necessary conditions for agricultural income because land and
agriculture incomes only are not sufficient to support household food security
in developing countries. The eight variables utilised affect one another. The PLS-
SEM technique allows these variables to estimate coefficients connecting these
complex relations. Three main hypotheses can be developed—i.e., (1) Land
property rights and crop biodiversity influence crops productivity; (2) Land
property rights and crop productivity influence agriculture income; and (3)
Land property right, agriculture income, fishery income, and nonfarm income
influence food security.
Figure 1. Research Model for Land Property Rights, Income, and Food
Security in Small Islands
Table 1 describes variables (latent constructs), dimensions, and indicators

included in the research model. Each variable consists of two or more indicators,
which are scaled using a Likert’s scale type of measure.
Table 1. Research Variables, Dimensions, and Indicators
Constructs Dimensions and Indicators Scale of Indicators
1. Land Property • LPR1: Perception about • 1 to 9, from very weak (1) to
Rights (LPR) the secure level of the land very strong (9), land certificate,
property rights borders, and ownership status
• LPR2: The size of land • 1 to 9, from very small (<0.25) to
(hectare/ household) very large (>2)
• LPR3: The number of land • 1 to 4, low (1) to high (4)
plots per household
2. Crops • CB1: Total of clove trees • 1 to 10, from very low (<10) to
Biodiversity (trees/household) very high (>100)
• CB2: Total of nutmeg trees • 1 to 10, from very low (<10) to
(trees/ household) very high (>100)
• CB3: Total edible fruits (trees/ • 1 to 10, from very low (<10) to
household) very high (>100)
• CB4= Total productive • 1 to 10, from very low (<10) to
plantations very high (>100)
3. Crops • CP1= Production of clove (kg/ • 1 to 6, from very low (<20) to
Productivity household/ year) very high (≥200)
• CP2=Production of nutmeg • 1 to 6, very low (<20) to very high
(kg/household/ year) (≥200)
• CP3=Productivity of clove (kg/ • 1 to 6, from very low (<1) to very
tree) high (≥5)
• CP4=Productivity of nutmeg • 1 to 5, from very low (<1) to very
(kg/tree) high (≥4)
4. Farm Income • Farminc1 = Total income of • 1 to 7, from very low (<5) to very
plantations (IDR million/year) high (>35)
• Farminc2 = Total income of • 1 to 6, from very low (<2) to very
edible fruits (IDR million/year) high (>10)
• Farminc3 = Total income of • 1 to 6, from very low to very high
Farm (Farm and livestock) (IDR per year)
5. Fisheries • FP1= Total fish catch per unit • 1 to 6, from very low (<100) from
Productivity effort (kg/ year) to very high (>1000)
• FP2= Tuna fish production • 1 to 6, from very low (500) to
(head/year) very high (3000)
• FP3= Types of catching gear • 1 to 4, No boat, boat without
technology machine, boat traditional machine
(Katinting), and boat with
modern machine 15 PK (+)
6. Fisheries • Fishinc1=Income from tuna • 1 (very low) to 40 (very high)
Income (IDR million/year)
• Fishinc2=Price of Tuna (IDR/ • 1 to 4, from very low (20000) to
Unit) very high (90000)
7. Non-Farm • NFI1=Income from trade (IDR • 1 to 7, from zero to very high
Income million/year) (≥30)
• NFI2= Total income from • 1 to 7, from zero to very high
services (≥30)
• NFI3= Total nonfarm income • 1 to 7, from zero to very high
(≥30)
8. Food Security • Foodsec1= Total rice • 1 to 7, from very low(<5 ) to very
consumption (kg/ capita/ high (≥10)
month)
• Foodsec2= Total of protein • 1 to 4, very low (<40) to very high
consumption (gram per day) (>60)
• Foodsec3=Total of food • 1 to 4, very low (<150000) to very
consumption per capita per high (>180000)
month (IDR/month)
• Foodsec4= Total tobacco/ • 0 to 9, from zero (0) to highest (8)
cigarettes consumption (IDR
million/year)
STUDY AREA
As an archipelagic region, 92% (52,7191 km2) of Maluku province is covered by
the sea. Out of 1,388 small islands, there are five big islands of which 22% are
inhabited, while the rest 78% are uninhabited (BPS, 2018). Ambon Island, about
700 km2, is one of the principal islands in the region. During the Portuguese and
Dutch colonisation between the 15th and 18th centuries, Ambon Island was
famous as the centre of global trade spice collection (Turner, 2011; Luiten and
Zanden, 2012). Post colonisation, Ambon Island was divided into Ambon City
and Central Maluku Regency (Resosudarmo, 2009). The sites of this study were
selected in four villages at Ambon Island, part of the Central Maluku Regency.
Those are Negeri Lima, Seith, Hila, and Morela villages (see Table 2).
Figure 2. Map of Maluku Province and Ambon Island
Figure 2 shows Ambon City and Central Maluku Regency. Ambon City is
the centre of government, education, trade, services, and business activities.
Population and economic growth in Ambon City have caused land conversion
to resettlements, buildings, and basic physical infrastructures. Excessive
urbanisation and high population density have caused land fragmentation,
land disputes, and environmental degradation. This will cause unsustainable
cities because of negating interactions between terrestrial and coastal areas
(Ririhena, 2015). Different from Ambon city, Central Maluku Regency has
applied the national rice field policy since the 1970s. However, the poverty rate
in the region is still high, about 20.1% or 78,720 people in total (BPS, 2020). In
2018, people of a remote indigenous community in the region were starving for
14 days because of rice food insecurity (Abbas, 2018).
DATA COLLECTION
This chapter was designed by combining survey and field observation to
understand the sustainable relationship between land property rights and food
security in the small islands. Research sites were determined in Ambon Island,
part of Central Maluku Regency. Samples were selected from four villages
that still maintain land property rights and practice traditional agroforestry.
The total population in four villages was 597 households. Based on Yamane’s
(1967) formula—i.e., n =N/(N.d2+1) where n is the number of samples, N is
the population, and d is the precision error at 10%--the minimum research
sample was 86 households. The total sample of this chapter was fostered to 139
households. Data were collected between March and June 2018 involving five
undergraduate and three postgraduate students. These enumerators have been
trained in advance to use research questionnaires and focus group discussions.
Land property rights are not single and have multiple dimensions,
including the size of land, ownership status, number of land plots, and secure
land property right in line with legal evidence of ownership status. Ideally,
the secure land property rights are indicated by whether the land has paid
tax and evidence of certificate. However, in this case, the main indicators are
distinct borders and ownership status, profit sharing with the owner of the land
(Supriatna, et al., 2022). Food security is based on household expenditures to
fulfil about 2200 Kcal per capita per day, equal to 200 kg of rice per capita per
year. It means 0.5 kg/capita/day, around 0.3 kg is carbohydrate from rice, and
the rest of 0.2 kg comprises protein, fat, mineral, and vitamins (Sajogyo, 1977;
Sajogyo, 1978). Food security indicators include the total rice consumption,
food, protein, and tobacco.
The higher number of plants and the higher number of productive
crops indicate a higher level of biodiversity. Crops productivity is defined
as production per year and production per tree. In the same way, fishery
productivity is the total production, number of trips, and the types of fishing
gear technology. Agriculture income is represented by income from the farm,
including plantations, edible fruits, total farm, and livestock (cattle, goat, and
chicken), whereas fisheries income is defined as the quality of fish, which is
represented by the price of fish and annual income from fish. Non-farm income
indicators are income from small-scale industries, trade, and services.
Household Characteristics
Most heads of households in the sample have multiple occupations. Definitely,
the number of farmers and fishers without secondary jobs was only 17.3% and
2.2%, respectively, whereas the number of farmers and fishers with secondary
jobs was 79.1% and 1.4%, respectively. Hence, the primary occupation of 96.4%
of household heads is based on farms, and the rest of 3.6% are fisheries-based
(Table 2).
Data showed that about 64.3% of heads of the households were old aged,
and education of 69.3% was below junior high school. Then, 58.4% of educated
children aged group of 15 years and over were unemployed. The rest, 26.4%
and 15.2% of educated children, worked on informal farms and non-farm.
Table 2. Profiles of Households by Villages
Villages (%)
Household Characteristics Negeri
Seith Morela Hila Total
Lima
1. Occupation of Household Head
Farmers/Livestock 17,1 10,0 30,6 10,5 17,3
Farmer-Fisher 0,0 16,7 5,6 5,3 6,5
Farmer-Non-Farm 80,0 63,3 58,3 71,1 68,3
Farmer-Fisher-Non-Farm 0,0 3,3 5,6 7,9 4,3
Fisher 2,9 3,3 0,0 2,6 2,2
Fisher-Non-farm 0,0 3,3 0,0 2,6 1,4
2. Age of Household Head
Young (<40) 16.7 26.7 8.3 5.3 13.6
Moderate (40-49) 25.0 13.3 27.8 21.1 22.1
Old (≥50) 58.3 60.0 63.9 73.7 64.3
Villages (%)
Household Characteristics Negeri
Seith Morela Hila Total
Lima
3. Head of Household Education
No School 0 0 2.8 2.6 1.4
Elementary 38.9 30 36.1 52.6 40.0
Junior HS 30.6 33.3 33.3 15.8 27.9
Senior HS 22.2 36.7 25.0 28.9 27.9
Academy/Graduate 8.3 0.0 2.8 0.0 2.9
4. Family Dependency (People) 5.1 4.5 4.3 4.7 4.6
5. No. of Children (People) 3.2 2.8 2.4 2.9 2.9
6. Marriage Status
Widower 13.9 6.7 2.8 5.3 7.1
Widow 11.1 6.7 0.0 0.0 4.3
Marriage Couple 75.0 86.7 97.2 94.7 88.6
7. Occupation of Wife
Farm 69.4 100.0 91.7 89.5 87.1
Farm and Non-farm 30.6 0.0 8.3 10.5 12.9
8. Children’s Occupation
No Job 71.2 52.2 61.9 48.1 58.4
Farm 19.2 26.7 31.0 29.8 26.4
Non-farm (Informal) 9.6 21.1 7.1 22.1 15.2
9. Children Education
No Education 14.4 6.7 9.5 9.6 10.2
Elementary School 26.0 12.2 16.7 19.2 18.8
Junior High School 13.5 25.6 20.2 10.6 17.0
Senior High School 29.8 33.3 36.9 36.5 34.0
Academy 16.3 22.2 16.7 24.0 19.9
Source: Fieldwork, 2018
Land is a basic invaluable asset in rural areas. Land ownership status in this
chapter includes tanah negeri (village land), tanah dati (communal-territorial
land), and tanah pusaka (private land). Tanah negeri belongs to the village
government and most of the land is village forest and coastal areas. Then, tanah
dati belongs to the descendants of the same family name (clan). Each family
member has the right to the yields that must be shared proportionally with the
whole family (mataruma). This culture is called makan bersama (food sharing)
to maintain responsibility, solidarity, and equality among the family members.
Next, tanah pusaka is inherited land by the family members, particularly
that belongs to the son. Basically, most of the land is part of the other land that
belongs to the big family. Therefore, family land cannot be for sale because it is
a legacy of benign family member relationships. The benefit is sustainable land
as the source of livelihood for the next generation.
The sign of land borders and size is crucial; otherwise, it will create land
disputes within and between families and people. The borders of the village
land are controlled/guarded by Kewang darat, while the borders of the coast
are controlled by Kewang laut. The main signs of land and coastal borders are
rivers, stones, and trees. The land size of almost half of the farmer households
tends to occupy a smaller size (< 1 ha). The number of households with small,
medium, and large sizes with an average of 0.68 ha, 1.56 ha, and 2.96 ha was
49.6%, 32.4%, and 18%, respectively (Table 3).
Table 3. The Distribution of Households and Average Number of Land
Plots by Land Size
Villages
Seith Morela Hila Negeri Lima
Land Size n n n n
Avg Plots Avg Plots Avg Plots Avg Plots
(ha) (%) (%) (%) (%)
Small 0.65 55.6 2.1 0.67 36.7 1.6 0.71 50.0 1.8 0.70 54.05 2.2
Medium 1.57 36.1 3.1 1.49 36.7 2.6 1.60 27.8 2.4 1.60 29.73 2.5
Large 2.67 8.3 3.3 3.00 26.7 2.9 3.00 22.2 3.4 3.00 16.22 3.3
Source: Field work, 2018.
Note: Small land size is less than 1 ha; Medium=1-1.99 ha; and Large is ≥ 2 ha. Avg= Average of
land size (ha); n= Total number of household (%); Plots=Number of farm locations (units)
The land plot spreads in 2-3 locations because a farmer will expand to new
and further land when soil fertility in the previous land has declined gradually.
The growth of the population and the number of new households cause higher
land demand for housing. As a result, the area of Dusun farmland will converse
to new housing, and at the same time, Tanah Dati land has changed to Tanah
Pusaka. Most farmers have small land sizes, but the larger the land size, the
bigger the number of land plots.
Household Income
The research found that the annual income average was about IDR26.58 million.
That is equal to USD1.25/capita/day. The contribution of agriculture (including
fishery) and the non-farm was 70.47% and 29.53%, respectively, meaning the
former was 2.4 folds higher than the latter (Table 4).
The higher contributors to agriculture income were from plantations,
followed by fisheries and edible fruits, which are 49.06%, 11.1%, and 7.64%,
respectively. The prime income from fisheries are Skipjack tuna (Katsuwanus
pelamis), followed by kawalinya (Caranx mate), tongkol/komu (Euthynnus
affinis), momar/layang (Detapterus pusailus), and lema/kembung (Rastrellinger
brachysoma).
As shown in Table 4, the higher contribution of non-farm sector was from
trade, services, and home industry. Types of the home industry include wood
processing products whilst the main sources of services income were from
salary (civil servants), remittances, rural ojeg transportation, and government
aid. The main income sources from trade were selling agriculture and fisheries
commodities (fish vendors), food, and beverages. Small income from sago and
tubers indicated that local food production and consumption are very small.
Table 4. Distribution of Household’s Annual Income by Sources in
Ambon Islands, Maluku
Household Income
Group Sources Farm/ Non- Total Farm and
IDR million
farm (%) Non-farm (%)
A AGRICULTURE 18.73 100 70.47
Plantation 13.04 69.6 49.06
Nutmeg 5.19 27.7 19.53
Clove 7.33 39.1 27.58
Coconut 0.16 0.9 0.60
Cacao 0.36 1.9 1.35
Edible Fruits 2.03 10.8 7.64
Sago and Tuber 0.26 1.4 0.98
Vegetables 0.01 0.1 0.04
Livestock 0.44 2.3 1.66
Fishery 2.95 15.8 11.10
B NON-FARM 7.85 100 29.53
Industry 0.94 12.0 3.54
Trade 2.78 35.4 10.46
Services 1.55 19.7 5.83
Civil Servant 1.49 19.0 5.61
Government Aid 1.09 13.9 4.10
C TOTAL (A+B) 26.58 100
OUTER AND INNER MODELS
The results of PLS-SEM are called outer and inner model. The outer model is
also called the measurement model, which describes that one latent variable
should have two or more dimensions and indicators. These indicators must
have a high correlation. That means that these indicators are reflective of the
latent variables. These are also called reflective indicators. The outer model is
valid if the correlation between indicators (loading factors and cross-loading) is
greater than 0.6 for exploratory research. Indicators that have values less than
0.6 will be omitted from the model. Another requirement is that the value of
average variance extracted (AVE) of each latent variables must have greater than
0.5. Besides, the outer model is called reliable when the values of Cronbach’s
Alpha and Composite reliability of the latent variable are greater than 0.7. In
short, the outer model measures the validity and reliability of latent variables.
The validity and reliability of the outer model are necessary conditions
for the inner model. If the outer model measures validity and reliability, the
inner model measures the significance and influence of exogen or independent
variables on endogen or dependent variables. This is also called a quality
model. The inner model consists of the path coefficient, R square, effect size,
and predictive relevance and also provides indicators of the fit model.
Figure 3. The Outer Model for Land Property Rights, Income, and Food
Figure 3 shows that the values of loading factors are greater than 0.7,
then the cross-loading values are greater than 0.7 and AVE values of all latent
variables are greater than 0.5 (Table 5). That means the model is valid. Then,
the model is also reliable. It can be seen from the value of Cronbach’s Alpha and
Composite reliability that is greater than 0.7 (Table 5).
Table 5. Loading Factors, Average Variance Extracted (AVE), Cronbach’s
Alpha, and Composite Reliability
Loading Cross- Cronbach's Composite
Constructs Indicators AVE
Factors loading Alpha Reliability
Land Property Rights LPR1 0.948 0.948 0.742 0.827 0.895
LPR2 0.901 0.901
LPR3 0.717 0.717
Crops Biodiversity CB1 0.861 0.861 0.671 0.754 0.859
CB2 0.737 0.737
CB3 0.853 0.853
Crops Productivity CP1 0.850 0.850 0.643 0.815 0.877
CP2 0.877 0.877
CP3 0.763 0.763
CP4 0.707 0.707
Farm Income FARMINC1 0.854 0.854 0.672 0.760 0.859
FARMINC2 0.724 0.724
FARMINC3 0.873 0.873
Fisheries Productivity FP1 0.957 0.957 0.895 0.942 0.963
FP2 0.944 0.944
FP3 0.937 0.932
Fisheries Income FISHINC1 0.960 0.960 0.899 0.890 0.899
FISHINC2 0.937 0.937
Nonfarm Income NFI1 0.746 0.746 0.633 0.714 0.837
NFI2 0.733 0.733
NFI3 0.898 0.898
Food Security FOODSEC1 0.636 0.635 0.554 0.610 0.787
FOODSEC2 0.793 0.793
FOODSEC3 0.793 0.793
Next, Figure 4 and Table 6 showed that the inner model depicted about
15 hypothesis testing. Significant influence of exogen variables on endogen
variables occurred when the T-statistic value is greater than 1.96 at ∝=5%.
This chapter will be focused on three groups of hypotheses from Table 6. First,
hypothesis number 1, 2, and 3. The results showed that land property rights
and crop biodiversity significantly influenced crop productivity. The value of R
square was weak, that is, 18.4%. However, land property rights have a big effect
size on crop biodiversity. The effect size is the change in R-square value when an
exogenous variable is removed from the model (R square ≥ 0.02 is small; ≥ 0.15
is medium; and ≥ 0.35 is large).
Figure 4. The Outer Model for Land Property Rights, Income, and Food
Furthermore, crop productivity has a big effect size ( f square = 2.5) and
predictive relevance on agriculture income (see Table 7). Next, the Q square is
predictive validity that is the same with the determination coefficient. All latent
variables' Q square value is good if the Q square value is greater than zero (Q
square > 0). Predictive relevance means that the model can explain variations of
existing information in the research data.
Second, hypothesis number 4 and 5. The result showed that crop
productivity and land property rights (LPR) significantly influenced agriculture
income. The value of R square was strong, which is 78.7%. Agriculture incomes,
however, have a small effect size but high predictive relevance on food security
(Table 7). The third group is hypothesis number 6, 7, and 8. The result showed
that land property rights positively influence fisheries productivity. R square
value was weak, that is 6.9%, small effect size but having predictive relevance.
That means landscape condition is still good to support fisheries productivity
in the coastal rural areas. Furthermore, agriculture income and fisheries
productivity significantly influenced fishery income (Table 7). The value of
R square was strong, 81.6% and fishery incomes have a big effect size and
predictive relevance on food security. These results showed that productivity is
the main determinant of agriculture and fishery income.
Table 6. The Hypothesis and Significant Relationship
between Latent Variables
Original
Sample Standard
sample T p
Hypothesis mean Deviation
(O)/Path Statistics Values
(M) (STDEV)
Coefficient
1. LPR → Crops Biodiversity 0.514 0.521 0.058 9.033 0.000
2. LPR → Crops Productivity 0.249 0.253 0.078 3.216 0.001
3. Crops Biodiversity → Crops Productivity 0.244 0.246 0.084 3.017 0.002
4. Crops Productivity → Agriculture
0.788 0.791 0.027 28.064 0.000
Income
5. LPR → Agriculture Income 0.208 0.204 0.045 4.755 0.000
6. LPR → Fishery Productivity 0.099 0.091 0.089 1.061 0.133
7. Fisheries Productivity → Fisheries
0.901 0.910 0.022 41.183 0.000
Income
8. Agriculture Income → Fishery
(0.312) (0.305) 0.081 3.842 0.000
Productivity
9. Agriculture Income → Food Security 0.294 0.304 0.086 3.317 0.000
10. LPR → Food Security 0.187 0.183 0.081 2.243 0.011
11. Fisheries Income → Food Security 0.510 0.506 0.089 5.393 0.000
12. Nonfarm Income → Food Security 0.220 0.225 0.059 3.616 0.000
13. Agriculture Income → Nonfarm
(0.365) (0.371) 0.076 4.563 0.000
Income
14. Fisheries Income → Nonfarm Income (0.226) (0.229) 0.079 3.026 0.000
15. LPR → Nonfarm Income 0.183 0.197 0.105 1.665 0.041
As indicated in Table 4, the higher contribution from non-farm sector was
trade, services, and home industry. Types of the home industry include wood
processing products whilst the main sources of services income were from
salary (civil servants), remittances, rural ojeg transportation, and government
aid. The main income sources from trade were selling agriculture and fisheries
commodities (fish vendors), food, and beverages. Small income from sago and
tubers indicated that local food production and consumption are very small.
Finally, the model is fit when the value of SRMR < 0.08, rms Theta value
is close to zero, and NFI>0.9. Results showed that the model fulfilled the
indicators of SRMR and rms Theta, but the value of NFI was smaller than 0.9.
This means the model is a marginal fit and needs improvement in future study.
For future research it is necessary to increase the number of household samples,
particularly fishers, and additional dimension and indicators of latent variables.
Table 7. R square, Effect Size (f 2), and Predictive Relevance (Q 2)
Variables R square f2 Q2
1. Crops Productivity 0.184 2.500 0.100
2. Agriculture Income 0.787 0.083 0.498
3. Fisheries Income 0.816 0.342 0.707
4. Food Security 0.326 - 0.147
5. Nonfarm Income 0.098 0.064 0.051
6. Crops Biodiversity 0.264 0.054 0.166
7. Fisheries Productivity 0.069 4.355 0.058
8. Land Property Rights
Agriculture income - 0.175
Crops Biodiversity - 0.358
Crops Productivity - 0.056
Fishery Productivity - 0.008
Food Security - 0.038
Nonfarm Income - 0.028
FACTORS DETERMINING HOUSEHOLD FOOD SECURITY

This chapter finds that land property rights and household income are two factors
determining household food security. Different from rice field and plantation
monoculture in the continental island, land property rights in the coastal rural
small island are an invaluable communal asset to the family and community that
has established the formation of the Dusung agroforestry systems. The Dusung
agroforestry systems are integrated land tenures and multiple tree cropping
(Kaya, et al., 2001; Salampessy, 2017) from the forest to the coastal (Wattimena,
2003; Wattimena, 2016; Hoshino et al., 2017), and controlled by local customary
law (Ayawaila, 1996). Land property rights consist of local norms and rules to
conserve forest and the Dusung agroforestry systems, which allows the land to
sustain. Stanton (1993) stated that the structure of the Dusung agroforestry
systems has stability and resilience towards external economic shocks, climate
change impacts, and natural disasters. The Dusung systems includes live food
security barns, particularly sago palms, breadfruits, tubers, and fisheries. Sago
forests, mangroves, and coral reefs are protected to reduce global warming.
One million ha of sago forest will reduce 0.03% of global warming (Ishizaki,
1997). Sago forest also reduces deforestation as wood is substituted by sago for
housing and buildings (Sasaoka et al., 2014; Angelsen and Kaimowittz, 2004).
Supriatna et al. (2022) stated that secure land property rights are the main
trigger of land conversion and environmental degradation. This chapter found
that strong communal land property rights—which are embedded in local
customary law—have prevented land conversion, environmental degradation,
and Dusung farmland transaction. To some extent, the local community converses
part of the Dusung to housing because of population growth. However, they still
maintain multiple tree crops in the form of agroforestry systems in the house
yard.
Our study supports Cheng and Resosudarmo (2021) who stated that
strong land property rights significantly and directly influenced food security.
However, this chapter finds that the strong communal land property rights of
the Dusung cannot be used as a loan guarantee to get financial capital from
a bank. The main reason is that most land ownership status is communal
land without a government land certificate. As a result, low financial capital
will cause low access to technology to improve low productivity. In the past,
commodities in the Dusung agroforestry system were the primary food security
source, particularly sago palm, tuber, and other food crops from the terrestrial
and fisheries.
Since the implementation of rice bias policy in the 1980s (Louhenapessy,
1992), it was exacerbated by national rice subsidy for the poor household
in the 2000s in which local staple foods have been substituted by imported
rice and flour (Girsang, 2014a; Girsang, 2014b). Furthermore, the rice bias
policy has also changed the perception and food habits of the local community,
particularly the young generation. They perceived local foods as inferior goods,
whereas imported rice, flour, bread, and fast food were superior. Later, the local
community became more dependent on imported rice and abandoned local
foods, particularly unleashing sago forest as a hidden treasure of the world
(Pranamuda, 2013).
Consequently, the local coastal rural community in small islands must have
the cash to buy imported rice and flour. That means income is the key variable
to determine food security. To obtain cash, the local community is dependent
on seasonal clove, nutmeg, edible fruits such as durian (Durio zibethinus),
mangosteen (Garcinia mangostana), duku/langsat (Lansum domesticum),
rambutans (Nephelium lappaceum), and gandaria (Bouea macrophylla).
Research showed that the annual rural household income average was around
IDR26.58 million. This equals USD1.25 per capita per day, below the poverty
line of about USD1.5 per capita per day. This chapter finds that land property
rights and productivity significantly improve agriculture and fishery income.
The coastal community will cover low and insufficient household income from
agriculture and fisheries from nonfarm economic activities (van Oostenbrugge,
et al., 2004; Babatunde, 2013).
Productivity is also determined by stronger land property rights and
higher biodiversity of prime commodities. Low productivity is also caused by
limited financial loan access to the bank as a necessary condition to improve
technology (Chheng and Resosudarmo, 2021). Technology needs to upgrade to
improve land fertility, biodiversity, and productivity of agriculture and fisheries.
The existing productivity of the prime commodities, particularly clove and
nutmeg, are 234 kg and 288 kg per ha, respectively. This is still half of the total
maximum productivity standard.
The other types of technology to improve agriculture are new varieties of
high-yield seeds, fertilizers to improve soil fertility, bio-pesticides to control
pests and diseases, and tools to nurture crops and trees. Low income is worsened
by existing farmers who are older and less educated. The rural youth—who are
more educated, stronger, and have access to digital technology resist working
in agriculture and fisheries—perceived farming and fishery as dirty jobs, hard
work, time-consuming, risky to external shocks, seasonal, uncertain, and low
income. They also have a strong aspiration of being a civil servant rather than
an entrepreneur, but it becomes more difficult to achieve because it is very
competitive.
As with farmers, the productivity of artisanal fisheries is also still low with
fishermen’s fish catching about 141 trips per year. This is due to the use of fishing
gear technology in the form of perahu (non-powered boat), small speedboat,
and big boat/ship-GT. The productivity of perahu or jukung (non-powered boat)
is around 10 kg per trip, whereas the productivity of semi-modern fishing gear
technology (speedboat: 15 HP Yamaha or Katinting) is estimated at 27 kg per
trip. Generally, traditional, and small motor fishing gear technology focused
to catch kawalinya (Caranx mate) and momar (Detapterus pusailus) whilst the
semi-modern technology focused to catch Skipjack tuna (Katsuwanus pelamis).
Technology types to improve fishery productivity are upgrading technology—
i.e., from non-powered boats to out boats, from out boat to 5 Gross ton (GT)
boat/ship and 30 GT boat/ship, and cold storage. The purpose is to support
fishermen to boost the number of trips, speed, catch capacity, access to the
deeper sea and further locations, and to keep the fish in cold storage.
According to Agus (2014), another alternative solution to improve low
productivity is to develop Integrated Bio-Cycle Farming Systems (IBFS). That is
integration between farm and non-farm. This is called good agricultural practice
and a low external input farming model. Farm sustainability is developed
through integrated crop-livestock and fishery farming systems, whereas non-
farm sustain-agility is developed by high-touch post-harvesting to improve
product added-value to expand the market network. Meanwhile, Ostrom (1992)
offers the concept of common pool resources (CPR) to avoid the tragedy of the
common through strengthening local institutions. In applying that concept,
local customary law is used to maintain strong communal land property rights
and to control the use of non-environmentally friendly fishing gear, oil spills,
and garbage from big and small ships, use the fish bomb, and devastating
mangrove and coral reef. The main reason for necessarily implementing the
CPR concept is that local communities have already established local wisdom,
everyday life interaction with nature, and developed mutual collective beneficial
action. Otherwise, uncontrolled development of new housing, real estate, basic
infrastructure development, and intensive monoculture crops will promote land
conversion that endangers the sustainability of the terrestrial and the coastal in
small islands (Ririhena et al., 2015).
CONCLUSION
This research shows that land property rights and household income are
determinants of household food security, whereas household income is also
determined by household farm productivity and land property rights. Income
has become a pivotal determinant of food security because local communities
have abandoned local food sources as live food security barns. They have
become more dependent on imported rice and flour that they must buy with
cash income. Substitution of local food with rice has been caused by the rice
bias policy since the 1960s, that is, to convert forest and sago palm land to rice
fields in small islands. This policy is supported by rice subsidy for the poor
programme at the beginning of the 2000s, and then it was exacerbated by
the local community's perception that rice is superior food, cheaper, easier to
access, and to cook.
This research also finds out that household income is still low because of
low productivity, which results from communal land property rights that have
no legal certificate. As such, farmers cannot use the certificate to get a financial
loan from the bank to upgrade agriculture and fishery innovation and technology.
These new technologies include soil fertility, high-yield seeds, bio-pesticide
to control crop pests and diseases, fishing gear technology, and fisher’s skill
and knowledge to improve catch per unit efforts. In addition, productivity can
also be enhanced by energising young and educated generations to substitute
existing old and uneducated farmers and fishers in rural areas. To sustain the
socio-ecology of small islands, local communities should strengthen communal
land property rights embedded with local customary law, get back to cultivate
and grow local foods, practice integrated bio-cycle farming, develop tourism,
and export local plantation and fisheries products to reduce dependency on
imported rice. This will increase the resilience of local communities to challenge
external shocks such as global and local food crises.
REFERENCES
Abbas, A. (2018). Dibalik Angka Kemiskinan. Dokumentasi BPBD Maluku
Tengah (Behind the numbers of poverty), https://epaper.dpr.go.id/index/
popup/id/10135. Accessed on 3 June 2022.
Agus, C. (2014). “Agribisnis Berbasis Pertanian Terpadu.” In Sunarminto, B.H.
(ed), Pertanian Terpadu untuk Mendukung Kedaulatan Pangan Nasional,
Gadjah Mada University Press. Yogyakarta.
Ayawaila, J.W. (1996). Tinjauan Sosial Budaya Agroforestry. Pusat Studi Maluku,
Universitas Pattimura. Ambon
BPS (2018). Maluku in Figures. Central Board of Statistic and Statistical Regional
Development Plan Board, Maluku Province, Ambon.
BPS. (2020). Leihitu Subdistrict in Figure 2021. Statistics of Maluku Tengah
Regency.
Angelsen, A. and D. Kaimowitz (2014). “Does agroforestry reduce deforestation?”
In G. Schroth, G.A.B. da Fanseca, C.A. Harvey, C.Gascon, H.L.Vasconcelos, and
A.M.N. Izac (eds). Agroforestry and Biodiversity Conservation in Tropical
Landscapes. London: Islands Press.
Babatunde, Raphael O. (2013). “On-Farm and Off-farm Works: Complement or
Substitute? Evidence from Rural Nigeria.” Conference Paper/ Presentation,
African Association of Agricultural Economists (AAAE),2013 Fourth
International Conference, September 22-25, 2013, Hammamet, Tunisia
Chheng, K. and B.P.Resosudarmo. (2021). Land Property Rights and Food
Insecurity in Rural Cambodia. Regional Science Policy and Practice.
Published by Wiley, April 2021. DOI:10.1111/rsp3.12418
Chin, W.W. (1998). ‘The Partial Least Squares Approach to Structural Equation
Modelling.’ Modern Methods for Business Research, 2: 295-336.
Dowling, N.G. and Greenfield, S.M. 1998. “Introduction and overview.” In N.G.
Dowling, S.M. Greenfield, and K.S. Fisher (eds). Sustainability of rice in the
Global Food System, 1-7, Pacific Basin Study Center, Davis, Calif, USA and
IRRI, Manila, Philippines.
Ducos, L. (2014). “Importance of the traditional land-use and land-tenure
systems of Waraka, Seram Island, Maluku.” Working Paper 144. Bogor,
Indonesia: CIFOR, 73. Hongkong: Periplus Editions
Girsang, W. (2014). ‘Socioeconomic Factors that have Influenced on the Decline
of sago Consumption in Small Islands: A Case in Rural Maluku, Indonesia.’
South Pacific Studies 34, no.2.
Girsang, W. (2014b). “Food security systems in Indonesia: Challenges and
implication to food security in Maluku.” Occasional Paper No.54, Kagoshima
University Research Centre for the Pacific Islands, Japan
Girsang, W., Nanere, M.G., Siregar, A and Hamid. 2019. “The contribution of
sustainable The Dusun farming to household income in small islands:
A case in Ambon island, Indonesia.” In T. Mulyaningsih, B.R. Samudro, V.
Hadiwiyono, H. Handra, D. Pratomo, A.A. Yusuf, and B.P. Resosudarmo
(eds). Strengthening Regional and Local Economy. IRSA Book Series on
Regional Development, No.17. UNS Press, Solo.
Ghozali, I. (2021). Partial Least Squares: Konsep, Teknisk dan Aplikasi
Menggunakan Program SmartPLS 3.2.9. untuk Penelitian Empiris. Badan
Penerbit Universitas Diponegoro. Semarang
Ishizaki, A. (1997). ‘Concluding remarks for the 6th International Sago
Symposium, Riau-Indonesia.’ Sago Communication 8(2):22-24
Hair, J.F., G.T.M. Hult, and C.M. Ringle (2017). A Primer on Partial Least Squares
Structural Equation Modeling (PLS-SEM). Thousand Oaks, CA: Sage.
Hoshino, E., E.I. van Putten, W. Girsang, B.P. Resosudarmo, and S. Yamazaki
(2017). ‘Fisher’s perceived objectives of community-based coastal resource
management in the Kei islands, Indonesia.’ Frontiers in Marine Science 4,
Article 141 (17 May).
Kastanya, A. (2019). ‘Forest landscape challenges in Wallace.’ Paper presented
at the International Seminar on Biodiversity in Wallacea: Linking-
Conservation Science with Practice, 9 December 2019, Pattimura
University-CIFOR and British Columbia University.
Kubitza, C., Krishna, V.V., Urban, K., Alamsyah, Z., and Qaim, M. (2018). ‘Land
Property Rights, Agricultural Intensifications, and Deforestation in
Indonesia.’ Ecological Economics 147 (2018): 312-321
Laimeheriwa, S., M. Pangaribuan, & M. Amba (2019). ‘Analisis Fenomena El
Nino dan Dampaknya Terhadap Neraca Air Lahan di Pulau Ambon.’ Jurnal
Budidaya Pertanian 15(2)
Legarda, B. (1984). Small Island Economies. Finance and Development,
International Monetary Fund.
Louhenapessy, J.E. (1992). “Sago in Maluku: Its potential, land condition and
its problems.” National Sago Symposium, Faculty of Agriculture, Pattimura
University (In Indonesian)
Luiten, Jan & Zanden D.M. 2012. Ekonomi Indonesia 1800-2010, Antara Drama
dan Keajaiban Pertumbuhan. Kompas-KITLV. Jakarta.
Matinahoru, J.M. (2014). “A Review on the dusun as an Indigenous Agroforestry
System Practice in Small Island.” In S. Yamamoto and S. Rahardjo (eds). A
New Horizon of Island Studies in the Asia-Pacific Region. Occasional Paper
No.54, December 2014, Kagoshima University Research Center for the
Pacific Islands. Japan
Mohanty, S. (2013). ‘Trends in global rice consumption, Rice Today.’ International
Rice Research Institute 12, no.1 ( Jan-March).
Ostrom, Elinor (1992). ‘Governing the Commons: The Evolution of Institutions
for Collective Action.’ Natural Resource Journal 32, no.2.
Pranamuda, H. (2013). “Preface.” In H. Prananmuda, Y. Toyoda, R.M. Osok, M.Y.K.
Chan, and N. Haska (eds). Unleashing Sago Hidden Treasure of the World.
Proceeding of the 11th International Sago Symposium, Manokwari, West
Papua, Indonesia, 6-8 November 2013. PPSI-PT Austindo Nusantara Jaya
Tbk and PT.ANJ Agri Papua.
Resosudarmo, B.P. & E. Thorbecke. (1998). ‘Reducing the Number of Pesticide
Related Illnesses: The Impact on Household in Indonesia.’ Bulletin of
Indonesian Economic Studies, 34(2): 143-57.
Resosudarmo, B.P. (2008). ‘The Economy-wide Impact of Integrated Pest
Management in Indonesia.’ ASEAN Economic Bulletin, 25(3):316-333.
Resosudarmo, B.P. 2009. “A Note on Socio-economic Development in Maluku.”
In B.P. Resosudarmo and F. Jotzo (eds.), Working with Nature against
Poverty: Development, Resources, and the Environment in Eastern Indonesia.
Singapore: Institute for Southeast Asian Studies, 74-93.
Retraubun, A.S.W. (2013). “Indonesia small island management.” Keynote paper
presented at International small islands conference at Tual and Langgur,
Indonesia, 10-13 July 2013, Ambon
Richter, N.F., G.Cepeda, J.L. Roldan & C.M. Ringle, C.M. (2016). ‘European
Management Research Using Partial Least Squares Structural Equation
Modelling (PLS-SEM).’ European Management Journal. Available at:
http://dx.doi.org/10.1016/ j.emj.2016.08.001
Ririhena, R.E., H.S. Alikodra, C. Kusmana, A.S.W. Retraubun (2015). ‘Contribution
of Dusun Agroforestry to Household Income on Ambon Island, Indonesia.’
International Journal of Sciences: Basic and Applied Research (IJSBAR),
23(1): 256–267. Available at: https://www.gssrr.org/index.php/
JournalOfBasicAndApplied/article/view/4138
Rosegrant, M. W., J. Koo, N.Cenacchi, C.Ringler, R.D. Robertson, M.Fisher, &
P.Sabbagh. (2014). Food security in a world of natural resource scarcity: The
role of agricultural technologies. Washington, DC; International Food Policy
Res Institute.
Salampessy, M.L., I.G. Febryano, & I. Bone (2017). ‘Pengetahuan ekologi
masyarakat local dalam pemilihan pohon pelindung pada system
agroforestry tradisional ‘the dusung’ pala di Ambon.’ Jurnal Penelitian
Sosial dan Ekonomi Kehutanan 14, No.2: 135-142.
Sasaoka, M., Y. Laumonier, & K. Sugimura (2014). ‘Influence of indigenous sago-
based agriculture on local forest landscape in Maluku, East Indonesia.’
Journal of Tropical Forest Science, 26 (1): 75-83, Forest Research Institute,
Malaysia.
Sajogyo (1977). Garis Kemiskinan dan Kebutuhan Minimum Pangan. LPSP-IPB.
Bogor.
Sajogyo (1978). ‘Lapisan Masyarakat yang Paling Lemah di Pedesaan Jawa.’
Prisma 3, (April): 3-14.
Stanton, W.R. (1993). ‘Perspectives on, and future prospects for, the sago palm.’
Agricultural Systems 1:2-7
Sudaryanto, T. (2003). Konversi Lahan dan Produksi Pangan Nasional (Land
Conversion and National Food Production). Prosiding Seminar Nasional
Multifungsi dan Konversi Lahan Pertanian. Balai Penelitian Tanah-
Penelitian dan Pengembangan Pertanian. https://balittanah.litbang.
pertanian.go.id/ind/dokumentasi/prosiding/mflp2003/tahlim04.pdf.
Asccessed on 2 June 2022.
Supriatna, S., F. Hashilah, M.K. Mukhtar, & K.K. Wardani (2022). ‘Determinant
of Land Use Change in Kalimantan: An Evidence from Banjarbaru City and
Banjar Regency.’ Forest and Society 6, no.1 (April): 422-435. https://doi.
org/10.24259/fs.v6i1.18469
Suswono (2011). “The Indonesia food security in the perspective of global
economy and national sovereignty.” Paper presented at the International
conference on ‘The future of global food security and safety: Issues and
justification.’ IPB Internasional Convention Centre, 27 October 2011,
Ministry of Agriculture. Jakarta
Turner, Jack (2011). Sejarah Rempah: Dari Erotisme Sampai Imperialisme.
[Spices history: From erotism to imperialism]. Depok: Komunitas Bambu.
Wattimena, G.A. (2003). “Contoh-contoh agroforestry di Maluku.” In H.S. Arifin,
M.A. Sardjono, T. Jogo, L. Sundawati, G.A. Wattimena, & Widianto (eds).
Materi Pelatihan Agroforestry di Indonesia,Bogor, World Agroforestry
Centre, ICRAF, 3-11.
Wattimena, G.A. (2016). “Moluccas archipelago agroforestry.” Keynote
presentation in The 1st International Seminar on Agroforestry (ISA-1),
Ambon, 2-3 November 2016
Weiss, J. (2009). The economics of climate change in Southeast Asia: A regional
review. Asian Development Bank.
van Oostenbrugge, J.A.E, W.L.T van Densen, & M.A.M. Machiels (2004). ‘How
the uncertain outcomes associated with aquatic and land resource use
affect livelihood strategies in coastal communities in the Central Moluccas,
Indonesia.’ Agricultural Systems 82:57-91
Yamazaki, S., B.P. Resosudarmo, W. Girsang, & E. Hoshino (2018). ‘Productivity,
Social capital and Perceived Environmental Threats in Small-Island
Fisheries: Insights from Indonesia.’ Ocean and Coastal Management, 152:
62-75, Elsevier
Yamane, T. (1967). Elementary Sampling Theory. No 04, QA276.5, Y3. New
Jersey, US: Prentice-Hall.
CHAPTER 11
Examining Community Perceptions

of PT. Newmont Nusa Tenggara’s
Corporate Social Responsibility
(CSR) Initiatives
Baiq Nurul Suryawati, Burhanudin, Zainal Abidin,
Muttaqillah, Isti Fadah, and Intan Nurul Awwaliyah
INTRODUCTION
Over times, the significant expansion in foreign investment in the nation’s
mining sector proves that Indonesia’s mining industry has significantly
impacted Indonesian society. As a country that is richly abundant in natural
resources, notably in natural products, Indonesia has been highly demanded
as investment destination for international mining corporations. According to
Suryantoro and Manaf (2002), Indonesia’s historical recognition as a mineral-
rich country with a long-standing tradition of exploiting natural resources—
such as gold, silver, precious stones, and building stones—has made it become
one of the world’s most significant mining investment hubs.
During the 18th century, European geologists, mainly the Dutch, conducted
scientific research in Indonesia and produced important discoveries about
energy and mineral resources, such as oil, gas, geothermal, coal, copper,
tin, nickel, iron, aluminium, diamond, and non-metallic minerals. This early
recognition of Indonesia’s geological reserves has sparked more investigation,
231
which is still ongoing today. The National Development Programme in the late
1960s was when exploration operations continued long after Indonesia attained
independence and led to the discovery of additional energy and mineral sources.
According to Chawa's (2014), the mining company that operates in
Indonesia has put in place a variety of community development initiatives,
including building capacity in the areas of infrastructure development, health,
education, agriculture, and local business, as well as charity programmemes
and revolving fund programmemes. The research has identified instances
of community resistance and unintended adverse effects arising from
implementing these programmemes—such as creating a dependency on the
mining company for financial support rather than achieving the intended goal of
community empowerment through development, although the local community
occasionally participates in decision-making. A study by Langston et al. (2015)
comparing findings from small-scale and large-scale mining demonstrates that
different forms of mining follow different developmental patterns. While large-
scale mining provides better job security and safer working conditions, the
advantages of capital transformation are not retained in the local community.
Small-scale mining contributes more to the local economy and encourages local
entrepreneurship but generates lower total financial returns. A “landscape
approach” to balancing conservation and development trade-offs would help
manage both small-and large-scale mining, and policymakers should emphasise
the formalisation of small-scale mining and carefully assess the impact of large-
scale mining on local populations.
The presence of mining corporations in several regions of Indonesia
has benefited the people and sparked conflict. Beyond previous analyses of
resource governance, extractive development, and environmental conflict,
Libassi's (2022) research on the gold conflict in Pongkor, Indonesia, examines
how the conflict between large-scale and small-scale gold miners generates
extractive subjectivities. The mining company’s efforts to eradicate small-scale
mining, the small-scale miners’ development of a collective “community miner”
identity, and the mining company’s pursuit of internal reforms for clean and
green development are identified as the three competing subject formation
processes at the centre of the territorial conflict. These processes complicate
narratives of industrial extractive dominance and local resistance, showing that
subjects inside and outside mining activities are co-constituted.
According to Burhanudin et al., investigation into Nusa Tenggara Barat
(NTB) in 2023 reports that PT. Newmont Nusa Tenggara (NNT) has lost
control of mining in Kabupaten Sumbawa Barat (KSB) to PT. Amman Mineral
Nusa Tenggara (AMNT). Through its CSR initiatives, PT. AMNT has pledged to
enhance the welfare of residents of KSB, notably in the Desa Sekongkang. This
study investigates the effects of community satisfaction as aid beneficiaries on
corporate commitment and the durability of CSR initiatives carried out by PT.
AMNT. Based on the results of a study with 56 respondents, it was found that
there is a low degree of community satisfaction with PT. AMNT’s commitment
to and sustainability of its CSR programmemes. PT. AMNT’s commitment
to to its CSR programmeme sustainability and the compatibility of the CSR
programmeme with the needs of the community have all been carried out
by PT. AMNT. However, there is a significant correlation between community
satisfaction with CSR assistance and its sustainability and compatibility.
As a reworked project, this paper aims to analyse the same research
instrument used to study the community beneficiaries of the Corporate
Social Responsibility (CSR) programmeme conducted by PT. Newmont Nusa
Tenggara (NNT), the predecessor of PT AMNT. This study follows Duarte's
(2011) investigation of CSR cultures. It asserts that businesses that actively
uphold ideals like social justice and environmental sustainability over times
form distinctive “CSR cultures” that are exemplified by specific structures,
behaviours, and symbolic expressions. These CSR cultures shape the company’s
identity, mission, and vision, giving rise to distinctive histories and meanings.
The study by Erwin Syarif and Hatori (2017) examined CSR initiatives by PT. Vale
Indonesia Tbk (PTVI) to promote regional sustainability following the closure
of the mine in Soroako, Indonesia. The research gives an integrated view of CSR
as a sustainable mine closure approach. Data were acquired through interviews
and reports. The report assesses the effectiveness of PTVI’s CSR strategies and
offers suggestions for improving regional sustainability.
PT. NNT, which has been active in NTB for many years, is currently
becoming PT. AMNT. According to Gifford, Kestler and Anand's (2010) research,
the transnational model of global strategy contends that multinational
companies (MNEs) often rely on well-established global capabilities to adapt
current business models. For MNEs in the gold mining industry operating in
underdeveloped countries, Gifford, Kestler and Anand (2010) argue that the
transnational model needs to be modified to allow for a hybrid strategy that
Examining Community Perceptions of PT. Newmont Nusa Tenggara’s 233

Corporate Social Responsibility (CSR) Initiatives
balances local and global tactics. This strategy is demonstrated by Newmont
Mining’s efforts to increase local legitimacy in Peru by contributing to
community development. According to Gifford, Kestler and Anand's (2010), an
institutional environment has been emerging across the industry that includes
regional CSR initiatives to balance capitalism’s effects on both developed and
developing countries. This study also compares Newmont’s case with CSR
practices at other MNEs.
This research examines mining businesses in Indonesia by focusing on
the perception of CSR assistance recipients from PT. NNT and examining their
association with the community’s perception of the commitment and alignment
of CSR programmes carried out by PT. NNT before the company was taken over
and changed its name into PT. AMNT to understand the relationship between
the perception of the community receiving CSR assistance from PT. Newmont
Nusa Tenggara (NNT).
PT. Newmont Nusa Tenggara (NNT) and Its Comdev Programmes

Since 1986, Kabupaten Sumbawa Barat (KSB) has been the site of copper
and gold mining operations by PT. NNT that is a global mining corporation
contracted with the government of Indonesia. By putting forward outstanding
performance in the mining sector, PT NNT aspires to become the most reputable
and recognised mining firm. This vision is further developed into a purpose that
strives to create a sustainable mining business that makes significant profits for
shareholders while setting the standard for social responsibility, environmental
protection, and workplace safety.
A five-year strategic plan for community development (Comdev)
encompassing 2014–2018 contains specific points concerning PT. NNT’s CSR
policy (Burhanudin, 2013). The Comdev strategic plan covers the following
topics: vision, mission, and mission objectives, foundations (values and core
principles adopted), strategic issues, and primary programme areas, such as
the economic, health, education, sociocultural, and environmental sectors. The
Comdev Programme is a live illustration of how PT. NNT’s CSR policy is being
carried out. Through several programmes in areas such as the environment,
socio-culture, religion, health, and education, it aims to improve the local
communities’ welfare and quality of life.
Yayasan Pembangunan Ekonomi Sumbawa Barat (YPESB) and Yayasan Olat
Parigi (YOP), two significant pillars, support the Comdev economic sector effort.
While YOP focuses on money or micro-enterprise financing, YPESB assists in
identifying, fostering, and expanding local businesses. A self-sufficient, long-
lasting organization with trained staff members who can design innovative,
participatory programmes and foster beneficial collaborations is what YOP
aspires to become.
PT. NNT’s CSR fund is used for programmes and operational purposes, and
from 1999 to 2014, YOP handled an average of IDR 5 billion a year. Additionally,
PT. NNT offers YOP personal guidance and support. Since 2011, YOP has
concentrated on administering two main programmes: microfinance and
strategic business unit development. Between 2011 and 2013, it distributed
IDR 5.8 billion for microfinance to 1,696 recipients dispersed throughout eight
subdistricts in the KSB region. As a result, this study will look at the perceptions
of CSR recipients who work under the supervision of YPSEB and YOP in the
example of PT NNT’s CSR programme, which is closely linked to community
development institutions.
CORPORATE SOCIAL RESPONSIBILITY

Firmansyah et al. (2020) sought to outline the landscape of corporate social
responsibility (CSR) disclosure research in Indonesia from 2008 to the beginning
of 2020 and offer suggestions for future research areas in their study. The
study used a qualitative bibliographic methodology and looked at 184 articles
from magazines accredited by Sinta and included in the Scopus database. The
majority of articles were printed in periodicals with Sinta 2 accreditation,
according to the data. The research’s primary focus was on the factors that
influence CSR disclosure, and legitimacy and stakeholder theories that were
employed as common theoretical frameworks. Based on the findings of this
study, the following discussion will examine the applicability and implications
of legitimacy and stakeholder theory.
According to Kumala and Siregar (2021), legitimacy and stakeholder
theories are the cornerstones for developing hypotheses regarding why
businesses engage in CSR. Stakeholders comprise shareholders, creditors,
suppliers, governments, consumers, and other parties that can influence a
company’s objectives. Gras-Gil, Palacios Manzano and Hernández Fernández
(2016) study on stakeholder theory emphasizes the need for businesses to

benefit stakeholders, maintain positive relationships with them, and incorporate
them into their corporate strategy. Then, as proposed by Brown and Deegan
(1998) and Suchman (2014), the legitimacy theory, frequently referenced in the
literature for Corporate Social Responsibility Disclosure (CSRD), asserts that
organizations operate within the constraints of society. According to Deegan,
Rankin and Tobin (2002), a corporation attains legitimacy when it conforms to
societal standards. This concept explains why enterprises inform the market
despite not being required to do so by law.
According to Solikhah's (2016) assertion of the legitimacy theory, firms
must persuade their stakeholders that their behavior is acceptable while
operating in a constantly changing external context. It was also said that the
legitimacy theory concentrates on how communities and businesses interact
and how a company’s legitimacy may be considered something given to or
desired by the community. It used legitimacy theory to provide a theoretical
framework for understanding why companies disclose their CSR initiatives and
how the size and sensitivity of the industry affect this involvement.
A study by Asis, Haq and Afiat, shows that CSR, which has historical roots in
the Cold War era of 1945–1960, focuses on corporations’ responsibilities to the
community and societal welfare. According to International Organization for
Standardization (ISO), CSR should include transparent and ethical behaviour
for sustainable development, health, and welfare that are integrated throughout
the organisation and should obligate businesses’ commitment to society’s well-
being and responsibility. As it affects corporate image and competitiveness in
the current global competitive environment, CSR is needed for organisations
to achieve certification standards for business entities, CSR standards, and
products. Regional variations exist in CSR methods in any case.
“Corporate Social Responsibility” or “Social Responsibility of a Corporation”
was first used in American business ethics literature. An organisation,
particularly a sizable organization, is called a corporation in Indonesia.
Despite being constitutionally founded to serve the public good and not for
profit, the company earns money as it grows (for profit). According to Kotler
and Lee (2004), a commitment to enhancing community well-being through
independent business practices and corporate resource donations is known as
“corporate social responsibility”. Large-scale companies involved in CSR must
be committed to enhance local welfare through independent business practices
and corporate resource aids that follow the abovementioned standards.
Additionally, Indonesian corporate social responsibility (CSR) is governed
by Law Number 40 of 2007 concerning limited liability companies. The definition
of CSR under the rule includes both social and environmental responsibility.
The law entails that companies dealing with or connected to natural resources
must uphold their social and environmental commitments (OJK, 2007).
Triple Bottom Line Concept

The term “corporate social responsibility” (CSR) has been used since 1970s. It
has become more well-known, particularly following the release of Elkington's
book entitled Cannibals with Forks: The Triple Bottom Line in 21st Century
Business (1997). The Triple Bottom Line concept was developed alongside the
notion of sustainable development, and Elkington framed CSR as the primary
focus of 3P—i.e., Profit, People, and the Planet. A business that is intended to
be sustainable will prioritise social welfare and environmental sustainability in
addition to economic rewards (profit) for its people.
The Triple Bottom Line is the foundation upon which today's concept of
CSR was created. The current notion of CSR is the foundation for the significant
concept of “sustainable development”. A long history of interdisciplinary
collaboration exists between CSR and sustainable development. Based on the
Triple Bottom Line idea (sustainability), businesses that consider the interests
of both groups of stakeholders—who frequently have conflicts of interest—
can generate long-term profit. According to the Triple Bottom Line hypothesis,
environmental conditions and existing and future stakeholder groups affect a
corporation’s profitability (Aziz et al., 2010).
Using the African setting as an example to demonstrate the limits of
Carroll’s CSR Pyramid as a framework for understanding CSR, Visser (2005)
used Archie Carroll’s Corporate Social Responsibility (CSR) Pyramid as a visual
depiction of CSR in the African environment. Four stages comprising Carroll’s
Pyramid of Corporate Social Responsibility released in 1991. The first tier—
i.e., economic responsibility—includes a company’s duty to give its workers a
stable job and fair pay and to create goods and services that satisfy society’s
needs while being profitable. The second tier—i.e., legal responsibility—calls on
business to operate according to all applicable laws and regulations. The third
layer—i.e., ethical responsibility—deals with a company’s moral obligation
to act justly and morally, regardless of outside pressure, and to solve social
and environmental challenges brought on by its economic operations. Lastly,

the highest level is philanthropic responsibility, which involves a company’s
voluntary commitment to help people through charitable endeavors without
compulsion or payment.
Figure 1. The Pyramid of Corporate Social Responsibility (Visser, 2005)
Source: Visser, W. (2005)

Furthermore, Kotler and Lee (2004) defined six distinct types of
programmes that companies may implement for social impact. They may
choose the types of these six distinct CSR programmes by evaluating their
objectives, programme structure, potential benefits, and activity phases.
These organizations comprise cause promotion, which employs persuasive
communications to generate interest in and awareness of emerging social issues.
A corporation’s commitment to donate a portion of its profits to charitable
organizations is an example of cause marketing. Supporting, developing, and
implementing behavior change initiatives to improve community welfare and
health are what corporate social marketing entails. Corporate philanthropy
involves company’s direct contributions to charitable organizations. As a form
of social engagement, community volunteering requires businesses to support
the community and environment through volunteer labour. Such support could
be in the form of knowledge, skills, concepts, or laboratory equipment.
Socially responsible business practices are businesses' actions and
financial commitments to improve human welfare and preserve or protect the
environment. Examples include the construction of waste treatment facilities,
the careful selection of suppliers, the use of eco-friendly packaging, and similar
measures.
Empirical Studies on CSR and Community Perception

According to the study by Swain and Babu's (2021) on CSR and perception
on it, the coal company in the Korba Coal Fields of Chhattisgarh (India) has
been providing various services to the people under the banner of corporate
social responsibility (CSR) to establish a feel-good factor and develop the
social relationship between the coal firm and the neighbouring villagers. The
organisation offers first-rate services related to village infrastructure and health.
The research issue, on the other hand, is about harmonising the company's
perception of the benefits appropriated by the villagers with the actual benefits
that the villagers receive through CSR initiatives. As a result, this essay aims to
investigate the gap, if any, between perception and reality about the benefits
of CSR initiatives. Primary data from 500 villagers and 17 industry leaders are
analysed using ANOVA and regression. The coal company's perception of the
benefits obtained by the villagers is discovered to be equal.
Shim and Kim's (2021) research analysed how perceptions of corporate
dishonesty from corporate social responsibility efforts link the public's ethical
philosophy to subsequent positive/negative opinion-sharing intentions. The
study empirically tests a theoretical model of perceived corporate hypocrisy
with two causal antecedents (i.e., the personal moral philosophy of deontology
and consequentialism) and the mediating role of corporate hypocrisy between
such antecedents and the public's subsequent communication intentions
(i.e., positive and negative opinion-sharing intentions) towards a firm. The
findings show that corporate hypocrisy strongly mediates between personal
ethical orientations and the public's communication intention based on ethical
attribution of crisis-related corporate social responsibility actions.

RESEARCH METHOD
The following constitutes the conceptual framework to address the research
questions raised in light of the topic, the background, and the theoretical review
(Figure 2).
Figure 2. Conceptual Framework of Research
Perception of
Company Commitment
to Sustainability
Programme CSR (Y1)
H1
Perception of
Beneficiaries' H2,3
Satisfactions (X1)
H4 Perception of the
Suitability of CSR
Programmes with
Community Needs (Y2)
Hypothesis
The hypothesis that this paper will test is as follows:
H1: The satisfaction of the community receiving CSR assistance affects the
company's commitment to the sustainability of PT. NNT
H2: The company's commitment to the sustainability of PT. NNT affects the
suitability of community needs.
H3: The company's commitment to the sustainability of the CSR programme
mediates the influence between the satisfaction of the community receiving
CSR assistance and the suitability of PT. NNT and community needs.
H4: The satisfaction of the community receiving CSR assistance affects the
suitability of the CSR programme of PT. NNT with community needs.
SEM-PLS Method
Raza et al. (2020) conducted an Structural Equation Modelling-Partial Least
Square (SEM-PLS) study to evaluate the association between adopting corporate
social responsibility (CSR) initiatives and numerous aspects of customer
behaviour in a developing nation. Consequently, since perceptions are latent
variables in this research, SEM-PLS is used to evaluate its hypotheses (Raza et
al., 2020). The method of data analysis and hypothesis testing used in this study
will be carried out using the estimation method using the Structural Equation
Model-Partial Least Squares (SEM-PLS). The path analysis model of all latent
variables in the PLS consists of three relationships:
1. The inner model specifies the relationship between latent variables
(structural model);
2. The outer model specifies the relationship between latent variables and
their indicators or manifest variables (measurement model); and
3. Weight relation in which case values of latent variables can be estimated.
DATA
This research was conducted at Kecamatan Sekongkang, near PT. Newmont Nusa
Tenggara’s mining region (PT. NNT). Five of the seven villages in Kecamatan
Sekongkang were selected due to the mining operations that directly impact
them. Besides, those selected villages are the closest to the mining site. The
villages that serve as research sites are: (1) Desa Sekongkang Atas; (2) Desa
Kemuning; (3) Desa Sekongkang Bawah; (4) Desa Tongo Sejorong; and (5) Desa
Aik Kangkung (Satuan Pemukiman (SP)-1 Tongo)
This study’s sample population represents Sekongkang neighbourhood
residents who have profited financially from PT. Newmont Nusa Tenggara.
Respondents were selected based on the predominant main occupations or
community livelihood groups in the study area through purposeful sampling.
Respondents were interviewed during a focus group discussion meeting in the
Comdev office. The number of respondents is 30 based on their volunteering
to fulfill the questionnaire and the Comdev supervisor’s advice and agreement
with the research team. Quota sampling was then used to determine the
number of respondents by assigning allocations to each cluster and stratum
(community group based on the location of the research village). Consequently,
the number of respondents was determined in clusters (each village), with the
number of respondents for each stratification of community groups averaging
5-7 individuals per village. Thus, the total number of respondents representing
five villages who participated in the survey is 30 people.

The poll resulted in the general public opinions regarding how PT. Newmont
Nusa Tenggara has conducted its CSR programme. The general public’s
perception of corporate social responsibility, specifically their feelings about
PT. Newmont Nusa Tenggara’s implementation of CSR programmes to improve
community welfare through business practices and voluntary contributions of
company resources, is the variable in this study.
Respondent Characteristics
The respondents in this study were comprised of 30 people who were related
to PT. NNT's CSR programme in Desa Sekongkang. The characteristics of the
respondents in this study can be differentiated based on gender, age, community
development, education and type of businesses.
Table 1. Characteristics of Respondents
No Characteristics People %
1 Gender
Man 16 53.00
woman 14 47.00
Total 30 100
2 Age (years old)
21-31 2 6.67
32-42 10 33.33
43-53 13 43.33
54-65 5 16.67
Total 30 100
3 Community Development (Lembaga Pendamping)
YOP 19 63.33
YPESB 11 36.67
Total 30 100
4 Education
Did not graduate from Elementary Level 2 6.67
Elementery Level (SD) 17 56.67
Junior High School (SMP) 4 13.33
Senior High School (SMA) 5 16.67
Diploma 1 2 6.66
Total 30 100
No Characteristics People %
5 Type of Businesses
Farmer 10 33.37
Grocery 7 23.33
Food stall 4 13.33
Tailor 1 3.33
Salon 1 3.33
Motorcycle Repairing 1 3.33
Cake Stall 1 3.33
Chicken Farmers 1 3.33
“Bakulan” 1 3.33
Contractor 1 3.33
Farm Tool Seller 1 3.33
Sellers of Agricultural Products 1 3.33
Total 30 100
Perceptions About PT NNT's CSR Programme

Based on the survey results, the respondents' perceptions of the PT NNT’s CSR
Programme are as follows:
Table 2. Respondents' Perceptions About PT NNT's CSR Programme
Response from
Percentage
Perception Respondents*
1 2 3 4 5 1 2 3 4 5
Community Satisfaction (X)
PT. NNT is in accordance with 1 0 0 21 8 3.33 0.00 0.00 70.00 26.67
community expectations. (X1.1)
PT. NNT according to community 0 1 2 25 2 0.00 3.33 6.67 83.33 6.67
needs (X1.2)
The community is satisfied with 0 0 0 10 20 0.00 0.00 0.00 33.33 66.67
PT. NNT. (X1.3)
PT NNT Corporate Commitment (Y1)
The CSR programme is very useful in 0 0 0 20 10 0.00 0.00 0.00 66.67 33.33
developing community businesses.
(Y1.1)

Response from
Percentage
Perception Respondents*
1 2 3 4 5 1 2 3 4 5
Implementation of CSR PT. NNT is 0 0 0 24 6 0.00 0.00 0.00 80.00 20.00
right on target. (Y1.3)
Suitability of PT NNT's CSR Programme with Community Needs (Y2)
PT. NNT can help the community's 0 0 1 21 8 0.00 0.00 3.33 70.00 26.67
economic development. (Y2.1)
Implementation of CSR PT. NNT in 0 0 0 21 9 0.00 0.00 0.00 70.00 30.00
the community's economic business
is already good so it needs to be
maintained and continued. (Y2.2)
* Scores for each question item are as follows: 1 = Disagree (T.S.), 5 = Strongly Agree (S.S.).
PARTIAL LEAST SQUARE (PLS) ANALYSIS

Measurement Model Evaluation Results (Outer Model)
The first PLS procedure is to evaluate the measurement model. The design of
the measurement model involves the relationship between constructs and their
variables to form latent variables. The first stage is to estimate the parameters
by calculating the PLS Algorithm. The PLS Algorithm calculation results are
shown in Figure 3.
Figure 3. PLS Algorithm Output
Figure 3 above explains the results of the PLS Algorithm which tests each
indicator against the construct, namely in terms of forming a good variable
construct. Whether the results of the PLS Algorithm can be said to be feasible
for further testing of the structural model (inner model), the measurement
model must be evaluated first as follows.
Convergent Validity Testing

The convergent validity test in this study refers to the opinion of Chin
(1998) that the rule of thumb criteria that can be used as a basic provision in
determining the convergent validity of PLS analysis is to look at the value of the
outer loadings, which is also called the loading factor of more than equal to 0.70
(outer loadings ≥ 0.70); and the Average Variance Extracted (AVE) value is more
than equal to 0.50 (AVE ≥ 0.50). The results of convergent validity testing based
on outer loadings and AVE values are presented in Table 3.
Table 3. Outer Loadings
Community PT NNT Corporate Suitability of PT NNT's CSR Programme
Satisfaction (X) Commitment (Y1) with Community Needs (Y2)
X1.1 0.811
X1.2 0.675
X1.3 0.722
Y1.1 0.418
Y1.2 0.816
Y1.3 0.906
Y2.1 0.921
Y2.2 0.933
The results of the convergent validity test show that each indicator used
in forming the latent variable of this study has an AVE value greater than 0.70,
except for indicator Y1.1 with a value of 0.418, which is less than 0.5 so that the
output results are printed in red. However, based on the criteria as quoted from
Hair Jr et al. (2016) "Generally, indicators with outer loadings between 0.40
and 0.70 should be considered for validity" and images of outer loadings for
variable X, Y1 and Y2 which show a value of more than 0.40 are all accepted for
further testing. These results indicate that the measurement model meets all
the convergent validity testing criteria and the measurement model can be said
to have good convergent validity.

Discriminant Validity Testing
The test for discriminant validity in this study refers to the opinion of Chin
(1998), which states that the value of discriminant validity can be known from
the results of the cross loadings of the construct itself with a comparison of the
values of cross loadings in other constructs. The model will have appropriate
discriminant validity if the value of the cross loadings in the construct itself is
greater (has a greater correlation value) than the cross loadings in the other
constructs dominantly. The two decisions related to discriminant validity are
explained in detail as follows:
1. If the cross loading value of each indicator in its own construct is dominantly
greater than the cross loading value of each indicator in another construct,
then the latent variable has fulfilled the discriminant validity test.
2. If the cross loading value of each indicator in its own construct is not
dominantly greater than the cross loading value of each indicator in
another construct, then the latent variable does not meet the discriminant
validity test. The results of the discriminant validity test based on the cross
loading values are shown in the following figure:
Table 4. Discriminant Validity Test with Cross Loadings Values for
Variables X, Y1 and Y2
Community PT. NNT Corporate Suitability of PT. NNT's CSR Programme
Satisfaction (X) Commitment (Y1) with Community Needs (Y2)
X1.1 0.811 0.338 0.329
X1.2 0.675 0.316 0.443
X1.3 0.722 0.252 0.258
Y1.1 0.095 0.418 0.198
Y1.2 0.350 0.816 0.457
Y1.3 0.393 0.906 0.829
Y2.1 0.454 0.675 0.921
Y2.2 0.444 0.743 0.933
Table 4 presents the result of the discriminant validity test based on the
cross loading criteria for each indicator in each construct. These results indicate
that the cross loading value of each indicator in its own construct is dominantly
greater than the cross loading value of each indicator in other constructs.
Decisions that can be made based on these results are that all latent variables in
this study have fulfilled the discriminant validity test.
PLS Reliability Testing

Reliability testing in PLS analysis can be measured using two methods, namely
Cronbach's Alpha and Composite Reliability. Cronbach's Alpha value must be ≥
0.50 and Composite Reliability must be ≥ 0.70 with a lower limit ≥ 0.60 in order
to be reliable (Hair et al., 2017). Constructs that meet these criteria indicate that
each indicator that forms the construct is reliable for further testing. The results
of reliability testing based on Cronbach's Alpha and Composite Reliability values
are shown in the following Table 5.
Table 5. Cronbach’s Alpha and Composite Reliability
Cronbach's Alpha Composite Reliability
Community Satisfaction (X1) 0.589 0.781
PT NNT Corporate Commitment (Y1) 0.620 0.774
Suitability of PT NNT's CSR Programme
0.837 0.924
with Community Needs (Y2)
Table 5 shows the result of the PLS reliability test based on Cronbach's
Alpha and Composite Reliability criteria. These results indicate that the value
of Cronbach's Alpha on variables X, Y1 and Y2 is greater than 0.50. Meanwhile,
the Composite Reliability value for each variable is greater than 0.70. Decisions
that can be made based on these results are that each indicator that forms the
construct is reliable and can be used for further testing.
RESULTS OF STRUCTURAL MODEL EVALUATION (INNER

MODEL) HYPOTHESIS TEST
The hypotheses in this study amounted to four hypotheses, in the PLS analysis
to test the hypothesis, the bootstrapping method was used to see the direct and
indirect effects of each latent variable that has been connected according to the
research hypothesis. The result of this bootstrapping is in the form of a PLS path
diagram. The bootstrapping results are shown in Figure 4.

Figure 4. Bootstrapping Output
Table 6. Path Coefficients Result

Original Sample Standard
T Statistics P
Sample Mean Deviation
(|O/STDEV|) Values
(O) (M) (STDEV)
Community Satisfaction (X) → PT
0.418 0.458 0.135 3.099 0.002
NNT Corporate Commitment (Y1)
Community Satisfaction (X)
→ Suitability of PT NNT's CSR
0.198 0.194 0.128 1.549 0.122
Programme with Community
Needs (Y2)
PT NNT Corporate Commitment
(Y1) → Suitability of PT NNT's
0.683 0.697 0.119 5.735 0.000
CSR Programme with Community
Needs (Y2)
Based on the picture above it appears that the first hypothesis (H1) and
second (H2) are accepted because the P Values are smaller than 0.05. But the
fourth hypothesis (H4) is rejected because the P Values are 0.122 greater than
0.05, meaning that there is no direct influence between the variables X and Y2.
Partial Least Square (PLS) analysis also has the ability to measure the effect of
latent variables indirectly (indirect effects). The indirect effect is shown in Table
7.
Table 7. Specific Indirect Results Output
Original Sample Standard
T Statistics P
Sample Mean Deviation
(|O/STDEV|) Values
(O) (M) (STDEV)
Community Satisfaction (X) ->
PT NNT Corporate Commitment
(Y1) -> Suitability of PT NNT's 0.286 0.317 0.107 2.677 0.008
CSR Programme with Community
Needs (Y2)
As indicated in Table 7, there are results of the indirect effect test (H3).
The indirect effect of the satisfaction of the community receiving CSR assistance
(X) on the suitability of the CSR programme of PT. NNT with community needs
(Y2), through the Company's Commitment to CSR Programme Sustainability
(Y1) produces a p value of 0.008 which is smaller than 0.05. These results
indicate that the Satisfaction of Community Recipients of CSR Assistance (X)
has a significant effect on the Conformity of CSR Programmes with Community
Needs (Y2) through the Company's Commitment to the Sustainability of CSR
Programmes (Y1). Based on the bootstrapping analysis results show that:
1. The first hypothesis (H1) is accepted, so it is stated that there is an influence
between the satisfaction of the community recipients of CSR assistance
from PT. NNT (X) with the company's commitment to the sustainability of
the CSR programme (Y1).
2. The second hypothesis (H2) is accepted, so it is stated that there is an
influence between the company's commitment to the sustainability of the
CSR programme (Y1) on the suitability of the CSR programme with the
needs of the community that has been carried out by PT. NNT (Y2).
3. The third hypothesis (H3) is accepted, so it is stated that there is an influence
between the satisfaction of the CSR beneficiaries of PT. NNT (X) with the
company's commitment to the sustainability of the CSR programme as
a mediating variable (Y1) on the suitability of PT. NNT with community
needs (Y2).
4. The fourth hypothesis (H4) is rejected, so it is stated that there is no
influence between perceptions about the satisfaction of the recipients
of CSR assistance from PT. NNT(X) with the suitability of PT. NNT with
community needs (Y2).

Coefficient of Determination (R2)
The results of the Partial Least Square (PLS) analysis for the coefficient of
determination are shown in Table 8.
Table 8. Coefficient of Determination (R2)
R Square
PT NNT Corporate Commitment (Y1) 0.175
Suitability of PT NNT's CSR Programme withCommunity Needs (Y2) 0.620
The R Square value indicates the magnitude of the influence variable in

explaining the affected variable. Endogenous variable Company commitment
to the sustainability of PT. NNT's CSR Programme (Y1) has a coefficient of
determination of 0.175 (17.5%) from the influence of the variable Perceptions
about Satisfaction of Community Recipients of PT. NNT's CSR Assistance (X).
These results indicate that X is only able to explain Y1 by 17.5%, while the
remaining 82.5% is explained by other variables that are not included in this
research model.
The endogenous variable Company Commitment to the Sustainability of
the CSR Programme (Y2) has a coefficient of determination of 0.620 (62.0%)
from the influence of the exogenous variable Perceptions about Satisfaction
of Community Recipients of PT. NNT CSR Assistance (X) and the endogenous
variable Perceptions about the Compatibility of PT. NNT's CSR Programme with
Community Needs (Y2). These results indicate that X and Y1 are able to explain
Y2 by 62%, while the remaining 38% is explained by other variables that are
not included in this research model.
Predictive Relevance (Q2)

Predictive relevance (Q2) is one of the tests in the structural model (inner
model) that serves to show how well the research model consisting of latent
variables Perception of Community Satisfaction of CSR Assistance Recipients
(X); Perception of Company Commitment to CSR Programme Sustainability
(Y1), and Perception of CSR Programme Conformity to Community Needs (Y2) in
forming a fit model (good model). A predictive relevance value greater than zero
(Q2 ≥ 0) indicates that the exogenous latent variable has predictive relevance on
the endogenous latent variable. In contrast, if the predictive relevance value is
less than zero (Q2 < 0), the exogenous latent variable does not have predictive
relevance to the endogenous latent variable. Chin (1998) suggested the formula
for determining predictive relevance in PLS analysis as follows:
Q2 = 1 – (1 – R2 endogen1) (1 - R2 endogen2) ... (1 - R2 endogenn)
Based on this formula, the predictive relevance (Q2) value in this study can
be calculated as follows:
Q2 = 1 - (1 - R2 endogen Y2) (1 - R2 endogen Y1)
Q2 = 1 - (1 - 0,620) (1 - 0,175)
Q2 = 1 - (0,380)(0,825)
Q2 = 1 - (0.3135)
Q2 = 0.6865
The results of the predictive relevance calculation show that the Q2 value is
0.6865, greater than 0 (Q2 ≥ 0). This result indicates that the exogenous latent
variables have predictive relevance on the endogenous latent variables and the
research model consisting of the latent variables of Perception of Satisfaction
of CSR Assistance Recipients (X), Perception of Company Commitment to CSR
Programme Sustainability (Y1), and Perception of CSR Programme Conformity
with Community Needs (Y2) can form a fit model (good model).
CONCLUSION
PT. NNT carries out its corporate social responsibility (CSR) programme in the
economic sector by working with two foundations it established, namely the
West Sumbawa Economic Development Foundation (YPESB) and the Olat Parigi
Foundation (YOP). The former supports community economic development
initiatives, such as aloe vera, coffee, and seaweed production and packaging,
providing agricultural training, developing red rice cultivation, promoting
organic farming, and constructing warehouses to support rice marketing. It also
collaborates with farmers in Desa Aik Kangkung to cultivate Situbagendit and
Inpari 13 rice types.
Meanwhile, YOP focuses on improving the welfare of the mining community
through community empowerment programmes that cover agriculture, animal
husbandry, fisheries, small business development, and microfinance. YOP offers
collateral-free and interest-free capital loans ranging from IDR1,500,000 to
IDR10,000,000 to farmers, breeders, and traders in the community.

The CSR programme of PT. NNT in Sekongkang has demonstrated progress
and development in crucial areas such as infrastructure, education, health,
agriculture, and small to medium-sized businesses. This positive growth is
reflected in the favourable perception of the programmes among the respondents
who have received assistance. The respondents appreciated the programme's
provision of interest-free funding and skill training, which they found extremely
helpful. In comparison, the respondents acknowledged the benefits of PT. NNT's
CSR programme and offered constructive criticism for future improvements.
They hoped the programme's implementation would continue with specific
enhancements. Based on the respondents' feedback, the Comdev management
of PT. NNT, YOP, and YPESB may consider the following suggestions:
To ensure that the community's viewpoints and requirements are
considered while creating the CSR programme, it is suggested that P.T. NNT
engage the community by appointing representatives from each group during
the planning process. Additionally, YOP should accelerate the establishment
of the Savings and Loan Cooperative, which would provide community micro-
credit loans exceeding IDR10,000,000. It is therefore necessary to improve
accessibility to the cooperative. It is also recommended that the service access
system be linked with the Post Office or BRI Bank, making it more accessible to
a broader community.
To ensure further losses from purchasing farmers' grain with sufficient
storage facilities, YPESB should prioritise constructing warehouses and drying
floors to accommodate the capacity of agricultural products. Additionally,
YOP management must investigate indications that some customers feel the
microcredit distribution is unfair as PT. NNT's CSR programme ends. Its partner
NGOs in the economic sector—i.e., YPESB, and YOP—must have a strategic
plan to continue implementing the programmes with the new company taking
over. This will ensure a smooth transition and readiness to sustain the progress
achieved during PT. NNT's CSR programme.
REFERENCES
Asis, P.H., M.N. A. Haq, & M.N. Afiat (no date) Corporate Social Responsibility
(CSR) and Sustainable Community Development in Mining Sector: Case
Study in Central Konawe and North Konawe Regency Southeast Sulawesi,
Indonesia, academia.edu. Available at: https://www.academia.edu/
download/56143031/Patta_Hindi.pdf.
Aziz, I.J. et al. (2010) Pembangunan Berkelanjutan Peran dan Kontribusi Emil
Salim, Kepustakaan Populer Gramedia.
Brown, N. & C. Deegan. (1998) ‘The public disclosure of environmental
performance information - A dual test of media agenda setting theory and
legitimacy theory.’ Accounting and Business Research, 29(1): 21–41. doi:10
.1080/00014788.1998.9729564.
Burhanudin, B. et al. (2023) ‘Corporate Social Responsibility of Mining
Companies in the District of West Sumbawa (Case Study of Mining Company
PT. Amman Mineral Nusa Tenggara, AMNT).’ International Journal of Social
Science and Education Research Studies, 03(01): 130–141. doi:10.55677/
ijssers/v03i1y2023-16.
Chawa, A.F. (2014). Mining and Community Development in Indonesia : A Case
Study. Victoria University.
Chin W, M.G. (1998). ‘The Partial Least Squares Approach to Structural
Formula Modeling.’ Advances in Hospitality and Leisure, 8(2)(January
1998): 5. Available at: https://books.google.com/books?hl=en&lr=&id
=EDZ5AgAAQBAJ&oi=fnd&pg=P A295&dq=The+partial+least+squares
+approach+to+st ructural+equ ation+mode ling&ots=49uH6qt2lk&sig=
Fwg2GGFWp3LUMMjxMu9h4jbOXnA.
Deegan, C., M. Rankin, & J. Tobin (2002). 'An examination of the corporate
social and environmental disclosures of BHP from 1983-1997: A test
of legitimacy theory.' Accounting, Auditing & Accountability Journal.
doi:10.1108/09513570210435861.
Duarte, F. (2011). ‘What Does A Culture of Corporate Social Responsibility “Look”
Like? A Glimpse into a Brazilian Mining Company.’ International Journal of
Business Anthropology, 2(1), 106–123. doi:10.33423/ijba.v2i1.1190.

Elkington, J. (1997). Cannibals with Forks The Triple Bottom Line of 21st Century
Business. Capstone Publishing Limited.
Erwin Syarif, A. & T. Hatori (2017). ‘Corporate social responsibility for regional
sustainability after mine closure: A case study of mining company in
Indonesia.’ IOP Conference Series: Earth and Environmental Science, 71(1):
1–13. doi:10.1088/1755-1315/71/1/012029.
Firmansyah, A. et al. (2020). ‘Corporate Social Responsibility Disclosure in
Indonesia: A Bibliographic Study.’ International Journal of Sustainable
Development & World Policy, 9(2): 91–121. doi:10.18488/journal.26.2020.
92.91.121.
Gifford, B., A. Kestler, & S. Anand (2010). ‘Building local legitimacy into corporate
social responsibility: Gold mining firms in developing nations.’ Journal of
World Business, 45(3): 304–311. doi:10.1016/j.jwb.2009.09.007.
Gras-Gil, E., M. Palacios Manzano, & J. Hernández Fernández (2016).
‘Investigating the relationship between corporate social responsibility
and earnings management: Evidence from Spain.’ BRQ Business Research
Quarterly, 19(4): 289–299. doi:10.1016/j.brq.2016.02.002.
Hair Jr, J.F. et al. (2016). A Primer on Partial Least Squares Structural Equation
Modeling (PLS-SEM). New York, US: Sage Publicatios, Inc.
Kotler, P. & N.R. Lee (2004). Corporate Social Responsibility: Doing the Most Good
for Your Company and Your Cause. New York, US: John Wiley & Sons Inc.
Kumala, R. & S.V. Siregar (2021). ‘Corporate social responsibility, family
ownership and earnings management: the case of Indonesia’, Social
Responsibility Journal, 17(1): 69–86. doi:10.1108/SRJ-09-2016-0156.
Langston, J.D. et al. (2015). ‘Comparative development benefits from small and
large scale mines in North Sulawesi, Indonesia.’ Extractive Industries and
Society, 2(3): 434–444. doi:10.1016/j.exis.2015.02.007.
Libassi, M. (2022). ‘Gold conflict and contested conduct: Large- and small-scale
mining subjectivities in Indonesia.’ Geoforum [Preprint], (October 2021).
doi:10.1016/j.geoforum.2022.10.005.
Raza, A. et al. (2020). ‘An assessment of corporate social responsibility on
customer company identification and loyalty in banking industry: a
PLS-SEM analysis.’ Management Research Review, 43(11): 1337–1370.
doi:10.1108/MRR-08-2019-0341.
Shim, K. & J.N. Kim (2021). ‘The Impacts of Ethical Philosophy on Corporate
Hypocrisy Perception and Communication Intentions Toward CSR.’
International Journal of Business Communication, 58(3): 386–409.
doi:10.1177/2329488417747597.
Solikhah, B. (2016). ‘An overview of legitimacy theory on the influence of company
size and industry sensitivity towards CSR disclosure.’ International Journal
of Applied Business and Economic Research, 14(5): 3013–3023.
Suchman, M.C. (2014). ‘Managing Legitimacy: Strategic and Approaches.’ Academy
of Management Review, 20(3): 571–610. Available at: http://citeseerx.ist.
psu.edu/viewdoc/download?doi=10.1.1.108.2768&rep=rep1&type=pdf.
Suryantoro, S. & M.H. Manaf (2002). ‘The Indonesian Energy And Mineral
Resources Development And Its Environmental Management To Support
Sutainable National Economic Development.’ Foreign Direct Investment
and the Environment, (February): 7–8.
Swain, S.C. & A.A.Babu (2021) ‘Perception vis-à-vis realisation of benefits from
CSR activities: a study of such activities in Korba Coalfields of Chhatishgarh
(India).’ International Journal of Mining and Mineral Engineering, 12(4):
251–265. doi:10.1504/IJMME.2021.121319.
Visser, W. (2005). ‘Revisiting Carroll’s CSR Pyramid: An African Perspective.’ In
Corporate Citizenship in a Development Perspective. Copenhagen Business
School Press.

CHAPTER 12
Public Financial Governance and

Socio-Economic Performance:
Evidence from Local Government in
Indonesia
I Kadek Dian Sutrisna Artha, Usman, Dewinta I Sumartono,
and Irwan Amarullah
INTRODUCTION
Transparency and accountability are fundamental pillars of good governance,
as it plays a vital role in enhancing the welfare of citizens. Maan (2009) defines
transparency as the unrestricted access to information, while accountability
refers to the government's obligation to the public for its actions. Kumorotomo
(2019) emphasizes transparency as a crucial element of good governance,
essential for combating corruption and increasing government accountability
to the public.
In many countries, there is a growing interest in strengthening government
accountability and transparency, particularly to the public. Enhancing
government accountability and transparency reinforces efforts to improve
policy effectiveness, ultimately leading to enhanced welfare for the people.
In the context of decentralised government where local governments have
authority over regional finances, transparency and accountability to the local
public become even more pertinent. Transparent and accountable regional
257
governments should openly provide financial information, involving the public
in the formulation, implementation, and evaluation of government budget
policies (Hehanussa, 2015).
In the context of Indonesia, the country has shifted from an authoritarian
and centralistic system of government to a much more democratic and
decentralised system since 1999. The public demand for a transparent and
accountable government is very high. To fulfil this demand, both national and
regional governments produce annual Regional Government Financial Report
(Laporan Keuangan Pemerintah Daerah/LKPD). To ensure accountability and
transparency, these LKPDs are audited by the Supreme Audit Agency (Badan
Pemerintah Keuangan/BPK).
BPK oversees and audits government financial reports at both the national
and regional levels, conducting regular inspections of LKPDs. The audit results
produce a report that includes the auditor's professional opinion on the fairness
of government financial information. The opinion is based on various criteria,
such as adherence to accounting standards, adequate disclosures, compliance
with laws and regulations, and the effectiveness of internal controls. BPK's audit
opinion on LKPD can fall into one of four categories: Unqualified Opinion (Wajar
Tanpa Pengecualian/WTP), Qualified Opinion (Wajar Dengan Pengecualian/
WDP), Disclaimer of Opinion (Tidak Memberikan Pendapat/TMT), and Adverse
Opinion (Tidak Wajar/TW). The opinion of the BPK holds significant weight
and the trust of the public.
When a regional government receives the best audit opinion on its
LKPD, namely WTP, it indicates that it has met the aforementioned criteria
and is assumed to manage regional finances properly and in accordance with
regulations. There is an argument that the opinion of the Supreme Audit
Agency (Badan Pemeriksa Keuangan/BPK) is believed to be closely associated
with the socio-economic performance of a region. Effective regional financial
management is likely to have a positive impact on the social and economic
performance of the region. However, this association is a topic of debate in the
literature.
Previous studies by Suryaningsih and Sisdyani (2016) and Masdiantini
and Erawati (2016) focused solely on the impact of BPK's opinion status on
local government financial performance, without taking its influence on social-
economic performance into account. Understanding the relationship between
BPK's opinion status and social-economic performance is crucial, as the
evaluation of fiscal policy relies on indicators such as income per capita, the
Human Development Index, poverty rates, and inequality. Therefore, this paper
aims to empirically explore the effect of BPK's opinion status on social-economic
performance at the regional level in Indonesia, addressing the existing gap in
studies that have not examined this relationship at the district or city level.
Examining the association between BPK's opinion and social-economic
performance is important to demonstrate the link between accountable and
transparent government and the welfare of its citizens. Focusing our analysis
on Indonesia is also important, as the country is the world's largest archipelago
nation, comprising thousands of islands in Southeast Asia. Indonesia boasts a
rich cultural heritage, diverse landscapes, and a population of over 270 million
people.
The estimation in this study incorporates data on BPK's opinion on LKPDs
from 514 districts/cities between 2015 and 2019, along with socio-economic
indicators of district/city governments, such as Gross Regional Domestic
Product (GRDP) per capita, Human Development Index (HDI), poverty rate, and
inequality. In addition, government fiscal data—including local government
expenditures per capita, local government revenues per capita, and regional
transfers per capita—are utilised in the estimation. The analysis employs a
panel regression model to estimate the effect of BPK's opinion on LKPD.
The main findings of this study are as follows: BPK's audit opinion status
on a region's LKPD has a statistically significant impact on the region's socio-
economic performance. Furthermore, when examining Java and non-Java
regions separately, similar patterns emerge. Regions in Java with a WTP audit
opinion exhibit better socio-economic performance compared to regions with
other audit opinion statuses. In non-Java regions, it was observed that regions
with WTP audit opinions have a significant impact on the Human Development
Index, GRDP per capita, and poverty level, while not having a significant impact
on the Gini Ratio.
In this paper, we will first present a review of previous literature on the
impact of audit opinion on economic performance. Subsequently, we will outline
our research framework, describe the data and methodology employed in our
analysis, and finally, present the results and conclusions of our study.
Public Financial Governance and Socio-Economic Performance: 259

Evidence from Local Government in Indonesia
PREVIOUS STUDIES ON THE IMPACT OF AUDIT OPINION
ON ECONOMIC PERFORMANCE
International literature examining the relationship between government
accountability, transparency, and development, while not abundant, is
relatively accessible. One notable study by Abdolmohammadi and Tucker
(2002) explores this association across various countries worldwide. While
Pina, Torres, and Acerete (2007) investigate the impact of government
accountability improvement through the use of ICTs on development in OECD
countries, Petkovic and Rahman (2021) provide evidence of a significant
association between government accountability and development in several
young democratic nations. Other noteworthy works in this field include those
by Mechkova, Bernhard, and Luhrmann (2019) and Smoke (2015). These
studies collectively contribute to our understanding of the linkages between
government accountability, transparency, and developmental outcomes.
In the case of Indonesia, several existing studies have examined the
relationship between BPK’s Opinion Status and government financial indicators.
Studies conducted by Suryaningsih and Sisdyani (2016) and Masdiantini and
Erawati (2016) focus on the impact of BPK’s opinion status on LKPD on local
government financial performance. In contrast, other studies, such as those
by Akhmad Hafidzan and Dwi Martani (2014), Nur Anita and Rudy Badrudin
(2017), and the World Bank (2017), explore the influence of social, economic,
and financial indicators on the status of BPK’s opinion on LKPD. This report
aims to observe the effect of BPK’s audit opinion on LKPD on the socio-economic
performance of regions in Indonesia.
Several studies have found a positive influence of BPK’s audit opinion on
LKPD on the financial performance of regional governments. Masdiantini and
Erawati (2016) examined LKPD data from cities and regencies in Bali from
2008-2018 and BPK’s audit opinion from 2009-2014. They employed simple
linear regression to analyse factors affecting the financial performance of
district and city governments in Bali Province. One of the factors considered
was BPK’s audit findings and opinions on the LKPD of each district and city.
The study measured regional financial performance using four financial ratios:
independence, economic, efficiency, and effectiveness. The results demonstrated
that BPK’s audit opinion on LKPD has a significant positive effect on local
government financial performance. Taking a broader scope, Suryaningsih and
Sisdyani (2016) conducted a similar study using data from all districts and
cities in Indonesia from 2013. Their study, employing simple linear regression,
also revealed a significant positive effect of BPK’s audit opinion on the financial
performance of district and city governments.
On the other hand, several studies have examined socio-economic factors
influencing the status of BPK’s audit opinion on LKPD. Hafidzan and Martani
(2014) explored the influence of human development (measured by the Human
Development Index, education level, health, and welfare level) and political
factors on the status of BPK’s opinion on LKPD. Using data from 107 regional
governments from 2009-2011, they conducted an ordered logit regression
analysis. The findings indicate that human development has a positive impact
on the status of BPK’s opinion on LKPD, while political dynasty and releasing
LKPD close to regional head elections have a negative impact.
Meanwhile, Nur Anita and Rudy Badrudin (2017) investigated the effect of
local government financial performance on the status of BPK’s audit opinion on
LKPD in the Special Region of Yogyakarta. They utilised LKPD data from six sub-
regions within DI Yogyakarta province from 2001-2013 and employed Structural
Equation Model – Partial Least Squares for analysis. The study revealed a
positive and significant effect of local government financial performance on the
status of BPK’s opinion on LKPD.
Shifting to a slightly different topic, a 2017 World Bank study discovered that
local governments participating in the Local Government and Decentralisation
pilot project (Proyek Pemerintah Daerah dan Desentralisasi or “P2D2”) or the
pilot project for fund allocation (Dana Alokasi Khusus or “DAK”) experienced a
positive and significant impact on increasing accountability. From 2011 to 2014,
the Government of Indonesia, with support from the World Bank, implemented
the P2D2 or DAK pilot project, which operated on a performance-based approach.
The project utilised an output-based disbursement scheme that incentivized
local governments to achieve improved targets in terms of providing quality
public services. Participating local governments were required to adhere to
compliance standards encompassing financial management, procurement,
environmental considerations, social factors, and technical aspects.
Initially, the programme involved five provinces and 78 districts/
cities. Several findings related to accountability indicators emerged from the
participation of local governments in the P2D2 project are: (1) Local governments

had a 5.5% probability of obtaining or maintaining a WTP opinion; (2) The
probability of local governments receiving a WDP opinion was 2.1%; (3) The
probability of local governments not receiving a TMP opinion was 1.1%; and (4)
The probability of local governments not receiving a TW opinion or disclaimer
was 5.6%. Based on this analysis, it can be concluded that participation in the
pilot project had a positive and significant impact on local governments.
This chapter makes a valuable contribution by presenting new evidence
regarding the significant association between government accountability,
as measured by BKP’s opinion, and regional development. The data used
in this chapter is distinct in its validity, and the methodology employed is
relatively robust. By examining this relationship, the study aims to enhance
our understanding of the dynamics between government accountability and
regional development, shedding light on important aspects of governance and
its impact on socioeconomic progress.
RELATIONSHIP BETWEEN GOOD GOVERNANCE AND

ECONOMIC PERFORMANCE
Definition of governance are plenty. However, this paper will use the definition
developed by the World Bank—i.e., “the manner in which power is exercised in
the management of a country’s economic and social resources for development”
(Sen, 1994). Meanwhile, the concept of good governance spans many aspects
including financial, economic, and social. This chapter follows UN ESCAP (2009)
definition that good governance is characterised by accountability, transparency,
responsiveness, equitable and inclusivity, effectiveness & efficiency, following
the rule of law, participatory, and consensus oriented.
Figure 1 illustrates the interconnectedness between good governance and
economic performance. Under the umbrella of good governance, key indicators
include accountability, transparency, effectiveness, and efficiency. These
governance indicators are closely linked to the successful implementation of
sound financial practices, resulting in the effective utilisation of budgets. In
turn, this efficient budget utilisation has a direct impact on various economic
performance indicators, such as the Human Development Index, Gini ratio,
income per capita, and poverty level, among others. The figure visualises the
integral relationship between good governance and its influence on economic
outcomes.
Figure 1. The Relationship between Governance of a Region and Its
Financial and Economic Performance
In the Indonesian context, receiving a WTP audit opinion from BPK

indicates that a region has adhered to government accounting standards,
provided sufficient disclosures, complied with applicable laws and regulations,
and maintained an effective internal control system. Therefore, it can be inferred
that a region with a WTP audit opinion has implemented good governance
practices, as its financial reporting is considered accountable, transparent, and
in line with national accounting standards. However, it is worth noting that the
relationship between the governance of a region, its financial and economic
performance, may not be unidirectional. The level of social and economic
performance in a region is indicative of its development, and a more developed
region is expected to possess greater capacity and resources to manage its own
budget effectively and efficiently, thereby influencing the quality of governance
positively.
Building upon the framework depicted in Figure 1, this chapter aims
to examine the impact of BPK's Opinion Status on LKPD, which serves as an
indicator of regional governance, on the social and economic performance at
the district and city level. To address the potential endogeneity issue in the
estimation, this paper will employ the Hausman test to ascertain whether
the two-way relationship between good governance and social-economic
performance, as illustrated in the aforementioned framework, is statistically
significant. By considering this potential endogeneity, the study aims to provide
a comprehensive analysis of the relationship between BPK's Opinion Status,
regional governance, and the social-economic performance of districts and
cities.

OVERVIEW OF BPK'S OPINION STATUS ON REGIONAL
LKPDS IN INDONESIA
The number of cities and regencies achieving a WTP (Wajar Tanpa Pengecualian/
Unqualified Opinion) audit opinion status has been steadily increasing from
2015 to 2019 (Table 1). This study utilises data on BPK's audit opinion on
LKPD (Regional Government Financial Reports) from 514 cities and regencies
in Indonesia during the period of 2015-2019. The table below highlights the
progression of audit opinion results for these regions. Notably, the number of
cities and regencies obtaining a WTP status has consistently risen.
Table 1. Development of BPK's Opinion on Regional Financial Statements
in Indonesia (2015–2019)
Source: Ikhtisar Hasil Pemeriksaan Semester (IHPS) Badan Pemeriksa Keuangan (2021)
In 2015, there were only 281 regions with a WTP status, whereas by 2019,
this number had increased to 451 regions, reflecting a notable increase of 170
regions over the course of four years. This finding aligns with a simultaneous
decrease in the number of regions holding a WDP (Wajar Dengan Pengecualian/
Qualified Opinion) or TMP (Tidak Memberikan Pendapat?Disclaimer of Opinion)
status on their LKPD during the same four-year period. Between 2015 and 2019,
the number of regions with a WDP status decreased from 192 to 50, while the
number of regions with a TMP status decreased from 35 to only 6.
To examine the distribution of regions with a WTP opinion status on their
LKPD (Regional Government Financial Reports), we categorized the 514 cities
and regencies into two groups: Java and outside Java. Our analysis revealed that
the percentage of regions with a WTP opinion status in Java is higher than that
outside Java (see Figure 20). In 2019, approximately 93% of cities and regencies
in Java obtained a WTP opinion status, while outside Java, the percentage was
slightly lower at 86%. When considering the number of regions with a WTP
status over the years, both Java and outside Java exhibited an increase in the
percentage of regions with this status from 2015 to 2019. Conversely, the
percentage of regions with a WDP (Wajar Dengan Pengecualian) opinion status
experienced a significant decline from 2015 to 2019, both in Java and outside
Java.
Figure 2. Development of BPK’s Audit Opinions on Regional Government
Financial Reports, Java vs. Non-Java regions. (Left: Java regions; Right:
Non-Java regions)
Furthermore, we conducted a comparison between the number of cities

and regencies with a WTP (Wajar Tanpa Pengecualian) status (Figure 3). Our
analysis revealed that the percentage of cities with a WTP status is slightly higher
than the percentage of districts with the same status. In 2019, approximately
89% of cities obtained a WTP status on their LKPD (Regional Government
Financial Reports), while the percentage of districts with a WTP status was
slightly lower at 87%. As the number of regions with a WTP status has been
increasing, the percentage of cities with a WDP (Wajar Dengan Pengecualian)
status experienced a significant decline from 31% in 2015 to merely 6% in 2019.
These findings indicate an improvement in the quality of regional financial
reporting across cities and regencies.

Figure 3. Development of BPK’s Audit Opinions on Regional Government
Financial Reports, Cities vs. Regencies. (Left: Cities; Right: Regencies)
In general, there was a notable increase in the number of regions that
achieved or maintained a WTP (Wajar Tanpa Pengecualian) opinion status from
2015 to 2019, with a total of 451 regions or 89.5% of the overall number of local
governments. Among the cities and regencies, 184 regions or 37% witnessed
an improvement in their BPK's (Supreme Audit Agency) audit opinion status
during the same period (see Figure 4).
Figure 4. Number of Regions Based on Changes in BPK Opinion Between
2015 and 2019
Notes: There are 7 missing regional governments
Conversely, 63% of cities and regencies, equivalent to 320 regions, did not
experience any changes in their BPK's audit opinion over the four-year period.
It is important to note that there were three regions that experienced a decline
in BPK's opinion status from 2015 to 2019: two regions saw a shift from WTP to
WDP (Wajar Dengan Pengecualian), and one region experienced a decline from
WTP to TMP (Tidak Memberikan Pendapat).
THE RELATIONSHIP BETWEEN BPK'S OPINION STATUS

ON LKPD AND SOCIO-ECONOMIC PERFORMANCE
As previously mentioned, our hypothesis posits that BPK's (Supreme Audit
Agency) audit opinion on a region's LKPD (Regional Government Financial
Reports) has a positive impact on the region's socio-economic performance.
In this study, we employ several indicators to measure socio-economic
performance, including the Human Development Index (HDI), GRDP (Gross
Regional Domestic Product) per capita, poverty level, and Gini Ratio.
Based on our analysis using combined data from regions across Indonesia
between 2015 and 2019, we find that regions with better BPK audit opinion
status tend to exhibit higher HDI scores (see Figure 5). The group of regions
with TMP (Tidak Memberikan Pendapat) status has the lowest average HDI score
at 59.8, while regions with WDP (Wajar Dengan Pengecualian) status have an
average HDI score of 64.8, and regions with WTP (Wajar Tanpa Pengecualian)
status have the highest average HDI score at 69.9.
Figure 5. Average HDI and Poverty Level Based on BPK’s Opinion
Source: Badan Pusat Statistik and Badan Pemeriksa Keuangan (2021)

Furthermore, regions with better BPK opinion status tend to have lower
poverty levels, as measured by the percentage of people living in poverty. The
average poverty rate varies across different BPK opinion statuses, with regions
having WTP status exhibiting a lower average poverty rate compared to regions
with WDP and TMP statuses. Specifically, regions with WTP status have an
average poverty level of 11.2%, while regions with WDP and TMP statuses
have average poverty levels of 16.8% and 22.1%, respectively. These findings
suggest a correlation between a region's BPK audit opinion status and its socio-
economic indicators, emphasizing the potential influence of good governance
on socio-economic outcomes.
In line with the findings for HDI and poverty level, our analysis indicates
that regions with better BPK's (Supreme Audit Agency) audit opinion status tend
to exhibit higher GRDP (Gross Regional Domestic Product) per capita values,
which often serve as a proxy for income per capita. The group of regions with
TMP (Tidak Memberikan Pendapat) and WDP (Wajar Dengan Pengecualian)
status have lower average GRDP per capita values of 24.9 million/capita/year
and 25.6 million/capita/year, respectively, compared to the group of regions
with WTP (Wajar Tanpa Pengecualian) status, which has an average GRDP per
capita value of 37.7 million/capita/year (see Figure 6).
While the relationship between BPK's audit opinion status and indicators
such as HDI, poverty level, and GRDP per capita is evident, the association
between BPK's audit opinion status and income inequality, as represented
by the Gini Ratio, is not as clear. The Gini Ratio, which measures economic
inequality on a scale of 0 to 1, with higher values indicating greater inequality,
does not exhibit a consistent pattern across different opinion status groups. The
average Gini Ratio for regions with TMP and WDP statuses tends to be smaller,
suggesting lower income inequality in these regions. Conversely, regions with
WTP status have a higher average Gini Ratio compared to the other two groups,
indicating relatively higher income inequality. These findings indicate that the
relationship between BPK's opinion status and income distribution inequality
in regions is not straightforward. In the next section, we will delve into a panel
regression model analysis to explore the impact of BPK's opinion status on
socio-economic indicators, including the Gini Ratio.
Figure 6. Average GDRP per Capita and Gini Ratio based on BPK’s Audit
Opinion

To further explore the impact of BPK's (Supreme Audit Agency) audit
opinion status on social and economic indicators, we examined the average
indicators for regions within two categories: Java and outside Java. Our
findings indicate a positive correlation between BPK's audit opinion status and
the Human Development Index (HDI) in both categories (see Figure 7). This
suggests that regions with higher BPK opinion status tend to exhibit better
social development outcomes.
In terms of poverty levels, the correlation between BPK's audit opinion
status and the percentage of people living in poverty (poverty level) in the
region is negative (Figure 7). This negative correlation is particularly evident in
regions outside Java, while the poverty gap in Java is relatively small. This implies
that regions with higher BPK opinion status tend to have lower poverty rates,
indicating a potential link between good governance and poverty reduction.
Figure 7. Average HDI and Poverty Level Based on BPK’s Audit Opinion
and Region

As with economic indicators, Figure 8 illustrates a positive relationship
between BPK's audit opinion status and Gross Regional Domestic Product
(GRDP) per capita in both Java and outside Java. This implies that regions with
better BPK opinion statuses tend to exhibit higher levels of economic output
per person. However, when examining the relationship between BPK's audit
opinion status and the Gini Ratio, which measures income inequality, the results
are less clear for both Java and outside Java regions (Figure 8). This suggests
that the influence of BPK's opinion status on income distribution equality is not
as pronounced or consistent.
In summary, our analysis reveals a positive correlation between BPK's
audit opinion status and indicators such as HDI and GRDP per capita, indicating
better social and economic performance in regions with higher BPK opinion
status. Furthermore, a negative relationship between BPK's audit opinion
status and poverty levels, particularly in regions outside Java, is also revealed.
However, the impact of BPK's opinion status on income inequality, as reflected
by the Gini Ratio, is less discernible.
Figure 8. Average GDRP per capita and Gini Ratio based on BPK’s Audit
Opinion by Region
PANEL DATA REGRESSION

We employed the panel data regression method to examine the relationship
between BPK's audit opinion on LKPD and regional social and economic
performance. The dataset used in this analysis includes information from 514
cities and regencies in Indonesia within the period from 2015 to 2019. Our
study focuses on several dependent variables, namely the Human Development
Index, Poverty rate, Gini Ratio, and GRDP per capita, which serve as indicators of
economic and social performance. In addition, we considered the independent
variables, including a dummy variable representing the status of the BPK's
audit opinion (WTP = 1; other than WTP = 0), as well as regional financial data
such as capital expenditures per capita, direct expenditures per capita, regional
income per capita, and transfer funds per capita. The panel estimation model is
represented by the following formula:
Yij = β0+β1 OPINIONij +β2 Xij+ +εij
i : index for the 514 districts and cities
j : index for year (from 2015 to 2019).
Y : dependent variables which consist of the social economic indicators, such
as Human Development Index (IPM), Poverty rate, Gini Ratio, and GRD P per
capita.
OPINION: dummy variable of the status of the BPK’s audit opinion (WTP = 1;
other than WTP = 0)
X : control variables which consist of government per capita spending,
government per capita capital spending, government per capita revenue,
and government per capita transfer.
By employing this panel estimation model, we aim to gain insights into the
relationship between BPK's audit opinion status and regional social and economic
performance, while controlling for various regional financial factors. The results
of this analysis will be discussed in the subsequent section of this report.
Once we have established the robustness of our main results, we proceed to
conduct several heterogeneity analyses. These analyses aim to explore potential
variations and differences in the relationship between BPK's audit opinion on
LKPD and regional social and economic performance across different subgroups
or regions. By conducting these heterogeneity analyses, we can gain a deeper
understanding of how the association may vary based on specific characteristics
or contexts.
In these analyses, we consider factors such as geographical regions,
population size, economic development level, or other relevant characteristics
that may influence the relationship between BPK's audit opinion and regional
performance. By examining these heterogeneity factors, we can identify any
variations or nuances in the relationship and assess whether the findings hold
consistently across different subgroups.

The heterogeneity analyses provide valuable insights into the complex
dynamics between BPK's audit opinion and regional social and economic
performance, contributing to a more comprehensive understanding of the
underlying factors and potential policy implications. The findings from these
analyses are discussed in detail in the subsequent sections, shedding light on
the diverse nature of the relationship and its implications for different regions
or contexts.
ESTIMATION RESULTS
Our key findings are summarised in Table 2. These findings are derived
from a fixed effect regression model estimation using panel data. The results
remain robust even when controlling for variables such as local government
expenditure per capita, regional government capital expenditure per capita,
local government revenue per capita, and regional transfers per capita (Table 2).
The primary outcome derived from these results is that the opinion status
from BPK (Supreme Audit Agency) is significantly associated with regional
social and economic performance. An Unqualified Opinion (Wajar Tanpa
Pengecualian/WTP) status of a region significantly correlates with higher
Human Development Index and GRDP per capita compared to regions receiving
other audit opinion status. Additionally, regions with WTP opinion status exhibit
significantly lower levels of poverty and inequality.
It is crucial to acknowledge that our findings do not establish a causal
relationship between receiving a positive opinion status from BPK, which serves
as a proxy for a strong accountable government, and regional development.
We recognise that our results may be subject to endogeneity issues, as the
governance of a region and its economic and financial performance could have
a reverse causality relationship.
Table 2. Summary of Result from Panel Regression Model at the National Level
Relationship between BPK’s Audit Opinion and Regional Socio-Economic Performance
Sumber: DRE, 2021.

Note: ***p < 0.001 ; **p <0.01 ; *p<0.05
To address this concern, we conducted a panel regression model using
lagged independent variables (refer to Appendix 3). The results of this panel
regression model remain consistent and statistically significant, providing
further support to our findings.
Additionally, the results presented in Table 2 indicate that the Hausman
test yields small values (lower than 0.05), indicating a relatively consistent
relationship. Therefore, we can argue that our findings consistently demonstrate
an association between accountable government and regional development.
Even when we divide the regions into Java and those outside Java, our
estimation results remain robust, as shown in Appendix 1 and Appendix
2. The status of BPK's audit opinion continues to demonstrate a significant
relationship with the socio-economic performance of the regions. Both within
Java and outside Java, regions with a WTP (Unqualified Opinion) audit status
exhibit higher Human Development Index and GRDP per capita compared to
regions with other opinion statuses.
Furthermore, within the Java region, those with a WTP audit status
experience lower levels of poverty and income inequality compared to those
with other opinion status. For regions outside of Java, the audit opinion status
has a significant impact on the levels of Human Development Index, GRDP per
capita, as well as the poverty level. However, the results from the regression
model indicate that the audit opinion status in regions outside of Java does not
have a significant impact on the Gini Ratio of those regions.

CONCLUSION
This chapter aims to explore the association between accountable and
transparent government and development performance, using Indonesia as
a case study. It examines the relationship between audit opinion status on a
region's LKPD/Laporan Keuangan Pemerintah Daerah (Regional Government
Financial Statement) and regional socio-economic performance. The analysis
utilises data from 514 regions in Indonesia within the period 2015-2019,
employing a fixed effect estimation model for panel data. The main findings are
as follows:
At the national level, a statistically significant association is found between
BPK audit opinion status and regional socio-economic performance. Regions
with a WTP (Unqualified Opinion) audit status demonstrate significantly higher
Human Development Index and GRDP per capita compared to those with
other opinion status. Furthermore, regions with a WTP opinion status exhibit
significantly lower levels of poverty and inequality.
The analysis further divides the regions into two segments: Java and non-
Java regions. In Java regions, a significant impact of WTP audit status on socio-
economic performance is observed. Conversely, in non-Java regions, WTP audit
status is significantly associated with Human Development Index, GRDP, and
poverty level within the same regions. However, WTP audit status does not have
a significant impact on the Gini Ratio.
The findings of this study contribute to the existing literature and offer
insights for relevant government policies to enhance the quality of local
government financial management and, ultimately, the welfare of the people.
Based on the estimation results, it is recommended that local governments
in Indonesia improve the quality of their financial management, as reflected
in a better audit opinion status from BPK, in order to enhance the social and
economic performance of their regions.
However, it is important to caution readers that our results do not provide
evidence of a causal relationship indicating that a strong accountable and
transparent government directly leads to better regional development. Further
studies are needed to establish this causal link.
REFERENCES
Abdolmohammadi, M. J. and R.R. Tucker. (2002). ‘The Influence of Accounting
and Auditing on a Country’s Economic Development.’ Review of Accounting
and Finance, 1(3): 42–53. Available at: https://doi.org/10.1108/eb026990
Akhmad Hafidzan Adzani & Dwi Martani. (2014). “Pengaruh Kesejahteraan
Masyarakat, Faktor Politik dan Ketidakpatuhan Regulasi Terhadap Opini
Audit Laporan Keuangan Pemerintah Daerah.” Universitas Indonesia.
Hehanussa, S. J. (2015). “Pengaruh Penyajian Laporan Keuangan Daerah dan
Aksesibilitas Laporan Keuangan Daerah Terhadap Transparansi dan
Akuntabilitas Pengelolaan Keuangan Daerah Kota Ambon.” Universitas
Islam Sultan Agung.
Kumorotomo, Ngatikoh, & Retnandari (2019). ‘Transparency in Government: A
Review of the Failures of Corruption Prevention in Indonesia.’ Advances in
Economics, Business, and Management Research 122.
Lewis, Blane (2014). ‘Indonesian Intergovernmental Performance Grants: An
Empirical Assessment of Impact.’ Bulletin of Indonesian Economic Studies
50, no.3: 415 – 433.
Maani, K. (2009). Transparansi dan Akuntabilitas dalam Pelayanan Publik.
Demokrasi VIII, no. 1.
Mechkova, Valeriya, Michael Bernhard, and Anna Luhrmann (2019). “Diagonal
Accountability and Development Outcomes.” Open Government Partnership,
V-Dem Institute. Available at: https://www.opengovpartnership.org/
wp-content/uploads/2019/09/Diagonal-Accountability-Development-
Outcomes20190909.pdf.
Musahadah & Rustan Amarullah (2018). “Sebuah Paradoks: Opini WTP Dan
Kesejahteraan Rakyat. Inspektorat Kabupaten Kotawaringin Barat.” Pusat
Kajian dan Pendidikan dan Pelatihan Aparatur III-Lembaga Administrasi
Negara. Samarinda-Kalimantan Timur
Ni Made Suryaningsih & Eka Ardhani Sisdyani (2016). ‘Karakteristik Pemerintah
Daerah dan Opini Audit Pada Kinerja Keuangan Pemerintah Daerah.’
E-Jurnal Akuntansi Universitas Udayana 15, no. 2 (Mei).

Nur Anita Chandra Putry & Rudy Badrudin (2017). ‘Pengaruh Kinerja Keuangan
Daerah Terhadap Opini Audit dan Kesejahteraan Masyarakat Di Daerah
Istimewa Yogyakarta.’ JRMB 12, no. 1 (Juni). STIE YKPN Yogyakarta.
Petkovic, Aleksandra & Tauhidur Rahman (2021). ‘Government Accountability
and Development: Evidence from Young Democracies.’ Available at SSRN:
https://ssrn.com/abstract=3971900 or http://dx.doi.org/10.2139/
ssrn.3971900
Pina, Vicente, Lourdes Torres, & Basilio Acerete (2007). ‘Are ICTs promoting
government accountability?: A comparative analysis of e-governance
developments in 19 OECD countries.’ Critical Perspectives on Accounting,
18(5): 583-602.
Putu Riesty Masdiantini dan Ni Made Adi Erawati (2016). ‘Pengaruh Ukuran
Pemerintah Daerah, Kemakmuran, Intergovernmental Revenue, Temuan,
dan Opini Audit BPK Pada Kinerja Keuangan.’ E-Jurnal Akuntansi Universitas
Udayana 14, no.2 (Februari).
Sen, Kunal (2014). “Governance and Development Outcomes in Asia.” ADB
Economics Working Paper Series (January).
Smoke, Paul (2015). “Accountability and Service Delivery in Decentralizing
Environments: Understanding Context and Strategically Advancing
Reform.” In Alan Whaites, Eduardo Gonzalez, Sara Fyson, Graham Teskey
(eds). A Governance Practitioner’s Notebook. Paris: OECD. 219-232
United Nations Economic and Social Commissions for Asia and the Pacific
(US ESCAP) (2009). “What is Good Governance?” Policy Brief (10 July).
Available at: https://www.unescap.org/resources/what-good-governance
World Bank (2017). Implementation Completion and Results Report Local
Government and Development Project, Jakarta.
APPENDICES
Appendix 1. Summary of Result from Panel Regression Model for Regions
within Java Region
Sumber: DRE, 2021.

Note: ***p < 0.001 ; **p <0.01 ; *p<0.05
Appendix 2. Summary of Result from Panel Regression Model for Regions

Outside of Java
Sumber: DRE, 2021.

Note: ***p < 0.001 ; **p <0.01 ; *p<0.05

Appendix 3. Result from Panel Regression Model of Impact of BPK’s Audit
Opinion on Economic Performance
Note: ***p < 0.001 ; **p <0.01 ; *p<0.05
CHAPTER 13
Major Determinants of the

Diarrhoea Incidence among
Children Under-Five (U5) in
Indonesia
Adrian Chrisnahutama and Ni Made Sukartini
INTRODUCTION
Diarrhoea remains one of the major health issues in the world, particularly in
developing countries (Omona et al., 2020; Shine et al., 2020). Approximately
1.7 billion cases and 525,000 children under 5-year-old (U5) deaths due to
diarrhoea have taken place every year. Diarrhoea has been classified as a the
leading cause of death for U5 in most developing countries, including Indonesia
(World Health Organization, 2017). Based on the estimation of Kementerian
Kesehatan (2019), about 11.5% or 73,188 U5 suffered from diarrhoea in 2018.
In addition, 2,948 villages/kelurahan experienced diarrhoea outbreak between
2017 and 2018 (Badan Pusat Statistik, 2018).
According to the Indonesian Ministry of Health (MoH), the incidence of
diseases such as diarrhoea that occurs longer than two weeks is considered
as an extraordinary event or Kejadian Luar Biasa (KLB) in Indonesian, which
is mainly characterised by situations such as (1) emergence of an infectious
disease that did not exist previously in the area; (2) an increase in incidence
of sickness in a certain period of time; (3) an increase in the number of new
sufferers in one month either doubling or significantly higher compared to
279
the previous period; (4) increasing number of sick individuals that is twice
higher or more compared to the previous period; (5) an increase in cases of
fatality in one period by 50% or more compared to the previous period; and (6)
increasing number new cases of disease in one period, which has doubled or
much higher compared to the previous period. The absence of proper medical
treatment to infants or U5 children who experience frequent diarrhoea might
result in nutritional deficiency or even the loss of it, which may cause them to
also suffer from a slow physical growth or known as stunting (Bosch et al. 2000;
Brenneman & Kerf, 2002; Grossman, 1972; Todaro & Smith, 2012).
Another factor that contributes to diarrhoea and indirectly increases
the probability of stunting is lack of access for safe drinking water (Budge
et al., 2019; Cumming & Cairncross, 2016; Kwami et al., 2019). Other studies
reported that children suffering from stunting is associated with lower
academic performance (Wahed et al., 2017), lower health status in adult life,
and consequently it results in lower paid job (Müller & Krawinkel, 2005).
These findings implies that the KLB of diarrhoea can destroy human capital
development and labour participation, which in the long run may result in low
output and low income. The consequences of this will eventually lower the
national output and increasing poverty rate.
Lack of the provision of safe drinking water, improved sanitation, and
handwashing facilities contributed to the diarrhoea and KLB diarrhoea.
Frequent incidence of diarrhoea with extraordinary cases might result in infant
health status as well as fatality risk (Brenneman & Kerf, 2002; Cameron &
Olivia, 2011; Esrey et al., 1991; Komarulzaman et al., 2017; Noguchi et al., 2021;
Piper et al., 2017; Wolf et al., 2018). United Nations Children’s Fund (UNICEF)
and World Health Organization (WHO) (2019) reported that there are 785
million people who do not have access to improved source of drinking water,
144 million people who still use surface water (lake, pond, river, etc.) as source
of drinking water, and about 701 million people do not have access to improved
sanitation. Despite progressive programme for safe drinking water and proper
sanitation in Indonesia, there remain 10% of population who have no access
to improved source of safe drinking water, and 20% of population who did not
have improved sanitation in 2019 (Badan Pusat Statistik, 2019).
Lack of access to improved sanitation is associated with larger incidence
of open defecation practice (Bosch et al., 2000; Brenneman & Kerf, 2002).
High incidence of open defecation practice within community is positively
correlated with water, soil, and air pollution (Komarulzaman et al., 2017), and
this condition leads to higher risk of diarrhoea incidence. Cameron and Olivia
(2011) and Patunru (2015) reported that the proportion of open defecation
within community is also positively correlated with diarrhoea incidence. In
the case of Indonesia, the larger proportion of household practice free open
defecation, the larger diarrhoea incidence in Indonesia. Andres et al. (2014)
and Komarulzaman et al. (2017) reported that diarrhoea incidence significantly
decreases in regions that provide safe water and clean sanitation programme.
The three studies that had been carried out in Indonesia used different
approaches and definitions in analysing the determinants of diarrhoea.
Cameron & Olivia (2011) utilised pooled cross section data from Survey Sosial
Ekonomi Nasional (Susenas) for the period of 1993-2009. This study focused on
types of sanitation, such as the ownership of household toilet, type of the closet,
and type of excreta storage after defecating. However, most of the Susenas data
were presented as available or not available, i.e., nominal scale of measurement.
Patunru (2015) utilised three difference data sets, namely Susenas 2011, Podes
2011 and INDO-DAPOER data to investigate the relation of access of safe
drinking water and improve sanitation with the incidence of diarrhoea. The
lower access for both safe drinking water and improved sanitation, the higher
the incidence of diarrhoea. Meanwhile, the study of Komarulzaman et al. (2017)
investigated the incidence of diarrhoea among U5 by utilising Survei Demografi
dan Kesehatan Indonesia (SDKI) data of 2007 and 2012. These three studies
all focus on the role of basic infrastructure, i.e., access to safe drinking water
and improved sanitation as determinant of diarrhoea incidence among U5 in
Indonesia. Due to data limitation, none of these studies’ utilised the behavioural
aspects and quality of environmental as contributors to the risk of infant
experiencing diarrhoea.
In addition to the access of safe drinking water and improved sanitation
variables, this study also considers variables of household’s water treatment
before consuming it. The treatment includes boiling, chlorinating, filtering,
and solar disinfectant. Second, in this study a variable of whether or not the
household has hand wash facilities is also added. This variable is classified as
Not available, Facility on premises without water/soap presence, and Facility
Major Determinants of the Diarrhoea Incidence among Children Under-Five (U5) 281
in Indonesia
on premises with water and soap presence. Third, various proxies are used
to measure environmental quality, such as percentage of household in the
community practicing free open defecation, treatment for after used baby’s
hygienic disposal of stools (HDS), and cattle or livestock growing around the
house or not.
Adding variable of environmental quality—especially the after-use babies’
HDS and the way cattle is farmed—could proxy the behavioural aspect of the
household. Even though the basic infrastructure such as safe drinking water
and improved sanitation are already accessible, but when healthy behaviour is
not applied, the risk of U5 being infected by bacteria is still high. This effort will
give insight of which aspect to be prioritise as policy instrument for promoting
and educating community about healthy lifestyle. Hence, the risk of diarrhoea
KLB in Indonesia could expectedly be minimized eventually.
GROSSMAN HEALTH DEMAND MODEL

Grossman Health Demand Model views health as durable capital stock that
generates healthy time. This model assumes every individual inherits health
capital. This capital can depreciate over time, but it can be improved through
investment—such as healthy food, exercise, diet, or health knowledge
improvement. Grossman (1972) argued that health capital differs from human
capital model by Becker (1962) and Ben-Porath (1967). In Grossman (1972)
view, individual’s stock of knowledge influences his/her market and non-
market productivity. Individual’s health capital stock determines the amount
of time that he or she can spend to work and produce money, earnings, and
commodities. He further stated there are two reasons why individuals demand
for health. First, individual’s health quality directly influences their preference
functions. Low health quality in the utility function can be interpreted that
illness is a source of disutility. When illness results in the absence for working
that takes longer time, then it will influence labour productivity and income
accordingly. The longer illness the lower income generates from working and
therefore less income to be generated and lower goods to be consumed. The
monetary value of this lost is an index of return of investment in health. Second,
illness determines individual’s time availability to do production activities on
market. This can be interpreted that people better health quality means more
time to produce money, earnings, and commodities. Investment for health
quality means increasing their health stock.
Regarding the incidence of diarrhoea among infant or U5 in Indonesia,
Grossman Demand for Health Model is applied as follows. Firstly, for the case of
diarrhoea, the demand for health is household demand not the infant demand
(for health) directly. The better quality of infant’s health the more productive
times available for the parents to work and generate income. When the infant
is not well and required further medication, parent could absence from work.
As Diarrhoea is one of contagious diseases, risk of diarrhoea can therefore be
transmitted from adult to infant or vice versa (Haricharan et al., 2013; Seksaria
& Sheth, 2014). Diarrhoea occurs more frequent and tend to be classified as
Kejadian Luar Biasa (KLB) or extraordinary cases in many regions in Indonesia
(Islamiati, 2022). To prevent further probability of KLB, everyone must practice
clean and healthy lifestyle, which is an implementation of health investment, as
stated in Grossman Demand for Health.
Correlation between Drinking Water, Sanitation, Handwashing

Facility, and Environmental Hygiene on Diarrhoea among U5
Numerous studies reported that provision of clean drinking water, improved
sanitation and basic handwashing facility prevents people from various
waterborne diseases, such as diarrhoea. Esrey et al. (1991) investigated 144
papers through meta-analysis study and they reported that access to clean
drinking water and improved sanitation directly reduced median mortality rate
due to diarrhoea. Their findings show that on average about 26% diarrhoea
incidence can be minimized as access to safe drinking water and sanitation is
improved by 1%.
Another study by Brenneman & Kerf (2002) concluded that clean drinking
water and improved sanitation provision lead to lower diarrhoea incidence
among children. This study reported that better access to safe drinking water
not only improves community health status and reduces incidence of diarrhoea
among U5, but also reducing poverty through labour market channels. Most
countries being reported were the slum areas in African countries, South and
Southeast Asia countries, and Indonesia particularly slum areas in Jakarta and
Malang. Both authors reported that in community where access to safe drinking
water is limited, household wife has to sacrificed at least two hours per day to
collect water. This condition made female labour participation lower than the
male, which consequently causing the family to lose their potential income. Low-
income family is less likely to invest in water and sanitation, making diarrhoea
more likely to occur.
in Indonesia
Wolf et al. (2018) also conducted a meta-analysed study. They focused
on 19 out of 80 papers that investigated the correlation between access of safe
drinking water and improved sanitation and incidence of diarrhoea among
U5 infants. The authors classified 14 papers that investigated safe drinking
water, 8 papers investigated improved sanitation, and 3 papers investigated
both safe drinking water and improved sanitation. They concluded that better
access to clean drinking water, improved sanitation, and handwashing facility
which equipped with soap were associated with children diarrhoea risk
reduction. Patunru (2015) utilised combination of micro data and aggregate
data to investigate diarrhoea incidence among U5 in Indonesia. He reported that
those who did not have improved source of drinking water and sanitation had
12 % and 27 % higher risk of suffering diarrhoea in Indonesia, respectively.
Komarulzaman et al. (2017) reported that typical families with U5 which did not
use piped water on premises as source of drinking water were associated with
higher risk of suffering diarrhoea in Indonesia. Similar study by Noguchi et al.
(2021) reported that access to handwashing facility on premises was associated
with lower risk of having diarrhoea among U5 in Lao People’s Democratic Republic.
International standard and classification for sources of drinking water as
water is contaminated free if the design and construction, method of collection,
transport, storage, and serving protects the water from human hands, bacteria,
insects, utensils, etc. (Shaheed et al., 2014; United Nations Children’s Fund
(UNICEF) & World Health Organization (WHO), 2019). This implies that due
to its design and construction until delivery, piped water is expected to be the
safest source of drinking water, which potentially could minimise the risk of
contamination (Jalan & Ravallion, 2003; World Health Organization & United
Nations Children’s Fund (UNICEF), 2013).
On the other hand, the quality of piped water in developing countries does
not guarantee the delivery process is free contamination. Shaheed et al. (2014)
reported that water that delivered through piped water in some countries in
Southeast Asia, South Asia, Sub-Saharan Africa, Latin America, and Caribbean
were polluted by factory wastes. Furthermore, Shaheed et al. (2014) stated that
the existing pipelines in those countries were under-maintained and under-
renewed. These conditions led to high contamination, disruptions of water
flows, and water shortages (Komarulzaman et al., 2017). Therefore, definition
of improved source of drinking water cannot be universally applied, especially
in the cases of developing countries like Indonesia (Patunru, 2015).
Lack of access and unhygienic drinking water provision in developing
countries has been getting worse due to the use of water containers or storing
devices (Komarulzaman et al., 2017). The containers were often unhygienic,
resulting in higher risk of contamination (Brenneman & Kerf, 2002; Fewtrell
et al., 2005). To reduce the risk of contaminated water, households typically
treat the water by boiling, chlorinating, filtering, or other methods (Clasen et
al., 2007). This strategy is effective to reduce the risk of diarrhoea outbreak
(Fiebelkorn et al., 2012; Sodha et al., 2011; Wolf et al., 2018). However, these
findings were not consistent with the household level study by Komarulzaman
et al. (2017). These authors reported that these types of interventions did not
associate with diarrhoea among U5 in Indonesia.
The availability of sanitation and handwashing facilities are required to
eliminate diarrhoea. Both access to safe drinking water and improved sanitation
are related to hygiene improvement. Cameron & Olivia (2011) reported that
typical household that utilised flush toilet had significant lower diarrhoea cases
in Indonesia. Andres et al. (2014) reported facilities that hygienically separate
excreta from human contact was associated with risk reduction of suffering
from diarrhoea among U5 in India. A meta-analyses study by Wolf et al. (2018)
concluded that the availability of handwashing facility with hand soap reduced
the odds of having diarrhoea by 30 % among children. This finding is supported
in recent study by Noguchi et al. (2021) in Lao People’s Democratic Republic.
However, the opposite finding were reported by Komarulzaman et al. (2017)
study in Indonesia, who reported that improved sanitation did not correlate
with diarrhoea among U5.
Low provision of improved sanitation associate with high incidence of
open defecation practice. Open defecation practice not only increases water, soil,
and air pollution within community, but also leads to higher risk of diarrhoea
incidence (Komarulzaman et al., 2017). Cameron & Olivia (2011) and Patunru
(2015) analysed the relationship between proportion of open defecation within
community and diarrhoea in Indonesia. As expected, the higher proportion
of open defecation practice the higher diarrhoea incidence. Other studies by
Andres et al. (2014) in India and Komarulzaman et al. (2017) in Indonesia
used proportion of having improved drinking water and sanitation within
community as environmental hygiene variables. These studies concluded that
higher proportion of improved drinking water and sanitation associated with
in Indonesia
lower diarrhoea incidence. These findings illustrate that environmental hygiene
indicator as one of the determinants of diarrhoea incidence as well as drinking
water, sanitation, and handwashing facility provisions.
Correlation between Other Variables on Diarrhoea among Us

This study follows several previous studies in determining determinants of
diarrhoea. Demography variables used in this study were age and sex of the
children, mother’s education, and mother’s electronic mass media exposure. The
households’ characteristics floor type, variation of assets owned by household,
and type of residential. As reported in several studies that claimed age and
sex of the children varies significantly. Younger children were found more
likely suffered diarrhoea than older one, especially after the kids reach 5 years
old. Walker et al. (2013) reported about 80 % of global diarrhoea incidence
occurred in children under two years old. On the other hand, sex of the children
is still inconclusive. Andres et al. (2014) reported that male children had lower
odds of having diarrhoea, while Komarulzaman et al. (2017) reported that male
children were more likely to experience diarrhoea.
Regarding U5’s mother education level is found positively related with
better awareness for reducing the risk of children infected by diarrhoea.
Furthermore, knowledge about diarrhoea is not only obtained through formal
education, but also obtained by information which typically quickly spread
through electronic mass media . Electronic mass media had been found effective
increasing parent’s awareness of oral rehydration therapy (ORT) prevention,
and encouraged to administer zinc to treat children’s diarrhoea (Lenters et al.,
2013; Naugle & Hornik, 2014; Rao et al., 1998). Following these studies finding,
this study also added potential source of information for typical household and
used information the frequency of watching television as electronic mass media
exposure variable.
In terms of household characteristics, housing floor type was reported
associated with high diarrhoea incidence. This is because bacteria potentially
would stay on the floor surface (Sahiledengle et al., 2021). Variation of assets
owned by household illustrates household wealth. Higher variation of assets
owned means that the household is better off. Better off households have more
resources to fulfil their basic needs and better living conditions and tend to
have better health. These conditions can deter diarrhoea incidence among U5.
Furthermore, children who live in urban area have advantages regarding health
infrastructure provisions, which leads to better health status.
USE OF LOGISTIC REGRESSION
This study utilised quantitative analysis to investigate the relationship
between factors—such as drinking water, sanitation, handwashing facility,
and proportion of open defecation—and the incidence of diarrhoea. Logistic
regression is chosen as method of analyses. This method is accepted as one
alternative model to deal with situation where the outcome variable is stated
in binary variable (Hosmer & Lemeshow, 2004; Wooldridge, 2013). The model
is designed following the framework presented in Figure 1.
Figure 1. Design of Empirical Model
Individual Characteristics: Household’s Infrastructure Community Characteristics:
• Age of children Characteristics: • Proportion of open
• Sex of children • Source of drinking water defecation
• Mother’s education level • Sanitation • Disposal hygienic disposal
• Mother’s frequency of • Handwashing facility of stools (HDS) treatment
watching television • Point-of-use • Residential surrounded by
• Residential location • Type of region cattle and poultry shed.
• Floor type
Diarrhoea on U5
Source: Author’s own modification following Cameron & Olivia (2011)

and Komarulzaman et al. (2017)
Determinants of diarrhoea are classified into 3 variables—i.e., household’s
infrastructure characteristics, healthy behaviour, and environmental hygiene,
while a set of household characteristics are used as control variables. Based on
this framework, the equation can be written as follows:
( )
This study follows Andres et al. (2014), Cameron & Olivia (2011),
Komarulzaman et al. (2017), and Patunru (2015) to determine a set of
independent variables. In Equation (1) notation Yi denotes outcome variable
indicating whether the child has suffered from diarrhoea for past two weeks,
with the option: yes = 1 and no = 0.
in Indonesia
Notation X1 represents vectors of household’s infrastructure characteristics,
which are: (1) Access to drinking water—which is classified into surface water,
unprotected dug well or spring, bottled/refilled water, protected dug well or
spring, and piped water; (2) Access to improved sanitation—which is classified
into no facility, pit latrine, flush toilet which is shared/public, flush toilet with
no septic tank, and flush toilet using septic tank; (3) Availability of handwashing
facility—which is classified into not available, facility on premises without
water/soap presence, and facility on premises with water and soap presence;
and (4) Drinking water treatment or point-of-use treatment.
Notation X2 is vector that represents healthy behaviour and environmental
hygiene, which includes among others: (1). Drinking water treatment—such as
boiling, chlorinating, filtering, and solar disinfection. This is recoded as answer
yes = 1 and no = 0; (2). Proportion of open defecation within community—i.e.,
percentage of total households; (3). Proportion of household in the communities
who manage their after used diapers or also known as hygienic disposal of
stools (HDS); and (d) Whether household raises cattle and poultry and their
cattle shed located around the yard. This proxy variable is also recorded with
the answer yes = 1 and no = 0.
Notation X3 in Equation (1) represents a set of demography and household
characteristics, this includes children characteristics (age and sex), his/her
mother’s characteristics (education level and watching TV as proxy of source of
information exposure), household floor type, and asset variation.
Age of children is expressed as ratio scale. Sex of the children is expressed
as nominal scales (1 = male; 0 = female). Mother’s education level is measured
using years of schooling. Mother’s frequency of watching television is expressed
as nominal scales (1 = at least once a week; 0 = less than once a week). Floor
type is expressed as nominal scales (1 = earth/sand or other; 0=otherwise).
Household income is proxied by variation of assets owned by household is sum
of electricity, radio, television, refrigerator, bicycle, motorcycle/scooter, car/
truck, separate room used as kitchen, mobile telephone, bank account, and
proper cooking fuel ownerships. Type of region is expressed as nominal scales
(1 = urban; 0 = rural). Notation ai is province fixed effect, and ui is residual
disturbance.
It is realised that Equation (1) suffered from omitted confounding variable
bias. Confounding variable bias is unmeasured variable in the model that
potentially correlate with both the dependent, i.e., incidence of diarrhoea and
at least one of the set independent variables (Islam et al., 2018; Ugboko et al.,
2020). Besides, unmeasured variable confounding variable bias can also result
from data unavailability. According to Islam et al., and Ugboko, the potential
confounding factors in the case of determinants of diarrhoea among U5, such
as model in Equation (1) are: (1) infants have never been breastfed: and (2)
infants suffering from food intolerance or allergy.
Children are especially at risk for diarrhoea as they tend to put objects
in their mouths and may not have developed good handwashing habits
(Haricharan et al., 2013). Haricharan et al., claimed that as children started
learning to stand, from sitting to walk, the higher probability they get food
infected. They eat everything found on the floor and these things might be
contaminated by bacteria. Children below 2 or 3 years-old will typically take
higher risks of getting food infected compared to children over 3 years old. The
two possible confounding factors discussed above—i.e., data for children who
have never been breastfed and who suffered from food intolerance—are not
available in Survei Demografi dan Kesehatan Indonesia (SDKI) data. To check
the robustness results, this study provides two comparison models, namely
model with disaggregated data by child’s age and regression with instrumental
variables.
THE USE OF SDKI AS DATA SOURCES

This study utilised Survei Demografi dan Kesehatan Indonesia (SDKI) data of
2012 and 2017. SDKI offers a nationally representative household level data that
contain various information on population, socioeconomic, health and nutrition
indicators (ICF, 2013, 2018). This study chooses SDKI as source of data analysis
for the following reasons. First, although there are many sources of micro data
set, such as Survei Sosial Ekonomi Nasional or Susenas; Riset Kesehatan Dasar
or Riskesdas; Indonesia Family Life Survey or IFLS; and SDKI, anly the last two
data that provide information about incidence of diarrhoea and households
access to infrastructure. In terms of period, SDKI provides latest data (2017)
compared with IFLS (2014). Second, in terms of coverage, both IFLS and SDKI
represent national coverage, however IFLS data only cover 13 provinces while
SDKI data cover 33 provinces. Based on these two considerations, this study
in Indonesia
utilises SDKI data. SDKI survey results is constructed as six modules. Those
modules are as follows: (1) Household Records (HR); (2) People Records (PR);
(3) Individual Records (IR); (4) Children Records (KR); (5) Birth Records (BR);
and (6) Men Records (MR). All SDKI protocols are reviewed by ethics reviewer
or institutional reviewing board in the survey country to ensure the protection
of human subjects (ICF, 2013, 2018).
The data were constructed as pooled cross section. This is because SDKI
survey was not designed as longitudinal data. Using panel data will make lower
sample size. Wooldridge (2013) argued that reason for using pooled cross
section was to increase the sample, hence got more precise estimators and test
statistics with more power. In order to capture the difference effect between
regions, we use provincial dummy. Furthermore, analysis is also disaggregated
by island groups within Indonesia. Combined samples used in this study were
34,177 under-five (U5) children. Those samples were acquired by interviewing
28,787 mothers. Moreover, this study did not include 1,673 missing data or do
not know response on dependent or independent variables.
Statistics Data Description

Description of data representation of Equation (1) is presented in the following
Table 1. Data descriptions are classified as: full sample, sample year 2012 and
sample year 2017. Characteristic of the data as nominal, ordinal, interval, and
ratio scale of measurement. Therefore, the statistics data are presented as
Mean, Proportion, and Standard Deviation (SD). It can be seen from Table 1 that
proportion of diarrhoea incidence was about 14-15% among U5 in Indonesia. In
terms of access and source of drinking water, there was significance progress on
sources of safe drinking water, especially classification of bottled/refilled water
and protected dug well or spring. About 33% of household in Indonesia consume
bottled/refilled water and about 37% consuming water from protected dug well
or spring. According to WHO, the highest classification of safe drinking water
is piped water. The SDKI data indicated about 10% of Indonesia households
utilise piped water and the rest still rely on surface water and unprotected dug
well or spring.
In terms of sanitation, especially defecation place, majority (63%)
of households already use flush toilet with no septic tank. Furthermore,
about 77% of households have facility on premises with water and soap,
with significant improvement of the facility in 2017 (86%) compared to the
facility in 2012 (69%). Regarding vector of variables of healthy behaviour and
environmental hygiene, the data indicated between 2012 and 2017 there were
no changes in proportion of households conducting drinking water treatment
before consuming. About 15% of households in the community still practice
open defecation, with significant decreased in proportion, about 19% (2012)
to 11% (2017). However, in terms of proportion of households practiced unsafe
hygienic disposal stools increase significantly, about 48% in 2012 rose into 58%
in 2017. Fortunately, the proportion of households who kept their cattle and
livestock’s around their yards has decreased from 37% in 2012 to 35% in 2017.
Table 1. Descriptive Statistics, by Survey Wave
Full Sample Year 2012 Year 2017
Variables Mean/ Mean/ Mean/
SD SD SD
Prop. Prop. Prop.
Diarrhoea (Yes = 1) 0.14 0.15 0.14
Household’s infrastructure Characteristics
Drinking Water Classifications
Surface Water 0.04 0.05 0.03
Unprotected Dug Well or Spring 0.16 0.22 0.10
Bottled/Refilled Water 0.33 0.30 0.37
Protected Dug Well or Spring 0.37 0.33 0.41
Piped Water 0.10 0.11 0.10
Sanitation Classifications
No Facility 0.15 0.19 0.11
Pit Latrine 0.05 0.05 0.04
Flush Toilet which is Shared/Public 0.10 0.11 0.09
Flush Toilet with No Septic Tank 0.07 0.06 0.08
Flush Toilet using Septic Tank 0.63 0.58 0.68
Handwashing Facility Classifications
Not Available 0.16 0.25 0.06
Facility on Premises without Water/ 0.07 0.07 0.08
Soap
Facility on Premises with Water and 0.77 0.69 0.86
Soap
in Indonesia
Variables Mean/ Mean/ Mean/
SD SD SD
Prop. Prop. Prop.
Healthy Behaviour & Environmental Hygiene
Drinking water treatment (Yes = 1) 0.67 0.67 0.67
Proportion of Open Defecation 0.15 0.23 0.19 0.26 0.11 0.18
Proportion of Unsafe Hygienic disposal 0.53 0.35 0.48 0.28 0.58 0.39
of stools (HDS))
Owns Poultry and Cattle around 0.36 0.37 0.35
residential (Yes = 1)
Household Characteristics as Control Variables
Age of the Children (Month) 29.17 17.26 29.13 17.32 29.21 17.20
Sex of the Children (Male = 1) 0.52 0.52 0.51
Mother's Education Attainment, no 0.02 0.03 0.01
education
Mother's Education Attainment, 0.28 0.30 0.25
primary
Mother's Education Attainment, 0.54 0.53 0.56
secondary
Mother's Education Attainment, higher 0.16 0.14 0.18
Mother's Frequency of Watching 0.82 0.82 0.82
Television (At Least Once per Week = 1)
Floor Type (Earth/Sand/Other = 1) 0.05 0.06 0.04
Total Variation of Assets 6.34 2.44 5.93 2.53 6.76
Type of Region (Urban = 1) 0.47 0.46 0.49 2.27
Observations 33,803 17,102 16,701
Table 2. Descriptive Statistics, by Island Groups
Full Sample Java-Bali Sumatra Kalimantan Sulawesi Others
Variables Mean/ Mean/ Mean/ Mean/ Mean/ Mean/
SD SD SD SD SD SD
Prop. Prop. Prop. Prop. Prop. Prop.
Drinking Water Classifications

Diarrhoea (Yes = 1) 0.14 0.13 0.15 0.18 0.17 0.13
Surface Water 0.04 0.01 0.03 0.12 0.03 0.05

Unprotected Dug Well or Spring 0.16 0.10 0.15 0.10 0.21 0.23
Bottled/Refilled Water 0.33 0.40 0.39 0.40 0.28 0.16
Protected Dug Well or Spring 0.37 0.40 0.35 0.26 0.37 0.42
Sanitation Classifications
Piped Water 0.10 0.10 0.08 0.13 0.11 0.14
No Facility 0.15 0.10 0.14 0.20 0.20 0.18

Pit Latrine 0.05 0.02 0.06 0.02 0.04 0.07
Flush Toilet which is Shared/Public 0.10 0.08 0.06 0.06 0.15 0.17
Flush Toilet with No Septic Tank 0.07 0.08 0.09 0.09 0.03 0.07
Handwashing Facility Classifications

Flush Toilet using Septic Tank 0.63 0.73 0.65 0.62 0.57 0.51
Not Available 0.16 0.07 0.15 0.12 0.19 0.29

Facility on Premises without 0.07 0.04 0.08 0.06 0.08 0.11
Water/Soap
Facility on Premises with Water 0.77 0.88 0.78 0.82 0.73 0.61
and Soap
in Indonesia
Major Determinants of the Diarrhoea Incidence among Children Under-Five (U5)
293
294
Full Sample Java-Bali Sumatra Kalimantan Sulawesi Others
Variables Mean/ Mean/ Mean/ Mean/ Mean/ Mean/
SD SD SD SD SD SD
Prop. Prop. Prop. Prop. Prop. Prop.
Healthy Behaviour & Environmental Hygiene
Point-of-Use Treatment (Yes = 1) 0.67 0.65 0.65 0.62 0.70 0.74
Proportion of Open Defecation 0.15 0.23 0.10 0.18 0.14 0.21 0.20 0.29 0.20 0.25 0.18 .26
Proportion of Unsafe Stool 0.53 0.35 0.40 0.35 0.49 0.33 0.60 0.33 0.65 0.30 0.65 .32
Disposal of Youngest Child
Owns Poultry/Chicken (Yes = 1) 0.36 0.31 0.37 0.34 0.44 0.36
Control Variables
Type of Region (Urban = 1) 0.47 0.67 0.43 0.47 0.37 0.33
Age of the Children (Month) 29.17 17.26 29.08 17.18 29.33 17.32 29.22 17.35 29.13 17.36 29.06 17.18
Sex of the Children (Male = 1) 0.52 0.50 0.52 0.52 0.51 0.52
Mother's Education Attainment, no 0.02 0.01 0.01 0.02 0.03 0.06
education
Mother's Education Attainment, 0.28 0.26 0.25 0.32 0.31 0.29
primary
secondary
higher
Mother's Frequency of Watching 0.82 0.88 0.87 0.88 0.82 0.61
Television (At Least Once per Week
COVID-19 Disruption, Small Island Economies and Regional Development in Indonesia

= 1)
Floor Type (Earth/Sand/Other = 1) 0.05 0.05 0.03 0.01 0.03 0.13
Observations 33,803 9,375 9,548 3,307 5,490 6,083

Total Variation of Assets 6.34 2.44 7.32 2.00 6.82 2.10 6.78 2.12 5.79 2.41 4.34 2.49
Table 2 presents statistical description of the variables used in this
study. Diarrhoea incidence between island groups varied (see Appendix 4) .
Kalimantan had the highest diarrhoea incidence amounting to 18%, which was
followed by Sulawesi with proportion of 17 %, and Sumatra with proportion of
15 %. Java-Bali and Nusa Tenggara, Maluku, and Papua (shown as others) had
the lowest diarrhoea incidence with approximately 13% on average.
Drinking water was classified into surface water, unprotected dug well
or spring, bottled or refilled water, protected dug well or spring, and piped
water. Estimation results indicate that on full sample, most of the respondents
used protected dug well or spring, bottled/refilled, and piped as source of
drinking water. This indicates that in general most of under-five (U5) children
in Indonesia drink improved drinking water. On the other hand, the proportion
of piped water was 10 %, meaning that Indonesia was yet to provide safe and
free drinking water for most of the population. Following Java-Bali’s highest
proportion of improved drinking water at 90 %, Sumatra’s drinking water has
improved by 81%, Kalimantan’s by 79%, Sulawesi’s at 77%, and Nusa Tenggara,
Maluku, and Papua’s (expressed as ‘others’) by 72%.
Sanitation facilities were classified into no facility, pit latrine, shared/public
flush toilet, flush toilet with no septic tank, and flush toilet with septic tank.
Around 69 % of the respondents used flush toilet with/without septic tank,
meaning that most of U5 children used improved sanitation (see Appendix 5).
However, access to improved sanitation varies between island groups. Region
Java-Bali had the highest proportion of improved sanitation at 81%. Proportion
of improved sanitation in Sumatra, Kalimantan, and Sulawesi was 75, 71, and
60, respectively. Nusa Tenggara, Maluku, and Papua (expressed as others) had
the lowest proportion of improved sanitation at 59.
Handwashing facilities were divided into not available, facility on premises
without water/soap, and facility on premises with water and soap. On full
sample, the proportion of basic handwashing facility were 77%. There is no
surprise that Java-Bali had the highest proportion of handwashing facility on
premises with water and soap at 88%. Furthermore, the proportion of type of
handwashing facility in Sumatra, Sulawesi, Kalimantan, and Nusa Tenggara,
Maluku, and Papua (expressed as ‘others’) were 82%, 78%, 73%, and 61%,
respectively. This reflects that there was inequality of basic handwashing facility
ownership between households with U5 children in Indonesia.
in Indonesia
Point-of-use treatment had the proportion of 67% on full sample, meaning
that most of households with U5 in Indonesia considered the quality of their
drinking water was not ideal. The proportion of point-of-use treatment among
households with U5 in Java-Bali, Sumatra, Kalimantan, Sulawesi, and Nusa
Tenggara, Maluku, and Papua (expressed as others) were 65%, 65%, 63%, 70%,
and 74%, respectively. The mean of proportion of open defecation was 0.15.
Kalimantan and Sulawesi had higher mean of proportion of open defecation
compared to other island groups at 0.20. On the other hand, Java-Bali had the
lowest mean of proportion of open defecation with 0.10.
The mean of respondent’s age in this study was 1.99 years old. About
52% of the respondents were male. The mean of mother’s years of schooling
was 9.71. Moreover, 82% of mother respondents watched television at least
once per week. Only 5% of the respondents used earth/sand/other as floor
material at their house. The mean of variation of assets were 6.35, meaning that
respondent’s households had 6 types of assets on average. About 47% of the
respondents lived in urban areas, meaning that most of the respondents lived
in rural areas.
Figure 2 presents progress on improved drinking water, improved
sanitation, and basic hygiene among under-five (U5) households in Indonesia.
The Y-axis shows the percentage of respective indicators, while X-axis shows
the survey round. Figure 2 shows that there was substantial improvement of
those infrastructure’s provision between 2012 and 2017. The average increase
of improved drinking water, improved sanitation, and basic hygiene were 12%,
11.6%, and 13%, respectively. This illustrates that Indonesian government
had successfully improved access on clean drinking water, sanitation, and
handwashing facility for U5 households.
Figure 2 also shows that Nusa Tenggara, Maluku, and Papua (expressed
as ‘others’) had better progress than other regions. In 2012, the proportion of
improved drinking water, improved sanitation, and basic handwashing facility
in those island groups were 56%, 47%, and 41%, respectively. In 2017, the
proportion of those infrastructures in those regions were 82%, 66%, and 72%,
respectively. On the other hand, Kalimantan had worse progress than other
island groups. In 2012, the proportion of improved drinking water, improved
sanitation and basic handwashing facility in that region was 70%, 63%, and
77%, respectively. In 2017, the proportion of those infrastructures in that region
was 82%, 76%, and 83%, respectively.
Figure 2. Progress on Improved Drinking Water, Improved Sanitation,
and Basic Hygiene Provisions among U5 Households in 2012 and 2017, by
Island Groups
Note: Improved drinking water includes piped water, protected dug well or spring, and bottled/
refilled water. Improved sanitation includes flush toilet with/without septic tank. Basic
hygiene includes facility on premises with water and soap.
Source: Author’s estimations using Survei Demografi & Kesehatan Indonesia (SDKI) or Indonesia
Demographic and Health Survey (IDHS) 2012 and 2017.
Indonesian government had established Penyediaan Air Minum dan
Sanitasi Berbasis Masyarakat (Pamsimas) programme to improve access on
drinking water, sanitation, and handwashing facility. This programme involves
local people to be stakeholders within their respective village/kelurahan
(Kementerian Pekerjaan Umum dan Perumahan Rakyat, 2018). Pamsimas has
been implemented for 14 years. Pamsimas implementation was divided into
three stages. The first stage is Pamsimas I. It was implemented in 2008-2012,
followed by the second stage (Pamsimas II) implemented in 2013-2015. The
third stage (Pamsimas III) was implemented in 2016-2020. Pamsimas I and
Pamsimas II had provided access to improved drinking water and improved
sanitation for about 10.4 million people in 12 thousand villages/kelurahan.
Pamsimas III was targeted to provide those access in another 15 thousand
villages/kelurahan. 1
1 https://pamsimas.pu.go.id/profil/#
in Indonesia
Table 3 presents progress on diarrhoea incidence among under-five
(U5) between 2012 and 2017 in Indonesia. Eventhough clean drinking water,
sanitation, and handwashing facility provisions were already substantially
improved, diarrhoea incidence did not decrease significantly.
Table 3. Diarrhoea Incidence among U5 in 2012 and 2017, by Island
Groups (%)
Survey Wave
Island Groups Change
2012 2017
Java-Bali 12.9 12.8 -0.1
Sumatra 14.1 15.3 1.2
Kalimantan 18.4 16.6 -1.8
Sulawesi 17.6 15.6 -2.0
Others 12.3 13.3 1.0
Average 15.1 14.7 -0.4
Source: Author’s estimations using Survei Demografi Kesehatan Indonesia (SDKI) of 2012
and 2017.
Diarrhoea incidence among U5 in Indonesia was reported by 15.1% in
2012 and 14.7% in 2017. This indicated that diarrhoea is a complex issue
resulting from multiple factors (Bosch et al., 2000; Brenneman & Kerf, 2002;
Komarulzaman et al., 2017). Table 3 also shows that Sumatra and Nusa
Tenggara, Maluku, and Papua (expressed as ‘others’) experienced an increase
of diarrhoea incidence, while other regions experienced otherwise. Java-Bali
had lower diarrhoea incidence of U5 children compared to other regions, which
are 12.9% in 2012 and 12.8% in 2017. On the other hand, Kalimantan had the
highest diarrhoea incidence, which are 18.4% in 2012 and 16.6% in 2017.
RESULTS
As indicated in Table 4, the better the quality of safe drinking water the lower
the incidence of diarrhoea among children. Typical households that consumed
drinking water from bottle/refilled water and piped water were found less likely
to have infant or U5 suffering from diarrhoea compared with those consuming
surface water. Estimation of full sample is consistent with estimation for sub
sample year 2012. Sanitation infrastructure in the form of defecation type
significantly reduces incident diarrhoea among infant, especially type pit latrine
and flush toilet using septic tank compared with typical households without
sanitation facility. This finding only applies to sub sample year 2017.
In terms of handwashing facility, the finding is quite surprising. Typical
households that have facility on premises with water and soap was found
associated with larger probability incidence of diarrhoea among infant U5. This
is found for full sample and sub sample year 2012. For variables drinking water
treatment and environmental hygiene, only proportion of free open defecation
on full sample model indicated significant association. The larger the practice of
free open defecation, the larger probability infant suffer for diarrhoea. In terms
of individual characteristic, this study found as follows. The older the infant in
typical households the less likely they suffer for diarrhoea. Boys infants were
more likely got diarrhoea compared with girls infant. The more variety of
assets owned by family, the less likely U5 suffer from diarrhoea. Finally, the more
frequent baby’s mother watching TV, the larger probability the U5 experience
diarrhoea.
Table 4. Logistic Regressions Results on Diarrhoea, by Full Sample and
Survey Wave (Odds Ratio)
(1) (2) (3)
VARIABLES
Constant 0.29*** 0.30*** 0.28***
(0.05) (0.08) (0.09)
A. BASIC INFRASTRUCTURE
Drinking Water Classifications (Ref: Surface Water)
Unprotected Dug Well or Spring 0.81** 0.75** 0.94
(0.08) (0.10) (0.15)
Bottled/Refilled Water 0.80* 0.67** 0.97
(0.09) (0.11) (0.16)
Protected Dug Well or Spring 0.89 0.91 0.94
(0.09) (0.12) (0.14)
Piped Water 0.78** 0.63*** 1.02
(0.09) (0.09) (0.17)
Sanitation Classifications (Ref: No Facility)
Pit Latrine 0.91 1.01 0.77*
(0.09) (0.12) (0.12)
Flush Toilet which is Shared/Public 1.08 1.10 1.05
(0.08) (0.11) (0.12)
Flush Toilet with No Septic Tank 1.12 1.08 1.13
(0.09) (0.13) (0.13)
in Indonesia
Flush Toilet using Septic Tank 0.91 0.99 0.84*
(0.06) (0.08) (0.08)
Handwashing Facility Classifications (Ref: Not Available)
Facility on Premises without Water/Soap 1.12 1.15 1.05
(0.08) (0.11) (0.13)
Facility on Premises with Water and Soap 1.09* 1.13* 1.02
(0.06) (0.07) (0.10)
B. DRINKING WATER TREATMENT & ENVIRONMENTAL HYGIENE
Point-of-Use Treatment (Yes = 1) 0.96 0.88 1.03
(0.06) (0.08) (0.08)
Proportion of Open Defecation 1.22* 1.16 1.31
(0.13) (0.18) (0.22)
C. HOUSEHOLD CHARACTERISTIC /CONTROL VARIABLES
Age of the Children (Month) 0.99*** 0.98*** 0.99***
(0.00) (0.00) (0.00)
Sex of the Children (Male = 1) 1.23*** 1.30*** 1.16***
(0.04) (0.06) (0.05)
Mother's Education Attainment 0.85 0.94 0.78
(0.11) (0.17) (0.18)
Mother's Frequency of Watching Television 1.18*** 1.20** 1.15**
(At Least Once per Week = 1) (0.06) (0.09) (0.08)
Floor Type (Earth/Sand/Other = 1) 0.94 1.09 0.78*
(0.08) (0.11) (0.10)
Total Variation of Assets 0.95*** 0.94*** 0.97*
(0.01) (0.01) (0.02)
Survey Wave Dummy Yes Yes Yes
Provincial Dummy Yes Yes Yes
Prob < chi2 0.00 0.00 0.00
Pseudo R2 0.0235 0.0339 0.0208
Observations 33,803 17,102 16,701
Notes: ***, **, and * denote statistically significant at 0.01, 0.05, and 0.1, respectively. Robust
seen form in parentheses.
Source: Author’s estimations using SDKI (Indonesia Demographic and Health Survey/IDHS)
data of 2012 and 2017.
Table 5 below provided similar results, but they were disaggregated by
groups of islands.
Table 5. Logistic Regressions Results on Diarrhoea, Disaggregated by
Island Groups
(1) (2) (3) (4) (5) (6)
VARIABLES Full Java-Bali Sumatra Kalimantan Sulawesi Others
Sample
Constant 0.29*** 0.10*** 0.31*** 0.26*** 0.36*** 0.22***
(0.05) (0.07) (0.11) (0.12) (0.14) (0.07)
Unprotected Dug Well or 0.81** 1.13 1.19 0.65** 0.69* 0.96
Spring (0.08) (0.50) (0.21) (0.13) (0.14) (0.17)
Bottled/Refilled Water 0.80* 1.45 1.07 0.65* 0.76 0.58**
(0.09) (0.65) (0.22) (0.15) (0.18) (0.14)
Protected Dug Well or Spring 0.89 1.24 1.25 0.70** 1.02 0.87
(0.09) (0.54) (0.22) (0.12) (0.21) (0.15)
Piped Water 0.78** 1.12 1.14 0.54*** 0.73 0.86
(0.09) (0.50) (0.23) (0.12) (0.17) (0.18)
Pit Latrine 0.91 1.27 0.83 1.22 0.72 0.92
(0.09) (0.28) (0.12) (0.39) (0.15) (0.19)
Flush Toilet which is 1.08 0.93 1.15 1.76*** 0.81 1.27
Shared/Public (0.08) (0.15) (0.17) (0.39) (0.11) (0.20)
Flush Toilet with No Septic 1.12 1.24 1.03 1.30 0.94 1.27
Tank (0.09) (0.19) (0.14) (0.28) (0.21) (0.25)
Flush Toilet using Septic 0.91 0.86 0.91 1.30 0.76** 1.03
Tank (0.06) (0.12) (0.10) (0.24) (0.10) (0.16)
Facility on Premises without 1.12 1.30 1.06 0.83 1.10 1.16
Water/Soap
(0.08) (0.24) (0.13) (0.21) (0.16) (0.16)
Facility on Premises with 1.09* 1.09 1.00 1.34* 1.09 1.07
Water and Soap
(0.06) (0.14) (0.09) (0.22) (0.11) (0.11)
in Indonesia
Point-of-Use Treatment 0.96 1.37*** 0.84 0.91 0.92 0.78*
(Yes = 1) (0.06) (0.15) (0.10) (0.14) (0.12) (0.11)
Proportion of Open 1.22* 1.34 1.64*** 1.39 0.92 0.79
Defecation (0.13) (0.31) (0.31) (0.36) (0.20) (0.19)
Age of the Children (Month) 0.99*** 0.98*** 0.98*** 0.99*** 0.99*** 0.99***
(0.00) (0.00) (0.00) (0.00) (0.00) (0.00)
Sex of the Children 1.23*** 1.13* 1.19*** 1.20** 1.44*** 1.28***
(Male = 1) (0.04) (0.07) (0.07) (0.11) (0.11) (0.10)
Mother's Education 0.85 1.13 0.72 0.67 0.83 1.19
Attainment = 3, higher (0.11) (0.54) (0.19) (0.25) (0.22) (0.28)
Mother's Frequency of 1.18*** 0.92 1.32*** 1.27 1.37*** 1.10
Watching Television (At (0.06) (0.09) (0.13) (0.20) (0.15) (0.11)
Least Once per Week = 1)
Floor Type (Earth/Sand/ 0.94 1.03 0.97 0.35 1.09 0.88
Other = 1) (0.08) (0.15) (0.16) (0.27) (0.23) (0.11)
Total Variation of Assets 0.95*** 1.00 0.94*** 0.93** 0.94*** 0.95**
(0.01) (0.02) (0.02) (0.03) (0.02) (0.02)
Survey Wave Dummy Yes Yes Yes Yes Yes Yes
Provincial Dummy Yes Yes Yes Yes Yes Yes
Prob < chi2 0.00 0.00 0.00 0.00 0.00 0.00
Pseudo R2 0.0235 0.0257 0.0267 0.0398 0.0261 0.0200
Observations 33,803 9,375 9,548 3,307 5,490 6,083
Notes: ***, **, and * denote statistically significant at 0.01, 0.05, and 0.1, respectively. Robust SE
in parentheses.
Source: Author’s estimations using SDKI (Indonesia Demographic and Health Survey/IDHS)
data of 2012 and 2017.
The summary of regression results presented in Table 5 is similar with
the one presented in Table 4, especially coefficients estimation under column
heading (1). Coefficients under heading column (2) until (6) are estimation
for subsample island as indicated by the name below the heading column
numbers. National or full sample data indicated access to safe drinking water
that is associated with lower incidence of diarrhoea. Disaggregating the sample
by island (see Appendix 3) found that coefficient for subregion Kalimantan’s
estimation quite consistent with the finding of national sample. The better
the drinking water access coefficient odd ratio of diarrhoea. The coefficient
indicates that infant in typical households in Kalimantan that main source of
drinking water from protected dug well or spring about less likely experiencing
diarrhoea about 30%2 compared to those family with infant that consume
drinking water from surface water.
The result for access to handwashing facility is surprising. This study found
that typical households having handwash facility equipped with soap have
higher probability of the risks of their infant suffer from diarrhoea, i.e., about
9% (full sample) and about 34% (for the subregion Kalimantan). The argument
for this unexpected finding is that although handwashing is a healthy lifestyle or
habitual process, but if the family members of the household that has the facility
are not aware of it, it will have no impact on health outcome.
Regarding drinking water treatment and environmental hygiene, this
study found mixed evidence. The larger proportion of households practice open
defecation, the more likely diarrhoea incidence occurs, and this is found both
for full sample estimation and in the subregion of Sumatera. On the other hand,
the habit of throwing away HDS without proper treatment is also positively
correlated with higher probability of diarrhoea. This finding was consistent
for estimation coefficients in the region Jawa-Bali, Kalimantan and the rest of
the region including Papua. In terms of infant and household characteristics,
the estimation results of subsample by region or island have consistent finding
compared to the coefficient with disaggregation by year of survey.
Before proceeding to the robustness check due to omitted confounding
variable bias, violation of classical assumption will first be discussed. Following
Gujarati & Porter (2009:157), 3 of 5 classical assumptions need to be evaluated
due to its consequence for biased estimation. The autocorrelation assumption.
This problem typically occurs when the model that is used deal with time series
data. This study utilises pooled cross section data, therefore autocorrelation
problem is less likely to occur. Homoschedasticity assumption typically occurs
in cross sectional data, such as SDKI data. To minimizes the consequences of
heteroscedasticity problem, this study added provincial and year fixed effects
in the estimation. Multicollinearity assumption. This classical violation is more
likely to occur when the estimation results have the conditions such as: (1) the
2 Following discussion in https://stats.oarc.ucla.edu/stata/faq/how-do-i-interpret-odds-ratios-in-

logistic-regression/; https://www.statology.org/interpret-odds-ratio-less-than-1/; the interpretation
of odds ratio coefficient is Change in Odds %: (OR-1) * 100; For example, the odds ratio (OR) for age is
0.92. Thus, we could calculate: Change in Odds %: (0.92 – 1) * 100 = -8%.
in Indonesia
value of R2 or adjusted R2 is low but mostly significant partial statistical test; and
(2) most of the estimation coefficients have the opposite sign than is expected
theoretically or empirically.
Goodness of Fit of estimation model in Table 4 and 5 indicated the
P-value of Chi Square Test for joint test is accepted, implying that the overall
independent variables jointly associated with the incidence of diarrhoea among
U5 in Indonesia. The Pseudo R2 for qualitative regression model in Table 4 and
5 are exceptionally low, just around 0.02-0.03 for full and disaggregated sample
by survey wave (Table 4), and around 0.02-0.04 for disaggregated sample by
region (Table 5). All of subindicators of independent variable have at least one
variable significant and consistent with the previous finding, except for the
sanitation. These findings give us chance to conclude that the model is less
likely to suffer from multicollinearity problem.
Robustness Test
Focusing on the results of Table 4 and 5, it is interesting to figure out that
better and safe drinking water is somehow significantly associated with lower
diarrhoea incidence. The results, however, show that improved sanitation for
most cases (national or subregions; full sample or dis-aggregated by year of
survey) is not significantly associated with the incidence of diarrhoea.
The first effort for testing the robustness result is conducted by estimating
the effects of age of U5 that classify the impact of independent variable to the
dependent variable. The estimation is then provided in Table 6. Table 6 and Table
4 and 5 can be seen and compared in which the coefficients—both in terms of
magnitude and sign—have similarity in the three specifications. Better access
to safe drinking water, improved sanitation, and living in hygienic environment
do not guarantee lower incidence of diarrhoea among U5. Typical households
that have piped running water in their households are less likely to have their
infant got diarrhoea, i.e., 22% (full sample) and about 33% (families with infant
36-59 months). However, the magnitude was not significantly different with
families that consume drinking water from unprotected dug or spring water
(surface water). This finding is consistent with Table 4, 5 and 6.
Something interesting to note from Table 6 is that the correlation between
drinking water quality and improved sanitation is only statistically significant
reducing diarrhoea incidence for typical households with infant older than 36
months. Furthermore, results presented in Table 4,5, and 6 indicate that better
basic infrastructures—i.e., improved sanitation and handwashing facility—
are not correlated with lower diarrhoea incidence. It is realised that omitted
confounding variables bias still exists in the empirical model (Table 6), therefore
in the next exercise, hygiene behaviour and environmental hygiene will be
considered as instrumental variable for the quality of safe drinking water, as
the latter is directly consumed by the household members.
Table 6. Diarrhoea Incidence and U5 Age
(1) (2) (3) (4)
VARIABLES Full 0-23 24-35 36-59
Sample Months Months Months
Constant 0.29*** 0.15*** 1.06 0.38***
(0.05) (0.04) (0.51) (0.14)
Unprotected Dug Well or Spring 0.81** 0.86 0.75 0.75*
(0.08) (0.12) (0.15) (0.11)
Bottled/Refilled Water 0.80* 0.88 0.78 0.66**
(0.09) (0.14) (0.17) (0.12)
Protected Dug Well or Spring 0.89 0.96 0.85 0.83
(0.09) (0.13) (0.16) (0.12)
Piped Water 0.78** 0.85 0.81 0.67**
(0.09) (0.13) (0.18) (0.12)
Pit Latrine 0.91 0.89 0.90 0.90
(0.09) (0.12) (0.18) (0.16)
Flush Toilet which is Shared/Public 1.08 1.04 1.13 1.07
(0.08) (0.11) (0.18) (0.14)
Flush Toilet with No Septic Tank 1.12 1.09 0.95 1.30*
(0.09) (0.13) (0.16) (0.18)
Flush Toilet using Septic Tank 0.91 0.90 0.84 0.95
(0.06) (0.08) (0.12) (0.11)
Facility on Premises without Water/Soap 1.12 1.11 1.12 1.19
(0.08) (0.11) (0.18) (0.16)
Facility on Premises with Water and Soap 1.09* 1.03 1.24* 1.14
(0.06) (0.07) (0.14) (0.11)
in Indonesia
(1) (2) (3) (4)
VARIABLES Full 0-23 24-35 36-59
Sample Months Months Months
Point-of-Use Treatment (Yes = 1) 0.96 0.94 1.10 0.87
(0.06) (0.08) (0.13) (0.09)
Proportion of Open Defecation 1.22* 1.09 1.40 1.20
(0.13) (0.17) (0.31) (0.22)
Proportion of Unsafe Stool Disposal of Youngest 1.08 1.27*** 0.84 1.13
Child
(0.07) (0.11) (0.10) (0.12)
Owns Poultry/Chicken (Yes = 1) 0.98 1.01 0.93 0.97
(0.04) (0.05) (0.07) (0.06)
Type of Region (Urban = 1) 0.96 0.96 1.00 0.93
(0.04) (0.06) (0.09) (0.07)
Age of the Children (Month) 0.99*** 1.04*** 0.96*** 0.98***
(0.00) (0.00) (0.01) (0.00)
Sex of the Children (Male = 1) 1.23*** 1.20*** 1.36*** 1.20***
(0.04) (0.06) (0.09) (0.07)
Mother's Education Attainment = 3, higher 0.85 0.90 1.08 0.62**
(0.11) (0.17) (0.31) (0.13)
Mother's Frequency of Watching Television (At 1.18*** 1.27*** 1.13 1.06
Least Once per Week = 1) (0.06) (0.09) (0.11) (0.09)
Floor Type (Earth/Sand/Other = 1) 0.94 0.91 0.93 1.01
(0.08) (0.11) (0.16) (0.14)
Total Variation of Assets 0.95*** 0.94*** 0.92*** 1.01
(0.01) (0.01) (0.02) (0.02)
Survey Wave Dummy Yes Yes Yes Yes
Provincial Dummy Yes Yes Yes Yes
Pseudo R 2
0.0235 0.0329 0.0299 0.0221
Observations 33,803 13,899 6,601 13,303
Notes: ***, **, and * denote statistically significant at 0.01, 0.05, and 0.1, respectively. Robust SE
in parentheses.
Source: Author’s estimations using Indonesia Demographic and Health Survey (IDHS or SDKI)
2012 and 2017.
As discussed previously, infected food and drinking water could be the
main media of bacteria transmission for diarrhoea cases. Therefore, the next
empirical model should attempt to utilise source of drinking water equipped
with behavioural and environmental hygiene to estimate the probability of U5
experiencing diarrhoea. The empirical model to be estimated can be formulated
as follows:
Yi = Drinking_wateri
= α0 + α1 Unsafe_Disposali + β2 Raise_Cattlei
+ β3 Open_Defeci + ai + ui …(2)
Yi = Diarrhoeai
= α0 + β1 Infant_Agei + β2 Drinking-Water-Sourcei + ai + ui …(2)
The estimation result is presented in Table 7. The outcome variables,
source of drinking water in the first stage regression and incidence of diarrhoea
in the second stages presented under full sample (see Appendix 5) , infants’
age below 24 months and infants’ age over 24 months but below 60 months.
Coefficients under column (1a), (2a), and (3a) are the first stage estimation
resulting from Equation (2); whereas the rest column (1b), (2b), and (3b) are
the second stages estimation resulting from Equation (3). As shown in the first
stage regression, the three factors that are related to non-hygiene—i.e., unsafe
stool disposal, raised cattle in backyard without proper cattle shed, and practice
open defecation—in typical households have caused the latter to be less likely
to have proper source of drinking water. The magnitude of non-hygiene is about
11% for unsafe disposal and raising cattle around the house.
in Indonesia
308
Table 7. Hygiene , Drinking Water Source, and Incidence of Diarrhoea
Full Sample (1) Infants’ Age <24 months (2) Infants’ Age 25<Xi<60 months (3)
First Stage Second Stage First Stage Second Stage First Stage Second Stage
Independent Variables Regression (1a) Regression (1b) Regression (2a) Regression (2b) Regression (3a) Regression (3b)
Sources of Experienced Sources of Experienced Sources of Experienced
Drinking Water Diarrhoea Drinking Water Diarrhoea Drinking Water Diarrhoea
(Yes=1) (Yes=1) (Yes=1)
Constant 3.287*** -0.314*** 3.293*** 0.165*** 3.268*** -0.041
(0.0283) (0.0469) (0.036) (0.188) (0.072) (0.273)
Children’s age -0.003 -0.088*** -0.007*** 0.044*** 0.0002 -0.144***
(0.004) (0.006) (0.008) (0.013) (0.016) (0.030)
Unsafe Stool Disposal -0.085*** -0.072*** -0.110***
(yes=1) (0.011) (0.015) (0.018)
Own and Raise poultry in -0.049*** -0.045*** -0.056***
backyard (yes=1) (0.0111) (0.0141) (0.017)
Open defecation (yes=1) -0.459*** -0.468** -0.441***
Sources of Drinking Water -0.313*** -0.347*** -0.262**

(0.015) (0.019) (0.025)
(0.046) (0.058) (0.078)

Province dummy FE Yes Yes Yes Yes Yes Yes
Goodness of Fit Model

Year dummy FE Yes Yes Yes Yes Yes Yes
Wald Test of exogeneity Chi sq. 42.13*** 34.23*** 9.50**

Chi sq. (dof) (0.000) (0.00) (0.001)
Number of obs 33,691 20,460 13,231

F(38, 33652) 80.99*** 48.69** 107.58***
(0.000) (0.000) (0.000)
Adj R-squared 0.0828 0.0831 0.0847
Notes: ***, **, and * denote statistically significant at 0.01, 0.05, and 0.1, respectively. Robust SE in parentheses.
Source: Author’s estimations using Indonesia Demographic and Health Survey (IDHS or SDKI) 2012 and 2017.
Open defecation practice is associated with lower probability of typical
households that have proper source of drinking water. Furthermore, this
finding is reflected in the second stages regression. The coefficients of source
of drinking water are varied between -0.262 (households with infant more than
two-years old) to -0.347 (households with infant
less than two-years old). The coefficient is the odd ratio in the probit
regression. These coefficients imply that typical household with low hygienist
practices could contaminatethe drinking water quality and furthermore
increase the probability the growth of bacteria. This condition can increase the
probability of infant suffer for diarrhoea.
ESTIMATION RESULTS
Diarrhoea incidence among U5 was lower when the quality of source of drinking
water was improved. This estimations results are in line with Esrey et al.
(1991), Komarulzaman et al. (2017), Patunru (2015), and Wolf et al. (2018).
Moreover, the estimation result shows that piped water reduces the odds of U5
having diarrhoea by 23%, which is higher compared to other source of drinking
water. This percentage is similar to Komarulzaman et al. (2017) which is 24%
and Fewtrell et al. (2005) which is 22%. This result shows that piped water
provisions need to be improved since the proportion of piped water as source
of drinking water was lower than other improved source.
Better quality of drinking water was also associated with diarrhoea
incidence reduction in Kalimantan and Nusa Tenggara, Maluku, and Papua
(shown as others). These results indicate that U5 that use improved source
drinking water has higher health benefits in low infrastructure regions, where
the circumstances are more difficult (Komarulzaman et al., 2017). On the other
hand, drinking water did not directly affect diarrhoea among U5 in Java-Bali,
Sumatra, and Sulawesi. This might be caused by two factors. First, the quality
of piped water in developing countries were decreased due to pollution and
under-maintenance (Shaheed et al., 2014). Second, there were other factors that
influence diarrhoea, indicate the complexity of diarrhoea disease. Agustina et al.
(2013) reported that diarrhoea can be transmitted through food contamination
caused by poor food-hygiene practices.
in Indonesia
Estimation results on sanitation classifications indicate that only the
cleanest sanitation type that could directly reduce diarrhoea incidence among
U5 in Indonesia. This result is in line with Andres et al, (2014), Cameron & Olivia
(2011), Esrey et al. (1991), Patunru (2015), and Wolf et al. (2018). On the other
hand, disaggregation results by island groups show that improved sanitation
did not necessarily affect diarrhoea among U5 in Indonesia. These results are
similar to Komarulzaman et al. (2017) finding.
In contrast to previous studies (Noguchi et al., 2021), the availability of
handwashing facility increases the odds of having diarrhoea among U5 in
Indonesia. It might indicate that handwashing behaviour is more important
on diarrhoea reduction. A meta-analysis on 29 Randomized Controlled
Trials studies by Ejemot-Nwadiaro et al. (2021) concluded that handwashing
promotion intervention reduced diarrhoea incidence by 30%. Khan et al.,
(2021) reported that children aged 8-12 years who washed their hands before
and after preparing foods and after defecated had 70% lower risk of suffering
from diarrhoea in India.
Meanwhile, Patunru (2015) reported that provinces with higher
handwashing practice percentage tend to have lower diarrhoea prevalence in
eastern part of Indonesia. Generally speaking, however, Indonesians have low
awareness on the importance of handwashing practice. Data from Kementerian
Kesehatan (2019) show that only 49.8% of population over 10 years old washed
their hands properly in 2018. This finding illustrates that the availability of
handwashing facility does not reflect Indonesian household’s clean and healthy
behaviour.
Point-of-use treatment was found not associated with diarrhoea incidence.
This condition might indicate that Indonesian households do not treat their
water effectively (Komarulzaman et al., 2017). Sodha et al. (2011) and Vollaard
et al. (2004) reported that E. Coli bacteria was still found on treated water in
Indonesia. On the other hand, estimation result on point-use-treatment shows
that this behaviour is significantly negatively associated with diarrhoea in Nusa
Tenggara, Maluku, and Papua (expressed as ‘others’). This finding shows that
behavioural change in that island groups has prevented children from diarrhoea
(see Appendix 6).
The proportion of open defecation within community was found
associated with diarrhoea incidence escalation among U5 in Indonesia. This can
be interpreted that the risk of suffering from diarrhoea for children who live in
dirty and polluted community remains high even though they have improved
drinking water and sanitation in their house. Therefore, environmental hygiene
is indicated as one of the main indicators of diarrhoea among U5 children in
Indonesia.
CONCLUSION
This study found that better quality of source of drinking water and sanitation
is associated with lower diarrhoea incidence among U5. The SDKI data indicated
that the majority of households in Indonesia consume drinking water either
from protected dug/well/spring or refilled bottle water. These types of drinking
water are safer and better protected from being contaminated by bacteria. On
the other hand, government of Indonesia is also on smart progress promoting
and delivering PAMSIMAS to provide safe drinking water stretching to remote
area in Eastern area of Indonesia since 2010s. The estimation results in this
study found that the safer the drinking water, the lower the risk of suffering
from diarrhoea among infant.
This study utilised hand wash facilities as one variable that associate
with diarrhoea incidence. The availability of hand wash facility is found not
significant statistically reduce incidence of diarrhoea. This might reflect the
behavioural aspect, practice washing hand soon after doing something still not
as good habit. Despite satisfactory progress on provision of infrastructure for
safe drinking water and improved sanitation, healthy lifestyle indicated by less
open defecation free, managed stool disposal, as well as the livestock manure
still in question.
This study did empirical estimation for the purposes of robustness test.
It is found that less hygienist practice and behaviour associate with typical
household with lower access for safe drinking water. Non-hygienist household
condition due to open defecation, cattle raised around the yard, and no treatment
for stool disposal could contaminated the drinking water storage. Storage water
easily contaminated by bacteria. This condition can increase the probability of
infant infected and suffer for diarrhoea. Policy relevance regarding this study
finding is that policy makers should prioritize promoting and educating the
community the importance of healthy lifestyle to prevent diarrhoea incidence
and other communicable diseases.
in Indonesia
This study also suggested that the protective effect of piped water would
be effective to prevent diarrhoea, particularly in poor situations. However, this
suggestion was not necessarily true due to poor condition of piped water in
developing countries. Therefore, increasing piped water provisions as well
as improving sanitation facilities, promoting clean and healthy behaviour,
and preserving natural environment are imperative to alleviate diarrhoea in
Indonesia.
REFERENCES
Agustina, R., T.P. Sari, S. Satroamidjojo, I.M. Bovee-Oudenhoven, E.J. Feskens,
& F.J. Kok (2013). 'Association of food-hygiene practices and diarrhea
prevalence among Indonesian young children from low socioeconomic
urban areas.' BMC Public Health, 13(1): 977. Available at: https://doi.
org/10.1186/1471-2458-13-977
Andres, L. A., B. Briceno, C.Chase, & J.A. Echenique (2014). "Sanitation and
externalities: Evidence from early childhood health in rural India." Policy
Research Working Paper Series (No. 6737; Policy Research Working Paper
Series). The World Bank. Available at: https://ideas.repec.org/p/wbk/
wbrwps/6737.html
Ayalew, A., W., W.T.Mekonnen, S. W.Abaya, & Z.A.Mekonnen (2018). 'Assessment
of Diarrhea and Its Associated Factors in Under-Five Children among Open
Defecation and Open Defecation-Free Rural Settings of Dangla District,
Northwest Ethiopia.' Journal Environmental Public Health, 4271915.
Available at: https://doi.org/doi: 10.1155/2018/4271915
Badan Pusat Statistik. (2018). Potensi Desa (PODES) Maret 2018. Jakarta: Badan
Pusat Statistik.
Badan Pusat Statistik. (2019). Survei Sosial Ekonomi Nasional (SUSENAS) Maret
2019. Jakarta: Badan Pusat Statistik.
Becker, G. S. (1962). 'Investment in Human Capital: A Theoretical Analysis.'
Journal of Political Economy, 70(5): 9–49. JSTOR.
Ben-Porath, Y. (1967). The Production of Human Capital and the Life Cycle
of Earnings. Journal of Political Economy, 75(4): 352–365. Available at:
https://doi.org/10.1086/259291
Bosch, C., K.Hommann, G. Rubio, G., Sadoff, C., & Travers, L. (2000). Water,
Sanitation and Poverty. PRSP Sourcebook.
Brenneman, A., & Kerf, M. (2002). Infrastructure & Poverty Linkages: A
Literature Review. World Bank. https://www.ilo.org/emppolicy/pubs/
WCMS_ASIST_8281/lang--en/index.htm
Budge, S., A,H.Parker, P.T. Hutchings, & C. Garbutt (2019). 'Environmental enteric
dysfunction and child stunting.' Nutrition Review, 77: 240-253.
Cameron, L., & S. Olivia (2011). "Sanitation and Health: The Past, The Future and
Working Out what Works." In Employment, Living Standards and Poverty in
Contemporary Indonesia. Institute of Southeast Asian Studies. 245–272.
Clasen, T., W.P. Schmidt, T.Rabie, I. Roberts, & S. Cairncross (2007). 'Interventions
to improve water quality for preventing diarrhoea: Systematic review and
meta-analysis.' BMJ, 334(7597), 782. Available at: https://doi.org/10.1136/
bmj.39118.489931.BE
Cumming, O., & S.Cairncross. (2016). 'Can water, sanitation and hygiene help
eliminate stunting? Current Evidence Policy Implication.' Maternality Child
Nutrition 12 (Supplement 1), 91-105.
Ejemot-Nwadiaro, R. I., J.E. Ehiri, D. Arikpo, W.M. Meremikwu, & J.A.
Critchley (2021). 'Hand‐washing promotion for preventing diarrhoea.'
Cochrane Database of Systematic Reviews, 1. Available at: https://doi.
org/10.1002/14651858.CD004265.pub4
Esrey, S. A., J.B. Potash, L. Roberts, & C. Shiff (1991). 'Effects of improved water
supply and sanitation on ascariasis, diarrhoea, dracunculiasis, hookworm
infection, schistosomiasis, and trachoma.' Bulletin of the World Health
Organization, 69(5): 609–621.
Fewtrell, L., R.B. Kaufmann, D. Kay, W. Enanoria, L. Haller, & J.M. Colford (2005).
'Water, sanitation, and hygiene interventions to reduce diarrhoea in less
developed countries: A systematic review and meta-analysis.' The Lancet
Infectious Diseases, 5(1): 42–52. Available at: https://doi.org/10.1016/
S1473-3099(04)01253-8
in Indonesia
Fiebelkorn, A. P., B. Person, R.E. Quick, S.M. Vindigni, M. Jhung, A. Bowen, &
P.L. Riley (2012). 'Systematic review of behavior change research on
point-of-use water treatment interventions in countries categorized
as low- to medium-development on the human development index.'
Social Science & Medicine, 75(4): 622–633. https://doi.org/10.1016/j.
socscimed.2012.02.011
Grossman, M. (1972). 'On the Concept of Health Capital and the Demand for
Health.' Journal of Political Economy, 80 (2): 223–255. JSTOR.
Haricharan, K. R., B.M. Shrinivasa, & V. Kumari (2013). 'Clinical and bacteriological
study of acute diarrhoea in children.' Journal of Evolution of Medical and
Dental Sciences, 2(23): 4229-4238.
Hosmer, D. W., & S. Lemeshow (2004). Applied Logistic Regression. New Jersey,
US: John Wiley & Sons.
ICF. (2013). Demographic and Health Surveys Standard Recode Manual for
DHS6. The Demographic and Health Surveys Program. Rockville, Maryland,
US: ICF.
ICF. (2018). Demographic and Health Surveys Standard Recode Manual for
DHS7. The Demographic and Health Surveys Program. Rockville, Maryland,
US: ICF.
Islamiati, W. (2022). Sejarah 15 Penyakit yang Pernah Picu KLB di Indonesia.
Available at: https://lifestyle.bisnis.com/read/20221130/106/1603940/
sejarah-15-penyakit-yang-pernah-picu-klb-di-indonesia
Jalan, J., & M. Ravallion (2003). 'Does piped water reduce diarrhea for children
in rural India?' Journal of Econometrics, 112 (1): 153–173. Available at:
https://doi.org/10.1016/S0304-4076(02)00158-6
Kementerian Kesehatan. (2019). Laporan nasional Riskesdas 2018. Lembaga
Penerbit Balitbangkes, Kementerian Kesehatan, Republik Indonesia, Badan
Penelitian dan Pengembangan Kesehatan.
Kementerian Pekerjaan Umum dan Perumahan Rakyat. (2018). Petunjuk
Teknis Perencanaan Kegiatan Tingkat Masyarakat. Available at: https://
pamsimas.pu.go.id/konten/pustaka/pedum-juknis/juknis-perencanaan-
tingkat-masyarakat-program-pamsimas-2018-edit.pdf
Khan, K. M., R. Chakraborty, S. Brown, R. Sultana, A. Colon, D. Toor, P. Upreti, &
B. Sen (2021). 'Association between Handwashing Behavior and Infectious
Diseases among Low-Income Community Children in Urban New Delhi,
India: A Cross-Sectional Study.' International Journal of Environmental
Research and Public Health, 18(23). Available at: https://doi.org/10.3390/
ijerph182312535
Komarulzaman, A., Smits, J., & de Jong, E. (2017). 'Clean water, sanitation, and
diarrhoea in Indonesia: Effects of household and community factors.'
Global Public Health, 12(9): 1141–1155. Available at: https://doi.org/10.
1080/17441692.2015.1127985
Kwami, C. S., Godfrey, S., Gavilan, H., Lakhanpaul, M., & Parikh, P. (2019).
Water, Sanitation, and Hygiene: Linkages with Stunting in Rural Ethiopia.
International Journal of Environmental Research and Public Health, 16(20),
3793.
Mebrahtom, S., A. Worku, & D.J. Gage (2022). The risk of water, sanitation
and hygiene on diarrhea-related infant mortality in eastern Ethiopia: a
population-based nested case-control. BMC Public Health, 343(2022).
Available at: https://bmcpublichealth.biomedcentral.com/articles/10.
1186/s12889-022-12735-7
Müller, O., & Krawinkel, M. (2005). Malnutrition and Health in Developing
Countries. Canadian Medical Association Journal, 173(3): 279-286.
Available at: https://doi.org/https://doi.org/10.1503/cmaj.050342
Lenters, L. M., J.K. Das, & Z.A. Bhutta (2013). Systematic review of strategies
to increase use of oral rehydration solution at the household level. BMC
Public Health, 13(3): S28. Available at: https://doi.org/10.1186/1471-
2458-13-S3-S28
Naugle, D. A., & C.R. Hornik (2014). 'Systematic Review of the Effectiveness of
Mass Media Interventions for Child Survival in Low- and Middle-Income
Countries.' Journal of Health Communication, 19(sup1): 190–215. Available
at: https://doi.org/10.1080/10810730.2014.918217
Noguchi, Y., D.Nonaka, S. Kounnavong, & J. Kobayashi (2021). 'Effects of Hand-
Washing Facilities with Water and Soap on Diarrhea Incidence among
Children under Five Years in Lao People’s Democratic Republic: A Cross-
Sectional Study.' International Journal of Environmental Research and Public
Health, 18(2): 687. Available at: https://doi.org/10.3390/ijerph18020687
in Indonesia
Omona, S., G.M. Malinga, R. Opoke, G. Openy, & R. Opiro (2020). 'Prevalence of
diarrhoea and associated risk factors among children under five years old
in Pader District, northern Uganda.' BMC Infectious Diseases, 20(1): 37.
Available at: https://doi.org/10.1186/s12879-020-4770-0
Patunru, A. A. (2015). Access to Safe Drinking Water and Sanitation in Indonesia.
Asia & the Pacific Policy Studies, 2(2): 234–244. Available at: https://doi.
org/10.1002/app5.81
Piper, J. D., J. Chandna, E. Allen, K. Linkman, O. Cumming, A,J.Prendergast, &
M.J.Gladstone (2017). 'Water, sanitation, and hygiene (WASH) interventions:
Effects on child development in low‐ and middle‐income countries.' The
Cochrane Database of Systematic Reviews, (3), CD012613. Available at:
https://doi.org/10.1002/14651858.CD012613
Rao, K. V., Mishra, V. K., & Retherford, R. D. (1998). 'Mass media can help improve
treatment of childhood diarrhoea.' National Family Health Survey Bulletin,
11: 1–4.
Sahiledengle, B., A. Kumie, D. Atlaw, Y. Te kalegn, D. Woldeyohannes, D. Zenbaba,
& T. Awoke (2021). Factors affecting under-five diarrhea in households
with dirt floors: A multilevel mixed-effects analysis from a national survey
of Ethiopia [Preprint]. In Review. Available at: https://doi.org/10.21203/
rs.3.rs-646901/v1
Seksaria, S. A., & M.K. Sheth (2014). 'Maternal Knowledge and Practices
Towards Sanitation and Their Relationships with Occurrence of Diarrhoea
in Children.' International Journal of Public Health Science (IJPHS), 3(3),
206-212.
Shaheed, A., J. Orgill, M.A. Montgomery, M.A. Jeuland, & J. Brown (2014). 'Why
“improved” water sources are not always safe.' Bulletin of the World Health
Organization, 92(4): 283–289. Available at: https://doi.org/10.2471/
BLT.13.119594
Shine, S., Muhamud, S., Adanew, A. Demelash, & M. Abate (2020). 'Prevalence
and associated factors of diarrhea among under-five children in Debre
Berhan town, Ethiopia 2018: A cross sectional study.' BMC Infectious
Diseases, 20(1): 174–174. PubMed. Available at: https://doi.org/10.1186/
s12879-020-4905-3
Sodha, S. V., M. Menon, K. Trivedi, A. Ati, M.E. Figueroa, R. Ainslie, K. Wannemuehler,
& R. Quick (2011). 'Microbiologic effectiveness of boiling and safe water
storage in South Sulawesi, Indonesia.' Journal of Water and Health, 9(3):
577–585. Available at: https://doi.org/10.2166/wh.2011.255
Todaro, M. P., & S. Smith (2012). Economic development (11th ed). Addison-
Wesley.
Tetteh, J., Adomako, I. U., E.,A., E., Y., H. Quansah, A.O. , Yawson, Essuman, A.,
& Yawson, A., E. (2022). Hygienic disposal of stools and risk of diarrheal
episodes among children aged under two years: Evidence from the
Ghana Demographic Health Survey, 2003–2014. PLoS ONE, 17(4),
e0266681. Available at: https://doi.org/https://doi.org/10.1371/journal.
pone.0266681
United Nations Children’s Fund (UNICEF), & World Health Organization (WHO).
(2019). Progress on household drinking water, sanitation, and hygiene 2000-
2017. Special focus on inequalities.
Vollaard, A. M., S. Ali, H.A.G.H van Asten, S. Widjaja, L.G. Visser, C. Surjadi, & J.T.
van Dissel (2004). 'Risk Factors for Typhoid and Paratyphoid Fever in
Jakarta, Indonesia.' JAMA, 291(21): 2607–2615. Available at: https://doi.
org/10.1001/jama.291.21.2607
Wahed, W., Y.,A., K. S. Hassan, & Eldessouki, R. (2017). 'Malnutrition and Its
Associated Factors among Rural School Children in Fayoum Governorate,
Egypt.' Journal of Environmental and Public Health, 2017. https://doi.org/
https://doi.org/10.1155/2017/4783791
Walker, C. L. F., Rudan, I., Liu, L., Nair, H., Theodoratou, E., Bhutta, Z. A., O’Brien,
K. L., Campbell, H., & Black, R. E. (2013). Global burden of childhood
pneumonia and diarrhoea. The Lancet, 381(9875), 1405–1416. https://
doi.org/10.1016/S0140-6736(13)60222-6
Wolf, J., Hunter, P. R., Freeman, M. C., Cumming, O., Clasen, T., Bartram, J., Higgins,
J. P. T., Johnston, R., Medlicott, K., Boisson, S., & Prüss-Ustün, A. (2018).
Impact of drinking water, sanitation, and handwashing with soap on
childhood diarrhoeal disease: Updated meta-analysis and meta-regression.
Tropical Medicine & International Health, 23(5), 508–525. https://doi.
org/10.1111/tmi.13051
in Indonesia
Wooldridge, J. M. (2013). Introductory econometrics: A modern approach (5th
ed). South-Western, Cengage Learning.
World Health Organization. (2017). Diarrhoeal disease. https://www.who.int/
news-room/fact-sheets/detail/diarrhoeal-disease
World Health Organization & United Nations Children’s Fund (UNICEF). (2013).
Progress on sanitation and drinking water—2013 update. World Health
Organization. https://apps.who.int/iris/handle/10665/81245
APPENDICES
This section presents some descriptions of variables that are used in this study.
Appendix 1. Indonesia Demographic Health Survey (IDHS) 6 – 2012
No Variables Questionnaires Answers Recode
1 Diarrhoea Had diarrhoea recently yes, last two weeks Yes
(Yes = 1) no No
don't know .
2 Drinking Source of drinking water piped into dwelling
Water Piped water
piped to yard/plot
Classifications
public tap/standpipe
protected well in
dwelling
Protected dug well
protected well in yard/
or springs
plot
protected public well
rainwater
bottled water Bottled/refilled
refill water water
open well in dwelling
open well in yard/plot
Unprotected dug
open public well
well or springs
spring
tanker truck
river/stream
pond/lake
Surface water
dam
other
3 Sanitation type of toilet facility private - with septic Flush toilet with
Classifications tank septic tank
private - without septic Flush toilet with no
tank septic tank
shared/public Flush toilet which is
shared/public
pit latrine Pit latrine
yard/bush/forest
river/stream/creek No facility
other
4 Handwashing place where household observed Facility on premises
Facility members wash their hands with water and
Classifications presence of water at water is available soap
handwashing place
items present: soap or yes
detergent
place where household observed
members wash their hands
presence of water at water not available
handwashing place
detergent
presence of water at water is available
handwashing place
items present: soap or no
detergent
presence of water at water not available
handwashing place
detergent
place where household not observed: not in
members wash their hands dwelling
place where household not observed: no
Not available
members wash their hands permission to see
place where household not observed: other
members wash their hands reason
in Indonesia
Appendix 2. Indonesia Demographic Health Survey (IDHS) 7 – 2017
1 Diarrhoea Had diarrhoea recently yes, last two weeks Yes
(Yes = 1) no No
don't know .
2 Drinking Source of drinking water piped into dwelling
Water piped to yard/plot Piped water
Classifications
piped to neighbour
protected well in
dwelling
tube well or borehole Protected dug well
protected well or springs
protected spring
rainwater
bottled water Bottled/refilled
refill water water
unprotected well
unprotected spring Unprotected dug
tanker truck well or springs
cart with small tank
river/dam/lake/
ponds/stream/canal/
irrigation Surface water
other
3 Sanitation type of toilet facility flush to septic tank Flush toilet with
Classifications septic tank
flush with no septic Flush toilet with no
tank septic tank
flush toilet : shared / Flush toilet which is
public shared/public
ventilated improved pit Pit latrine
(vip) latrine
no facility/bush/field/
river/ beach/pond No facility
other
4 Handwashing place where household observed, fixed place/ Facility on premises
Facility members wash their hands observed, mobile place with water and
Classifications presence of water at hand water is available soap
washing place
detergent
place where household observed, fixed place/ Facility on premises
members wash their hands observed, mobile place without water/soap
presence of water at hand water not available
washing place
detergent
place where household observed, fixed place/
members wash their hands observed, mobile place
presence of water at hand water is available
washing place
detergent
place where household observed, fixed place/
members wash their hands observed, mobile place
presence of water at hand water not available
washing place
detergent
place where household not observed: not in
members wash their hands dwelling
place where household not observed: no
Not available
members wash their hands permission to see
place where household not observed: other
members wash their hands reason
in Indonesia
Appendix 3. Island Groups Disaggregation
Appendix 4. Diarrhoea Prevalence among U5, by Province
Appendix 5. Coefficient Plot, Full Sample
Appendix 6. Coefficient Plot, by Island Groups
in Indonesia
CHAPTER 14
Credit Channelling to MSMEs in

Indonesia: The Role of Regional
Digital Competitiveness in Pre and
During the COVID-19 Pandemic
Militcyano Samuel Sapulette, Nury Effendi, Vera Intanie Dewi,
Fitri Hastuti, Eva Ervani, and Anhar Fauzan Priyono
BACKGROUND AND OBJECTIVES

The COVID-19 Pandemic has brought new challenges and opportunities to
the economy. One of which is the acceleration of economic digitalization. To
businesses who are already “tech-savvy,” this may come as a great opportunity.
However, the opposite may be true, especially for the micro, small, and medium
enterprises that lack the resources to quickly transform their businesses to fit
the needs of the digital era (Klein and Todesco, 2021).
MSMEs contribute a significant share of Indonesia’s GDP and employment
(Martdianty, Coetzer, and Susomrith, 2020; Sanny, Angelina, and Christian,
2021). Thus, the failure to enable the digitalization of MSMEs post-COVID-19
may thwart Indonesia’s economic recovery efforts. Additionally, MSMEs can also
be seen as seeds for future economic growth. Indonesia’s notorious regional
inequality problem can be addressed by encouraging MSMEs growth in regions
currently “left behind” (Nanziri and Wamalwa, 2021; Lu, Wu, and Liu, 2020).
325
As proven in past empirical studies, affordable financing is one of the
prerequisites for MSMEs’ growth (Lu, Wu, and Liu, 2020; Lu et al., 2022; Rao et
al., 2021). The Pandemic and the acceleration of the digital transformation that
comes with it, on the other hand, may have caused unequal credit channelling
to businesses operating in an environment with a low level of digitalization or
businesses that are not digitalized yet. Thus, further hampering recovery efforts
from the Pandemic and widening the inequality gap (Chan et al., 2019; Jiang,
2020). To date, there has not been a study that explicitly investigates the role of
the digital environment on MSME credit channelling in Indonesia.
Since the growth of MSMEs is crucial for the Indonesian economy,
especially in its recovery from the devastating COVID-19 Pandemic, immense
attention must be given to ensure that MSMEs have sufficient environmental
enabling factors to grow, one of which is the financings they receive. In the wake
of recent attention on the digital divide post Pandemic, not much is known
about how this aspect affects loan channelling to MSMEs, particularly in the
regional context of Indonesia, which is characterised by socio-economic and
technological inequality.
Against these backdrops, this study aims to analyse credit channelling to
MSMEs before and during the Pandemic, emphasizing the effect of the level of
digital technology competitiveness in the regions where the MSMEs operate.
We hypothesize that MSMEs operating in regions with higher digital technology
competitiveness have better access to credit/loans based on credit rationing
theories. Furthermore, this relationship gets even stronger during the Pandemic
period as digital technology becomes more critical.
To test these hypotheses, we rely on province-level monthly MSMEs loan
channelling data from 2019 (pre-Pandemic) to December 2021. We mainly use
the one-step system GMM to estimate our models. To ensure the robustness
of our findings, we use multiple model specifications and estimation methods.
Our findings support the hypothesis that the digital environment (proxied by
digital competitiveness index rank-DCI rank) positively and significantly affects
loan channelling to MSMEs and that the effect is stronger after the start of the
COVID-19 Pandemic.
LOAN CHANNELING TO MSMES AND DIGITAL GAP:
INDONESIAN REGIONAL PERSPECTIVE
Regional inequality in Indonesia is not a novel issue. For a long time, regions
outside of Java, especially in East Indonesia, are known to be significantly less
developed. This is true in terms of (Information and Communication Technology)
ICT infrastructures and financing, the focus of this study. Untari et al. (2019)
studied the impact of ICT infrastructure on economic growth and inequality in
Indonesia. Their study shows that ICT infrastructures impact both economic
growth and inequality; as the ICT infrastructure becomes more developed,
economic growth is likely to increase and inequality is likely to decrease. Similar
evidence is also put forward in a study by Sujarwoto & Tampubolon (2016).
They show how regional inequality in Indonesia is highly linked to the “internet
divide.” In this section, we look at the evolution of the digital environment and
MSMEs financing from before the Pandemic (2019) to the early, towards-post
Pandemic recovery period (2021).
The East Ventures-Digital Competitiveness Index (DCI) is an index that
shows the digital competitiveness of a region. It consists of three sub-indexes that
describe the state of digital development of a province: Input (human resource,
ICT usage, ICT expenditure); output (economy, partnership and productivity,
and manpower); and, support (infrastructure, finance, and regulation and
local government capacity). The sub-indexes forming the DCI framework are
calculated based on interrelated pillars, which are calculated from indicators
of which data are collected from various sources (Table 6). The DCI calculation
was applied to 34 provinces annually since 2019 (East Ventures, 2022).
The calculation of DCI follows a nested approach. The score of each indicator
are compiled using a simple average into the pillar score. The pillar scores are
also compiled using a simple average to form the sub-index score. The DCI is
then computed using the weighted average of the sub-indexes with the input
sub-index weighting 40%, output weighting 40%, and support weighting 20%.
Since the value of each of the 49 indicators is measured in different units,
the (raw) score of each indicator xi is normalised into a scale of 0 (worst) to
100 (best) as shown in Equation 1. This scoring shows the relative comparison
among provinces. We use the (ascending) rank of DCI for our estimation.
𝑥𝑥𝑖𝑖 − 𝑥𝑥𝑚𝑚𝑚𝑚𝑚𝑚
𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆 𝑜𝑜𝑜𝑜 𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝑖𝑖 = 100 (1)
𝑥𝑥𝑚𝑚𝑚𝑚𝑚𝑚 − 𝑥𝑥𝑚𝑚𝑚𝑚𝑚𝑚
Credit Channelling to MSMEs in Indonesia: The Role of Regional Digital Competitiveness 327
in Pre and During the COVID-19 Pandemic
Figure 1. Average Digital Competitiveness Index Ranking 2019–2021
Source: Compiled from East Ventures (2020, 2021, 2022)

Note: This figure presents the average rank (ascending) of the East Venture
Digital Competitive Index (DCI) for provinces in each of the six regions in
Indonesia. The rank is computed for each year from 2019 to 2921.
As seen in Figure 1, there is a clear divide among the six main regions of
Indonesia. The Java region ranks significantly better than the rest (the ranking
follows an ascending order). This is in stark contrast to the East Indonesia
region. The other four regions tend to cluster together at the centre.
In regard to the Pandemic which started in 2020, these four regions
(Sumatera, Kalimantan, Sulawesi, and Bali & Nusa Tenggara) experienced a
relative increase, while East Indonesia and Java experienced a decrease in their
rankings (East Indonesia had a larger decrease), indicating the inequality of
the digital environment in Indonesia, especially in the post-Pandemic period
for provinces in East Indonesia. Figure 2 shows the average per capita MSME
loan (million IDR per thousand people measured in constant 2018 price) for
provinces in each of the six regions in Indonesia. The black, solid, vertical line
signifies the start of the COVID-19 Pandemic in Indonesia (March 2020).
Figure 2. Per Capita MSME Loan in 2019-2021
Source: Bank Indonesia

A similar pattern to the one displayed in Figure 1 can also be seen in Figure
2, which shows per capita MSME loans. Again, there is a stark contrast between
the Java region and regions outside of Java, especially in East Indonesia. The
Figure also shows that the per capita MSME loan channelling has returned
(surpassed in some regions) to its pre-Pandemic level.
The credit rationing theories, which were formally compiled and defined
by Jaffee & Stiglitz (1990), explain how loans are distributed in the case of
excess demand for credit/loans. Thus, in this chapter, loans received by MSMEs
are always assumed to be below their desired level from the borrowers’
perspectives. We believe this assumption holds as it has been proven in previous
studies that MSMEs in Indonesia tend to receive loans that are lower than what
they apply for (International Labour Organization, 2019). There are four types
of credit rationing according to Jaffee & Stiglitz (1990):
1. Interest rate (or price) rationing--At a specific lending rate, a borrower
could get a loan that is smaller than they expected. The borrower must pay
a higher rate to get a bigger loan.
2. Divergent views rationing—Some individuals cannot borrow money at the
interest rate they consider reasonable based on what they perceive to be
their probability of default.
3. Redlining—A lender will decline to extend credit to a borrower based on
the risk level if they cannot make their needed return at any interest rate.
4. Pure credit rationing—Some individuals may get loans while others who
appear to be the same and would be eager to borrow money under the
same conditions do not.
This study focuses on the third type of credit rationing (redlining). This
is because we assume that banks perceived the level of digital technology
proliferation surrounding MSMEs as a significant factor in the banks’ decisions
on whether to extend loans to MSMEs (Lu et al., 2022). Loans channelled to
MSMEs and their digital technology adoption are the focus of this study. We test
whether the digital environment in a given province is a significant factor that
can affect credit received by MSMEs in that province. The hypothesis implies
that banks perceive provinces with a better digital environment as ones that
can give relatively higher returns.
Furthermore, regarding the Pandemic, our second hypothesis implies that
banks perceive the digital environment as being a more important factor after
the start of the Pandemic. Thus, banks will tend to channel loans to provinces
with a better digital environment (ones that give higher returns).
EMPIRICAL STRATEGY
We select the variable to be included in our model based on a previous similar
study of MSME loan channelling in Indonesia by Trinugroho et al. (2015). The
variables are mainly macroeconomic variables, adjusted to per capita and
transformed to growth. A full description of the variables included in our model
can be seen in Table 1.
The variables that are transformed into growth are calculated based on
their constant, 2018-price values. The population data used to construct the
per capita variables are based on BPS’s mid-year population projection data,
interpolated to monthly (or quarterly for GDP and its component) based
on linear trends. Variables that are originally not in monthly frequency are
presented in a constant value for each month during the quarter/semester/
year (based on the original data frequency). The GDP and its component are
originally available quarterly, unemployment half-yearly, HDI and DCI annually.
Table 1 contains the description of each variable used in this study along with
their sources.
Table 1. Variable Description and Data Source
Variable Description Source
gr_msme_loan_percap MTM growth of per capita MSME loan Bank Indonesia
(outstanding).
gr_gdp_percap QTQ growth of per capita GDP. BPS
gr_gfcf_percap QTQ growth of per capita gross fixed capital BPS
formation.
gr_gv_cons_percap QTQ growth of per capita government spending. BPS
gr_cpi MTM inflation. BPS
unempr Unemployment rate. BPS
hdi Human Development Index. BPS
dci_rank The rank of DCI. East Ventures
Note: Except for the dci_rank variable, all data were sourced using the CEIC Indonesia
Premium Database platform (available at https://insights.ceicdata.com/node/INDONESIA).
QTQ=Quarter-to-Quarter; MTM=Month-to-Month.
The inclusion of the unemployment rate and HDI variables is key here.
These two variables represent the level of development in each province. Thus,
we can control for other developmental factors that may cause different levels of
loan channelling other than the level of digital competitiveness in that province.
The data span from January 2019 to December 2021 for the 34 Indonesian
provinces. Table 2 presents the statistics summary of variables used for the
estimations; and they are presented for each year from 2019-2021 since the
estimation is done for each year.
332
2021 2020 2019
Variable Std. Std. Std. Observations
Mean Min Max Mean Min Max Mean Min Max
Dev. Dev. Dev.
gr_msme_loan_ overall
percap
between
0.0090 0.0375 -0.2534 0.3297 -0.0023 0.0371 -0.3289 0.5971 0.0032 0.0213 -0.1397 0.1214 N= 408
within
0.0070 -0.0063 0.0257 0.0061 -0.0229 0.0076 0.0046 -0.0095 0.0123 n= 34
gr_gdp_percap overall
0.0369 -0.2553 0.3279 0.0366 -0.3388 0.5872 0.0208 -0.1270 0.1188 T= 12
between
0.0215 0.0418 -0.0855 0.2004 -0.0096 0.0534 -0.1451 0.1310 0.0086 0.0362 -0.1364 0.1079 N= 408
within
0.0230 -0.0135 0.1052 0.0157 -0.0506 0.0171 0.0062 -0.0076 0.0224 n= 34
gr_gfcf_percap overall
0.0351 -0.0864 0.1500 0.0511 -0.1331 0.1145 0.0357 -0.1420 0.1218 T= 12
between
0.0346 0.1599 -0.5222 1.1191 -0.0160 0.1129 -0.5027 0.5321 0.0290 0.1993 -0.6866 1.9003 N= 408
within
0.0825 -0.0264 0.4505 0.0203 -0.0709 0.0222 0.0858 -0.0146 0.5029 n= 34
gr_gv_cons_percap overall
0.1376 -0.9380 0.7033 0.1111 -0.4565 0.5783 0.1804 -1.1606 1.4263 T= 12
between
0.0833 0.3981 -0.7088 1.1407 0.0455 0.3281 -0.6661 1.0166 0.0819 0.3717 -0.6720 1.2574 N= 408
within
0.0614 0.0035 0.2419 0.0573 -0.0588 0.1884 0.0624 -0.0186 0.2239 n= 34
gr_cpi overall
0.3935 -0.8674 1.0266 0.3232 -0.8091 0.9214 0.3665 -0.8140 1.1232 T= 12
between
0.0019 0.0039 -0.0090 0.0155 0.0014 0.0037 -0.0125 0.0126 0.0018 0.0047 -0.0181 0.0257 N= 408
within
0.0006 0.0010 0.0036 0.0006 0.0001 0.0031 0.0006 -0.0005 0.0028 n= 34
unempr overall
0.0039 -0.0097 0.0159 0.0036 -0.0127 0.0123 0.0046 -0.0182 0.0248 T= 12
between
0.0550 0.0175 0.0301 0.1012 0.0528 0.0192 0.0121 0.1095 0.0461 0.0154 0.0119 0.0811 N= 408
within
0.0176 0.0320 0.1001 0.0165 0.0297 0.0932 0.0153 0.0136 0.0786 n= 34
hdi overall
0.0018 0.0502 0.0598 0.0101 0.0227 0.0828 0.0032 0.0374 0.0547 T= 12
between
71.3612 3.8839 60.6200 81.1100 71.0809 3.8488 60.4400 80.7700 71.0403 3.8600 60.8400 80.7600 N= 408
within
3.9374 60.6200 81.1100 3.9019 60.4400 80.7700 3.9132 60.8400 80.7600 n= 34

dci_rank overall
0.0000 71.3612 71.3612 0.0000 71.0809 71.0809 0.0000 71.0403 71.0403 T= 12
between
17.5000 9.8183 1.0000 34.0000 17.5000 9.8205 1.0000 34.0000 17.5000 9.8228 1.0000 34.0000 N= 408
within
9.9537 1.0000 34.0000 9.9560 1.0000 34.0000 9.9582 1.0000 34.0000 n= 34
0.0000 17.5000 17.5000 0.0000 17.5000 17.5000 0.0000 17.5000 17.5000 T= 12
We estimate our models (Equation 2) using monthly data for each year.
The estimation for the whole period (2019-2021) is not conducted owing to
our key variable DCI only showing relative levels among provinces for each year.
gr_msme_loan_percapi,t = α0 + β1 gr_msme_loan_percapi,t–1
+ β2dci_ranki + δControli,t + vi,t ...(2)
The variables from Table 1 are arranged as shown in Equation 2 where
the subscripts i and t indicate the provinces and months, respectively. The
dependent variable gr_msme_loan_percap is the month-to-month growth of per
capita MSME loan. One-period lag of this dependent variable is also one of the
independent variables. dci_rank is the key independent variable, which shows a
province’s rank (ascending: bigger is better) compared to other provinces—i.e.,
the higher the province’s rank is, the more competitive its state of digitalisation
will be. Note that dci_rank is originally available in an annual frequency. Hence,
for our estimation, the value of dci_rank is constant for the 12 month-period of
estimation for each year from 2019 to 2021 (note the absence of subscript t to
this variable in Equation 2).
Control in Equation 2 represents a set of key macroeconomic variables: GDP
(gr_gdp_percap); gross fixed capital formation (gr_gfcf_percap); government
spending (gr_gv_cons_percap)—in which these first three are adjusted to
per capita and transformed to growth; inflation (gr_cpi), unemployment
rate (unempr); and, human development index (hdi). In some estimations of
Equation 2, several variables in Control are omitted to shows the robustness of
the estimation results.
We use the system GMM estimator as our main estimator citing the
dynamic model specification, i that is larger than t (34 and 12 for each year),
the persistence trend of our dependent variable, and the possible endogeneity
of loan and macroeconomic variable as shown in previous literature (Sapulette,
Effendi, and Santoso, 2021; Sapulette and Dyana, 2020; Tunay and Kurt, 2015;
Ullah, Akhtar, and Zaefarian, 2018). The model is estimated each year for 2019,
2020, and 2021 so that we can see the changes in the value and significance
of β_2 (coefficient of the DCI rank variable). This will answer our research
objectives.
For our one-step system GMM estimations, we checked for the number of
instruments which must not exceed our N (34). We also performed the Hansen
test of overidentifying restrictions to confirm the validity of our instruments;
and the Arellano-Bond test to check the presence of second-order serial
correlation (Roodman, 2009b, 2009a).
To provide robust results, we also estimate the model using pooled OLS
and the random-effect model following the empirical strategy used by Sapulette
et al. (2021). A fixed-effect model cannot be used since dci_rank has an annual
frequency (constant for all months in each year estimation). Furthermore, we
also omit either gr_gdp_percap, gr_gfcf_percap, or gr_gv_cons_percap in several
alternative estimated model specifications to show how the result changes
owing to the exclusion of certain macroeconomic variables (these variables are
chosen since GFCF and government spending both compose GDP, thus tend to
be multicollinear).
ESTIMATION RESULT AND ANALYSIS

Results of Hypothesis Testing
Our system GMM estimation shows support for our hypotheses: Digital
competitiveness became more significant in determining the amount of loans
MSMEs receive after the start of the COVID-19 Pandemic. Furthermore, the
result is also robust to changes in model specification and estimation method.
The estimation results are shown in Tables 3 to 5. Table 3 presents the system
GMM estimation results with the main models being in columns (1) to (3), and
the rest are results with different model specifications. Similar results are true
for OLS estimation in Table 4 and REM estimation in Table 5.
Table 3. System GMM Estimation Results
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12)
2021 2020 2019 2021 2020 2019 2021 2020 2019 2021 2020 2019
L.gr_msme_loan_
percap
-0.2943*** -0.3302*** -0.2810*** -0.2916*** -0.3262*** -0.2776*** -0.2925*** -0.3230*** -0.2754*** -0.2912*** -0.3249*** -0.2853***
gr_gdp_percap
-0.0468 -0.0314 -0.0892 -0.0457 -0.0404 -0.0875 -0.0445 -0.0431 -0.0879 -0.0446 -0.0324 -0.0895
-0.0598 0.1237*** 0.0075 -0.0373 0.1746*** 0.0299
gr_gfcf_percap
-0.0396 -0.0315 -0.0328 -0.0341 -0.0359 -0.0341
0.0232** 0.0210* 0.0045** 0.0206** 0.0526** 0.0063
gr_gv_cons_percap
-0.0096 -0.0108 -0.0023 -0.0092 -0.0227 -0.0041
0.0056 0.0190*** 0.0149*** 0.0044 0.0286*** 0.0154***
gr_cpi
-0.0045 -0.0045 -0.0041 -0.0043 -0.0054 -0.0042
-1.4024*** -1.7981** -1.2866*** -1.4392*** -1.9904** -1.3059*** -1.4004*** -1.9789** -1.2948*** -1.4121*** -1.8503** -1.3041***
unempr
-0.3098 -0.8445 -0.2366 -0.3047 -0.8936 -0.2570 -0.3117 -0.9465 -0.2493 -0.3155 -0.8431 -0.2383
-0.1781*** -0.1574** -0.0125 -0.1704** -0.1201* 0.0363 -0.1644** -0.0660 0.0341 -0.1748*** -0.1247* -0.0097
hdi
-0.0619 -0.0663 -0.0505 -0.0663 -0.0648 -0.0538 -0.0641 -0.0684 -0.0537 -0.0644 -0.0735 -0.0499
-0.0017*** 0.0006 0.0000 -0.0018*** 0.0007 0.0000 -0.0016*** 0.0010 0.0000 -0.0017*** 0.0008 0.0000
dci_rank
-0.0004 -0.0006 -0.0003 -0.0004 -0.0006 -0.0003 -0.0004 -0.0007 -0.0003 -0.0004 -0.0007 -0.0003
0.0004*** 0.0000 -0.0001 0.0004* 0.0000 -0.0001 0.0004*** -0.0001 -0.0002 0.0004** 0.0000 -0.0001
constant
-0.0002 -0.0002 -0.0001 -0.0002 -0.0002 -0.0001 -0.0002 -0.0003 -0.0001 -0.0002 -0.0003 -0.0001
0.1127*** -0.0302 0.0051 0.1136*** -0.0346 0.0033 0.1034*** -0.0592 0.0005 0.1109*** -0.0453 0.0069
month_dummy
-0.0268 -0.0389 -0.0234 -0.0271 -0.0415 -0.0235 -0.0236 -0.0460 -0.0242 -0.0238 -0.0466 -0.0236
yes yes yes yes yes yes yes yes yes yes yes yes
Obs. 408 408 408 408 408 408 408 408 408 408 408 408
Groups 34 34 34 34 34 34 34 34 34 34 34 34
Instruments 34 31 25 32 29 23 32 29 23 32 29 23
AB AR (2) in FD 0.357 0.094 0.256 0.601 0.241 0.53 0.495 0.315 0.561 0.45 0.094 0.244

Credit Channelling to MSMEs in Indonesia: The Role of Regional Digital Competitiveness
Hansen Test 0.732 0.33 0.424 0.696 0.0435 0.12 0.727 0.0501 0.116 0.719 0.299 0.41
* p<0.1 ** p<0.05 *** p<0.01
335
Note: This table presents the estimation result using the one-step system GMM estimation method. Different model specification/estimation
period is presented in each separate column from column one to 12.
336
Table 4. OLS Estimation Results
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12)
2021 2020 2019 2021 2020 2019 2021 2020 2019 2021 2020 2019
L.gr_msme_loan_
percap
-0.3642* -0.3663 -0.2811*** -0.3563* -0.3708 -0.2815*** -0.3651* -0.3669 -0.2821*** -0.3576* -0.3603 -0.2814***
gr_gdp_percap
-0.2098 -0.3724 -0.0902 -0.2101 -0.3791 -0.0896 -0.2129 -0.3771 -0.0899 -0.2109 -0.3683 -0.0902
-0.0397 0.0853** -0.0249 -0.0092 0.0923** -0.0241
gr_gfcf_percap
-0.0507 -0.0368 -0.0277 -0.0433 -0.0412 -0.0277
0.0265 0.0081 0.0018 0.0242 0.0197 0.0017
gr_gv_cons_percap
-0.0214 -0.0091 -0.0064 -0.0203 -0.0142 -0.0064
0.0059 0.0126 0.0017 0.0049 0.0136 0.0016
gr_cpi
-0.0052 -0.0217 -0.0034 -0.0048 -0.0219 -0.0035
-1.3246*** -1.8044*** -1.2178*** -1.3645*** -1.8480*** -1.2210*** -1.3293*** -1.8263*** -1.2105*** -1.3447*** -1.8015*** -1.2171***
unempr
-0.3298 -0.5967 -0.3095 -0.3200 -0.6439 -0.3087 -0.3274 -0.6436 -0.3090 -0.3217 -0.6012 -0.3092
-0.1867* -0.1709* -0.0867 -0.1738* -0.1669 -0.0844 -0.1749* -0.1533 -0.0840 -0.1818* -0.1610 -0.0847
hdi
-0.1013 -0.1005 -0.0641 -0.0989 -0.1062 -0.0650 -0.0978 -0.1075 -0.0651 -0.0999 -0.1007 -0.0641
-0.0017*** 0.0009 0.0000 -0.0017*** 0.0009 0.0000 -0.0016*** 0.0011 0.0000 -0.0017*** 0.0011 0.0000
dci_rank
-0.0004 -0.0010 -0.0003 -0.0004 -0.0011 -0.0003 -0.0005 -0.0012 -0.0003 -0.0004 -0.0011 -0.0003
0.0004* 0.0000 -0.0001 0.0004* 0.0000 -0.0001 0.0004* -0.0001 -0.0001 0.0004* -0.0001 -0.0001
constant
-0.0002 -0.0005 -0.0001 -0.0002 -0.0006 -0.0001 -0.0002 -0.0006 -0.0001 -0.0002 -0.0006 -0.0001
0.1094*** -0.0566 -0.0132 0.1102*** -0.0670 -0.0137 0.1026*** -0.0825 -0.0139 0.1113*** -0.0759 -0.0127
month_dummy
-0.0298 -0.0593 -0.0211 -0.0283 -0.0747 -0.0211 -0.0306 -0.0793 -0.0209 -0.0288 -0.0653 -0.0210
Obs. 408 408 408 408 408 408 408 408 408 408 408 408

Adj. R-Sq. 0.446 0.229 0.373 0.438 0.228 0.376 0.447 0.223 0.375 0.439 0.225 0.375
* p<0.1 ** p<0.05 *** p<0.01

Note: This table presents the estimation result using pooled OLS estimation method. Different model specification/estimation period is
presented in each separate column from column one to 12.
Table 5. REM Estimation Results
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12)
2021 2020 2019 2021 2020 2019 2021 2020 2019 2021 2020 2019
L.gr_msme_loan_
percap
-0.3642*** -0.3663*** -0.2811*** -0.3563*** -0.3708*** -0.2815*** -0.3651*** -0.3669*** -0.2821*** -0.3576*** -0.3603*** -0.2814***
gr_gdp_percap
(0.0683) (0.0455) (0.0850) (0.0697) (0.0500) (0.0846) (0.0687) (0.0527) (0.0842) (0.0701) (0.0498) (0.0846)
-0.0397 0.0853*** -0.0249 -0.0092 0.0923*** -0.0241
gr_gfcf_percap
(0.0412) (0.0286) (0.0358) (0.0336) (0.0298) (0.0360)
0.0265*** 0.0081 0.0018 0.0242*** 0.0197 0.0017
gr_gv_cons_percap
(0.0097) (0.0099) (0.0016) (0.0088) (0.0153) (0.0017)
0.0059 0.0126* 0.0017 0.0049 0.0136* 0.0016
gr_cpi
(0.0048) (0.0068) (0.0047) (0.0046) (0.0069) (0.0048)
-1.3246*** -1.8044* -1.2178*** -1.3645*** -1.8480* -1.2210*** -1.3293*** -1.8263* -1.2105*** -1.3447*** -1.8015* -1.2171***
unempr
(0.3109) (0.9362) (0.2313) (0.3067) (0.9690) (0.2331) (0.3085) (0.9864) (0.2315) (0.3148) (0.9368) (0.2345)
-0.1867*** -0.1709** -0.0867 -0.1738** -0.1669** -0.0844 -0.1749** -0.1533** -0.0840 -0.1818*** -0.1610** -0.0847
hdi
(0.0659) (0.0734) (0.0658) (0.0709) (0.0719) (0.0678) (0.0683) (0.0756) (0.0677) (0.0689) (0.0778) (0.0654)
-0.0017*** 0.0009 0.0000 -0.0017*** 0.0009 0.0000 -0.0016*** 0.0011 0.0000 -0.0017*** 0.0011 0.0000
dci_rank
(0.0004) (0.0007) (0.0004) (0.0004) (0.0008) (0.0004) (0.0004) (0.0008) (0.0004) (0.0004) (0.0008) (0.0004)
0.0004*** 0.0000 -0.0001 0.0004** 0.0000 -0.0001 0.0004*** -0.0001 -0.0001 0.0004** -0.0001 -0.0001
constant
(0.0002) (0.0002) (0.0001) (0.0002) (0.0003) (0.0001) (0.0002) (0.0003) (0.0001) (0.0002) (0.0003) (0.0001)
0.1094*** -0.0566 -0.0132 0.1102*** -0.0670 -0.0137 0.1026*** -0.0825 -0.0139 0.1113*** -0.0759 -0.0127
month_dummy
(0.0253) (0.0464) (0.0281) (0.0259) (0.0504) (0.0282) (0.0234) (0.0551) (0.0276) (0.0240) (0.0535) (0.0275)
Obs. 408 408 408 408 408 408 408 408 408 408 408 408
Groups 34 34 34 34 34 34 34 34 34 34 34 34
R-Sq. Overall 0.472 0.265 0.402 0.462 0.261 0.402 0.47 0.255 0.401 0.462 0.258 0.401
R-Sq. Within 0.479 0.28 0.437 0.474 0.277 0.437 0.476 0.275 0.435 0.474 0.278 0.435

Credit Channelling to MSMEs in Indonesia: The Role of Regional Digital Competitiveness
R-Sq. Between 0.302 0.00785 0.0162 0.194 0.00186 0.0176 0.324 0.00453 0.0117 0.21 0.00385 0.0151
* p<0.1 ** p<0.05 *** p<0.01
337
Note: This table presents the estimation result using the random effect model estimation method. Different model specification/estimation
period is presented in each separate column from column one to 12.
All of our one-step system GMM estimations (Table 3) have numbers of
instruments that do not exceed our N (34). The Arellano-Bond test for second-
order serial correlation in the first difference (AB AR (2) in FD) also confirms
the absence of second-order serial correlation in our estimated models. With
regard to instrument validity, the Hasen test of overidentifying restrictions
for our main models (columns 1-3 in Table 3) confirms the validity of our
instruments even at a 10% significance level; similar results can also be seen
except for the estimated model in column 5 and 8 of Table 3 which we kept
nonetheless due to their purpose of showing robustness only. Overall, our post-
estimation testing shows the validity of our estimated one-step system GMM
models.
As shown in Tables 3 through 5, the coefficient of the dci_rank variable
consistently, across models and estimations, has an increasing value from 2019
to 2021. The coefficient is also positive and significant in 2021. This proves that
the digital environment has evolved to become a significant factor affecting
MSME loan channelling. This means that, as we move away from the Pandemic
into an unprecedented digital era, regions with a significantly poor digital
environment will face hardships in securing MSME credit, thus hampering its
future growth, starting from the growth of MSMEs.
As shown in Figures 1 and 2, regions with a significantly poor digital
environment will likely be the provinces outside of Java, thus worsening
Indonesia’s inequality problem. Another interesting finding from our
estimations is the increased significance of the GFCF and government spending
variables after 2019, signifying the important role of increased investment and
government spending during the Pandemic in ensuring the financial sector’s
trust in MSMEs. Furthermore, the human development index variable is negative
and significant for the 2021-period estimation, signifying that less-developed
regions tend to display higher loan growth post-pandemic, as shown in Figure
2.
Our main findings confirm that the type three credit rationing (redlining)
theory exists and that the digital environment is a significant factor. This
supports the findings of Klein & Todesco (2021). Our novelty compared to
this study is that we prove that MSMEs operating in areas that are not digitally
competitive will be lagging from the financial perspective compared to their
counterparts in more “tech-savvy” areas.
Furthermore, our findings can also be seen as an early warning of a possible
scarring effect of the Pandemic and the digital divide issues that come with it in
terms of MSMEs’ financing, thus complementing the findings of Jiang (2020).
Moreover, since MSMEs contribute a significant share of Indonesia’s GDP and
employment (Martdianty et al., 2020; Sanny et al., 2021), the future may see
the role of MSMEs diminishing for regions that are not digitally competitive,
especially in East Indonesia.
Next, we look at each component of the digital competitiveness index
and compare it by region. This comparison can help identify key areas that
are required to solve the inequality problem regarding the supporting digital
environment.
Analysis of the Digital Environment

There are three main components of the digital competitiveness index. Each
main component has its set of sub-components and indicators (Table 6). The
input component of the DCI is composed of the human resource, ICT usage,
and ICT expenditure sub-components. The human resource sub-component is
concerned with efforts related to improvements in the level of digital capabilities
and literacy of the people. The ICT usage sub-component is concerned with the
use of ICT, while the ICT expenditure sub-component focuses on the spending
on ICT-related products.
The output component of DCI is concerned with the output or share of
the ICT and ICT-related sectors in the economy (economy sub-component),
the level of digital entrepreneurship (entrepreneurship and productivity sub-
component), and the share or number of digital-related workers in the economy
(manpower sub-component).
The third component (support) focuses on the availability of infrastructure
necessary to support digital transformation (infrastructure sub-component),
fintech adoption and inclusion in the economy (finance sub-component), and
the state of development of the local policy infrastructure (regulation and local
government capacity).
Table 6. Components of the East Venture Digital Competitiveness Index
Main Sub-
Num. Num. Num. Indicators
Component Component
1 Input 1 Human 1 Number of Students with Digital Capabilities
Resource
2 Growth of Students with Digital Capabilities
3 Number of Lecturers in Digitalization-Related
Study Programs
4 Number of Digitalization-Related Study
Programs
5 Digital Literacy Index
2 ICT Usage 1 The ratio of Citizens that Have Cellular Phones
2 The ratio of Citizens that Have Computer
3 The ratio of Citizens that Have Access to the
Internet
4 The ratio of Citizens that Access the Internet
from Home
from the Office
from School
with a Laptop
with a Cellular Phone
3 ICT 1 The ratio of Households that Have Expenditure
Expenditure on ICT
2 Average Expenditure of Households for ICT
3 Total Remuneration and Wage of Information
and Communication Sector Workers
4 Average Remuneration and Wage of
Information and Communication Sector
Workers
2 Output 1 Economy 1 GRDP of the Information and Communication
Sector
2 GRDP Contribution to the Information and
Communication Sector
3 GRDP Growth of the Information and
Communication Sector
4 GRDP of Warehousing, Transportation Support,
Post & Courier Subsectors
5 GRDP Contribution to Warehousing,
Transportation Support, Post & Courier
Subsectors
Main Sub-
Component Component
6 GRDP Growth of Warehousing, Transportation
Supporter, Post & Courier Subsectors
7 GRDP of the Financial Services Sector
8 GRDP Contribution to the Financial Services
Sector
9 GRDP Growth of the Financial Services Sector
2 Entre- 1 The ratio of Workers Using the Internet in their
preneurship Main Job
and
2 The ratio of Workers Using the Internet at
Productivity
Work for Communication
Work for Promotion/Marketing
Work for Sales via Social Media
Work for Website/E-commerce Sales
6 Loan Using Fintech
3 Manpower 1 Number of Workers in Digitalization-Related
Sectors
2 The ratio of Workers in Digitalization-Related
Sectors
3 Growth of Workers in Digitalization-Related
Sectors
4 Number of Workers in Digitalization-Prone
Categories (Reverse Indicator)
5 The ratio of Workers in Digitalization-Prone
6 Growth of Workers in Digitalization-Prone
3 Support 1 Infrastructure 1 Level of Electricity Disturbance (Reverse
Indicator)
2 The ratio of Villages that Get Strong and Very
Strong Signal
3 The ratio of Villages that Get 3G Signal
4 The ratio of Villages that Get a 4G Signal
5 The ratio of Households with Fixed Lines
Connection
2 Finance 1 Financial Inclusion Index
2 Number of Officeless Financial Services Agents
(Laku Pandai Agents)
3 E-wallet Adoption Rate as Payment Method by
E-commerce
Main Sub-
Component Component
3 Regulation 1 Gross Enrollment Rate of Senior High Schools/
and Local Vocational Schools
Government
2 Gross Enrollment Rate of Higher Education
Capacity
(Diploma-Bachelor)
3 Growth Rate of Life Expectancy
4 Reduction of Poverty Rate (Reverse Indicator)
Source: (East Ventures, 2022)
Note: This table presents the components, sub-components, and indicators of the East Venture
Digital Competitiveness Index.
For the input component (Figure 3), the human resource sub-component
shows the starkest disparity between Java and regions outside of Java. The
average score of the Java region is more than three times the average score of
the East Indonesia region and approximately doubles the national average. The
East Indonesia and Sulawesi regions are also particularly left behind in terms of
ICT usage. These findings highlight the need for better improvement of digital
literacy for digital transformation (Setiawan et al., 2020; Mohelska & Zubr,
2019). Figure 3 shows the average 2021 score of the Input Component of DCI
for provinces in six Indonesian regions and the national average.
Figure 3. Average Score of the 2021 DCI Survey for the Input Component
Source: East Ventures (2022)
Figure 4 shows the average score for each sub-component of the output
component of DCI. The economy sub-component, which represents the share of
the ICT and ICT-related sectors in the economy, shows that the Java region is well
ahead compared to other regions. Meanwhile, the most concerning disparity
is found in the entrepreneurship and productivity sub-component. This sub-
component shows the level of digital entrepreneurship. The score for the Java
region is almost six times the score of the East Indonesia region and roughly
twice the national average. This finding underscores the need for better digital
literacy for entrepreneurs, supporting the findings of Klein & Todesco (2021).
The indicators for this sub-component emphasize the use of digital
technology by workers (or entrepreneurs). Thus, the starkly lower scores
for regions outside of Java, especially in East Indonesia, represent how
entrepreneurs in these regions lack the capability to utilise digital technology
for their businesses. Furthermore, the fact that the manpower sub-component,
which mainly concerns the share of worker in digitalisation-related sectors, is
relatively equal across the six regions, meaning that even though workers in
regions outside of Java have transitioned to digital sectors, they still lack the
digital skills required.
Figure 4. Average Score of the 2021 DCI Survey for the Output Component
Source: East Ventures (2022).

Note: This figure shows the average 2021 score of the Output Component of DCI for provinces
in six Indonesian regions and the national average.
Lastly, this chapter takes a look at the third component of DCI support.
In the infrastructure sub-component, the East Indonesia and Kalimantan
regions have the lowest average score. This finding calls for investments in ICT
infrastructures in these two regions so that they will not be left behind (Untari,
Priyarsono, & Novianti, 2019b; Sujarwoto & Tampubolon, 2016b).
The second area of concern regarding this component is the finance
sub-component. The 2021 score for this sub-component shows that the East
Indonesia region is unequivocally left behind; the Java region’s score for this
sub-component is ten times that of East Indonesia and more than two times
that of the national average. This result shows how digital financial inclusion
and adoption are largely concentrated in Java, while regions outside of Java are
notably lagged behind.
Figure 5. Average Score of the 2021 DCI Survey for the Support
Component
Source: (East Ventures, 2022).

Note: This figure shows the average 2021 score of the Support Component of DCI for provinces
in six Indonesian regions and the national average.
CONCLUDING REMARKS
The digital divide is likely to become a more pressing issue post COVID-19
Pandemic, with businesses that are “tech-savvy” benefiting from this
technological innovation and changes in consumer behaviour. This issue is
relevant to MSMEs' growth in a country where many of its employees can be
found in MSMEs.
This study aims to analyse credit channelling to MSMEs before and during
the pandemic. We hypothesize that MSMEs operating in regions with higher
digital technology adoption have easier access to credit/loans. Furthermore,
this relationship is likely to be stronger during the Pandemic period as digital
technology becomes more critical.
Our estimation using monthly, province-level data confirms this hypothesis.
We show that the better a province’s digital competitiveness rank is, the higher
the loan channelling to MSMEs in that province will be. Furthermore, the effect
is stronger after the start of the COVID-19 Pandemic; the magnitude of impact
from digital competitiveness rank on loan channelling to MSMEs is higher in
2021 compared to 2020, and higher in 2020 compared to 2019. These results
are robust to changes in model specifications and estimation methods. The
findings of this chapter suggest the need for interventions in regions that
lack digitalisation, as without it the regional inequality problem will likely
be worsening. Our analysis of the 2021 East Venture DCI shows that the key
areas of concern are mostly digital literacy (especially for entrepreneurs) and
investment in ICT infrastructures.
Furthermore, in tackling these issues, policymakers need to consider how
regions outside of Java, particularly regions in East Indonesia, are unequivocally
left behind. The inability to consider this crucial aspect may result in the future
worsening of the existing regional inequality problem.
Future research can be built upon and expanded from the limitations of this
research. This research has not investigated the impact of the components of the
DCI variable individually. An investigation into this may provide more robust
policy recommendations regarding key areas of concern related to digitalisation
and MSMEs. Future research can also use other proxies of digitalisation given
data availability, and run the model developed in this chapter for each region in
Indonesia to better capture the heterogeneity among regions.
ACKNOWLEDGEMENT
The early draft of this chapter was presented at the 17th Indonesian Regional
Science Association (IRSA) conference in July 2022 and at the Forum Kajian
Pembangunan (‘Development Research Forum’) in August 2022. The authors
sincerely thank the participants of the conference and seminar who gave several
valuable inputs. The authors also thank Maisa Az-Zahra, Nur Azizah Arianggi,
and Sarah Taqiya for their valuable contributions to data collection.
DISCLOSURE STATEMENT
This research was supported by Universitas Padjadjaran under its Academic
Leadership Grant Program and by the Economic Research Institute for ASEAN
and East Asia (ERIA). We have no competing interests to declare.
DATA AVAILABILITY STATEMENT

The data used in this study are openly available at Bank Indonesia’s website
(https://www.bi.go.id/id/statistik/ekonomi-keuangan/sekda/Default.aspx)
Badan Pusat Statistik website (https://www.bps.go.id/) and East Venture’s
website (https://east.vc/).
REFERENCES
Chan, Calvin M.L., Say Yen Teoh, Adrian Yeow, & Gary Pan (2019). ‘Agility in
Responding to Disruptive Digital Innovation: Case Study of an SME.’
Information Systems Journal 29 (2): 436–55. Available at: https://doi.
org/10.1111/isj.12215.
East Ventures (2020). East Ventures Digital Competitiveness Index 2020.
East Ventures (2021). Digital Competitiveness Index 2021: Indonesia’s
Accelerating Digital Economic Transformation. East Ventures.
East Ventures (2022). Digital Competitiveness Index 2022: Towards Indonesia’
S Digital Golden Era.
International Labour Organization (2019). Challenges and Opportunities
Financing Small Businesses in Indonesia. www.ilo.org/publns.
Jaffee, Dwight & Joseph Stiglitz (1990). “Credit Rationing.” In Handbook of
Monetary Economics, 2:837–88.
Jiang, Xiaojuan (2020). ‘Digital Economy in the Post-Pandemic Era.’ Journal of
Chinese Economic and Business Studies 18 (4): 333–39. https://doi.org/10.
1080/14765284.2020.1855066.
Klein, Vinícius Barreto & José Leomar Todesco (2021). ‘COVID‐19 Crisis and
SMEs Responses: The Role of Digital Transformation.’ Knowledge and
Process Management 28 (2): 117–33. https://doi.org/10.1002/kpm.1660.
Lu, Zhiqiang, Junjie Wu, Hongyu Li, & Duc Khuong Nguyen (2022). ‘Local Bank,
Digital Financial Inclusion and SME Financing Constraints: Empirical
Evidence from China.’ Emerging Markets Finance and Trade 58 (6): 1712–
25. https://doi.org/10.1080/1540496X.2021.1923477.
Lu, Zhiqiang, Junjie Wu, & Jia Liu (2020). ‘Bank Concentration and SME
Financing Availability: The Impact of Promotion of Financial Inclusion in
China.’ International Journal of Bank Marketing 38 (6): 1329–49. Available
at: https://doi.org/10.1108/IJBM-01-2020-0007.
Martdianty, Fanny, Alan Coetzer, & Pattanee Susomrith (2020). ‘Job
Embeddedness of Manufacturing SME Employees in Indonesia.’ Employee
Relations: The International Journal 42 (1): 180–93. Available at: https://
doi.org/10.1108/ER-01-2019-0087.
Mohelska, Hana & Vaclav Zubr (2019). “Financial Literacy as a Part of Lifelong
Education.” In Proceeding of the16th International Conference on ‘Cognition
and Exploratory Learning in Digital Age.’ CELDA 2019, no. Celda: 225–32.
Available at: https://doi.org/10.33965/celda2019_201911l028.
Nanziri, Lwanga Elizabeth & Peter Simiyu Wamalwa (2021). ‘Finance for SMEs
and Its Effect on Growth and Inequality: Evidence from South Africa.’
Transnational Corporations Review 13 (4): 450–66. Available at: https://
doi.org/10.1080/19186444.2021.1925044.
Rao, Purnima, Satish Kumar, Meena Chavan, & Weng Marc Lim (2021). ‘A
Systematic Literature Review on SME Financing: Trends and Future
Directions.’ Journal of Small Business Management (August): 1–31. Available
at: https://doi.org/10.1080/00472778.2021.1955123.
Roodman, David (2009a). ‘A Note on the Theme of Too Many Instruments.’
Oxford Bulletin of Economics and Statistics, 71(1): 135–158. Available at:
https://doi.org/https://doi.org/10.1111/j.1468-0084.2008.00542.x
Roodman, David (2009b). ‘How to do xtabond2: An introduction to difference
and system GMM in Stata.’ The Stata Journal, 9(1): 86–136.
Sanny, Lim, Verencia Angelina, & Bernando Budi Christian (2021). ‘Innovation
of SME Service Industry in Indonesia in Improving Customer Satisfaction.’
Journal of Science and Technology Policy Management 12 (2): 351–70.
Available at: https://doi.org/10.1108/JSTPM-03-2020-0056.
Sapulette, Militcyano Samuel, & Dyana (2020). ‘Does Fintech Growth Affect
Bank Stability: An Empirical Evidence from Indonesia Banking Industry.’
Otoritas Jasa Keuangan Vol. 2 (1).
Sapulette, Militcyano Samuel, Nury Effendi, & Teguh Santoso (2021). ‘Fintech,
Banks, and the COVID-19 Pandemic: Evidence from Indonesia.’ Bulletin of
Monetary Economics and Banking 24 (4).
Setiawan, Maman, Nury Effendi, Teguh Santoso, & Vera Intanie Dewi (2020).
‘Digital Financial Literacy, Current Behavior of Saving and Spending and Its
Future Foresight.’ Economics of Innovation and New Technology. Available
at: https://doi.org/https://doi.org/10.1080/10438599.2020.1799142.
Sujarwoto, Sujarwoto & Gindo Tampubolon (2016a). ‘Spatial Inequality and the
Internet Divide in Indonesia 2010–2012.’ Telecommunications Policy 40
(7): 602–16. Available at: https://doi.org/10.1016/j.telpol.2015.08.008.
Trinugroho, Irwan, Agusman Agusman, Mochammad Doddy Ariefianto,
Darsono, & Amine Tarazi (2015). ‘Determinants of Cross Regional Disparity
in Financial Deepening: Evidence from Indonesian Provinces.’ Economics
Bulletin 35 (2): 896–910.
Tunay, K. Batu & Serkan Kurt (2015). “Modelling and Estimation of Credits with
Dynamic Space-Time Panel Data Methods on the Basis of the Provinces
in Turkish Commercial Banking Sector.” In Proceedings of the 26th
International Business Information Management Association Conference
on ‘Innovation Management and Sustainable Economic Competitive
Advantage: From Regional Development to Global Growth.’ IBIMA 2015.
Ullah, Subhan, Pervaiz Akhtar, & Ghasem Zaefarian (2018). ‘Dealing with
Endogeneity Bias: The Generalized Method of Moments (GMM) for Panel
Data.’ Industrial Marketing Management 71 (May): 69–78. Available at:
https://doi.org/10.1016/j.indmarman.2017.11.010.
Untari, Rima, Dominicus Savio Priyarsono, & Tanti Novianti (2019a). ‘Impact
of Information and Communication Technology (ICT) Infrastructure on
Economic Growth and Income Inequality in Indonesia.’ International
Journal of Scientific Research in Science, Engineering and Technology,
January, 109–16. Available at: https://doi.org/10.32628/IJSRSET196130.

COVID-19 Disruption, Small Island Economies and Regional Development in Indonesia

Uploaded by

Copyright:

Available Formats

You might also like

COVID-19 Disruption, Small Island Economies and Regional Development in Indonesia

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

COVID-19 Disruption, Small Island Economies and Regional Development in Indonesia

Uploaded by

Copyright:

Available Formats

COVID-19 Disruption,

Small Island Economies

IRSA Book Series On Regional Development No. 21

Hak Cipta dilindungi Undang-Undang

PART I. COVID-19 DISRUPTION

PART II. SMALL ISLAND ECONOMIES

PART III. REGIONAL DEVELOPMENT

Ahmad Syahrul FAUZI

Alan Dwi WIBOWO

Aloysius Gunadi BRATA

AM. Rini SETYASTUTI

Baiq Nurul SURYAWATI

The COVID-19 pandemic has caused a significant and unprecedented disruption

Bambang Hari Kusumo

Implications of COVID-19 Disruption

CONTENT OF THIS BOOK

2 COVID-19 Disruption, Small Island Economies and Regional Development in Indonesia

Implications of COVID-19 Disruption on Small Island Economies 3

4 COVID-19 Disruption, Small Island Economies and Regional Development in Indonesia

Implications of COVID-19 Disruption on Small Island Economies 5

6 COVID-19 Disruption, Small Island Economies and Regional Development in Indonesia

Implications of COVID-19 Disruption on Small Island Economies 7

8 COVID-19 Disruption, Small Island Economies and Regional Development in Indonesia

Infrastructure for Mobilities and

10 COVID-19 Disruption, Small Island Economies and Regional Development in Indonesia

12 COVID-19 Disruption, Small Island Economies and Regional Development in Indonesia

Beyond Movement: Approaching Mobility Through Relational Lens

14 COVID-19 Disruption, Small Island Economies and Regional Development in Indonesia

8 Ho proposes a concept “emotional citizenship” to explain the social construct of ‘citizenship’ as

16 COVID-19 Disruption, Small Island Economies and Regional Development in Indonesia

SITUATING SANGIHE ISLANDS: STUDY SITE AND METHODS

Source: Map prepared by authors.

Source: Map prepared by authors.

18 COVID-19 Disruption, Small Island Economies and Regional Development in Indonesia

TYPOLOGIES OF TRAVELLING AND COMMUTING IN THE

20 COVID-19 Disruption, Small Island Economies and Regional Development in Indonesia

The first pattern of travel route, characterised as a “commuting,” involves

Photo credit: Christian Manangsang (2022).

22 COVID-19 Disruption, Small Island Economies and Regional Development in Indonesia

SEA TRAVEL AND (IM)MOBILITIES

Photo credit: Naimah Lutfi Abdullah Talib (2022).

24 COVID-19 Disruption, Small Island Economies and Regional Development in Indonesia

Photo credit: Naimah Lutfi Abdullah Talib (2022).

26 COVID-19 Disruption, Small Island Economies and Regional Development in Indonesia

Photo credit: Naimah Lutfi Abdullah Talib (2022).

Photo credit: Naimah Lutfi Abdullah Talib (2022).

28 COVID-19 Disruption, Small Island Economies and Regional Development in Indonesia

Photo credit: Naimah Lutfi Abdullah Talib (2022).

(Un)familiarity as An Embodied Experience

30 COVID-19 Disruption, Small Island Economies and Regional Development in Indonesia

32 COVID-19 Disruption, Small Island Economies and Regional Development in Indonesia

34 COVID-19 Disruption, Small Island Economies and Regional Development in Indonesia

36 COVID-19 Disruption, Small Island Economies and Regional Development in Indonesia

38 COVID-19 Disruption, Small Island Economies and Regional Development in Indonesia

40 COVID-19 Disruption, Small Island Economies and Regional Development in Indonesia

42 COVID-19 Disruption, Small Island Economies and Regional Development in Indonesia

44 COVID-19 Disruption, Small Island Economies and Regional Development in Indonesia

Does Economic Diversity Help

46 COVID-19 Disruption, Small Island Economies and Regional Development in Indonesia