Option For Social Protection Reform in Indoensia - 2008

Options for Social Protection Reform
in Indonesia
BAPPENAS
Options for Social Protection Reform
in Indonesia
Options for Social Protection Reform in Indonesia
Deutsche Gesellschaft für Technische Zusammenarbeit (GTZ) GmbH

- German Technical Cooperation -
GTZ Office Jakarta

Deutsche Bank Building, 20th Floor
Jl. Imam Bonjol No. 80
Jakarta 10310, Indonesia
Responsible
Alfred Hannig
Editing Team
Katja Bender
Johanna Knoess
Sita Zimpel
Authors
Katja Bender
Eberhard Feess
Jens Holst
Johanna Knoess
AA Oka Mahendra
MW Manicki
Asih Eka Putri
Hans-Jürgen Rösner
Tobias Schramm
Soedarmono Soejitno
Piotr Suder
Frank Tibitanzl
Iris Vernekohl
Franz von Roenne
Sita Zimpel
Design & Production

Reza – Gradasi Aksara
Image Copyrights
All images are property of GTZ.
Copyright © 2008
Deutsche Gesellschaft für Technische Zusammenarbeit (GTZ) GmbH
German Technical Cooperation/Economic Policy Advice on Social Security
All Rights Reserved/Printed in Indonesia
ii
Foreword
Social protection reform is considered as one of the major challenges in public policy, not only in developed countries, but
also in developing countries. Shortcomings in our social security system became more evident after a significant increase in
unemployment and poverty in Indonesia as a result of the economic and financial crisis in 1997. There is now greater
awareness and acceptance among policy makers in the country that changes are needed in our social protection system.
Although efforts are already being made to develop and better implement social protection policies, there is still room for
improvement. In spite of the fiscal burden, the Government of Indonesia has put social protection as a priority, conforming to
the stipulation of Article 28H, 1945 in the Indonesian Constitution.
The Government has set up goals and instruments for social protection in the National Long Term Development Plan (RPJPN),
National Medium Term Development Plan (RPJMN), and Government Work Plan (RKP). However, more work on social
protection in Indonesia is needed to identify better implementation mechanism and to enhance the quality and coverage of
social protection services.
Countries have usually adopted models of social protection that have featured state involvement in varying degrees, ranging
from a fully welfare state to one where the role of the state in social protection is rather minimalistic. The challenge countries
face is to provide feasible services that not only reflect a pragmatic framework for implementation, but also include principles
of social justice and are responsive to the needs of a vulnerable population. The latter is especially true in developing
countries confronted with poverty and slow economic growth.
Considering the challenges, the publication of this study on social protection in Indonesia is indeed timely. It highlights
current problems and offers future options that can be pursued. I believe this study will serve as a useful reference for us to
formulate future policies and programs to reform the social protection system in the country.
Finally, I would like to extend my gratitude to everyone who was involved in making this study possible. In particular, I would
like to thank Mr. Alfred Hannig, Program Director for Sustainable Economic Development and the team who prepared this
report.
Bambang Widianto
Deputy Minister for the Evaluation
of Development Performance
iii
Preface
Social Protection refers to the entire system of protective measures against risks to assist individuals, households, and
communities to better manage risks and economic shocks, and to provide support to the critically vulnerable. It includes
public interventions, such as social insurance schemes, social assistance and social safety nets, and provides room for
private as well as community-based initiatives. At some point in life, everyone will be exposed to risks that can drastically
alter their social and economic standard of living. More than half of all people worldwide are uninsured against risks of this
nature. Insufficient social security can have disastrous consequences, especially for the poor. Moreover, social security
systems are often coupled with formal employment arrangements. Most people in developing countries, however, are not in
formal employment and thus do not have access to such systems.
Besides the development and reform of existing systems, GTZ, on behalf of the German Federal Ministry for Economic
Cooperation and Development, also promotes innovative approaches that enable previously excluded population groups to
gain access to health services and other benefits through social security in 29 countries worldwide, including Indonesia. The
objective of German support is to cooperate with partner countries in the establishment of integrated social protection
systems and the extension of coverage to those groups who have hitherto been excluded.
The Government of Indonesia has taken a major step towards this end by passing Law No. 40/2004 on the establishment of
a National Social Security System (Undang-Undang Sistem Jaminan Sosial Nasional, or SJSN). This law is an important
element in paving the path for reform of the existing social protection system towards universal coverage. Despite initial
progress, the conceptual design and timing of the overall social protection reform agenda is an enormous and complex task
for all parties involved. The exchange of successful models for social protection reform from other countries is therefore
being used to further shape the reform strategy and planning.
The Government of Indonesia and the German Government currently collaborate on Social Protection in three important
areas:
1) Specific technical advice to the Ministry of Health on the implementation of Social Health Insurance;
2) Policy advice and technical support to the National Social Security Task Force headed by the Coordinating Ministry of
People’s Welfare, which is working on the legislative reform and the introduction and harmonization of new and exist-
ing regulations relevant in the context of Law No. 40/2004;
3) Policy advice to the Indonesian National Planning Authority BAPPENAS to support the strategic planning required by
the overall reform process.
This report has been compiled to provide strategic policy advice to BAPPENAS in order to support it in its planning function
in the reform process. By providing international know-how, GTZ aims at facilitating lessons learnt on social protection
reform that can assist BAPPENAS in developing long-term strategies for a comprehensive system of social protection. The
objective is to present visionary, but feasible options for the future beyond the current reform and show how the reform steps
already undertaken will influence future options.
Alfred Hannig
GTZ Program Director for Sustainable
Economic Development
iv
Acknowledgements
The “Options for Social Protection in Indonesia” report is an output of the joint BAPPENAS-GTZ collaboration on Social
Protection that was initiated in 2006. We would like to express our sincere gratitude to our partner BAPPENAS for their
continuous commitment to this study. A special thanks to Bambang Widianto, Deputy Minister for the Evaluation of Development
Performance BAPPENAS, for his leadership and full support during the entire process. This study would have not been
possible without the dedicated support of Ellyna Chairani, Acting Head of the Social Insurance Sub-Directorate BAPPENAS,
who has provided us with important background information and technical guidance on Social Protection Reform in Indonesia.
As a response to BAPPENAS' invitation to discuss key issues on the reform agenda, Alfred Hannig, Program Director for
Sustainable Economic Development, has facilitated valuable intellectual responses and provided strategic input.
Furthermore, the report has extremely benefited from the literature review on Social Protection in Indonesia prepared by
SMERU.
We are grateful for the enriching comments and fruitful discussions throughout the entire development of this report and
during the workshop held in August 2007 for the first review of the report. We would also like to thank the international
organizations that attended the workshop for their contributions, and we hope that we can continue our exchange and
collaboration on this important issue.
We would especially like to thank Yves Guerard, Asep Suryahadi and Dharendra Wardhana for their extremely valuable
comments on the final draft of this report.
Indispensable organizational support was provided by Dian Vitriani and the entire RED Team.
v
vi
Executive Summary
Indonesia has achieved remarkable progress in its transition to democracy and its economic recovery from the Asian financial
crisis of the late 1990s. The country is now characterized by a young and growing democracy with a decentralized government
in a vast archipelago of different cultures and a population of 230 million people. Poverty levels that had increased by over
one-third during the crisis are now back to pre-crisis levels. Indonesia has once again become one of the world’s emergent
middle-income countries 1. The Indonesian Government estimates the country’s economic growth to be around 6.6 percent
for 2008.
Despite this encouraging success and the ambitious development goals of the Government of Indonesia, the number of
people living under 2 USD a day is at an alarming 42 percent of the total population. Approximately 40 million people, nearly
10 million households, are clustered around the lowest two of the three national poverty categories with an average income
between Rp. 125,000 to 150,000 per month per person. These figures are exacerbated by the fact that many households are
clustered around the national income poverty line and defined as near poor, indicating an even larger number of Indonesians
who are vulnerable to poverty. Second, poor human development outcomes and a lack of access to basic services for many
Indonesians make non-income poverty as serious, if not more so, than income poverty. Third, given the vast size and varying
conditions in Indonesia, regional disparities are a main challenge for the Indonesian Government. Poverty incidence is far
higher in Eastern Indonesia, but most of the population lives in the densely populated western regions. Recent poverty
developments in Indonesia have been significantly affected by large increases in the price of rice and a rise of fuel prices in
2005. The latter has had a mixed effect, as the price for fuel rose because subsidies on it were reduced to mobilize the
funding for Askeskin, the new government program to provide basic health services free of charge to the poor strata of the
population.
This indicates that persistent efforts are required to ensure that economic growth benefits the majority of the population, the
poor and the near-poor. The latter crucial in order to achieve the ambitious goals laid down by the Government for social
development. For example the mid-term development plan (RPJM) for 2005-2009 aims to reduce poverty significantly in
2009. The Government of Indonesia is also dedicated to achieving the Millennium Development Goals (MDGs) by 2015.
To increase coverage specifically for the poor, the Government of Indonesia has embarked on several steps towards this
end: in 2005, despite strong political controversy, the fuel-subsidy, which disproportionately benefited the medium-income
and rich households and at the same time was one of the centerpieces of the Indonesian social protection system until that
time, was reduced 2. Furthermore, the Government of Indonesia has launched several pro-poor programs recently to buffer
the increased cost of living for the poor.
1
World Bank (2006b:181).
2
Ibid.
vii
These efforts to provide increased support to the poor were complemented by the Government’s decision in October 2004 to
implement Law No. 40 concerning a National Social Security System (Udang-Undang Sistem Jaminan Sosial Nasional, or
SJSN), which foresees reforming the existing social protection system towards universal coverage. A task force appointed
through the Vice-President’s Office, first created by the Presidential Decree (Kepres) No. 22/2002, drafted the law. The key
feature of the new law is that it mandates the creation of several social security schemes for citizens: old-age pension, old-
age savings, national health insurance, work injury insurance, and death benefits for survivors of deceased workers. The
schemes would be financed by a payroll tax imposed on workers’ wages, collected equally from employers and workers,
mostly in the formal sector. Law No. 40/2004 is an important milestone because it stipulates that the existing social security
programs in Indonesia will be expanded to cover all Indonesian citizens, including those who are working in the informal
sector, the unemployed and the poor.
The implementation of Law No. 40/2004 on the National Social Security System is currently bringing about a comprehensive
reform of the existing system. Law No. 40/2004 provides the legal basis for the above stated objective of social protection
coverage for all. A Task Force, headed by the Coordinating Ministry of People’s Welfare, is working on the introduction of
new regulations as well as the mapping and harmonization of existing regulations relevant in the context of Law No. 40/2004.
It has made considerable achievements since its appointment: finalizing mapping and extensive review of existing laws, as
well as a comprehensive design of a harmonized legal and regulatory framework. Based on the latter, there is now a
roadmap for implementation in place, and technical sub-groups are working on new laws - most importantly, the ‘carrier law’,
and government regulations. The whole process has been aligned with Law No. 10/2004 that stipulates how laws are to be
drafted. It has recently been formally endorsed by the president and now appears set to be completed by the legal deadline
in October 2009.
public interventions such as social insurances, social assistance and social safety nets, and provides room for private and
community-based initiatives. Social Security is a component of social protection and comprises the key social insurances,
including health insurance, pension insurance, unemployment benefits, work injuries, and disability grants (mainly contribution-
financed). Basic Social Protection or Social Assistance comprises non-contributory measures targeted at vulnerable groups/
individuals, e.g. the poor, elderly, disabled persons (mainly tax financed).
As important initial reform steps of the social protection system have been taken, major challenges lie ahead:
• Low coverage: Coverage by the existing contributory social insurance system is still low 3: only about 16 million workers,
out of a total labor force of about 101 million people (including job seekers), are covered by TASPEN, ASABRI and
JAMSOSTEK schemes. This means that only 16 percent of workers are currently covered by formal social security
schemes. Health insurance coverage by public schemes is slightly higher with approximately 18 million insured. Extending
coverage in terms of both organizational as well as sustainable financing options will be based on a comprehensive
strategy that is currently prepared by the ministries charged with building the new social security system (Jaminan Sosial
Nasional, JAMSOSNAS). This requires detailed conceptual work and decisions on, for example, benefit package(s),
contribution rates, and necessary institutional reforms. It also requires a decision on appropriate social protection
instruments with regards to specific risks, such as old age and unemployment.
• Fragmentation: One objective of the reform is to build a coherent system of social protection and reduce the high
degree of fragmentation that characterizes the current system. Benefits of streamlining include improved efficiency of
allocation of administrative resources. Coordination between the various programmes that are targeting the poor needs
to be enhanced. Cooperation and communication between the concerned government ministries and the service and
insurance providers will be facilitated by a new statutory body, the National Social Security Council, soon to be installed
by the President.
3
Angelini and Hirose (2004).
viii
Executive Summary
• How to cover the informal sector? The present system excludes the non-poor in the informal sector from any
arrangement: They are not covered by the contributory social insurance and are not eligible for any form of social
assistance. Considering the fact that the informal sector in Indonesia comprises a substantial part of the working population4
with a high proportion of them being considered non-poor, developing an effective strategy on behalf of this population
group is crucial for realizing the goal of universal coverage.
As mentioned previously, one crucial reform issue is the streamlining of the numerous bodies that comprise the current
system, and the defragmentation of its functions. At present, a relatively small part of the population is covered by public
social security, namely employees of the military (through PT Asabri, PT Askes), civil servants (through PT Taspen, PT
Askes) and to those employed in private enterprises above a certain staff size and total payroll (through PT Jamsostek). The
latter are insured against work accidents, illness, death and funeral risks, but not against unemployment. In case of retirement,
they receive a lump sum payment from a mandatory saving scheme (provident fund) related to their contributions, plus
interest. Members of the armed forces and civil servants are insured against illness (PT Askes), as well as old age in which
case they receive lump sum payments as well as annuity benefits for life after retirement (through PT Asabri or PT Taspen).
It can be seen that currently the institutional set-up of social security is organized primarily according to group affiliation. The
responsibilities for programs of social assistance are vested within different ministries and different administrative structures.
In addition to this, programs are operated at central and local levels.
Social Protection, as it is discussed in this report, is based on the Social State Model, and incorporates several elements of
the Welfare State Model.
In a Social State Model social protection is based (1) on the social insurance principle, meaning adequate, continuous
income replacement in case of the occurrence of certain existential risks (social security), and is financed by income-related
contributions; and (2) on complementary social assistance, meaning residual economic or social support for poor and needy
individuals or vulnerable groups of society, and is means-tested and financed by taxes. The Welfare State Model, in contrast,
aims at providing comprehensive social protection against all existential risks and is based on general provisions of public
care and financed by taxes, thus applying universality and without means-testing. The main difference between the two
approaches is, therefore, the extent of how responsibility for provision against common risks is shared between the individual
and society.
The guiding principle for the Social State Model is, thus, combining individual effort with solidarity - the normative basis for a
social protection policy in a social state concept based on the principle of subsidiarity, derived from the Latin “subsidium” (i.e.
backing, support or protection), according to which provisions against social risks should always be responsible of the most
basic unit possible, on a range from the individual through family, the voluntary sector and group solidarity up to the different
levels of the state. The state should only intervene when the more basic units are unable to support their own members. This
is especially the case for those individuals or groups who are in danger of being excluded from relations with the community
and its social or economic development. Those who are economically and socially better off provide those who are
disadvantaged with the social element of redistribution, or solidarity compensation.
This approach of combining the Social State Model with welfare elements seems to be the most suitable in Indonesia
because it is not easy to decide whether Indonesia should turn to the Welfare State or to the Social State approach to Social
Protection. When examining the present social protection policies in Indonesia, the impressions are mixed: on the one side,
there is strong political commitment towards universal social progress and poverty reduction, as expressed in Law No. 40/
2004, while on the other side the institutional framework for implementation is fractured and incomplete. While Law No. 40/
2004 stipulates that financing of social protection should be based on personal contributions in the field of social insurance
and on means-tested tax-financed elements in the field of social assistance, there are institutional and procedural deficits in
that responsibilities for financing and the necessary amount of contributions are not clearly specified. Furthermore, the
competencies for specific risks and for implementation of related programs between the administering bodies are not clearly
defined and often contradictory. This is further complicated by the high level of poverty and the size of the informal sector,
4
According to the National Labor Force Survey (2007), the informal sector in Indonesia comprises a substantial and growing part of the working population,
with currently 69 percent of the total labor work force.
ix
which seem to imply the principle of universality, as it would facilitate the integration of different spheres of society under the
umbrella of centralized and uniform protection measures. On the other hand, it is doubtful whether there is sufficient
administrative capacity to implement such universal programs and whether it would be financially sustainable at the present
stage of development in Indonesia.
Opting for Bismarck and the Social State Model under these different conditions involves certain risks in that it presupposes
an ideal situation of high economic growth and increasingly formal employment. This given, the social insurance principle
would not only gradually provide more and more benefits for the workforce but, and what is probably even more important,
this system finances itself automatically as wages rise and contributions are income related. On the other hand, opting for
Beveridge and the mainly tax-financed Welfare State Model requires not only effective fiscal policy, but also positive political
budget decisions which might be hard to get when times are economically or politically rough.
Taking these limitations into account, it might be advisable for Indonesia to integrate elements from both, the Bismarck and
the Beveridge Models, into a specific “hybrid” form better suited to the country’s present needs.
The future face of the Indonesian Social State Model of social protection could therefore consist of a composition of
public social insurance and social assistance schemes together with provisions for the private sector in three overlapping
structures.
The enactment of Law No. 40/2004 provides an important step and the legal basis for the development of a comprehensive
and coordinated social security system in Indonesia. Prior to that, the administration of social security was fractured and
incomplete, covering only a relatively small part of the population. Still, open questions remain concerning the administration
of a Social Security program.
From a technical point of view, there are two basic options for organizing the administration of social protection: (1) integration
of all risk-related activities under one specific administrative body for each risk, and (2) separation of all risk-related activities
for certain groups of society and their specific administrative bodies. There is also a third, more pragmatic, approach, i.e.,
harmonization and rationalization of regulations between existing administrative bodies.
Focusing on the technical efficiency of a public system of social protection with an integration of carriers is an appropriate
option for social insurances, whereas social assistance is best provided within a framework of (broad) separation: insurance
works by pooling risks across a large population. The larger the risk pool, the smaller the volatility of claims, thereby facilitating
management of the insurance. However, to effectively reduce volatility, an insurance risk pool must necessarily not be as
large as the entire population5, and the positive returns to size diminish beyond a certain threshold. Yet for social insurance
to work, a balanced composition of the risk pool is essential. This is achieved by integration, not by separation. In contrast,
social assistance is, generally speaking, best provided within the framework of a (broad) separation: First, by definition it
targets specific groups within a population, i.e. the poor and other specific groups in need. Secondly, again by definition, all
social assistance benefits are targeted benefits. As targeting methods involve non-negligible transaction costs, subsuming
social assistance under one administrative body provides the opportunity for exploiting economies of scope. In the third
option of harmonization, institutional path dependencies are considered, and the existing institutional framework for social
protection is gradually adapted as required by new or changing needs.
Related to the present situation in Indonesia, the first option of institutional integration of risk-related activities under one
specific administrative body for each risk would imply to make JAMSOSTEK the superstructure of a comprehensive system
of social insurances against work accidents, old age and, perhaps later, unemployment. Institutions, such as Asabri, Taspen
or Askes, would then have to open their membership or merge into the general social insurance system. The final result
would be a pillar-like structure of administering bodies with clear-cut responsibility for the provisions and regulations in their
special fields. The main advantage of this solution would be that a higher level of transparency, as well as effectiveness and
efficiency, could be achieved. Referring to the well-known resilience of administrative institutions, the second reform option
of institutional separation of risk related activities for certain groups of society would be easier to realize, as such more or
5
If this was the case, private insurance would be unable to operate.
x
Executive Summary
less strictly group-oriented provision of social security services is already in place. Hence, separation would only require to
follow this principle more consequently in that for informal workers (and probably for micro-entrepreneurs as well) a similar
special scheme and, respectively, a separate administration would have to be institutionalized following the specific needs
and requirements of this part of the workforce. The main problem connected with this reform option is that it would divide the
Indonesian society into privileged groups, such as the military, civil service and big company employees, with membership
in their specific administrative bodies (ASABRI, ASKES, ASPEN, and JAMSOSTEK) on the one side, while leaving the rest,
the majority, of the population in a precarious situation, as it will be difficult to find a financially viable solution for their social
security. The third and last reform option of basically leaving the institutional framework in place would limit reform needs to
only requiring harmonization and rationalization of sometimes contradictory regulations between administrative bodies.
Harmonization is the option that Indonesia has presently chosen with the aim of harmonizing all laws and regulations
through 2009. Harmonization represents an important first step. However, given limited technical benefits of harmonization
in comparison to the other two options, the country may consider proceeding further in the mid-term. Options for later
reforms are: (1) institutional integration of risk-related activities under one specific administrative body for each risk in standard
social insurance, and (2) institutional separation of activities for the poor and other vulnerable groups, i.e. social assistance.
In this context, responsibility for social assistance should then be centered on a newly founded institution, perhaps under the
umbrella of the Coordinating Ministry of Social Affairs6.
Social health insurance is one of the major components of the system of social security. This fact is also reflected by the
priority this component is given within the overall reform process in Indonesia. With regards to social health protection, the
systemic question of which instruments to apply appears to be answered by combining social insurance and tax-financed,
means-tested social assistance for the poor. At present, PT Askes and PT Jamsostek provide social health insurance to civil
servants and formal sector employees. In addition, the ASKESKIN scheme, managed by PT Askes, provides tax-financed
health care to poor households. The ratification of Law No. 40/2004 on the NSSS established the basic principles of universal
social security and a timeframe for the works necessary for the new system to be implemented throughout Indonesia. Within
the boundaries already set by the new NSSS law and the reform and implementation steps taken since then, however, many
decisions still need to be made. For example, the choice on how to pool funds, i.e. single vs. multiple funds, is currently
under scrutiny.
Besides social health protection, the issue of old age protection is gaining increasing attention within the Indonesian context.
Formal sector workers receive a single lump sum payment from a mandatory saving scheme (provident fund) related to their
contributions, plus interest. Longevity risks are not covered. Members of the armed forces and civil servants receive a single
lump sum payment as well as an annuity benefit for life after retirement. In the absence of a comprehensive social insurance
scheme providing life-long old-age benefits, the reform agenda for pensions in Indonesia is still open for innovative solutions.
For prevention against demographic risks one option would be to combine funding elements with pension insurance in a new
national pension scheme. Subsequently, once the new national pension scheme is fully established, the share of pay-as-
you-go financing could be reduced gradually in favor of the funding principle. Low-income earners could be supported to a
certain extent by tax-financed instruments. A second option would be to introduce a tax-financed universal basic pension
scheme with pensions based on a fixed percentage of local minimum wages. However, in this option there are certain
drawbacks to consider: (1) like all universal schemes, it could be very costly to guarantee such a basic pension, considering
the high percentage of the poor and near poor population; (2) a non-contributory pension scheme could create negative
incentives for participation in contributory social insurance, so this solution could endanger movement from informal to
formal employment. Another approach to address the problem of old age poverty could be seen in the future role of social
assistance, which would be subject to means-testing and targeted to all persons with income below the national poverty line.
This option, however, is not a substitute to the previous options, but rather a complementary one.
6
With respect to certain social insurances, the question arises whether the poor or other specific groups of need should be included within the general
insurance framework or covered separately. This is particularly relevant for social health insurance. In this case, the poor should be included within the
specific social insurance, and the administrative body responsible for social assistance could remit the contributions on behalf of the poor to be covered to
the respective social insurance administrative body, which would then be responsible for delivering the benefits or services to the poor.
xi
Unlike social health protection, the issue of unemployment compensation and the instruments to choose for it is still strongly
debated at present. Currently only severance pay plays an important role in Indonesia. Alternative instruments, such as
unemployment insurance, unemployment assistance, individual saving accounts for unemployment benefits, and mandatory
saving accounts for severance payments, do not exist in Indonesia. Preliminary recommendations for unemployment
compensation in Indonesia include a combined system of mandatory individual saving accounts and mandatory savings for
severance pay, both of them complemented by solidarity funds.
Social Assistance comprises the non-contributory provision of public aid to targeted poor and/or certain groups considered
to be vulnerable. Social assistance measures in Indonesia display a high degree of fragmentation. Providing effective social
assistance requires horizontal and vertical integration. Therefore, as a first step in the direction of further harmonization, the
complex web of inherited social assistance programs should be consolidated and simplified, and administration delegated to
a newly founded institution operating under the umbrella of the Ministry of Social Affairs. In contrast to social security, which
is financed via individual contributions of participants, taxes provide the most important financing source for social assistance.
Strengthening the outreach of social assistance programs involves the need to increase available tax revenues or to reallocate
expenditures among sectors. Several options exist to enhance the tax base without increasing taxes. Considering the
unfortunate relationship between the number of poor and near poor persons in Indonesia and the available tax resources,
the potential outreach of social assistance is limited and any ad hoc measures should be avoided. Reform options require
a decision as to whether to provide a relatively broad range of social assistance measures to very limited groups of persons
or to focus on a more limited number of measures (such as health services or modest forms of income support via cash
transfers) provided to a wider share of the poor population. Last, but not least, sustained economic growth combined with a
clear pro-poor focus is clearly the eminent factor. It increases tax revenues, while simultaneously decreasing the number of
potential beneficiaries.
One of the crucial questions remaining concerns how to extend social protection coverage to the non-poor in Indonesia’s
large informal sector. The non-poor among informal sector workers are the population segment currently suffering the highest
degree of exclusion from social protection: they are not covered by contributory social insurance and are not eligible for any
form of social assistance. The legal postulate of universal coverage will have to bring about fundamental changes in order to
provide social security to this group. Whereas standard contributory social security is designed for formal sector employees
and the poor are targeted by tax-financed and means-tested social assistance mechanisms, only a few programs are directed
towards the non-poor in the informal sector. Considering the enormous size of the informal sector in Indonesia, the overall
integration of the informal into the formal economy should be the ultimate long-term objective. To include this group within a
mid-term perspective, specific social security programs will have to be designed and implemented, as standard social
insurance with its defining characteristics of statutory membership and income-related contributions is not appropriate for
reaching out to the informal sector. Due to the special characteristics of the heterogeneous and poorly defined group of
people living in the informal economy and the resulting specificities of social security schemes for the informal sector,
schemes for this group carry a variety of risks that endanger their effectiveness in terms of coverage and financial sustainability.
Reliable data for planning, a careful design and skilled administrative personnel for implementation are, therefore, necessary
preconditions.
Questions with respect to social health protection reform refer primarily to technical problems within an previously defined
system and are presented earlier in this report (see above) 7. The question as to whether the Government contributions for
health services for the poor in community health facilities (PUSKESMAS) should be paid via the regular health insurance
carrier or directly to the PUSKESMAS using block grants is reviewed separately.
Both financing modalities have been applied since 2005: insurance-based funding via an insurance carrier (PT Askes) and
direct funding of PUSKESMAS using block grants from the central and local government. Considering the specific incentive
structures and different requirements involved, insurance-based financing seems to be the more sustainable and effective
option in the long term.
7
See Chapter 4.3. in this report, “The National Social Health Insurance Program: Work in Progress”, prepared by the GTZ Health Policy Project/Social
Health Insurance Team.
xii
Executive Summary
Reviewing the current situation and analyzing the options for the various social protection schemes for citizens revealed that
the efficacy of ongoing and future reforms will depend highly on developments and progress in other sectors. Most prominently,
it will depend on private sector development and the expansion of the formal sector, and, respectively, reduction of the
informal sector, tax system reforms, and sustained efforts in decentralization reforms. As economic growth clearly impacts
the feasibility and outreach of publicly provided social protection, continuing efforts to promote an enabling environment for
economic growth are essential. This includes, among others, the creation of a conducive and investment-friendly business
environment.
Further, reform options also need to be carefully designed in order to prevent adverse impact on the labor market. Another
important aspect is that although social protection in the long run has the potential of enhancing social stability and social
cohesion by achieving equity, it demands solidarity and ownership from the entire population. Communication and marketing
of reform efforts can aid in achieving support. However, in the end, people will only support a social protection system if they
receive benefits and if it operates in a credible way. Most likely, the society, specifically the economically stable, will expect
solid evidence that they are also benefiting from the system and not just primarily financing it.
All in all, institutions need to be strengthened, and technical support and capacity development need to be available at all
stages of the reform.
Further, reform efforts should be realistic and consider prevailing constraints. The questions regarding the sequencing and
timing of reform steps are beyond the scope of this paper. However, the continuous fine-tuning and further development of
the Road Map for Social Protection Reform in Indonesia are, naturally, very important issues.
Last but not least, it needs to be emphasized that achieving universal coverage, as intended by the Government of Indonesia,
is a lengthy and incremental process, especially with a vast and heterogeneous population such as that of Indonesia. With
its long-term political commitment to achieving social security for all, the Government of Indonesia has met an essential pre-
condition for success.
xiii
Abbreviations and Acronyms

ADB Asian Development Bank
AF Asia Foundation
APBD Regional Budget (Anggaran Pendapatan dan Belanja Daerah)
APBN National Budget (Anggaran Pendapatan dan Belanja Negara)
ASABRI Insurance for the Armed Forces (Asuransi Angkatan Bersenjata Republik Indonesia)
ASKES Health Insurance (Asuransi Kesehatan)
ASKESOS Social Welfare Insurance (Asuransi Kesejahteraan Sosial)
ATP Ability-To-Pay
BAPPENAS National Development Planning Agency (Badan Perencanaan dan Pembangunan Nasional)
BKSP Social Welfare Program (Bantuan Kesejahteraan Sosial Perrmanen)
BKM Special Aid Program for Students (Bantuan Khusus Murid)
BPS Central Bureau of Statistics (Badan Pusat Statistik)
BOS Operational Aid Program to Primary and Secondary Schools (Bantuan Operasional Sekolah)
CBO Community-Based Organization
CCT Conditional Cash Transfer
CGI Consultative Group on Indonesia
DAK Special Allocation Fund (Dana Alokasi Khusus)
DAU General Allocation Fund (Dana Lokasi Umum)
DE Directory Establishments
DEPSOS Ministry of Social Affairs (Departemen Sosial)
EC Economic Census
GDP Gross Domestic Product
GTZ Deutsche Gesellschaft für Technische Zusammenarbeit
HFA Health for All
IFC-PENSA International Finance Cooperation - Program for Eastern Indonesia SME Assistance
IFLS Indonesia Family Life Survey
ILO International Labor Organization
INPRES Presidential Instruction (Instruksi Presiden)
IPP Individual Paying Program
ISA Individual Savings Account
JAMSOSNAS National Social Security Program (Jaminan Sosial Nasional)
JAMSOSTEK Social Security Program for Employees (Jaminan Sosial Tenaga Kerja)
JPK-Gakin Health Insurance for Poor Families (Jaminan Pemeliharaan Kesehatan Keluarga Miskin)
JPS Social Safety Net Program (Jaring Pengaman Sosial)
JPS-BK Social Safety Net - Health Care Program (Jaring Pengaman Sosial - Bidang Kesehatan)
JSR Job Security Regulation
Kepres Presidential Decree (Keputusan Presiden)
KPPOD Regional Autonomy Watch (Komite Pemantauan Pelaksanaan Otonomi Daerah)
xiv
Abbreviations and Acronyms
KUD Rural Cooperatives (Koperasi Unit Desa)

LT Layoff Taxes
MDGs Millennium Development Goals
MOF Ministry of Finance
MOH Ministry of Health
MOHA Ministry of Home Affairs
NDE Non-Directory Establishments
NHIP National Health Insurance Program
NLFS National Labor Force Survey
NSHIP National Social Health Insurance Program
NSSS National Social Security System
OECD Organisation for Economic Cooperation and Development
OG Organized Groups
OSS One-Stop-Shop
PAYG Pay-As-You-Go
Perda Regional Regulation (Peraturan Daerah)
PKPS-BBM Compensation Program for Reduced Subsidies (Program Kompensasi Pengurangan Subsisi Bahan
Bakar Minyak)
PNPM National Program for Community Empowerment (Program Nasional Pemberdayaan Masyarakat)
POU Program-Operating Unit
Puskesmas Community Health Center (Pusat Kesehatan Masyarakat)
RPJM Mid-term Development Plan (Rencana Pembangunan Jangka Menengah)
SA Social Assistance
Sakernas National Labor Force Survey
SfDM Support for Decentralization Measures (project supported by GTZ)
SHI Social Health Insurance (project supported by GTZ)
SJSN National Social Security System (Sistem Jaminan Sosial Nasional)
SME Small and Medium-Sized Enterprises
SNPK National Poverty Reduction Strategy (Strategi Nasional Penanggulangan Kemiskinan)
SP Severance Pay
SUSENAS National Socio-Economic Survey (Survei Sosial Ekonomi Nasional)
UA Unemployment Assistance
UCT Unconditional Cash Transfer
UI Unemployment Insurance
UNDP United Nations Development Programme
UNESCAP United Nations Economic and Social Commission for Asia and the Pacific
UNICEF United Nations International Children’s Emergency Fund
USD United States Dollar
WTP Willingness-To-Pay
xv
Table of Contents
Foreword iii
Preface iv
Acknowledgements v
Executive Summary vii
Abbreviations and Acronyms xiv
1. Introduction 1
2. Background Information 7
2.1. Economic Growth and Poverty Reduction 7
2.2. Decentralization 9
3. The Social State Model: a Model for Indonesia? 13

3.1. Models for Social Protection 14
3.1.1. The Bismarck Model 14
3.1.2. The Beveridge Model 15
3.1.3. Comparison of Bismarck and Beveridge 16
3.1.4. The Welfare State Model vs. the Social State Model 17
3.2. Guiding Principles of the Social State Model 18
3.3. The Social State Model within the Indonesian Context 19
3.3.1. The Current Situation of Social Protection Policy 19
3.3.2. The Vision: Indonesia - a Social State with Welfare Elements 20
4. Social Insurance: Reforming Contributory Social Security 23

4.1. Social Insurance: Conceptual Framework 23
4.1.1. Organizational Principles of Social Insurance 25
4.1.2. Financing Social Insurance 26
4.2. The Administration of Social Security 27
4.2.1. The Current Situation 27
4.2.2. Reform Options 28
4.3. The National Social Health Insurance Program: Work in Progress 32
4.3.1. The Current Status of the System 32
4.3.2. NSSS - the Law on the National Social Security System 37
4.3.3. The Proposed Presidential Regulation on the National Health Insurance System 38
4.3.4. Closing Remarks 46
4.4. Pensions 47
4.4.1. The Current Situation 47
4.4.2. Reform Options 48
5. Unemployment Compensation 51
5.1. Instruments for Unemployment Compensation and Implementation in Different Countries 52
5.1.1. Unemployment Insurance (UI) 52
5.1.2. Unemployment Assistance (UA) 53
5.1.3. Social Assistance (SA) 54
xvi
Table of Contents
5.1.4. Individual Saving Accounts (ISA) 54

5.1.5. Severance Pay 55
5.1.6. Layoff Taxes 55
5.2. Incentive Effects of Different Instruments of Unemployment Compensation 56
5.2.1. Unemployment Insurance, Unemployment Assistance and Social Assistance 56
5.2.2. Individual Saving Accounts (ISA) 60
5.2.3. Severance Pay and Layoff Taxes 61
5.3. The Current Situation in Indonesia 64
5.3.1. Social Security in Indonesia - Some Brief Remarks from the Perspective of 64
Unemployment Compensation
5.3.2. Some Brief Remarks on Social Assistance 65
5.3.3. Some Brief Remarks on Severance Pay 67
5.4. Preliminary Recommendations and Open Questions 69
6. Social Assistance: Reforming Non-Contributory Social Protection 73

6.1. Social Assistance: Conceptual Framework 74
6.2. The Current Situation 77
6.3. Reform Options 80
6.3.1. Organization of Social Assistance 80
6.3.2. Financing Social Assistance 81
7. Extending Social Security to the Informal Sector 85

7.1. The Informal Sector: Conceptual Framework 86
7.1.1. Defining the Informal Sector 86
7.1.2. Proposed Working Definition of the Informal Sector 87
7.1.3. Explaining Informality or Why Individuals Operate Informally 88
7.2. The Current Situation 90
7.2.1. Identification and Characterization of Excluded Individuals: 92
Socio-Economic Characteristics of the Informal Sector in Indonesia
7.2.2. Problem Statement 94
7.3. Reform Options 95
7.3.1. General Reform Options 95
7.3.2. Integration of the Informal Sector: the Formalization Approach 96
7.3.3. Extending Social Security Coverage to the Informal Sector: the Direct Inclusion Approach 100
7.3.3.1. Specific Characteristics of Social Security Programs for the Informal Economy 100
7.3.3.2. Specific Risks of Social Security Programs for the Informal Economy 101
7.3.3.3. Program Development 103
7.3.3.4. Contracting with the Private Sector to Target the Informal Economy 107
7.3.3.5. Country Examples 110
7.4. Summary 113
8. The Role and Financing of Puskesmas in the Indonesian Health System 115
8.1. Puskesmas: Service Delivery and Financing 116
8.2. Sustainable Financing for Puskesmas and Social Health Insurance 117
8.3. Conclusions regarding the Financing of Puskesmas 118
8.4. Options for the Future Inclusion of Puskesmas in a SHI Scheme 120
9. Conclusions 123
xvii
Annex 127
Annex 1: Characteristics of Social Security prior and according to Law No. 40/2004 128
Annex 2: Various Funding Sources of Puskesmas according to Routine and Non-Routine Income 132
Annex 3: Social Health Protection in Latin America 133
Annex 4: Case Studies: Protecting the Elderly Poor 138
Annex 5: Overview of Different Systems for Unemployment Compensation 143
Annex 6: Months of Severance and Long Service Pay 144
Annex 7: Severance Pay as Tax on Monthly Wages 145
Bibliography 146
xviii
List of Tables and Figures
Tables Page
Table 1 Number and Percentage of Population below the Poverty Line 8
Table 2 Main Principles of the Bismarck and Beveridge Models 14
Table 3 Characterizing Risks 25
Table 4 Administrative Options 30
Table 5 Synopsis of the Existing State-Supported Health Insurance Programs in Indonesia, 35
prior to the NSSS Law
Table 6 Further Legislative and Regulative Requirements Stipulated by the NSSS Law 38
Table 7 How the NSSS Law Delegates NHIP Regulation to a Presidential Regulation 39
Table 8 Relationships of PresReg NHIP and Other National Laws 40
Table 9 Contributions/Premiums for the JAMSOSTEK Program (Percentage of Wages) 65
Table 10 Employment and JAMSOSTEK‘s Members 66
Table 11 Estimated JAMSOSNAS Contributions/Premiums for Formal Sector Workers 67
(Percentage of Wages)
Table 12 Types of Social Assistance in Indonesia 80
Table 13 Actual Government Revenues (in Billion Rp) 82
Table 14 ILO Conceptual Framework: Informal Employment 88
Table 15 Conceptualizations, Characteristics and Views on the Informal Sector 89
Table 16 Categorization of the Indonesian Labor Force 91
Table 17 Key Differences between the Informal and Formal Sector 100
Table 18 Specific Risks of a Social Security Program for the Non-Poor in the Informal Sector 103
Table 19 Contracting with CBOs 109
Figures Page
Figure 1 Administrative Divisions in Indonesia 12

Figure 2 Core Elements of the Social State Model 21
Figure 3 Timeline of Health Sector Reform in Indonesia 33
Figure 4 Laws and the Presidential Regulation in the Making Affect Different Parts 41
and Functions of the System
Figure 5 Options for the Pooling of SHI Funds 42
Figure 6 Severance and Long Service Pay for Lay-Offs and Plant Closings 68
xix
xx
1. Introduction
Indonesia has achieved remarkable progress in its transition to democracy and its economic recovery from the Asian financial
crisis of the late 1990s.
The country is now characterized by a young and growing democracy with a decentralized government in a vast archipelago
of different cultures and a population of 230 million people. Poverty levels that had increased by over one-third during the
crisis are now back to pre-crisis levels. Indonesia has once again become one of the world’s emergent middle-income
countries 8. The Indonesian Ministry of Finance estimates the country’s economic growth to be around 6.6 percent for 2008.
Despite this encouraging success and the ambitious development goals of the Government of Indonesia, the number of
people living under 2 USD a day is at an alarming 49 percent of the total population. Approximately 40 million people, nearly
10 million households, are clustered around the lowest two of the three national poverty categories with an average income
between Rp. 125,000 to 150,000 per month per person. These figures are exacerbated by the fact that many households are
clustered around the national income poverty line and defined as near-poor, indicating an even larger number of Indonesians
who are vulnerable to poverty. Moreover, poor human development outcomes and lack of access to basic services for many
Indonesians make non-income poverty as serious, if not more so, as income poverty. Given the vast size and varying
conditions in Indonesia, regional disparities are a main challenge for the Indonesian Government. Poverty incidence is far
higher in Eastern Indonesia, but most of the population lives in the densely populated western regions. Recent poverty
developments in Indonesia have been significantly affected by large increases in the price of rice and a rise of fuel prices in
2005. The latter has had a mixed effect, as the price for fuel rose because subsidies on it were reduced to mobilize the
funding for ASKESKIN, the new government program aimed at providing basic health services free of charge to the poor
strata of the population.
Persistent efforts are required to ensure that economic growth benefits the majority of the population, the poor and the near-
poor. This is crucial in order to achieve the ambitious goals laid down by the Government for social development. For
example, the mid-term development plan (RPJM) for 2005-2009 aims to reduce poverty significantly in 2009. The Government
of Indonesia is also dedicated to achieving the Millennium Development Goals (MDGs) by 2015.
8
World Bank (2006b:181).
1
To increase coverage specifically for the poor, the Government of Indonesia has embarked on several steps towards this
end. In 2005, despite strong political controversy, the fuel subsidy, which disproportionately benefited the medium-income
and rich households and at the same time was one of the centerpieces of the Indonesian social protection system until that
time, was reduced 9. Further, the Government has launched several pro-poor programs recently to buffer the increased cost
of living for the poor, reallocating Rp. 17 trillion for health, education and village infrastructure programs and cash transfers
to 19.2 million poor and near-poor households from 2005 to 2006. The Government has further declared to import rice if the
domestic harvest is not sufficient, which has led to a stabilization of prices for this important commodity 10. It also has recently
launched a pilot household and community conditional cash transfer scheme targeting the extreme poor to improve human
development outcomes. In addition, the long existing Kecamatan Development Program (KDP) will be scaled up to a National
Community Empowerment Program, or Program Nasional Pemberdayaan Masyarakat (PNPM), to provide block grants to
some 2,800 rural and urban sub-districts, with full national coverage of 5,360 sub-districts by 2009 11.
These efforts to provide increased support to the poor were complemented by the Government’s decision in October 2004 to
implement Law No. 40 concerning a National Social Security System (Undang-Undang Sistem Jaminan Sosial Nasional, or
SJSN), which foresees reforming the existing social protection system towards universal coverage. A task force appointed
through the Vice President’s Office, first created by the Presidential Decree (Kepres) No. 22/2002, drafted the law. The key
feature of the new law is that it mandates the creation of several social security schemes for citizens: old-age pension, old-
age savings, national health insurance, work injury insurance, and death benefits for survivors of deceased workers. Law
No. 40/2004 is an important milestone because it stipulates that the existing social security programs in Indonesia will be
expanded to cover all Indonesian citizens, including those who are working in the informal sector, the unemployed and the
poor.
public interventions such as social insurances, social assistance and social safety nets, and provides room for private and
community-based initiatives. Social Security is a component of social protection and comprises the key social insurances,
including health insurance, pension insurance, unemployment benefits, work injuries, and disability grants (mainly contribution
financed). Basic Social Protection or Social Assistance comprises non-contributory measures targeted at vulnerable
groups/individuals, such as poor, elderly or disabled persons (mainly tax-financed).
A social protection system that enables the population to cope with risks and vulnerabilities is especially important in Indonesia
due to the high movement in and out of poverty among households and the large number of households clustered around
the poverty line. These specific characteristics of Indonesia justify the decision of policy-makers to develop a well-designed
social protection system to safeguard citizens from risks and vulnerabilities, such as income loss, illness and out-of-pocket
health expenditures, a poor harvest, or the rise in commodity prices.
Thus, given the well-established links between human capital and economic well-being, the Government of Indonesia
recognizes social protection as a core part of development policies to combat existing poverty, prevent people from falling
into poverty, and increase equity.
Improving social protection coverage has the potential to contribute to poverty reduction in many ways: social protection
protects individuals from the adverse financial consequences linked to the occurrence of the above-mentioned risks. It thus
prevents people from becoming poor or falling deeper into poverty (protection function). In addition, social protection entails
an income-generating effect and reduces prevailing poverty by enabling productive investments in human capital (investment
function). For example, social health protection allows individuals to invest in health by facilitating access to health services.
Another example is cash transfers, a form of basic social protection, which helps to break the vicious intergenerational cycle
of poverty by improving the health and nutritional status, as well as education level, of beneficiaries, thereby increasing the
likelihood of better future employment opportunities for children. By reducing existential fears of the members of society,
9
Ibid.
10
http://news.tradingcharts.com/futures/4/8/91679784.html (accessed on 19 April 2007).
11
World Bank (2007d).
2
1 • Introduction
social protection policy encourages individuals to take risks that they would not otherwise be willing to take, such as investing
in new business opportunities. Further important benefits of social protection accrue from a reduction in the use of inefficient
risk-coping methods such as selling productive assets or withdrawing children from school. Finally, mutual protection against
social risks strengthens the bonds of cooperation and reciprocity, thereby enhancing social stability and social cohesion.
The current form of publicly provided social protection in Indonesia is based on two kinds of public intervention, namely (a)
contributory social insurance or mandatory savings for civil servants (covered by PT Taspen and PT Askes), armed forces
(covered by PT Asabri) and formal sector employees (covered by JAMSOSTEK), and (b) targeted efforts to provide economic
or social support for poor or vulnerable groups of society.
The implementation of Law No. 40/2004 on the National Social Security System is presently bringing about a comprehensive
reform of the existing system. Law No. 40/2004 provides the legal basis for the above-stated objective of social protection
coverage for all. A task force, headed by the Coordinating Ministry of People’s Welfare, is working on the introduction of new
regulations as well as the mapping and harmonization of existing regulations relevant in the context of Law No. 40/2004. It
has made considerable achievements since its appointment: finalizing mapping and extensive review of existing laws, as
well as a comprehensive design of a harmonized legal and regulatory framework. Based on the latter, there is now a
roadmap for implementation in place, and technical sub-groups are working on new laws - most importantly, the ‘carrier law’,
and government regulations. The whole process has been aligned with Law No. 10/2004 that stipulates how laws are to be
drafted. It has recently been formally endorsed by the president and now appears set to be completed by the legal deadline
in October 2009.
As important initial reform steps have been taken, major challenges lie ahead:
• Low coverage: Coverage by the existing contributory social insurance system is still low 12: only about 16 million workers
out of a total labor force of about 101 million people (including job seekers) are covered by TASPEN, ASABRI and
JAMSOSTEK schemes. This means that only 16 percent of workers are currently covered by formal social security
schemes. Health insurance coverage by public schemes is slightly higher with approximately 18 million insured. Extending
coverage in terms of both organizational as well as sustainable financing options will be based on a comprehensive
strategy that is currently prepared by the ministries charged with building the new social security system (Jaminan Sosial
Nasional, JAMSOSNAS). This requires detailed conceptual work and decisions on, for example, benefit package(s),
contribution rates, and necessary institutional reforms. It also requires a decision on appropriate social protection
instruments with regard to specific risks, such as old age and unemployment.
• Fragmentation: One objective of the reform is to build a coherent system of social protection and reduce the high
degree of fragmentation that characterizes the current system. Benefits of streamlining include improved efficiency of
allocation of administrative resources. Coordination between the various programs that are targeting the poor needs to
be enhanced. Cooperation and communication between the concerned government ministries and the service and
insurance providers will be facilitated by a new statutory body, the National Social Security Council, soon to be installed
by the President.
• How to cover the informal sector? The present system excludes the non-poor in the informal sector from any
arrangement; they are not covered by the contributory social insurance and not eligible for any form of social assistance.
Considering the fact that the informal sector in Indonesia comprises a substantial part of the working population 13 with a
high proportion of them being considered non-poor, developing an effective strategy on behalf of this population group is
crucial for realizing the goal of universal coverage.
12
13
According to the National Labor Force Survey (2007), the informal sector in Indonesia comprises a substantial and growing part of the working population,
with currently 69 percent of the total labor work force.
3
With regard to the above mentioned problems the following strategic issues are important for further planning the reform
agenda:
• Analyzing institutional options for administering a comprehensive system of social protection
• Organizing the provision of social assistance to the poor, including the development of appropriate financing options
• Identifying and deciding on options for extending coverage to the informal sector
• Deciding which underlying social protection mechanism should be applied in some areas, such as unemployment
compensation
• Acknowledging that reforming social protection and achieving universal coverage is a lengthy process and requires
long-term commitment
Within the framework of development cooperation between BAPPENAS, the Indonesian National Planning Authority, and
the GTZ Economic Reform Project supported by the German Government, an expert mission from Germany took place in
2005 to discuss possible collaboration in three areas: promotion of Small & Medium-sized Enterprises (SMEs), the
establishment of an Economic Advisory Council and options for reforming Social Security/Protection in Indonesia. Upon
request by BAPPENAS, an initial mission on Social Protection was conducted in December 2006 with the objective of
identifying key areas for the strategic planning of the overall reform process, which is still in need of fundamental policy
decisions, as well as strategic options within these areas, and outlining potential roles for BAPPENAS within the overall
reform process. Following this mission, BAPPENAS expressed a clear demand for further policy advice. In a joint GTZ-
BAPPENAS workshop in February 2007, it was decided to put together a report defining various options for Social Protection
Reform in Indonesia. It was requested that these options be based on the Social State Model.
This report has been compiled to deliver strategic policy advice to BAPPENAS in order to support it in its planning function
in the reform process. By providing international expertise, GTZ aims at assisting BAPPENAS in developing long-term
strategies for a comprehensive system of social protection. The objective is to present visionary, but feasible, options for the
future that go beyond the current reforms and legislation, while taking the important steps already taken into consideration.
The report is divided into eight sections:
Following the introduction in chapter one, chapter two provides a brief overview of economic growth and poverty reduction
indicators is provided as well as an outline of the current decentralization process, which serves as an important backdrop to
the implementation of social protection in Indonesia.
Chapter three presents models of social protection and guiding principles of the social state model. It further discusses
whether the social state model is the right model for Indonesia.
In chapter four, the focus lies on reforming contributory social security. First, the conceptual framework of social insurance
and basic insurance principles, including organization and financing, are presented. This is followed by a review of policy
options for reforming the administration of social security. As the question of how to organize the administration of social
security is difficult to disentangle from the question of how to organize social assistance, both will be dealt with in this
chapter. Also, the question of how to include the provision of social security for the informal sector will be discussed here.
This general part of chapter four concludes with recommendations for reforming the administration in the medium and long
term.
The remainder of this chapter deals with social health insurance, the current situation and reform options for pension schemes.
Due to the fact that social health insurance has been a priority of the social protection reform process, the character of this
contribution varies from the remaining report. The focus lies on a description of the status quo of the national health care
system prior to reform and a description of the National Social Security System (NSSS) Law, followed by an analysis of the
ongoing legislative and regulatory process. Policy options for social health insurance are discussed within the boundaries
already set by the NSSS Law.
4
1 • Introduction
Unemployment compensation is discussed in chapter five. After providing an overview of instruments for unemployment
compensation and their incentives, the status quo in Indonesia is described. The chapter concludes with recommendations
for policy reform.
As the administration of social assistance is dealt with in chapter four, chapter six provides and overview of the conceptual
framework for social assistance and describes the current situation. This is followed by a focus on the decentralization and
coordination aspects of social assistance.
Extending social security coverage to the informal sector is dealt with in chapter seven. Following a discussion of the
different definitions of the informal sector, facts and figures on the informal sector in Indonesia are presented. The problems
associated with extending social security to the informal sector are discussed, and two approaches that can be used for their
inclusion are then examined in the Indonesian context.
After providing an overview of the health service providers in Indonesia, the question as to whether the Government
contributions for health services for the poor in community health facilities (PUSKESMAS) should be paid via the regular
health insurance carrier or directly to the PUSKESMAS using block grants is reviewed separately in chapter eight.
A summary of conclusions and open issues is presented in chapter nine.
5
6
2. Background Information
2.1. Economic Growth and Poverty Reduction
With sustained modest growth rates, Indonesia seems to have recovered from the Asian financial crisis of 1997/1998.
According to estimates from the ADB, the GDP growth rate is expected to rise from 5.5 percent in 2006 to rates of 6.0 percent
and 6.3 percent for 2007 and 2008, respectively 14. National public expenditure at all levels of government rose to 20 percent
of GDP in 2006 from an average of 19 percent in previous years. Finally, after a high of over 100 percent of GDP following the
crisis, government external debt is now projected to fall from 41 percent of GDP in 2006 to 37 percent of GDP in 2007 15.
While the reduction of fuel subsidies in October 2005 freed up a fiscal space of some 15 billion USD, it also prompted the
government to enact a massive fuel price hike, with the result that inflation in January 2006 rose to above 17 percent. It was
only towards the end of the year that inflation fell back to a 7 percent mark. There are better prospects for 2007 and 2008
though, with a projected inflation rate of a little over 6 percent16.
The overall economic growth in Indonesia is mainly based on private consumption and a strong export performance, which
surged by 9.2 percent due to high global prices for commodities, such as natural gas and crude oil. The contribution of
investment to economic growth is still small. Within this context, the newly enacted investment law is particularly promising
as it provides the legal framework for better investment regulations.
However, the macroeconomic stabilization since 2001 should not mask the fact that in the same period, estimated
unemployment rates grew from below 9 percent to 10.3 percent in 2006. Job creation and poverty reduction continue to pose
major challenges. About 42 percent of households are clustered around the national income poverty line and have to
live on only 1-2 USD a day which renders them particularly vulnerable at various levels: at the micro- and household level
(e.g. illness in the family), at the meso- or local level (e.g. natural disasters) and at the macro- or national level (e.g. rise in
fuel prices). In 2004, Indonesia ranked 69 worldwide according to the Gini index with a distribution of family income of 34.8.
14
ADB (2006b).
15
Ibid.
16
ADB (2006b) projects 6.2 percent and 6.1 percent for 2007 and 2008, respectively.
7
Table 1: Number and Percentage of Population below the Poverty Line
Area / Year Poverty Line (Rp / Capita / Month) Number of Percentage of

Population below Population below
the Poverty Line (mio.) the Poverty Line
Food Non Food Total

Urban
March 2006 126,163 48,127 174,290 14.49 13.47
March 2007 132,258 55,683 187,942 13.56 12.52
Rural
March 2006 102,907 27,677 130,584 24.81 21.81
March 2007 116,265 30,572 146,837 23.61 20.27
Urban + Rural
March 2006 114,125 37,872 151,997 39.30 17.75
March 2007 123,992 42,704 166,697 37.17 16.58
Source: BPS 2006/2007
As about 37 million people live below the poverty line (see table 1), it becomes clear that the higher levels of growth
experienced in recent years are not adequate. At present, the national poverty incidence is estimated at 16.6 percent
rising from 15.9 percent in 2005. This is, among others, caused by a continued lack of access to basic services, such
as health, education, infrastructure, housing, clean water and sanitation. The government is presently trying to counter
the lack of access to education by allocating 17.2 percent of total public expenditure for education alone. This is the highest
8
2 • Background Information
sectorial spending and a rise from only 2.4 percent of GDP in 2001 to 3.8 percent of GDP in 2006. However, despite this
positive development in terms of public spending, according to the Public Expenditure Review 2007, the government's
current spending is not efficient enough in addressing the country's pivotal challenges to development. About a third of total
expenditures still go to subsidies and administration, with 11.9 percent of total expenditures for core government administration
(excluding salaries for teachers, doctors and nurses). Meanwhile, spending on infrastructure development constitutes only
10.2 percent of total expenditures, which is relatively low by international standards, whereas less than 1 percent of the GDP
or a mere 4.2 percent of total expenditures is spent on health.
The economic data indicate a steady growth of GDP as well as relative fiscal and macroeconomic consolidation. Still, the
prime challenge for the Indonesian government lies in coupling this growth with the creation of sufficient jobs in order to
reduce unemployment, underemployment and the vulnerability of the poor and near-poor.
2.2. Decentralization
Since the end of President Soeharto’s regime in 1998, Indonesia has undertaken many steps towards a more decentralized
governmental system 17. The main objectives of the government's decentralization policies can be summarized as follows: maintaining
national territorial integrity, improving public service delivery and creating a more accountable regional government 18. The motivation
for decentralization is based on the belief that local authorities are more familiar with the needs of the local population and
can adjust the provision of public services accordingly. In the post-Soeharto era, many responsibilities with respect to financing
and delivering public services and a large extent of autonomy have been passed to the districts and provinces.
Furthermore, as more and more regions are claiming greater economic and political autonomy, the decentralization policies
of the Indonesian government are considered as an appropriate instrument for achieving this goal. People in resource-rich
regions, such as East Kalimantan, Papua or Aceh, in particular, have long held aversions towards centralized power 19.
The main pillars of the legal framework for the decentralization process are Law No. 32/2004 and Law No. 25/1999. Whereas
Law No. 32/2004 regulates regional administration and the distribution of political competencies, Law No. 25/1999 is concerned
with the distribution of fiscal competencies. Both have gradually been implemented since 2001 and profoundly changed the
country. A further major step towards decentralization is the recent decision of the Constitutional Court to remove articles in
Law No. 32/2004 which stipulated that only candidates endorsed by political parties can run for local elections 20.
At present, Indonesia consists of 33 provinces (Propinsi) and 483 districts (Kabupaten) or cities (Kota) 21. Their sub-units are
the sub-districts (Kecamatan), which again consist of villages (Desa or Kelurahan), the smallest units in the political system
of Indonesia. Each of the provinces is headed by a governor and has a legislative body, the Regional People’s Representatives
Assembly (Dewan Perwakilan Rakyat Daerah). The governor is the head of an autonomous region and the same time the
17
Whereas Indonesia during the period of liberal democracy in the 1950s was characterized by regions provided with a wide degree of power, the Guided
Democracy under President Soekarno already showed clear tendencies toward centralization. The centralism was strengthened during the New Order
Regime under President Soeharto. The central government exercised control over the different levels of the political system in almost all aspects of political
life through a highly centralized bureaucracy and military strength (The Asia Foundation 2003:4-5). Law No. 5/1974 on the basic principles of Administration
in the Regions provided the legal framework for decentralization (UNESCAP 2003:9) and implied a delegation of certain tasks from the central government
to lower levels of the political system (Suwando 2002:1). However, decentralization in terms of a ”transfer of authority and responsibility for public functions
from the central government to subordinate or quasi-independent government organizations (...)” (World Bank Decentralization Thematic Team) did not
take place (AF 2003:5). Following the crisis of 1997 and Soeharto's fall, Indonesia underwent a profound reshaping of its political structure. Today,
democratization and decentralization are the priorities on Indonesia’s reform agenda.
18
The Asia Foundation (2003:4).
19
The Asia Foundation (2003:6).
20
However, the ruling of the court will have to go through a long process at the Ministry of Home Affairs and the House of Representatives before taking
effect and it may be months before it is fully implemented.
21
There are four provinces that have a greater autonomy than the others: Nanggroe Aceh Darussalam, Yogyakarta Special Region, Papua, and Jakarta
Special Capital Region. Aceh, for example, has the right to introduce the Shariah as its legal basis (Wagener, GTZ 2005: 3).
9
representative of the central government, subordinated under the president through the Ministry of Home Affairs. The provinces
are responsible for ”macro-planning, human resource development and research, the management of regional ports,
environmental protection, trade and tourism promotion, pest control/quarantine and spatial planning” 22. Tasks that require
the coordination of different districts also belong to the responsibilities of a province. The districts likewise have their local
parliaments and own governments, which are headed by regents or mayors. They are no longer subordinated under the
President, as was the case during the New Order regime 23.
Regional governments ”shall cover the authorities in all fields of governance, except authorities in the fields of international
policies, defense and security, judicature, monetary and fiscal, religion and authorities in other fields 24” (Law No. 22/1999,
Article 7). Their fields of governance include ”public works, health, education and culture, agriculture, communication, industry
and trade, capital investment, environment, land, co-operative and manpower affairs” (Law No. 22/1999, Article 11). These
activities should be supervised by the central government, which has the power of nullifying regional decisions (Law No. 22/
1999, Article 112, 114). In order to manage these new responsibilities, approximately 2.2 million civil servants and 16,000
service facilities were transferred from the central government to the regions 25.
To avoid ”unfunded mandates” the system of local revenues was changed in accordance with Law No. 25/1999. There are
different sources of income: local taxes, charges and fees from local enterprises and borrowings (own source income).
Additionally, the provinces and districts receive money from different equalization funds (Dana Perimbangan), which play an
important role for the regional budgets. A central issue of the fiscal decentralization is the general purpose grant (DAU –
Dana Alokasi Umum). Local governments can freely decide on spending and managing the grants.
The specific purpose grant (DAK – Dana Alokasi Khusus) is supposed to ensure a minimum service standard 26. Further
funds for equalization are the regional share of the property tax (PBB) and property transfer tax (BPHTB) and the regional
share of natural resources revenue (bagi hasil). Apart from these funds, a certain amount from the national budget (APBN)
is destined for the provinces and regions 27.
Various problems emerged during the implementation period. According to many experts, one problem is the lack of clarity
of the laws concerning decentralization. A great number of laws were prepared without a blue print which led to a lot of
”confusion about the law itself” and rendered a ”valid reference” impossible 28. Furthermore, there seemed to be rivalries
between the Ministry of Home Affairs (MOHA) and the Ministry of Finance (MOF) concerning the authority of the decentralization
process. The National Human Development Report 2004 diagnosed an ”unclear distribution of functions between the central
government and the regions” 29.
Another problem was that both provinces and districts ”had to absorb more government workers than they could immediately
make use of”. As a result, high routine expenditures and little service delivery occurred. To have a real influence on these
expenditures and political outcomes, the local population did not seem to be sufficiently organized and experienced. Some
argued that after such a long period of ”being perceived as only beneficiaries”, it would take time for them to assume a more
active role in local policies 30. Further problems regarding finances were the own-source revenues, which were hardly high
enough to finance the district's tasks, since all major taxes were kept by the central government 32.
22
GTZ SfDM (2003).
23
During the New Order Regime, all levels of government were in one chain of command that reached from the president down to the village level (GTZ
SfDM 2003).
24
Other fields are defined are as ìthe policies on national planning and macro-national development control, financial balance fund, state administration and
state economic institutional systems, human resources development, natural resources utilization as well as strategic high technology, conservation and
national standardizationî (Law 22/1999).
25
UNDP, BAPPENAS (2004:19).
26
The local governments create the programs or projects and the central government decides which activities will be financed (Brodjonegoro 2004:10)
27
22.5 percent of domestic revenues should be transferred to the local level and 2.5 percent to the provincial level (GTZ SfDM 2003).
28
Brodjonegoro (2004:2).
29
UNDP, BAPPENAS (2004: 19).
30
UNESCAP (2003: 17).
31
According to SMERU, the municipal and district governments are responsible for more than 700 tasks (SMERU 2001:8).
10
2 • Background Information
This led to the introduction of new local taxes, which in turn

had a negative influence on the investment climate 32. UNDP
argued that the fiscal structure favors the resource-
rich regions and could lead to stronger regional
disparities 33. Apart from these disparities, there were
worries about the dominance of natural resources as
the main income sources for local governments, which
could lead to an over-exploitation and negative effects
for the environment. Another problem was the
unintended incentives for local elites to form new local
governments, even if this was not necessary and
proved inefficient 34.
In order to solve the above-mentioned problems and

improve the decentralization process, many laws and
regulations were endorsed or amended after 2001. The
laws 32/2004 and 33/2004 replaced the older laws from
1999 and provide clearer and more detailed regulations
on the implementation of regional autonomy. In
addition, direct elections of regents, mayors and
governors were introduced in June 2005 35.
The government has also begun to draft a ”Grand

Strategy for Decentralization” addressing the following
7 fields: ”(i) the functions of government, (ii) the optimal
structure of government, (iii) civil service reform, (iv)
financial governance/ administration, (v) representation
of all components of society, (vi) modes of public service
delivery, and (vii) ensuring that the control and
supervision system locks in accountability and reform” 36.
In order to solve the problem of lack of resources, the government initiated a National Action Plan for Fiscal Decentralization.
Law No. 32/2004 re-introduced preventive legal control for regulations on local budgets (APBD) and aimed to reduce the
impact of irregular local taxes on private investments. As a consequence, many local government regulations concerning
local taxes and charges, which were considered harmful to the business climate, were cancelled by the central government.
The challenge for the Indonesian government lies in finding an effective way to control regions and ensure national objectives
without excessively limiting local autonomy. The rules and procedures necessary for this process are provided by the
Government Regulation No. 79/2005, which implies a stricter treatment of faulty regional regulations concerning the local
economy, but also concerning the exploitation of natural resources, for a sustainable development.
In order to ensure that new regional entities are ”in line with the principles of efficient, effective and accountable regional and
local governance”, stricter rules regarding their formation are required. To this end, the central government has to find the
right balance between too many regional and local entities, causing inefficiencies, and too few entities, leading to a less
recipient government for local interests.
32
A survey of the Regional Autonomy Watch (KPPOD) indicates that the main victim of illegal and excessive charges is manufacturing (Brodjonegoro
2004:2).
33
UNDP, BAPPENAS (2004: 20).
34
The annual growth rate of new districts and municipalities amounts to more than 10 percent (Brodjonegoro 2004: 3).
35
CGI (2006).
36
CGI (2006).
11
In 2004, various guidelines concerning Minimum Service Standards were introduced. To ensure certain standards, the
Minister of Home Affairs and the Minister of National Development Planning at BAPPENAS issued the National Framework
for Capacity Building to Support Decentralization. This program is aimed at strengthening the capacity of institutions at all
levels 37.
Figure 1: Administrative Divisions in Indonesia
Republic of Indonesia
Head : President
Province (Propinsi), 33,

Head : Governor
Districts:
Regency (Kabupaten), 349,
and City (Kota), 91
Head : Regent (Bupati)
or Mayor (Walikota)
Sub-district (Kecamatan)
Head : Civil Servant (Camat)
Village (Desa and Kelurahan)

Head : Head of Desa, Lurah
The challenges brought about by the decentralization process are a major factor for the quality of services provided by the
government and are thus an important aspect of Social Protection. Effective decentralization with clear division of roles and
responsibilities of the central and local government is essential in designing and implementing protective measures that
support individuals, households and communities to better manage risks and economic shocks, and to provide support to
the critically vulnerable. Public interventions, such as social insurance, social assistance and social safety nets, including
basic service delivery in health and education rely heavily on well-functioning mechanisms of government policy, planning
and budgeting systems. Institutional accountability and civil society participation are also key elements of decentralization to
ensure that the need of all citizens, especially the vulnerable, are accounted for. Many of the issues discussed in this paper
require high levels of participation and ownership by society as a whole. Hence, the decentralization process constitutes an
important factor for sustainable social protection.
37
Ibid.
12
3. The Social State Model:
a Model for Indonesia?
- There is strong political commitment towards universal social protection and poverty reduction, as expressed in Law
No. 40/2004, while the institutional framework for implementation is still incomplete.
- Taking institutional and procedural deficits into account, it is not easy to decide whether Indonesia should focus more
on the Welfare State or the Social State approach of Social Protection.
- The high level of poverty and the size of the informal sector, which seems to imply the principle of universality, as it
would facilitate the integration of different spheres of society under the umbrella of uniform protection measures,
further complicate this. On the other hand, it is doubtful whether there is sufficient administrative capacity to implement
such universal programs, and whether it would be financially sustainable at the present stage of development in
Indonesia.
- It would be advisable for Indonesia to integrate elements from both, the Bismarck and the Beveridge Model, into
some specific “hybrid” form better suited for the country’s present needs.
- The future face of the Indonesian Social State Model for social protection could therefore consist of a combination of
public social insurance and social assistance schemes together with provisions of the private sector in three overlapping
structures.
13
3.1. Models of Social Protection

In 1981, social insurance was one hundred years old. There were of course, already provisions for social insurance in
Europe prior to this, for example introduced by trade unions. Worker’s funds were based on the insurance principle that
members had to pay contributions, and only contributors were eligible for benefits after the materialization of the risks
insured. These early social insurances were, however, not strictly actuarially-based, but more rooted in the solidarity principle.
Hence, their outreach and viability remained rather limited. Contributors were not strictly related, on a basis of equivalence,
to the dimensions of the insured risk, and benefits were not strictly related to the contributions paid. These elements of
“organized solidarity” were meant to mitigate the equivalence aspects of strict application of insurance principles and had,
intentionally, redistributive effects. The first insurance schemes of this kind only concerned contingencies of occupational
accidents and diseases, invalidity, old age and death, and coverage was mostly limited to workers in the steel and mining
sector. Triggered by industrialization and urbanization, more and more citizens became dependent on formal employment,
changing family profiles from three to two generation households as well as increasing female employment.
As a consequence, new approaches to social insurance were necessary. The first prominent models of social protection
policy in Europe were developed by Bismarck and Beveridge in Germany and Britain, respectively. These early models were
primarily concerned with social insurance schemes. Figure 2 highlights the main principles of these models.
Table 2: Main Principles of the Bismarck and Beveridge Model
Bismarck Model Beveridge Model
Objectives Status-oriented Subsistence-oriented

Scope Categorial (workers) Universal (residents)
Funding Income-related contributions Flat-rate taxes
Benefits Income-related National minimum

Administration Private law Public authority
3.1.1. The Bismarck Model

The first country to adopt statutory social insurance was Imperial Germany. In 1881, Chancellor Otto von Bismarck was the
initiator of legislation that used the design of the many existing mutual help organizations and established compulsory social
insurance. The legislation started with the Health Insurance Act (1883), followed in 1884 by the Work Accidents Insurance
Act with contributory compulsory membership for employers. In 1889 the Pension Insurance Act passed legislation, mainly
centered on social security in case of invalidity, including old age but no survivor pensions. Bismarck accepted statutory
social insurance not primarily for humanitarian reasons, but mainly to curb socialist and communist political agitation. Indeed,
his social policies helped calm down social unrest and contributed substantially to the pacification of the labor movement in
Germany. Therefore, while personal coverage was mainly targeted at blue collar workers, it also included white collar workers
with smaller income. Thus, although general income replacement rates remained on a low level of 30 to 40 percent of
wages, only slightly above provisions by the poor law, the first nearly complete legislation on social insurance worldwide was
implemented. Unemployment insurance was not included, as it was not deemed financially viable. At that time, a modest
unemployment compensation only was provided by trade unions for their members. After the First World War (1914-1918)
left many returning workers jobless, the first means-tested benefits from the state were introduced, and in 1927 the legislation
for Unemployment Social Insurance followed, integrating vocational guidance and job placement services with the payment
of a modest unemployment compensation.
From an organizational point of view, Bismarck’s social insurance was not unlike the earlier non-statutory worker insurances.
They were industrial schemes, though more employer- than employee-based. Elements of organized solidarity mitigated
14
3 • The Social State Model : a Model for Indonesia ?
equivalence aspects, and while fundamental decisions were made by state legislation, the principle of self-administration of
the social insurance bodies was taken over from the mutual help organizations and applied in an altered form to the state-
organized system. Although Bismarck’s original intentions were to strictly limit the financial responsibilities of the state, the
old age and disability scheme already included a modest subsidy, and it soon became clear that the state would have to bear
the responsibility if social insurances ran into trouble. Still, the insurance principle was predominant: the schemes were
financed by contributions, and only members, i.e. employees, were eligible for benefits. Benefits could be paid only for
specified risks, as mentioned in the laws. While originally insurance benefits were more directed towards moderating the
dependency of the elderly, incapable and unemployed workers on the poor law and on reducing poor law expenditures, the
insurances’ objective soon was to guarantee the economic and social status of the insured in the event of materialization of
these risks. Benefits were, therefore, income-related.
This process, although interrupted by the two World Wars, went hand in hand with an impressive personal and financial expansion
of the original social insurances towards a more general and comprehensive system of social protection in Germany. The
leading idea was the formation of a society where the economic status should mainly be decided by individual economic effort.
Within social insurance, the corresponding normative basis was “contributive justice” which implied a decisive role of the
principle of actuarial equivalence according to individual income and contribution biography. The combination of a comprehensive
system of social protection with a full employment policy, responsible collective bargaining by entrepreneurs and trade unions
acting together as “social partners” later became the fundamentals of Germany as a modern “Social State” 38.
3.1.2. The Beveridge Model

A distinctively different approach to social insurance was formulated in Great Britain in 1942 during World War II with the so-
called Beveridge Report 39. William Beveridge was head of a state commission to reform Britain’s deficient social security
system. Heavily influenced by a feeling of “one society” which arose during the war, the ambitious intentions of this report
were not only to end poverty but as well to perform a total reconstruction of the diversified and fragmented former system in
following the principles of unification, universality, adequacy and comprehensiveness.
Unification meant to unify the many existing different branches of social insurance into one centralized institution under the
Ministry of Social Security. All contributions should be paid into one single fund and all benefits and other insurance payments
should be paid from this fund. The central agency should provide general coverage against all social risks (principle of
universality) and not only for workers, as is the case in the Bismarck model, but for all members of society.
Adequacy meant that the level of all social benefits should guarantee sufficient provisions against all risks, but only at a
minimum level, in order to stimulate incentives for supplementary private efforts. Because of his dislike of a means test,
Beveridge intended to replace discretionary benefits with contractual entitlements, but since he was also convinced that an
adequate system could hardly operate without a framework of disciplinary constraints, he wanted to ensure that public
assistance was supplemented by private provisions and insurance schemes as well.
Comprehensiveness meant that the system of social insurance should be operated in a transparent and comprehensive
way for all British residents to cover all their needs and guarantee freedom from want on an accepted social minimum level
and for unlimited duration. Precisely defined non-means-tested flat-rate insurance benefits should correspond with one
uniform weekly-paid flat-rate contribution. Paradoxically, although Beveridge was a strong supporter of the insurance principle,
this way of financing gave his reform the characteristics of a general per-capita tax.
The importance of the Beveridge Report lies in the comprehensiveness of the social security system it proposed, and in the
elaboration of the principles underlying this system. The proposals were subsequently enacted in the National Insurance Act
of 1946. In accordance with the report, this act created a public universal social insurance system, providing coverage
38
Döring (1997: 45).
39
Beveridge (1942).
15
against the loss or interruption of earnings due to unemployment, sickness or old age. However, although still in accordance
with the proposals of the report, the right to unemployment benefits was restricted, and a main departure was that benefits
were not set at the subsistence level. The British government contended that the costs of subsistence varied widely from one
claimant to another depending on the amount of housing rent claimants had to pay. This meant that for claimants who had to
pay high rents, the benefits could reach the subsistence level only if excessive additions were paid. Such a varying benefit
rate would infringe on the contributory principle. Beveridge’s proposals regarding the duration of benefits were not accepted.
The government argued that an unlimited duration of benefit entitlement would give rise to abuse.
Furthermore, the British system, though universal in its objectives, did not follow the principle of universality expressed in the
Beveridge Report as not all residents were insured against all social risks. As stated above, coverage in the event of
unemployment was not extended to all residents. Neither did the system of means-tested social assistance benefits vanish
completely in Great Britain. On the contrary: as social insurance benefits were flat-rated, they had to be set at a low level in
order not to exceed the lowest wage levels. In fact, insurance benefits were often too low to support a family. As a result,
although the idea of a national insurance was very attractive, after the reform was elaborated and measures put into practice,
not much was left of Beveridge’s original intentions. Nevertheless, his vision that a comprehensive system of social protection
against existential risks should be provided and financed by the state laid the conceptual foundation for the modern Welfare
State and can be regarded as a counter model to the Bismarckian type of self-administered contributory social insurance 40.
While full personal responsibility and full state provision are equally abstract extreme positions, a sound social protection
policy should strive to achieve a division of responsibilities between the individual and society. Consequently, many countries
throughout the world have adopted elements related to both of these two basic approaches. However, the following comparative
analysis will show that there are substantial differences between the two concepts with regard to the application and combination
of certain measures.
3.1.3. Comparison of Bismarck and Beveridge

The main conceptual differences between the Beveridge and the Bismarck models of social insurance can be seen, first, in
that Beveridge wanted to introduce a strongly centralized basic insurance scheme in the hands of the state aimed at the
prevention of poverty and apt to provide security against all kinds of social risks and for all members of the society. In
contrast, the intentions of Bismarck were not as ambitious and should be measured against the historical background of the
miserable living and working conditions during the first stages of industrialization which in Germany started later than in
many other European countries, such as in Britain. His intentions were to provide insurance coverage for all those who
belonged to the working classes and were able to pay the necessary premiums for statutory social insurances against well-
defined risks of age and disablement, work accidents, illness and - many years later - unemployment. Thus, this historically
first complete system of mandatory social insurance was not really aimed at the poor but at the rising middle classes of
industrial blue and white collar employees with the intended political side effect of overcoming Marxist class struggle theories
by integrating the rebellious industrial proletariate.
In general, one can say that the basic intentions of Beveridge and Bismarck were different in their perception of a social
protection policy and of the role of the state. For Beveridge, it was the task of the state to supply basic provisions to prevent
poverty and to secure against social risks, while for Bismarck the state only had to provide the institutional framework in
order to enable citizens to take care of themselves. Thus, in the German model, state and non-state activity are put in a more
additive position producing living standard security and combining individual efforts for social risk provision with state policies
in those cases where responsibility for poverty prevention cannot be delegated. The task of guaranteeing minimum security
is left to means-tested assistance schemes outside social insurance, while in the Beveridge conception guaranteeing minimum
security is the center task of state policy and living standard security is delegated entirely to non-state activity. In other words,
“prevention of poverty versus income maintenance” is the main and dividing characteristic between the two models.
40
One can even go as far to say that the main principles of the Beveridge concept in most perfect form have been institutionalized in the former socialist
countries, especially in Middle and Eastern Europe, but also throughout the world.
41
Koslowski (1997: 351).
16
3.1.4. The Welfare State Model versus the Social State Model
While one cannot say that the Beveridge Report of 1942 set the starting point for the development of modern Welfare State
regimes, it is certainly correct that the basic principles have remained the same. Furthermore, while in the English speaking
world the term Welfare State is normally used to describe national regimes for social protection 41, it would be more
appropriate to differentiate the German term “Sozialstaat” or, respectively, Social State from this general meaning.
Following the well-known definition of Asa Briggs, the Welfare State is characterized as a state, in which organized power
is deliberately used (through politics and administration) in an effort to modify the play of market forces in at least three
directions: first, by guaranteeing individuals and families a minimum income irrespective of the market value of their work or
their property; second, by narrowing the extent of insecurity by enabling individuals and families to meet certain “social
contingencies” (for example, sickness, old age and unemployment) which otherwise lead to individual and family crisis; and
third, by ensuring that all citizens without distinction of status or class are offered the best standards available in relation to
a certain agreed range of social services 42.
In distinctive contrast to this perception, the German social state model refers to the role of the state in the procurement of
the necessary means of subsistence (Daseinsvorsorge) for its citizens and not to achieve a general redistribution of income
or to provide for comprehensive social welfare in the sense of “decommodification” 43, but to set up and secure a system of
mandatory social insurances against certain definable social risks based on actuarial equivalence between benefits and
income-related contributions 44. Thus, contrary to minimizing dependency on the labor market and overall socialization of
risks like in the welfare model, the social state approach accentuates the necessity for the individual to activate their own
capabilities for risk provision in the first place and that falling back on the help of others is only a last resort when the potential
of self-help is exhausted. The resulting differentiation according to individual earned income biography and contributions
paid causes a strong inequality of monetary claims but this is accepted as “contributive justice”, meaning that inequality of
economic status is justified as long as it is defined by economic effort and should strengthen incentives. However, this
orientation includes a certain minimum income guarantee if claims from social insurance would be too low to meet the needs
42
Briggs (1961: 228).
43
Decommodification is the process of viewing social utilities as an entitlement rather than as a commodity that must be paid for. In this case, it refers to
activities and efforts by the welfare state to reduce citizen’s reliance to exploit their labor on the market in order to secure their protection against social
contingencies like (for example) unemployment and sickness (Esping-Andersen 1990).
44
Koslowski (1997: 351).
17
of the concerned household. The recipient then has to fall back on income support from means-tested social assistance,
which is financed from taxes. In the social state concept, redistribution of income is not an objective by itself but a measure
of solidarity in times of need and for those who cannot sufficiently help themselves.
3.2. Guiding Principles of the Social State Model
Combining individual effort with solidarity, the normative basis for social protection policy in a social state concept
is the principle of subsidiarity, derived from the Latin “subsidium” (backing, support or protection), according to
which provisions against social risks should always be the responsibility of the most basic unit possible, on a
range from the individual through family, the voluntary sector and group solidarity up to the different levels of the
state. The state should only intervene when the more basic units are unable to support their own members. This is especially
the case for those individuals or groups who are in danger of being excluded from relations with the community and its social
or economic development. What is thus given, by those who are economically and socially better off to those who are
disadvantaged, is the social element of redistribution or of solidarity compensation.
It is also true, however, that any solidarity relationship is a complex network of mutual obligations and considerations so that
a community may demonstrate solidarity towards its members only to the extent that members demonstrate solidarity towards
their community. The principle of subsidiarity regulates these relationships between individual and collective responsibilities
as well as the interaction of collective responsibilities among themselves by following three postulates 45:
Subsidiary delegation means that the potential for self-help found among the population should be encouraged to develop
and not hampered by government interference.
Subsidiary assistance means to make provisions only where and insofar as a specific risk situation objectively overtaxes
the self-help capacity of individuals or groups. Assistance should not replace individual efforts and private provisions but
complement them.
Subsidiary reduction requires withdrawing support when it is no longer needed as the potential for self-help has recovered
through measures of an activating social policy.
From these postulates, three criteria may be derived for the conceptualization and allocation of social protection policies,
namely functional, territorial, and administrative subsidiarity 46.
Functional subsidiarity demands that the state should arrogate to itself functions of social protection only if it is capable of
handling these functions more efficiently, more effectively and, consequently, more smoothly than any private institution.
This, for example, applies whenever large groups of society are to be covered by a system of social insurance. The same
holds true with regard to guaranteeing a subsistence level of income through social assistance.
Conversely, non-governmental organizations, such as religious institutions, professional associations, cooperative societies,
local solidarity networks and other self-help organizations have more affinity towards their specific members. In these cases,
private initiatives may complement or even replace governmental risk provision and the state should confine itself to creating
the necessary framework conditions and to lending legal or financial support where it is necessary.
Territorial subsidiarity demands that decision-making and executive competences should be shared out among the state
as the central authority, the subordinate regional authorities, and other public social policy institutions on the basis of factual
or technical competence. Thus, territorial subsidiarity calls for decentralization through delegation and/or devolution and,
particularly, for the allocation of independent financial resources through a system of vertical and horizontal fiscal equalization.
The aim is to mobilize local self-help potential and to strengthen self-administration capabilities so as to improve the local
presence and outreach of social protection measures, as well as to stimulate a demand-led social policy formulation
45
Rösner (1995).
46
Rösner (1995).
18
characterized at the grass-roots level. The resulting improvement in information and consultation facilities may help not only
to enhance the transparency of the system but also to overcome discrepancies between formal entitlement for social services
and chances for implementation.
Administrative subsidiarity makes it incumbent on the state to scale down the density of formal regulation wherever this
makes sense, thus ensuring that the provision of social protection services remains in keeping with the demand. This
implies, on the one hand, a need to create a legal basis for the administration of (for example) social insurance agencies by
those immediately involved. On the other hand, it implies a willingness to permit, at the various levels and in the manifold
branches of social protection, a diversity of autonomous or at least partially autonomous agencies to develop on the basis of
self-organization.
At the level of rural communities, these could be local self-help organizations developing social services, like micro insurance
schemes, specifically designed for particular target groups (tradesmen, farmers, rural workers, women), thus focusing and
strengthening the potential of informal risk management.
On the middle level, self-administered non-governmental organizations should be encouraged to function as an important
link for the integration of formal and informal social protection measures. Finally, at the level of the central government, we
find the formal social protection agencies, like social insurances and provident funds. Next to their analytical and safeguarding
functions in those sectors of the national economy where formal employment prevails, these agencies could provide
supplementary services for non-governmental agencies on the middle level so as to support these in their consultation and
promotion activities.
Putting it all together: Freedom achieved through social justice, based on the idea of solidarity and organized under
the principle of subsidiarity - would describe fairly well the ethical substructure of the modern social state principle.
In the German tradition of a Social Market Economy, social policy and economic policy are complementary in that economic
policy also pursues general socio-political objectives, while social policy, in turn, also supports economic policy goals.
3.3. The Social State Model within the Indonesian context
3.3.1. Present Situation of Social Protection Policy

When looking at the present situation of social protection policies in Indonesia, the impressions are mixed: on the one side,
there seems to be a strong political commitment towards universal social progress and poverty reduction, as was expressed
in Law 40/2004, while on the other side the institutional framework for implementation looks fractured and is incomplete.
While Law 40 stipulates that financing of social protection should be based on personal contributions in the field of social
insurance and on means-tested tax financed elements in the field of social assistance, there are institutional and procedural
deficits in that responsibilities for financing and the necessary amount of contributions are not clearly specified. Furthermore,
the competencies for specific risks and for implementation of related programs between the administering bodies are both
not clearly defined and often contradictory. Current eligibility for social security extends mainly to employees of the military,
civil servants and to those employed in private enterprises above a certain staff size and total payroll, while the large
remainder of the population has to rely on out-of-pocket payments, family support and private insurance 47.
Notwithstanding that the informal sector thus accounts for a large majority of the total working population, programs to
extend social security coverage and for the enforcement of compliance of employers have not been a major focus of the
government. This has changed so far as Law 40 aims to provide a guarantee of social protection and welfare for all the
people of Indonesia, but, with the exception of access to health care, there are no clear provisions about how the general
integration of informal workers could be achieved. It is hard to see, however, how the social reform project could be successful
without including this important part of the population.
47
Lindenthal (2004).
19
In the field of poverty alleviation or even reduction, numerous ad hoc social safety net programs have been set under way by
the central government since the crisis of 1997, but these programs have shown only mixed results concerning targeting and
leakages and are managed by scattered institutions and not in integrated form. There is a lack of any national administrative
structure that could adequately control and coordinate implementation of social assistance measures horizontally between
public institutions involved on the macro level as well as vertically between central government and governments on regional
or local level.
The public health care system is generally under-funded and weakened by opting-out opportunities towards private insurance
so that the gap between health indicators of people living in richer and in poorer provinces as well as in urban and in rural
areas is widening 48. Although Law 40/2004 provides the foundation of a National Health Insurance Program there is still
considerable resistance from the side of stakeholders so that a consensus (for example) concerning the allocation and
balance of funds for implementation, especially for local health offices, still has to be found.
3.3.2. The Vision: Indonesia - a Social State with Welfare Elements

Taking these institutional and procedural deficits into account, it is not easy to decide whether Indonesia should turn more to
the side of the Welfare State or to the Social State approach to social protection. The high level of poverty and the size of the
informal sector make it look advisable to apply the principle of universality, as it would facilitate the integration of different
spheres of society under the umbrella of centralized and uniform protection measures. On the other hand, it is doubtful if
there is the administrative capacity to implement such universal programs and if it would be financially sustainable at the
present stage of development. Furthermore, since in the Social State Model access to the risk pooling and group solidarity
of social insurance is based on formal employment, providing a guaranteed minimum income or social security status would
probably set wrong incentives in that it would be even less attractive then to seek formal employment and join contributory
social insurance. One advantage of applying subsidiarity principles lies in the fact that entitlements and obligations between
individual and society are clearly defined.
Such considerations notwithstanding, the design of social protection policies also has to be adapted to the historic background,
the dominating religious and cultural factors, and the specific social needs of society. Finally, there is a certain societal path
dependency to respect as a limiting frame condition in the institutional design of any kind of reform. This is actually the case
in Indonesia concerning the stratification of social security schemes between different groups of society (military, civil service),
and this has also been the case in Europe where ways and means for the “invention” of social security have been dominated
by the consequences of industrialization. In this concern, the present situation in Indonesia is different in at least two dimensions:
First, it is still not quite certain, whether Indonesia will be as successful as some neighboring countries in performing the
necessary structural changes from an economy dominated by agriculture to comprehensive industrialization. Second, as a
latecomer, Indonesia will have to undergo this transition process under conditions of intensified global competition.
Opting for Bismarck and the Social State Model under these different conditions involves certain risks in that it presupposes
an ideal situation of high economic growth and increasingly formal employment. This given, the social insurance principle
would not only gradually provide more and more benefits for the workforce but, and what is probably even more important,
this system finances itself automatically as wages rise and contributions are income related. On the other hand, voting for
Beveridge and the mainly tax-financed Welfare State Model requires not only effective fiscal policy but also positive political
budget decisions which might be hard to get when times are economically or politically rough.
Taking these limitations into account, it might be advisable for Indonesia to integrate elements from both the Bismarck and
the Beveridge Model into some specific “hybrid” form better suited to the country’s present needs. Above all, in the
conceptualization of social protection two basic conditions should be considered: First, the specific national profile of the
risks to be covered; and second the state of development in which the future schemes should be applied. All in all, the future
face of the Indonesian Social State Model of social protection could consist of a composition of public social insurance and
social assistance schemes together with provisions of the private sector in three overlapping structures as shown below.
48
Scheil-Adlung (2004).
20
Figure 2: Core Elements of the Social State Model
Public
Social
Insurance
1 2
4
Private Public
Sector Social
Provisions 3 Assistance
The intersections between social insurance, assistance and the private sector should leave room for complementing or
supplementing services following the principle of functional subsidiarity. For instance in the first intersection, public social
pension or health insurance could be completed by employer-sponsored pension and health plans or related schemes from
private companies (life insurance, managed care etc.).
In the second intersection, services from social insurance could be made available for the poor via subsidized contributions
or even free of charge. This could be the case, for example, concerning primary health care and maternity services, but also
in the field of employment or occupational training programs and other measures of active labor market policy.
Intersection three might serve for third sector welfare initiatives or for the implementation of programs for poverty alleviation
financed from external donors. While public social assistance should, in general, be more concentrated on direct income
support and conditional cash transfer programs (health, education) 49, this intersection could leave room for a multitude and
variety of more or less informal self-help initiatives concentrated (mostly) on target groups at the micro level 50. The promotion
of such and related forms of social risk management on district or community level follows the principles of territorial and
administrative subsidiarity within the Social State Model and may include local micro finance and micro insurance schemes,
cooperatives, women associations as well as initiatives from private companies and donor organizations.
Finally, intersection four should serve for the integration of the informal sector. This cross sectional task requires a tripartite
cooperation from the side of private business, self help groups of informal workers and small-scale entrepreneurs as well as
support from local social security agencies. Technical and financial support to extend coverage of the informal economy
workers could consist in programs for the integration in social insurance, in creating a special scheme for the informal
workers or in encouraging the development of mutual assistance or micro insurance schemes 51. Related to European
experiences with social health insurance, a gradual extension in the scale and scope of provisions for informal workers
would be financially easier to perform than full membership at the beginning 52. In the case of Indonesia, however, poverty
level and size of informality could make it advisable to prioritize general access to health services as a demonstration of the
political willingness of all spheres of society to cooperate and to optimize the use of resources in the realization of economic
and social progress as an entity and as accessible for all.
49
World Bank (2006b: 195).
50
Gsänger (1995).
51
Lindenthal (2004).
52
Bänighausen and Sauerborn (2002: 1568).
21
22
4. Social Insurance:
Reforming Contributory
Social Security
4.1. Social Insurance: Conceptual Framework
- In general, insurance reduces vulnerability by replacing the uncertain prospect of losses with the certainty of making
small regular premium/contribution payments.
- Mandatory membership for specific legally defined population groups, the independence of contribution levels of
individual risk exposure (income related or to a lesser extent flat-rate contributions) and self-administration are
constituent features of social insurance.
- Both, a Pay-As-You-Go (PAYG) or a funded system are possible financing procedures of social insurance. None of
both systems is per se superior. Factors such as demographic changes, prevailing norms of fairness, or financial
market characteristics impact on the relative effectiveness of both systems.
- Combining both types of financing procedures is possible. In addition, other social protection mechanisms, such as
tax-financed components for the poor, mandatory (subsidized or non-subsidized) individual saving accounts, or private
voluntary measures, such as insurance or savings, can complement social insurance mechanisms.
23
Insurance offers the service of financial protection to its clients by reimbursing an individual for some or all of his financial losses
that are linked to the occurrence of a risky event. The expression “risk” refers to the chance or likelihood that an event will cause
damage or loss. In its simplest feature, insurance has two fundamental characteristics: risks are transferred from an individual
to a group (Risk Transfer) and losses are shared on a predefined basis by all members of the group (Risk Sharing) 53.
Financial protection is accomplished through a pooling mechanism, which constitutes the basic underlying principle of
insurance. The pooling mechanism comprises two elements:
1. People who face the same particular risk are joined together into a risk pool (Risk Pooling).
2. Each person pays a small amount of money (the premium or contribution) into the pool, which is then used to compensate
those individuals who do actually suffer a loss (Resource Pooling).
Insurance works by sharing the risk across a larger number of people. The larger the risk pool the more efficient risk pooling is.
Insurance reduces vulnerability by replacing the uncertain prospect of losses with the certainty of making small
regular premium/contribution payments. Furthermore, it enables an individual to protect himself from financial losses that
exceed his own income. Therefore, from a micro level perspective, insurance is more effective than savings or credit,
because these instruments can only protect from losses, which do not go beyond the individual wealth.
People are exposed to a wide array of risks throughout their lives. With regard to social insurance, illness, old age, and
unemployment are the most prominent risks. A risk can be classified according to the following three categories:
1. The type and degree of uncertainty caused by the risk
2. The relative size of the loss and
3. Whether the risk is idiosyncratic or covariate
The uncertainty may refer to the timing or the magnitude of the event. For example, health risks are particularly uncertain
regarding timing: The severity of a certain illness might be known, but it might not be known if and when this illness might
occur. The occurrence of a risk may lead to high or low costs. Health risks are very dissimilar in nature. They may vary from
rather low costs (e.g. basic medicines) to very high costs (e.g. major surgeries). The distinction between idiosyncratic and
covariate risks points at the number of people affected by the occurrence of a risk. It may affect an individual or many people
at one time. In the first case, the risk is called “idiosyncratic”, in the latter case it is called “covariate”, which means that the
risk is correlated between different people. For example, health risks can be idiosyncratic (e.g. illness) as well as covariate
(e.g. epidemics).
Table 2 categorizes the three main risks of social insurance according to the above-mentioned categories.
In addition, the prospect of insurance or the state of being insured might alter individual behavior in different ways. These are
the so-called “behavioral risks” associated with insurance, which impact negatively on the financial situation of an insurance
provider. Adverse selection refers to the risk that only persons with a high-risk profile join an insurance program, whereas
low-risk profile persons do not. Adverse selection induces high claim ratios, which can lead to a continued need for increasing
premium/contribution levels. As a result, people with relatively low-risk profiles opt out of the insurance. If premiums/contributions
levels are not increased, the financial condition of the insurance program deteriorates. In both cases, the insurance breaks
down. Moral hazard on behalf of the clients occurs if the insurer is not able to monitor the behavior of the insured. Being
insured introduces incentives for the insured to alter their behavior towards being less careful to avoid the risk (as they do
expect financial compensation), which might not be easily detectible by the insurance provider. This is called ex ante moral
hazard. In addition, if insurers cannot verify if a submitted claim did really occur, clients face an incentive to produce false
claims. This problem is called ex post moral hazard. Free rider behavior refers to the risk that people join an insurance
program when in need and opt out after directly having received compensation. Another form of moral hazard might occur if
a third party, i.e. besides the insurance and the insured, is involved and the insurance provider is not able to fully monitor or
53
Vaughan and Vaughan (2003).
24
4 • Social Insurance: Reforming Contributory Social Security
verify the actions of this party (third party moral hazard). This is the case with health insurance: it is not easy to verify if a
certain treatment was necessary or appropriate. Thus, health providers face an incentive to exploit this fact and treat people
more often or with higher cost care than is actually necessary.
Table 3: Characterizing Risks
Illness Old Age Unemployment
Characteristics of insured risk

Type of uncertainty
Timing Uncertain (Mostly) certain Uncertain
Magnitude (Mostly) certain Uncertain Certain
Relative size of the loss Low – very high High High
Idiosyncratic - covariate Both Idiosyncratic Both
Behavioral risks associated with
insured risk
Adverse selection Yes Yes Yes
Moral hazard (clients) Yes (ex ante, ex post) Yes (ex ante) Yes (ex ante, ex post)
Free rider Yes — Yes
Moral hazard (3rd party) Yes No No
However, two aspects need to be stressed:
1) “All behavioral risks are risks”. This means that they are not necessarily present in any context. For example, the
presence of moral hazard in health insurance for poor persons working on a daily basis is highly unlikely, as
spending time to seek health services involves high opportunity costs.
2) An insurance program disposes several instruments to deal with those risks. As will be shown below, certain
constituent features of social insurance reduce the risks of adverse selection and free riding behavior.
Table 2 shows how specific behavioral risks are relevant for each risk. The specific characteristics of each risk impact on the
design of appropriate social protection instruments.
4.1.1. Organizational Principles of Social Insurance

There are two constituent features of social insurance schemes: first, membership is mandatory for specific legally defined
population groups with limited opting-out clauses (e.g. persons above a certain income threshold). Secondly, the level of
contribution is independent of individual risk exposure (income related or to a lesser extent flat-rate). Both features
contrast with private commercial insurance. Income-related contributions and the resulting redistribution of income are
expressions of the principle of solidarity, which is one of the guiding principles of social insurance.
Due to mandatory membership social insurance is not prone to adverse selection or free riding behavior. Of course,
the enforcement of mandatory membership needs to be ensured.
Mandatory insurance membership is stipulated and enforced by the government. However, the government is not necessarily
the insurance provider. The insurance carrier is either a private- or a public-law organization. Self-administration is another
central principle of social insurance. Government assumes a supervising and regulatory role. It is not common that the
insurance carrier is a public authority 54. Social insurance systems vary with regard to the number of insurance carriers
involved in insuring a specific risk. Either monopolistic structures prevail with one insurance body providing social insurance
54
Public authorities are typical for tax funded universal systems, for example the British National Health Service.
25
(e.g. old age in Germany, health in France) or several insurance carriers compete with each other (e.g. health in Germany).
4.1.2. Financing Social Insurance

The main sources for resource mobilization are contributions. These can further be distinguished between
* Type of contributions: Income-related contribution (quasi income tax) vs. flat-rates, and
* Who is paying contributions: Employee, employer, or shared between both.
Although in principle, social insurance relies on contribution financing only, in reality it is often supplemented by tax-financing.
Regarding the financing or risk pooling procedures, there are two options - Pay-as-you-go (PAYG) or a Funded System 55:
* PAYG is the original social insurance financing principle. The risk pool comprises the entire population. Current
expenditures for benefits are financed by members within each accounting period (usually one year). No capital stock is
built and only small liquidity reserves are held. Another main feature is the so called “intergenerational contract”: the
working age population finances children and old people (health: ‘3-generation-contract’) or old people only (old age: ‘2-
generation-contract’). Only with regard to unemployment, PAYG involves no intergenerational consequences
(unemployment: ‘1-generation-contract’).
* Funded System: Within a funded system risk pooling occurs within a certain predefined age cohort. Contributions of
each age cohort are used for building up a capital stock to finance expenditures for benefits for each cohort. The present
value of all contributions of an age cohort (year of birth) equals the expected present value of expenditures for benefits
of each age cohort (year of birth) based on actuarial calculations 56.
Specifically with regard to pensions, there is a wide international debate on the relative appropriateness of these two systems.
Social health insurances most commonly operate through PAYG, but -borrowing from commercial insurance experiences -
additional funded components are possible (specifically for financing higher health expenditures in old age).
It needs to be stressed that no system is superior per se. A variety of factors influence the relative efficiency of both
systems. One main argument in favor of a funded system is its stronger independence of demographic changes such as
changes in fertility rates or an increasing life expectancy which lead to a higher share of elderly within a given population.
However, it is not fully independent of those changes: if life expectancy increases, contribution level within a funded system
must increase, too, in order to secure a given level of pensions. From a normative perspective, people might rate the fairness
of a funded system higher than a funded system is more reciprocity oriented than a PAYG.
Compared to a PAYG, a funded system also exhibits disadvantages. In contrast to a PAYG, a funded system involves the
building up of a capital stock. This capital accumulation enables a funded system to increase its financial resources by
investing the capital stock and earning interest. However, the existence of a capital stock might endanger the stability of a
social security system at the same time: Depending on the type of investment, it either renders the system more vulnerable
to inflation or to financial market volatility 57. Thus, financial market characteristics influence the relative effectiveness of
both types of systems. Further, a funded system requires skilled investment managers to minimize the risks of a depletion of
the capital stock. In addition, in the absence of effectively enforced legal provisions the existence of a capital stock increases
the risks of abuse.
With regard to a specific risk it is possible to combine both types of financing procedures. In addition, other social
protection mechanisms such as tax-financed components for the poor, mandatory (subsidized or non-subsidized)
individual saving accounts or private voluntary measures such as insurance or savings can complement social
insurance mechanisms.
55
Schulenburg and Greiner (2007:31-32).
56
Confusion sometimes surrounds the notion of a “funded system” which is often mixed up or used equivalently to individual saving accounts leading to
complaints on the lack of solidarity within funded systems. However, from the above definition it is obvious that this is not the case. The difference between
a PAYG or a Funded System is not the difference between risk pooling and no risk pooling, but rather the underlying reference group for risk pooling.
57
Lampert and Althammer (2001:234).
26
4.2. The Administration of Social Security

From a theoretical point of view there are two basic options for the future reorganization of the Administration of
Social Security; integration and separation. In addition, there is a third, more pragmatic option: harmonization:
• Integration of all risk related activities under specific administrative bodies for each risk.
• Separation of all risk related activities for certain groups of society and their specific administrative bodies.
• Harmonization and rationalization of regulations between existing administrative bodies.
Focussing on technical efficiency of a public system of social protection,
• integration of administration is an appropriate option for social insurances;
• whereas social assistance is best provided within a framework of (broad) separation.
• The third option, harmonization, takes institutional path dependencies into account so that the already
existing institutional framework for social protection is only gradually adapted to new or changing needs.
Conclusion:
• Relating these options to the Indonesian context, harmonization is chosen at present with the aim of
harmonizing all laws and regulations up to 2009. Besides the political need to mitigate conflicts, the current
early state with regards to basic decisions in many reform areas (e.g. pension system or unemployment
compensation) still limits the potential for comprehensive administrative reform.
• However, the limited technical benefits of harmonization versus the other two options demonstrate that the
objective of harmonization should be seen as a short-term solution within the reform of social protection in
Indonesia and not as the final result.
• Within a mid-term perspective, however, for the administrative set-up of social protection in Indonesia (a.)
institutional integration of risk-related activities under one specific administrative body for each risk in
standard social insurance and (b.) institutional separation of risk-related activities for the poor and other
vulnerable groups, i.e. social assistance is recommended as the overall guiding framework.
4.2.1. The Current Situation

There has never been a systematic approach in the conceptualization of social security (or social protection in general) in
Indonesia as has been the case in Europe following the Bismarck or the Beveridge Legislation. Therefore, the administration
of social security services has remained fractured and incomplete, covering only a relatively small part of the population.
Public administration of social security currently extends only to employees of the military (ASABRI, ASKES), civil servants
(TASPEN, ASKES) and to those employed in private enterprises above a certain staff size and total payroll (JAMSOSTEK).
The scale and scope of risk coverage is also rather limited in that for example private sector employees are only insured
against work accidents, illness, death and funeral risks, but not against unemployment. In the case of retirement, they
receive a lump sum payment from a mandatory saving scheme (provident fund) related to their contributions plus interest, so
that longevity risks are not covered. Furthermore, many have to draw from these savings even before retirement when
unemployment or other risks occur. Members of the armed forces and civil servants are better off as they are generally not
exposed to the risk of unemployment and receive, besides health care services, annuity benefits for life after retirement 58.
Considering that the informal sector accounts for about 65 to 70 percent of the total work force 59, only a relatively small (and
already privileged) part of the economically active population is covered by these social security schemes, while the majority
still has to rely on the family and other traditional social networks for risk protection. Furthermore, demographic as well as
economic and social changes are aggravating these underlying institutional deficits. Rapid urbanization caused by rural-
urban migration and by the changing status of villages from rural to urban areas as well as the decline in fertility and increase
in life expectancy have already caused the population to age and led to increasing dependency from public or private social
protection measures. However, while mostly informal employment in the agricultural sector has been steadily declining, the
58
Lindenthal (2004).
59
Lindenthal (2004:18).
27
growth of formal employment in the rapidly expanding service and manufacturing sectors of the economy has enlarged, but out
of step with changes of the workplace, so that the gap between dependency and access to social security has widened 60.
It has to be mentioned, however, that in the wake of the 1997 Asian Crisis, a number of Social Safety Net Programs
concerning health, nutrition, education, rice and fuel subsidies, have been launched by the Indonesian National Planning
Board (BAPPENAS) against rapidly rising poverty levels, especially in rural areas. The Coordinating Ministry of People’s
Welfare was assigned to coordinate and manage the implementation of these programs 61. In addition, at the level of districts
and communities in different provinces, numerous ad hoc measures for poverty alleviation have been set under way by local
self-help groups, non-governmental organizations and foreign donors, yet there is a fundamental deficit of any national
administrative structure which could comprehensively control and coordinate the implementation of these measures horizontally
between the institutions involved on the macro level, as well as vertically between the central government and governments
on the regional or local levels (see chapter 6).
4.2.2. Reform Options

Before the enactment of Law No. 40/2004, there was no adequate public response to these challenges. Law No. 40/2004
provides the legal basis for the development of a comprehensive and coordinated public social protection system. Still, open
questions remain concerning the operationalization of this objective.
In this context, the question how to organize the administration of social security is important. It impacts on the overall cost-
efficiency of the system as well as on coverage achieved. As the question, how to organize the administration of social security
is difficult to disentangle from the question of how to organize social assistance, both will be dealt with in this chapter 62. Also, the
question of how to include the provision of social security for the informal sector will be taken up here 63.
Law No. 40/2004 does not provide much guidance on this issue. It simply reconfirms the existent administrative bodies and
introduces the possibility of establishing new ones. In addition, it makes no provision regarding the administrative responsibilities
for the informal sector. As it is a law on social security it also makes no reference to the administration of social assistance for
the poor. But at the same time it states the objective to include the poor, starting with social health protection, thus tapping
into the realm of social assistance.
From a theoretical point of view, there are two basic options for a future reorganization of the administration: integration and
separation. In addition, there is also a third more pragmatic approach: harmonization. The following elucidations outline
major arguments for the relative effectiveness of each option. However, when taking a system-wide perspective the options
are not in any case mutually exclusive. Therefore, after having discussed the basic arguments for each option, the inter-
linkages between the options will be discussed. Table 3 summarizes the basic ideas.
First, ‘integration’ means that the design of social protection could in general follow the notion that administrative bodies
with responsibilities for all activities related to one specific risk such as illness, old age, or unemployment are set-up 64. This
option is favorable when size, i.e. number of people covered, tends to be an important factor in the provision of the social
protection benefit. Further, integration also enables favourable conditions regarding the composition of the risk pool: As the
risk pool (ideally) comprises the entire population, a potential societal segmentation of high and low risk groups is avoided.
Second and alternatively, separation refers to the integration of all activities related to the risks of a specific group of society
within one specific administrative body. Historically, such schemes are reaching far back in Europe even to the Middle Ages,
when local groups of miners and carpenters established mutual risk relief funds based on solidarity and trust 65. Later on,
60
Axel (2006).
61
Axel (2006:88).
62
For a more detailed treatment of social assistance, see chapter 6.
63
For a more detailed treatment of the inclusion of informal sector workers, see chapter 7.
64
This option does not preclude the possibility of establishing several administrative bodies for one risk - each with an identical task - and thus introducing
competition. Further, even with one central administrative body certain tasks can be carried out at regional or local levels.
65
Like Friendly Societies in Britain, Hilfskassen in Germany, Mutualités in France and other mutual benefit societies throughout Europe. Marcel van der
Linden (1996).
28
when industrialization generated large groups of blue and white collar workers employed under standardized conditions and
threatened by the same risks, in most countries, these first schemes were superseded by application of the social insurance
principle following the Bismarck Legislation. The mutual risk relief funds show some resemblance to today’s micro-insurance
schemes, which are often established on the basis of common socio-economic backgrounds of participants and personalized
relationships, in this way making use of mutual trust. However, with regard to public social protection interventions, which
generally involve anonymous society-wide relationships, separation of administration is advantageous since other criteria
are also relevant. If providing social protection involves high transaction costs for specific population groups whereas for
other population groups these are not present and these transaction costs replicate themselves per type of benefit provided,
separating this group(s) from the rest of the population and allocating administrative responsibilities within one specific
group-related administrative body can provide cost-saving economies of scope 66. Further, if specific groups within a society
face specific risks or needs not faced by the rest of the population and these risks or needs ought to be covered, subsuming
these groups under the responsibility of a specific administrative body is advisable or even inevitable (as only specific groups
are targeted).
Reconsidering the arguments described above, integration of administration is an appropriate option for social
insurances: Insurance works by pooling risks across a large population. The larger the risk pool, the smaller the volatility of
claims, thus facilitating the management of the insurance. However, an insurance risk pool must not necessarily be as large
as the entire population to effectively reduce volatility 67 and the positive returns to size diminish after a certain threshold. Yet,
for social insurance to work, a balanced composition of the risk pool is essential. This is achieved by integration but not by
separation. In contrast, following the arguments for separation, social assistance is generally speaking best provided
within the framework of a (broad) separation: firstly, it is per definition targeted at specific groups within a population, i.e.
the poor and other specific needy groups. Secondly, again per definition all social assistance benefits are targeted benefits.
As targeting methods involve non-neglectable transaction costs, subsuming social assistance under one administrative
body provides the opportunity for exploiting economies of scope.
Providing social protection to the informal sector, especially to those in the informal sector not covered by social assistance,
requires specific approaches and involves high transaction costs (see chapter 7 for more details). This would provide an
argument for separation. However, and in stark contrast to social security and social assistance, the provision of specific
social protection mechanisms to the informal sector should be considered as a temporary measure as the long-term objective
should focus on extending the size of the formal economy and reducing the informal economy per se. Thus, the need for
specific social protection for the informal sector ceases in time and all programs can be considered as temporary measures.
Therefore, all responsibilities regarding the informal sector are of a temporary nature. However, it is difficult to abolish public
bodies after they have come into existence. Therefore, “separation” does provide a general guiding framework for the
informal sector, but the administration should not be in the hands of a body without any other responsibilities.
66
Economies of scope are present if there are benefits to be realized from producing multiple goods or services, i.e. if it is cheaper to produce both good X
and good Y together rather than separately. Economies of scale in return occur when per unit cost savings are realized by producing more of a good or
service, i.e. if the average cost per unit decreases as output increases.
67
If this were the case, private insurances would not be able to operate.
29
Table 4: Administrative Options
Rationale Recommendation
Integration Risk pooling (size and composition Social insurance

of the risk pool)
Separation Economies of scope Social assistance

Specific risks/needs by certain Social security for the non-poor
population groups in the informal sector
Harmonization Avoiding political conflicts ‘1st step’ only

High uncertainty
Open policy questions within important
inter-related reform areas
Although the analysis above does provide a general guiding framework for the organization of administration, inter-linkages
within the social protection system exist. Basically, this means that although the overall administrative responsibility
either for one specific risk (integration) or one specific group (separation) is vested within a specific administrative body
certain tasks could be transferred to another body in order to exploit cost-saving synergies.
* Regarding social insurance irrespective of the type of risks covered, all social insurances rely on pay-roll deductible
contribution payments. To reduce transaction costs, the task of collecting contribution could be assigned to one
administrative body, which then transfers the respective sums of contributions to the other administrative bodies.
* With regard to certain social insurances the question arises if the poor or other specific needy groups should be included
within the general insurance framework or covered separately. This is particularly relevant for social health insurance.
The motivation might arise from either equity or efficiency concerns. As access to health is considered a basic human
right, including the poor in social health insurance precludes the provision of lower quality services to the poor resulting
from stigmatization effects (equity concerns). From an efficiency perspective the inclusion would be motivated by a
reduction in transaction costs: Contrary to other social insurances such as old age, social health insurance involves the
inclusion of a 3rd party (health services provider) and the related activities of accreditation, contracting and monitoring.
It is more efficient to allocate these responsibilities within one administrative body than replicating these transaction
costs by having more than one administrative body involved. If the poor should be included within the specific social
insurance, the administrative body responsible for social assistance should remit the contributions on behalf of the poor
to be covered to the respective social insurance administrative body, which would then be responsible for delivering the
benefits/services to the poor. The same argument holds for extending social health protection to the non-poor in the
informal sector.
Both options analyzed up to now, focus on aspects related to the technical efficiency of a public system of social protection.
The third option, “harmonization” takes institutional path dependencies into account for the design of social protection so
that the already existing institutional framework for social protection is only gradually adapted to new or changing needs. It
would mean maintaining a mixed-, multi-pillar system with divided administrative responsibilities, but harmonizing existing
responsibilities and eliminating contradicting regulations and laws. In terms of technical considerations this option is inferior
to the before mentioned options. The relevance of this option, however, results from a political economy perspective: Changing
an existing institutional framework most likely involves conflicts of interests and thus potential political conflicts hindering the
reform process. For example, changing the administrative set-up involves the reallocation and potential reduction of assigned
tasks for specific Social Security Administering Bodies (SSABs) and their respective line ministries and thus, might induce
their opposition. However, the presence of political conflicts is a natural element of any reform and political democratic
process. The challenge within each democracy is to find ways to moderate and cope with these conflicts. However, within
this context harmonization might provide a strategy that postpones critical changes within the present system, thus offering
30
a “grace period” for political stakeholders for necessary consensus and commitment building. In addition, harmonization
allows current social security provisions to be gradually modernized and adapted to changing needs in respect of demographic
trends and employment patterns. Slow adaptation is specifically important if uncertainty on future developments of basic
conditions is extremely high 68 or if policy decisions within important reform areas, which impact on the administrative set-up,
have not been made yet (such as types of benefits provided). In this context, harmonization can provide an intermediate
solution or a ‘1st step option’ within the entire reform process.
Looking at administrative set-ups in European countries where the social protection system is more or less organized
according to the social state model, i.e. statutory contributory social insurance combined with non-contributory means-tested
social assistance, the principles of integration for social insurance and separation for social assistance actually do provide
the overall framework 69. In Switzerland, Hungary, the Czech Republic as well as Latvia one finds these principles implemented
in a rather pure way, whereas other countries show certain deviations or exceptions from the general framework with regard
to social insurance with some occupational based funds in place. The existence of those occupational based funds can be
primarily traced back to historical path dependencies: In France separate administrative bodies for people working in the
agricultural sector as well as for the self-employed exist. In Austria specific administrative bodies targeted at the self-employed,
railway and mining sectors, farmers, notaries and public service employees are in place. In Germany occupational funds
were present within social health insurance, but recent reforms have abolished these separations and funds had to open up
their membership. Still, specific funds are in place for miners, farmers and seamen. Also for old age one specific administrative
body for miners, railway workers and seamen (The German Pension Insurance - Mining, Railways, Marine) is in place 70.
Relating those options to the Indonesian context, harmonization is the option that is chosen at present with the aim
of harmonizing all laws and regulations up to 2009. Besides the wish to mitigate political conflicts, the openness of many
reform areas (e.g. type of pension system or unemployment compensation, modalities for including the non-poor in the
informal sector) still limits the potential for a comprehensive administrative reform, at least to a certain extent. However, the
arguments outlined above demonstrate that the objective of harmonization should be seen as a short-term objective within
the reform of social protection in Indonesia and not as the final result.
Within a mid-term perspective, however, for the administrative set-up of social protection in Indonesia (a) institutional
integration of risk-related activities under one specific administrative body for each risk in standard social insurance
and (b) institutional separation of risk-related activities for the poor and other vulnerable groups, i.e. social assistance
is recommended as the overall guiding framework.
(a) Within the broad framework of integration several options for the allocation of responsibilities are possible. One option
is to make JAMSOSTEK the superstructure of a comprehensive system of social insurances against work accidents,
old age and, perhaps later, unemployment. Institutions, such as ASABRI, TASPEN or ASKES, would then have to open
their membership or merge into the general social insurance system. It is, for instance, imaginable that ASKES would
turn into the role of a general social health insurance. Concerning JAMSOSTEK, current provident fund provisions
could be complemented by pension insurance schemes (see chapter 4.4.). Besides risk provisions against old age as
well as work accidents, active labor market policy could also be implemented under the administrative roof of JAMSOSTEK,
including contributory unemployment compensation schemes. Another option would again focus on ASKES as social
health insurance and a division of tasks between JAMSOSTEK and TASPEN, for example JAMSOSTEK could become
the main body for unemployment compensation whereas TASPEN would be responsible for old age insurance.
(b) Furthermore, to concentrate resources for poverty reduction or alleviation more effectively, responsibility for social
assistance should then be centered on a newly founded institution under the umbrella of the Ministry of Social Affairs.
Social health protection contributions for the poor could also be transferred from the social assistance administrative
body to the respective social health insurance body, which would then be responsible for providing the benefits.
68
Naturally, uncertainty is omnipresent and inevitable. Thus, uncertainty per se is not an argument for harmonization.
69
These European countries comprise Germany, Austria, France, Switzerland, Hungary, Latvia, the Czech Republic, and partly the Netherlands, where a
rather unique mixed system - combining Beveridge and Bismarck principles - is in place.
70
Specific provisions are also in place for civil servants, judges and soldiers, but these are not associated with separate administrative bodies.
31
In this set-up the integration of the non-poor informal workers could either rely on a gradual transformation into formal
employment and contributory membership in social insurance or in the creation of a special program (see chapter 7). In the
latter case, the allocation of the overall administrative responsibility depends on the type of risks covered for informal sector
workers. Principally, it should be vested within one body but possible linkages to the other bodies could be explored.
Regarding social insurance and referring to the well-known resilience of administrative institutions, the second reform option
of institutional separation of risk-related activities for certain groups of society would probably be easier to realize, as
more or less strictly group-oriented provision of social security services is already in place. Hence, separation would apply
more to informal workers (and probably micro-entrepreneurs as well) under a similar special scheme and, respectively, a
separate administration would have to be institutionalized (but again within an existing one) following the specific needs and
requirements of this part of the workforce. The main problem connected with this reform option is that it would divide the
Indonesian society into privileged groups, such as the military, civil service and big company employees, with membership
in their specific administrative bodies (ASABRI, ASKES, ASPEN and JAMSOSTEK) while leaving the remaining part of the
population in a precarious situation, as it will be difficult to find a financially viable solution for their social security. Considering
the less favorable risk composition and limited ability to pay contributions of this part of society, only meager services and the
necessity for heavy subsidization from the central government can be expected. Although this option is politically attractive
as resistance from existing authorities will be rather low, it is not recommendable from a technical point of view.
Last but not least, irrespective of the administrative structure chosen effective governance including transparent
procedures and a functioning supervision of the administrative bodies are clearly essential ingredients for a
successful outcome of current reform endeavors.
4.3. The National Social Health Insurance Program: Work in Progress

4.3.1. The Current Status of the System
The Indonesian healthcare delivery system consists of both public and private providers. Its funding also draws from a
variety of sources, both public and private.
The backbone of the system is a government line item budget aiming to fulfill the constitutional mandate to provide statutory
health services to all citizens. The state budget is allocated for both public and personal healthcare. After the introduction of
administrative and government decentralization in 1999, local government is obliged to contribute financially to development,
including health.
The Indonesian Government installs and finances primary healthcare centers (Puskesmas) and sub-centers (Puskesmas
Pembantu and midwife/village clinics) at the sub-district and village level as points of healthcare delivery. The government
also installs and finances hospitals for referral to secondary care in all districts in Indonesia.
Currently, the government allows public providers to impose tariffs based on their costs, as mandated by Law No. 9/1968 jo
Law No. 20/1997 on non-tax state revenues, Law No. 5/1974 jo Law No. 18/1997 on local taxes and retribution, and Law No.
32/2004 on regional government. This policy also applies for members of existing health insurance schemes 71. The difference
between service charges levied by public providers and payment rates set up by health insurance agencies must be born by
the insurance members, so that cost sharing occurs in the financing of health services. There are two reasons for the
mismatch: on the one hand, the insurers have fixed their rates based on what is taken for an acceptable rate for the
members, not strictly on cost-recovery mechanisms. Furthermore, there are no functional standards for service quality and
matching costs under conditions of efficiency. As this de facto cost sharing is regressive to members’ income, tariffs create
an access barrier particularly for people with a low income. On the other hand, user fees account for an essential source of
revenue for public healthcare providers.
71
ASKES for government employees, and JAMSOSTEK for the formally employed.
32
Private health insurance only plays a role in a small segment of the market for affluent individuals, and as a substitute for the
health insurance scheme for formal sector employees provided by state-owned JAMSOSTEK. By statute, formal employees
must be insured either by JAMSOSTEK or by any scheme that matches or exceeds the coverage provided by this company.
Many enterprises chose either private insurers, or they offer coverage from their own resources.
The Indonesian government allows the private sector to install and provide healthcare services for a wide spectrum of
services. The private entities serve all levels of care, primary, secondary and tertiary. The major source of finance for private
healthcare providers is user fees. Private providers are mostly located in areas where people can afford their services.
In conclusion, the Indonesian personal healthcare system combines a government line item budget, out-of-pocket fees for
service, and health insurance. Out-of-pocket payments are a major source of revenue for Indonesian healthcare providers.
User fees have brought inequity to the healthcare system both in terms of financial and geographical access.
The National Social Security System (NSSS) Law opens an avenue to restructure the personal healthcare system. The way
forward requires careful design to assure that the future system can significantly diminish inequity, and improve its performance
and outcomes. The design also needs to carefully integrate personal healthcare with overall health policy with its important
elements of public health, and with other programs within the national social security system.
The reform process has started with the ratification of Law No. 40/2004 on the National Social Security System (NSSS Law).
This law sets the basic principles for universal social security, as well as a timeframe for the preparatory works necessary for
the new system to be implemented throughout Indonesia. The deadline for completion of the new system is August 2009
(see Figure 3).
Figure 3: Timeline of Health Sector Reform in Indonesia 72
legislation and
regulatios for
new system
old System new System
gradual
implementation of
new system
2004 2007 2009

This chapter contains a description of the status quo before the reform and of the NSSS Law (sections 4.3.1. and 4.3.2.),
followed by an analysis of the ongoing legislative and regulatory process in section 4.3.3. Within the boundaries already set
by the new NSSS law, many options are still open. The challenge is to choose the options that will best serve the purpose of
the health sector reform: to increase equity and efficiency in healthcare for all Indonesians. Section 4.3.4. shows these
options, in connection with the steps that need to be undertaken according to the roadmap of the ongoing reform process.
By necessity, this paper focuses more on the legal and regulatory aspects of the ongoing reform process. Less emphasis
has been put on wider aspects of policy questions and options in the broader and longer-term prospects of social protection.
There are two reasons for this approach:
1. The legal process is in the decisive phase of the implementation of Law 40/2004. Whatever the future options are, they
will be shaped by the outcome of this process. There are many open issues nested within a tight schedule of work on
new laws and regulations, and the adaptation of existing ones. We want to draw the attention of the readers to these
issues, which can easily be overlooked.
2. At present, the legal system that must be manipulated in order to succeed in carving out the first step towards universal
72
The interval in 2004 represents the date of ratification of Law No. 40/2004, and the interval in 2009 represents the deadline set by the same law to launch
the new system.
33
social protection is complicated, and some of the tasks to harmonize all its elements, and make the best choices at the
same time, are cumbersome to say the least. It is necessary to understand the details in order to appreciate the
requirements to shape any system of social protection in Indonesia. We feel that the understanding of this process is a
prerequisite for the discussions on the large policy questions that will arise once the foundations of the system are in
place.
4.3.2. The Current Status of the System

Prior to enactment of the new NSSS Law, Indonesia had established various health insurance programs, funded by employees,
employers, a mix of employees and employers and the government. Table 5 provides a summary of health insurance
programs that are in effect up to the present. The following observations add to the facts presented in this synopsis:
1. Health insurance programs were developed separately, based on different member groups and without coordination.
2. The regulatory foundation of the programs and their implementation is based on different hierarchies of laws and
regulations for each program, ranging from national laws to ministerial regulations.
3. Program implementation is fragmented along the lines of population segments, based on separate regulations applied
only to the respective individual population groups.
4. Program implementation is mainly focused on organized population groups in formal employment: civil servants, military,
police, and private formal employees.
5. Social security for informal workers, farmers and fishermen should have been provided by Laws No. 14/1969 and No.
13/2003 on Manpower, as well as Law No. 3/1992 on Manpower Social Security. However, these laws were not
implemented as envisioned, and their implementing regulations are not yet enacted despite explicit instructions given in
Law No. 3/1992 on Manpower Social Security.
6. Implementation of the health insurance program for active government employees varies according to their place of
work (civil government institutions and military-police), even though contributions are set uniformly for all government
employees at 2 percent of their basic salary; the health insurance program for retired government employees is managed
by PT. ASKES.
7. The benefit packages offered by health insurance programs vary greatly.
8. Healthcare services are mainly provided in public healthcare facilities, but different programs require different service
packages, so that insured service coverage is patchy both from the patients’ and the providers’ perspective.
9. Payment methods to healthcare facilities vary greatly, ranging from budgets to contracts. Contract payments to healthcare
facilities vary from claim payments through fee-for-service to prospective capitation payment in primary healthcare
facilities.
34
Table 5: Synopsis of the Existing State-Supported Health Insurance Programs in Indonesia, prior to the NSSS Law
State-provided health insurance program for
Active military
Civil servants and police Private sector workers Indigents Other population groups
personnel
Civil Military and War veterans

government police DHK 73 LHK 74 and pioneers of Others
institutions institutions independence
Based on law No. 8/1974 jo No. 8/1974 jo No. 34/2004 jo No. 14/69 jo No. 14/69 jo
No.43/1999 No.43/1999 No. 2/2002 No. 13/2003 No. 13/2003
Legal foundation GR 69/91 GR 69/91 GR 67/91 Law 3/92 Law 3/92 Law 23/1992 GR 69/91 Law 23/1992
for program Law 22/99 jo
32/2004
Legal foundation Minister of Minister of GR 14/93 jo GR not yet Minister of Health Minister of
for technical GR 6/1992 Defense Defense GR 64/2005 enacted (man- Regulation GR 69/91 Health
arrangements Regulation Regulation date of art. 4(2) No. 1241/2004 Regulation
Law No. 3/92) Local
Government
Regulation
Employees 2% 2% 2% 0 - 0 0 Fixed rate
dependent on
benefit
Employers / 2% (gradual) National govern-

Government GR 28/2003 ment Budget
Contribution
paid directly to 3% (single) - APBN: Rp 5000/ National Varies
PT ASKES or 6% (family) cap/month government
73
Di luar hubungan kerja = within the contract - membership is mandatory for employers who employ 10 employees or pays wages for all employees Rp. 1.000.000,00 at minimum.
74
Di luar hubungan kerja = Outside the contract; FFS=fee for service.
35
36
discount for Local government budget
tariffs at govern- budget APBD:
ment health varies for contri-
facilities butions and/or
cost sharing
Beneficiaries Nuclear family Nuclear family Nuclear family Nuclear family - Nuclear family & Nuclear family Nuclear family
max 5 members max 5 open for certificate
of poor holders
Conditionally Determined by Voluntary
compulsory (opt- government (workers in the
Membership Compulsory Compulsory Compulsory out for employers - (quota & eligibility informal
who can provide criteria) economy)
better benefits)
Outpatient care, Outpatient care, Outpatient care, Outpatient care, Outpatient care,
inpatient care, inpatient care, inpatient care, inpatient care, inpatient care,
surgery, surgery, surgery, surgery, surgery, The same as
Benefit package maternity care, maternity care, maternity care, maternity care, - maternity care, civil servants’ Varies
rehabilitation, rehabilitation, rehabilitation, rehabilitation, rehabilitation,
prostheses, prostheses, prostheses, prostheses, prostheses,
domestic domestic domestic domestic domestic
Contributions Membership Membership Membership Membership - Free for Membership Membership
payments payments payments payments members payments payments
Contracted Military and/or Military and/or Contracted Mandatory The same as Contracted
health facilities, police health Police Health health facilities provision at civil servants health facilities
Healthcare mandatory facilities Facilities - government
providers provision in owned facilities
the government (Puskesmas &
owned facilities hospitals), volun-
tary provision at
private facilities
Payment method Varies (FFS, FFS, capitation, The same as
to healthcare capitation, Budget Budget FFS, capitation - lump sum civil servants FFS
providers lump sum)
4.3.3. NSSS - the Law on the National Social Security System

Article 19 of Law No. 40/2004 on the National Social Security System, called the NSSS Law below, dictates that personal
healthcare is to be carried out through a national health insurance scheme that is based on the principles of social insurance
and equity.
Health insurance as a national program is built within national structures and functions to ensure social justice for all Indonesian
citizens 75. The program’s structure and functions are to be regulated by a new Presidential Regulation on the National
Health Insurance Program (PresReg NHIP). Furthermore, there is a need for a comprehensive and integrated regulatory
framework spanning from the NSSS Law to implementation level regulations. The comprehensiveness, alignment and concord
of regulations on the health insurance program will greatly influence the effectiveness of the PresReg NHIP as a regulatory
instrument in ensuring the attainment of benefits from health maintenance and protection for all the population in meeting
basic healthcare needs 76.
The fulfillment of the principles of social insurance depends on the financing system, and on the implementation system
framework that connects members, executing agencies of social security and social (health) service facilities. This framework
therefore includes both the flow of funding and that of benefits. In this context the following aspects need to be carefully
designed in order to ensure a functioning social health insurance scheme:
Allocation and flow of funding include the following functions: 1) determination of contributions and entitlements; 2) collection
of contributions; 3) pooling of funds collected; 4) purchasing of services; 5) provision of services.
The attainment of social insurance principles will also be influenced by the way the benefit flow is ensured. These are the
important elements of such a flow: 1) services received and fulfillment of basic healthcare needs; 2) availability of quality
healthcare and freedom for members to choose healthcare facilities; 3) the level of direct co-payments from members for
healthcare services covered by insurance.
The equity principle stated in the NSSS Law postulates that members receive services according to their medical needs,
not according to their contributions. This includes equity of financial and physical access to healthcare. It also means equal
protection from financial risks when in need of healthcare, standardized quality of healthcare, and relief of financial burdens
that may constrain members from benefiting from healthcare or even impoverish them. The respective wording of Article 19
of the NSSS Law is as follows:
(1) Health insurance shall be administered on a national scale based on the principles of social insurance and equity.
(2) Health insurance shall be administered with the aim of ensuring that members receive the benefits of healthcare and
protection so as to fulfill basic health needs.
The NSSS Law dictates that four points are to be further regulated in the new PresReg NHIP: 1) membership, 2) amount of
contribution, 3) benefit package, and 4) provision of services by service facilities.
These four points are very much interlinked with other issues stipulated by the NSSS Law that are listed in Table 5.
75
Article 2 of the NSSS Law No. 40 / 2004
76
Article 19 point 2 of the NSSS Law No. 40 / 2004
37
Table 6: Further Legislative and Regulative Requirements Stipulated by the NSSS Law
Issue Further legislation and regulation as stipulated Reference Current status

by the NSSS Law in NSSS Law
Social health A new law for the development of Executing Agencies Article 5 Formulation
insurance carrier(s) of NSS (BPJS) and the fund allocation flow (“carrier law”) stage
Enrollment Presidential Regulation on Enrollment of Members Article 13 Formulation

of members in Stages by Employers to Executing Agencies of point 2 stage
NSS / BPJS
Recipients of social Government Regulation on Recipients of Social Article 14 Formulation

assistance Security Assistance point 3 stage
Management Government Regulation on Procedures for Managing Article 47 Formulation

of funds and Developing Social Security Funds point 2 stage
All of these are currently in the formulation stage. Ideally, technical policies for the national health insurance program (NHIP)
should be formulated after the above law and 3 regulations are ratified, so that alignment and concord can be achieved more
easily. With this consideration, the formulation of the PresReg NHIP needs to cautiously take into account the options that
are being proposed in the formulation of the above law and 3 regulations.
The PresReg NHIP stipulated by the NSSS Law will affect all stakeholders involved in health financing and service provision
by changing their roles and relationship in a fundamental manner. Furthermore, the NSSS Law will bring new actors into the
management of health services, thereby redefining the working relationship among old and new institutions as well.
Consequently, the active participation of stakeholders in the formulation of the PresReg NHIP and the ensuing collective
decision making process is essential and directly influences the effectiveness of the PresReg NHIP in the implementation
stage. Consensus among stakeholders in the formulation may prevent the PresReg NHIP from having to face a judicial
review by the Supreme Court as regulated by Article 31 of the Supreme Court Law.
4.3.4. The Proposed Presidential Regulation on the National Health Insurance System
The purpose of the new regulation
The NSSS Law stipulates further regulation of the technical aspects of the implementation of the NHIP by a new PresReg
NHIP in the articles shown in the table below. The tasks of the new Regulation can be put into three groups: membership,
benefits, and healthcare provision.
38
Table 7: How the NSSS Law Delegates NHIP Regulation to a Presidential Regulation
Aspect NSSS law Items to be regulated by the PresReg NHIP

articles
delegating
regulation
Membership 21 point 4 Point (1): Membership of health insurance remains valid for a maximum of
and Contributions 6 months after member has been laid off.
Point (2): In the event as stated in Point (1), if the said member is still
unemployed after 6 months and judged to be economically handicapped,
his/her contributions will be paid by the government.
Point (3): For members with a complete and permanent physical disability,
and those who are economically handicapped, contributions will be paid by
the government.
27 point 5 Point (1): the amount of health insurance contributions for members receiving
wages is determined as a percentage of wages up to a certain limit, which,
in stages, shall be the joint responsibility of the employee and employer.
Point (2): the amount of health insurance contributions for members who
do not receive wages shall be determined based on a nominal amount that
will be reviewed periodically.
Point (3): the amount of health insurance contributions for recipients of
contribution assistance shall be determined based on a nominal amount
that will be reviewed periodically.
Point (4): the wage limit referred to in Point (1) shall be reviewed periodically.
28 point 2 Point (1): An employee who has more than 5 family members and who
wishes to include other family members is required to pay an additional
contribution.
Benefit Package 22 point 3 Point (1): Health insurance benefits that are personal healthcare in nature
take the form of health services, which include health promotion, preventive
and curative treatment, and rehabilitation, including medicines and medical
consumables used as necessary.
Point (2): For types of services that may give rise to misuse of services,
members shall be required to share costs (co-payments).
26 Types of services not covered by the Executing Agency of NSS (BPJS)

shall be further governed by a Presidential Regulation.
Healthcare Provision 23 point 5 Point (3): In the event that in a region there are no adequate health facilities
available to meet the medical needs of a number of members, the Executing
Agency of NSS (BPJS) is required to provide compensation.
Point (4): In the event that a member requires in-patient treatment in a
hospital, the class of service provided at the hospital shall be based on a
standard class.
These three aspects are to be regulated in consideration of the principles of equity and needs coverage of the NSSS Law,
Article 19.
39
The implementation of the NHIP as an integrated system of financing and personal healthcare provision involves structures
and functions that administratively encompass the whole nation. The PresReg NHIP connects the rights of Indonesian
citizens as members of the NSSS to the Executing Agency of NSS (BPJS) as the institution that is authorized by its members
to purchase health services from healthcare facilities. It also connects members to healthcare facilities that provide services
according to the agreement between the Executing Agency of NSS (BPJS) and healthcare facilities.
The importance of the relationship of PresReg NHIP with other laws and regulations
The PresReg NHIP should be formulated in an aligned and harmonized way with stipulations in the NSSS Law including its
implementing regulations (see Table 6 for directly related NSSS Law articles), and the Law on the Executing Agency of NSS
(BPJS) and the implementing regulations that follow from the latter.
In addition to this, the substance of the PresReg NHIP must be harmonized with existing regulations on technical arrangements
of health insurance (please refer to Table 8). According to Law No. 10 / 2004, the law that regulates the formulation and
establishment of laws and governmental implementation regulations, some of the existing technical regulations on social
health insurance rank higher in the legal hierarchy than the NSSS Law. The NSSS Law does not amend any of the other laws
related to social security programs. Therefore, to enable the new technical regulation on the NHIP to work effectively for
entire population groups, the substance of the new regulation must be much better than that of existing ones. This means
that both funding and benefit mechanisms of the new program must improve or even correct current practice. If this precondition
is not in place, existing programs will continue to serve their customary members, and fragmentation will continue to exist.
Other laws and regulations with an influence on social security are listed in the table below:
Table 8: Relationships of PresReg NHIP and Other National Laws
Aspect of the PresReg NHIP Related laws

Membership and members’ contributions Human Rights Law
Manpower Law
Indonesian Citizenship Law
Population Administration Law
Essential Personnel Law
Law on Indonesian Workers Abroad
Indonesian Military Law
Indonesian Police Law
Law on Social Welfare for the Elderly
The benefit package Proposed revised Health Law

JAMSOSTEK Law
Civil Service Law
Military and National Police Law
Healthcare provision Proposed revised Health Law

Medical Practice Law
Proposed Hospital Law
Law on Local Government
(At present, Indonesia does not have a Nursing Law)
40
Harmonizing laws and regulations related to the PresReg NHIP provides a better chance for the latter to be implemented
more effectively. This will open a way to improve, and where necessary correct, current practice in accordance with the
principles of social insurance and equity.
4.3.5. Questions to be addressed by the presidential regulation on national health insurance

The success of health financing through SHI will depend on a well-aligned system of laws and regulations
Figure 5 depicts the flow of funds and the flow of benefits in a system of social health insurance. Furthermore, it shows that
the new laws and the residential regulation that are in the making affect the system at different points.
Whereas the laws will shape the insurance funds and the set-up for service providers, the presidential regulation will determine
how money flows from the individuals through the funding mechanism, and how benefits from the system (coverage and
services) will flow back to the insured individuals.
The key message is that in order to ensure steady and effective flows, the utmost attention must be given to harmonization
of the laws and the presidential regulation, so that one functioning mechanism can be shaped with these different tools.
Another important message is that the benefit flow will depend on participation and choice of the insured at all stages of the
system. The presidential regulation has a key role in ensuring this participation and choice.
The following sections will look at the role of laws and regulations in more detail, highlighting the open questions that need
to be addressed in the context of each.
Figure 4: Laws and the Presidential Regulation in the making affect different parts and functions
of the system
fund flow
collecting funds
Participation
& choice
pooling funds benefit flow
Participation
coverage
& choice
purchasing services
Participation control regulation

coverage
& choice
providing services
health services
insurance payments
Presidential
Law
regulation
41
Regulation of funding flows - the role of the PresReg NHIP

Presently, the proposed Law on the Executing Agencies of NSS (RUU BPJS) that will regulate the function of funding flows
in the NSSS is still in the formulation stage. There are two main issues to be considered when formulating PresReg NHIP:
1) The reorganization of the pillars of social security as a whole, and
2) How the social security program will be implemented at the local government level 77.
The reorganization of the pillars of social security will provide clarity about the scope of the executing agencies’ (BPJS) role
and authority in the NSSS program. Particularly for the pillar of SHI, currently the authority is given to two executing agencies,
PT JAMSOSTEK and PT ASKES, to serve formal sector workers except for the military and police in active duty. Furthermore,
the decision - yet to be taken - to include or separate health insurance programs for active military and police personnel into
one pillar of social security will affect various technical policies concerning the NHIP.
The decision of the Constitutional Court on 31 August 2005 provides an opportunity for local governments to implement
social security programs within the corridor of the NSSS. Up to now, consensus has not been achieved whether local
executing agencies of NSSS should be a subordinate subsystem, or independent alternatives, or additional to the national
program that will be carried out by one or more NSSS executing agencies.
The choice is still to be made between two alternative models to regulate the function of funding flows: as a single
pool or as multiple pools.
Pooling of SHI funds in a single executing agency can be done according to one of two models, based on financial accounts:
(1) a single account for all member groups, and (2) multiple accounts for various groups of members.
Pooling of SHI funds by several executing agencies can also be achieved through two models, based on segmentation of
members: (1) vertical pooling, and (2) horizontal pooling. An example of vertical pooling is the current system of state-
provided health insurance: in vertical pooling the national executing agencies provide nationwide health insurance programs
to different member groups. Horizontal pooling, on the contrary, applies to member groups in a province, district or township.
Figure 5 shows the options for pooling in the form of a decision tree.
Figure 5: Options for Pooling of SHI Funds
universal SHI
to be
singular fund decided multiple funds
to be to be
decided decided
several accounts vertical pooling horizontal pooling

1 account for population nationwide by local pool
groups population groups
77
The decision of the Constitutional Court on 31 August, 2005, on the question of how the national and local government should partition their roles in SHI
42
In order to choose the most suitable of the alternative models, the ultimate goals of the NSSS should be kept in mind. The
new PresReg NHIP should be conceived in such a manner that it applies to all existing NHIP executing agencies and any
possible model of pooling. Regardless of the model that will be chosen, there should be uniformity of technical regulations
concerning planning, purchasing of services, supervision, recording, reporting and evaluation.
Funding flows result in healthcare provision, a function that does not fall under the immediate responsibility of the executing
agencies. There are five laws that govern health service provision: the Health Law, the Local Government Law, the Medical
Practice Law, the proposed Hospital Law and the proposed Nursing Law. All five are required to regulate the quality and
competency of healthcare facilities.
Among the issues to be considered when formulating the new PresReg NHIP related to membership and the function of
funding flows are the following aspects:
1. The principle of solidarity: fulfilling members’ rights to fair financial access (membership) by preventing the system
from adverse selection, cream skimming and exclusion of any group in the community
2. The principle of portability: continued SHI coverage (membership) regardless of change of work place, change of
residence, being laid off or retired
3. Members’ rights as dictated by Articles 2 and 19 of the NSSS Law: availability of standardized healthcare that is
physically and financially accessible)
4. Establishment of a correction factor (equalization factor) for pooling of funds, to ensure equity among regions in
Indonesia. Regional variety in financial ability and availability of services need to be corrected by this factor so that
funds can be more evenly distributed.
Membership and contributions - open questions in the NSSS law

Membership - Attention to detail is crucial.
Article 20, point 1 and 2 of the NSSS Law, dictates that:
(1) Participants in the health insurance scheme are all people who have paid contributions or whose contributions have
been paid by the Government and
(2) Family members of participants are entitled to receive health insurance benefits.
Limitation of membership regarding payment of mandatory contributions will be inconsistent with Article 2 of the NSSS Law:
‘The National Social Security System is administered on the basis of the principles of humanitarianism, benefit, and social
justice for all citizens of Indonesia’.
Efforts to overcome limitations, particularly regarding geography, economy, and steps toward the fulfillment of the right of all
Indonesians to NHIP should be included in short, medium and long-term implementation plans for NHIP. The formulation of
these plans should involve all the stakeholders who will take part in implementation. Mutual understanding and commitment
to follow the roadmap will prevent unnecessary future conflicts or judicial reviews.
The NSSS Law has dictated various stipulations to accelerate fulfillment of citizens’ rights to NHIP:
1. Using a family approach, ranging from the nuclear to the extended family, as stated in the Elucidation of the NSSS Law:
Article 20:
Elucidation of point (2)
Family members are legal wife/husband, children, stepchildren from a legal marriage, and officially adopted
children, up to a maximum of 5 persons.
Elucidation of point (3)
The meaning of other family members in this provision is the fourth child and successive children, father,
43
mother, and in-laws. To include other family members, an employee provides recommendations from attorney
to make additional contributions to the Social Security Executing Agency as established by this Law.
2. Making it obligatory that the government pays the contributions of the poor and the totally permanently handicapped.
3. Extending the membership period to six months after a member has been laid off. If after that period the member is still
unemployed and is poor, then the government will pay his/her contributions.
The three stipulations above are still very normative and need to be interpreted more operationally in the PresReg NHIP. To
operationalize these stipulations is not easy. Careful considerations and consensus among stakeholders are necessary.
In operationalizing these stipulations the following questions need to be answered, using precise calculations and analysis:
1. Family membership needs to be defined in operational terms. This may require careful attention given the fabric of the
Indonesian society. Examples of questions of detail with far-reaching consequences are:
• How many children are covered? What will be children’s rights when they have become adults but are still at school?
Are children who have entered working age but still unemployed to be considered as dependents, or categorized as
poor?
• If the number of nuclear family members is less than four, should then the father/mother/in-laws pay additional
contributions?
• What are the rights of working husband and wife (2 NHIP members in one family)? Should both pay contributions or
only one of them? If their wages/salaries are not the same, may the one with the lower or higher wage/salary pay the
contributions?
• May a working child under the maximum age of being considered dependent become a member in place of their
parents who are unemployed?
• Are retired parents living together with an under-adult working son/daughter to be considered as separate members
of the NHIP, or are they to be considered as dependent of the son/daughter?
• How about a member who has more than one wife? What if the wife/wives work and has/have a salary? How is the
calculation of the dependence in this kind of family?
2. Membership of the poor
• Who has the right to be categorized as poor? The new PresReg NHIP must be harmonized with the Government
Regulation on Recipients of Subsidized Contribution.
• Are indigent and isolated tribes categorized as poor?
Amount of contribution - Accurate calculations and consumer choice are crucial issues to be
discussed.
There are three stipulations that regulate the amount of contribution as dictated by Article 27 of the NSSS Law:
1. The amount of health insurance contributions proportional to wages/salary for members shall be the joint responsibility
of the employee and employer; the proportion of the minimal and maximum limits of the contribution will be regulated in
the PresReg NHIP.
2. The amount of health insurance contributions for members who do not receive a wage shall be determined based on a
nominal amount that will be reviewed periodically and regulated in the PresReg NHIP.
3. The amount of health insurance contributions for recipients of contribution subsidy will be regulated in the PresReg
NHIP and harmonized with the Government Regulation on Recipients of Subsidized Contributions.
Calculation of the amount of health insurance contributions should also consider stipulations in the proposed Law on the
Executing Agencies. Variations among NHIP executing agencies should also be carefully considered.
44
Is competition among executing agencies of NHIP allowed, and if yes, are variations in the
amount of contributions also allowed as a decisive factor for winning the competition?
The NSSS Law dictates periodic evaluation of the amount of NHIP contributions. The cost of healthcare services is greatly
influenced by technology and their cost is influenced by inflation (often even higher than the inflation rate of common
commodities), so that increases in contributions will frequently have to be considered. A high frequency of contribution
increases will not be popular among the public and the government. Accuracy in actuarial calculations of financial risks
and projected funding of NHIP will significantly stabilize the system.
Benefit packages - How to make funding reliable, and how to harmonize the notions of choice
and equity?
Articles 22 and 26 of the NSSS Law stipulate the following concerning benefits to be ensured by the NHIP:
1. Personal healthcare services are to be covered
2. Promotive, preventative, curative and rehabilitative health services, including medicine and consumables are to be covered
3. Some healthcare services are not covered
4. Members are required to make co-payments for services that can induce over-utilization
5. Standardized in-patient services in hospitals are covered.
Further stipulations of NSSS Law articles 22 and 26 need to be formulated in the PresReg NHIP, considering the following:
1. The NHIP system capacity to fund, and its capacity to ensure sustainability of the program
2. Fulfillment of the solidarity principle
3. Fulfillment of the principle of portability; this means that the concept of “money follows the member (patient)” must be
affordable.
The open questions emerging from these stipulations are not of principle, but nonetheless require much political attention and
sensitivity. IIt needs to be decided how adequate and continuous funding can be achieved, without which it will be impossible to
provide the reliability and credibility that a system of risk protection needs. Both the paying members and the state as the source
of important subsidies must be embraced by a fund collection and management system imposing persuasive discipline.
The question of whether to allow competition between providers, a principle that can much increase efficiency and consumer
choice, needs to be pondered. At the same time, this must be harmonized with the need of SHI as a system of universal
solidarity to be equitable and portable throughout the area covered by the system.
Provision of healthcare services - standardization, involvement of the private sector, local

government, and professional bodies, and mechanisms for negotiation
The NSSS Law ensures the same right of public and private healthcare facilities to gain service provision contracts from
executing agencies. This leads to a number of requirements:
1. Standardized medical, nursing and pharmaceutical procedures for all levels and types of service
2. Standardized capacity in terms of infrastructure, manpower and equipment as part of contract requirements
3. Standardized procedures and formulae for pricing healthcare services
4. Standardized methods of payment to providers that can foster equitable distribution of healthcare manpower all across
Indonesia
5. Requirement for executing agencies and healthcare facilities to plan services for prospective payment to ensure availability
of services to members
6. Involvement of central and local governments in developing quality healthcare facilities that are appropriately distributed
throughout the country.
45
Article 24 of the NSSS Law dictates that negotiation for contract is to be done between executing agencies and the associations
of healthcare facilities in each region. Implementation of this stipulation needs further formulation regarding the following:
1. Requirements of associations to be involved in the negotiation process

2. Requirements regarding regional territory
3. Requirements regarding negotiation mechanisms including final decision-making in case of a deadlock between
associations and executing agencies
4. Scope of negotiations
Negotiation between executing agencies and healthcare facilities may involve the following two alternatives:
a. Negotiations with executing agencies are done by representatives of the medical professions (doctors and dentists)
and healthcare facilities to set tariffs for healthcare services, and methods of payment by executing agencies.
b. Negotiations are limited to rules, norms, standards and tariffs for each kind of service. However, final negotiation concerning
tariff of services and the signing of contract are done directly between the executing agencies and individual healthcare
facilities.
The NSSS Law also stipulates mandatory compensation by executing agencies in cases where healthcare facilities are
unavailable to members who need them. This compensation takes the form of cash payment to the member. Further formulation
of a compensation mechanism and the amount of compensation to be given need to consider abuse of these stipulations by
members for purposes other than medical need. This compensation can be in the form of the costs of referral to a higher
healthcare facility outside the region and to ensure optimal healthcare services.
Discussing regulations for providing healthcare in facilities abroad for members living or travelling abroad is necessary and
needs to be negotiated with countries where many Indonesian citizens use health services. In the preliminary stages, payment
for healthcare services abroad can be limited to payment by executing agencies to healthcare facilities in Indonesia or by
establishing maximum limit of coverage. For communities living in regions bordering neighboring countries, mechanisms for
the provision of healthcare services in facilities in these countries also needs to be arranged.
Furthermore, the NSSS Law has changed the payment mechanism for healthcare services from traditional budgeting in
government healthcare facilities to contract mechanisms. The staging of these changes needs to be formulated, as it will
affect the payment of services in government healthcare facilities by the executing agencies.
4.3.6. Closing Remarks

The new PresReg NHIP will regulate the implementation and financing of personal healthcare services through contract
mechanisms. This new system of healthcare service provision will also create new structures, functions and relationships
among actors. Therefore, the formulation of the PresReg NHIP needs to be done carefully and meticulously, supported by
accurate analyses, and involving all stakeholders.
The effectiveness of this PresReg will much depend on:
1. Careful choice of actors and institutions who will actively participate in the new system
2. Availability of detailed and effective regulations
3. A policy framework integrating public health policy with other social security programs
4. Clearly and decisively defined relationships among actors
5. Clarity of information provided to actors
6. Clarity of performance indicators for each actor, including clarity of supervision mechanisms.
46
4.4. Pensions
In the absence of comprehensive social insurance providing lifelong old age benefits, the reform agenda in Indonesia is
open for innovative solutions. Yet, a limiting factor for the universal coverage of insurance principals is the sheer size of
the informal sector and the high level of poverty, as many workers will not be able to pay contributions regularly or any
at all. Private savings and pension plans will also have to play their role averting old age poverty in the future, but only
for those affluent enough to pay the necessary premiums. So the main question remains, which reform option could
solve this problem for the major part of the working population? In the end, a combination of various schemes for old
age protection is the most viable option for Indonesia:
- Promoting comprehensive industrialization and formal employment together with enforcing general compliance
would alleviate the participation deficit in the JAMSOSTEK provident fund scheme. Therefore it could be advisable
to complement JAMSOSTEK’s mandatory saving scheme with contributory social pension insurance. Social
insurance pension requires personal contributions from formal employment income. Benefits are related to
contributions, pension entitlement is not poverty resistant. Informal sector is difficult to integrate.
- National basic income requires no direct personal contributions as it is tax-financed. Benefits are oriented on
minimum wage income level. No means-testing, poverty resistant, no problem for informal sector integration, but
probably discouraging movements to the formal sector. Financing might prove difficult within the present context
(high number of poor, low tax base).
- Social Minimum requires no direct personal contributions as it is tax-financed and part of social assistance (public
social welfare) and complementary to social or private insurance provisions. Benefits are oriented on minimum
living standards of society. Poverty resistant if timely adapted to needs. As means-testing is required, informal
sector difficult to integrate.
4.4.1. Current Situation

Before the social protection system proposed by Law No. 40/2004 was drafted and the social insurance principle became
the cornerstone of future reform design, public options for risk provisions against old age were not only very limited in
participation but also in scope. A public pension system in the sense of social insurance, i.e. providing lifelong monthly
payments, only exists for government employees (civil servants, armed forces and police). JAMSOSTEK, the mandatory old
age benefit system for employees in private enterprises above a certain staff size and total payroll, provides only for
accumulation of defined contributions in individual
accounts of a provident fund scheme. In case of
retirement, members receive a lump sum payment from
their individual account related to their contributions plus
interest. Thus, the JAMSOSTEK scheme cannot be
considered as an insurance but as mandatory savings
only, since risk pooling does not take place and longevity
risks are not covered. Furthermore, many have to draw
from these savings even before retirement in case of
unemployment or when other risks occur 78. Besides
these public provisions, a large number of pension plans
from private insurance companies and banks exist
offering old age provisions in the form of defined benefits
or defined contribution schemes.
78
Lindenthal (2004: 23).
47
Finally, the mandatory severance benefits payable under Labor Law No. 13/ 2003 should be mentioned. The law requires
that dismissed workers in all industries, both foreign and locally owned, receive up to several years in severance pay; even
employees who quit are entitled to payouts 79. From the side of private business it is often argued that these provisions
discourage new investment, as foreign shoe, garment and other labor-intensive manufacturers would rather go looking for
alternative production sites, especially in Vietnam. From the side of employees, however, severance payments replace
absent social insurance benefits in case of unemployment or retirement, rendering it politically difficult to change this law.
Evaluating the general situation, the main deficit of current old age provisions in Indonesia consists in limited coverage. Only
around 7 million out of an estimated formally employed work force of 27 million or about 26 percent in the private sector
participate, together with 4.6 million enlisted government employees. This makes up only about 20 percent of the total
workforce, as many small and medium scale entrepreneurs do not participate in the JAMSOSTEK scheme. Private pension
plans play only a diminutive role with 1.7 million participants. When considering that the formal sector employs only about
one third of the total workforce of around 100 million, this would leave about 60 to 65 million informal workers outside
participation of these retirement schemes 80. The problem of non-eligibility of old age benefits for informal workers is further
aggravated by the absence of a comprehensive social assistance scheme providing a minimum basic income which could
(at least partially) replace social insurance entitlements.
4.4.2. Reform Options

Positively speaking, in the absence of comprehensive social insurance providing lifelong old age benefits, the
reform agenda in Indonesia is wide open for innovative solutions, as there are only few limiting path dependencies
to respect. In particular, reform options are not burdened with an expensive entitlement legacy, which has made reforms so
difficult in many countries of the OECD world. Nonetheless, the sheer size of the informal sector and the high level of poverty
in general severely endanger the application of insurance principles, as many workers will not be able to pay contributions
regularly or any at all. From an actuarial point of view, it would also be difficult to calculate contribution obligations and
entitlement for benefits, as informal workers often tend to work periodically and part-time only. Regular income from continuous
full time employment, the basis of social insurance development in most industrialized countries, is still rather the exception
in an economy dominated by agriculture and small-scale entrepreneurs, as is the case in Indonesia.
In consideration of these facts, an old age crisis is to be expected, especially, as - forced by rural exodus and
urbanization - traditional family assistance schemes for the elderly will gradually lose their former importance.
Furthermore, Indonesia will remain in a demographic “grace period” for only about 20 years before dependency rates (retired
persons in relation to workforce) will slowly but progressively increase. This period, however, could be slightly extended by
raising retirement from the current age of 55/56 to a higher age threshold.
Private savings and pension plans will also have to play their role averting old age poverty but only for those affluent enough
to pay the necessary premiums. So the question remains, which reform options could solve this problem for the major part
of the working population? The following options are possible:
1. Following the European example in promoting comprehensive industrialization and formal employment together with
enforcing general compliance would certainly alleviate the participation deficits in the JAMSOSTEK provident fund
scheme; but it would only partially solve the problem of old age poverty because lump-sum payments can be misused
by recipients or may not be sufficient regarding rising life expectancy. Therefore, it could be advisable to complement
JAMSOSTEK’s mandatory savings scheme with contributory social pension insurance. Contributions from low-wage
earners could be subsidized so that a certain level of pension entitlement would be achieved.
2. Alternatively, a tax-financed universal basic pension scheme could be introduced with pensions based on a fixed
percentage of local minimum wages. This solution would not only solve problems of integrating the informal sector but
also that of the lifetime poor 81. However, there are also certain drawbacks to consider. First, like all universal schemes,
79
Martineau und Guérard (2005: 17).
80
Martineau and Guérard (2005:13)
81
Lindenthal (2004: 36).
48
it could be very costly to guarantee such a basic pension considering the high percentage of poor and near-poor
population. Second, the option for a non-contributory pension would create negative incentives for participation in
contributory social insurance, so this solution could be a hindrance against turning informal into formal employment.
3. Another approach to solve the problem of old age poverty could be seen in the future role of social assistance. A
universal entitlement to a tax-financed social minimum, which could be introduced, subject to means-testing and targeted
to all persons (or households) with income below the national poverty line. This solution would solve the problem of
poverty on a general basis as well, but without violating incentives to join social insurance. However, this option is no
substitute to the first option but rather a complementary option.
Organization
As mentioned above, only civil service employees are covered by existing pension schemes complying with ILO social
security standards in that periodic and predictable payments against the loss of income after retirement are made. The
scheme for civil servants is provided by the GOI with Taspen being the paying agent. However, these schemes have been
notoriously underfinanced in the past, and considering the rapidly ageing membership among civil service, the financial
viability of future pension entitlements is doubtful. It is, therefore, imaginable to integrate Taspen and ASABRI into one
national pension scheme under administration of JAMSOSTEK. Alternatively, Taspen could become the responsible
administrative body for old age (see chapter 4.2). This general pension scheme could be composed of social insurance
and provident fund elements as well. These contributory old age provisions could be complemented with a universal social
minimum income for the poor requiring the setting up of a new institution on the state as well as the district level for the
integrated administration of all social assistance measures (see chapter 6).
Benefits and Financing

Due to high levels of informality and poverty among the Indonesian population on the one hand, and limited financial
resources of the state on the other, it will be of strategic importance to promote the capacity of wage earners for
paying contributions (and taxes) to support the future system of social security. As prevention against demographic
risks, pension insurance should be gradually complemented with funding elements. However, when starting, for a transitory
period the pay-as-you-go way of financing pensions will have to be dominant as otherwise there will be many participants
without sufficient entitlements for pensions from their own contributions. When the new national pension scheme will be fully
established, the share of pay-as-you-go financing should be gradually reduced in favor of the funding principle 82.
Basically, pension benefits should be related to earned income and former contributions. It will be necessary, however, to
subsidize pension entitlements of low-wage earners at least for a transitory period. This could be done in the form of cross-
subsidization from contributions of other members or from general taxation. In the case that a social minimum income is
provided from social assistance, there should be a clear distinction concerning the level of basic pensions from the contributory
scheme.
The Role of Private Insurance

As mentioned above, private insurance actually plays only a minor role for private old age provisions. This could change,
however, when the general income level of the middle class increases, more wage earners will be interested (and
capable) to invest in supplementary individual pension plans. Such efforts could be sponsored from the side of employers
via company pension plans or from the state via tax treatments as they render the public pension system financially more
stable and generate incentives for additional savings. It is thinkable, for instance, that participants may be opting out from
public provident funds in favor of private pension plans.
For the informal sector, private provisions against old age poverty are especially important as in this case they do
not only have a complementary role but have to substitute benefit entitlements for which informal workers are not
eligible (see chapter 7).
82
It should be borne in mind, however, that funding is no magic wand against all demographic risks (see chapter 4.1.).
49
50
5. Unemployment
Compensation
What Kind Of Unemployment Assistance Do You Want?
Unemployment Insurance (UI): Benefit of limited duration based on a compulsory para-governmental insurance requiring
employees/employer’s contributions and general public funds. Related to the last income. Requires a minimum
employment or contribution period. Applicants are without work and currently searching for work.
Unemployment Assistance (UA): Like UI, access to UA requires that applicants are without work and currently searching
for work. Amounts are income/wealth tested flat rates usually lower than under UI. Often fully financed by public funds.
Applicants have been employed for a certain period and have to be eligible for the labor market. Duration is unlimited or
depends on the existence of SA.
Social Assistance (SA) is an income/wealth tested financial assistance paid by taxpayers to people who are unable to
support themselves, hence not generally restricted to unemployed. SA thus comes close to UA, the amount is equal or
only a little lower than UA.
Individual Saving Accounts (ISA): Financed through mandatory contributions by employers and/or by employees.
Unlike UI and UA, ISA are personal based funds available only to the individual under consideration. Remaining parts
of the fund can be taken after retirement. Can be combined with the general principle of unemployment insurance in a
“Solidarity Fund”.
Severance Pay (SP): Lump-sum one time payments from employers to their laid-off employees. No redistribution
among different firms and different workers. Dismissed workers do not need to be eligible for the labor market. Severance
pay can either be mandatory or agreed upon in private labor contracts. Related to the number of working years in the
company (one month’s salary is paid for each year).
Layoff taxes (LT) are paid by firms in cases of dismissals of workers. Unlike SP, layoff taxes do not go to the workers
dismissed themselves, but to general funds used to finance UI or UA. Hence, the eligibility side of layoff taxes resembles
unemployment insurance systems, whereas the financing side resembles SP. LT does not exist in any country yet.
Summary and Conclusion:

- Unemployment insurance, unemployment assistance, individual saving accounts for unemployment benefits,
mandatory saving accounts for severance payments and layoff taxes do not exist to date in Indonesia and social
assistance plays only a minor role.
- Severance pay, however, is quite important in Indonesia.
- Preliminary recommendations for unemployment compensation in Indonesia include an integrated system of
mandatory individual saving accounts and mandatory savings for severance pay, both of them complemented by
solidarity funds.
51
This chapter discusses options to implement a system of unemployment compensation in Indonesia, i.e. any kind of support
that is somehow linked to former employment and/or to proving suitability for the labor market. Section 5.1. discusses the
most important instruments and provide a brief overview of their implementation in different countries. Section 5.2. is concerned
with various drawbacks of the different instruments. As for this, the focus lies on negative incentive effects and distinguish
between adverse effects for (potential) employees on the one hand and for firms on the other hand, and there is a summarize
of the main results of empirical studies. Finally, the situation in Indonesia is described in section 5.3. along with some
preliminary recommendations in section 5.4.
5.1. Instruments for Unemployment Compensation and Implementation in

Different Countries
This section focuses on six different instruments for unemployment compensation as summarized in Annex 5: Unemployment
insurance, unemployment assistance, individual saving accounts, severance pay, and layoff taxes. There are also some
brief remarks on social assistance (i.e. payments that are not linked to former employment but only to the applicant’s wealth)
because social assistance influences the incentive effects of any kind of unemployment compensation. The various incentive
effects of the different instruments will then be discussed in section 5.2.
5.1.1. Unemployment Insurance (UI)

Compared to other income support systems, two aspects are crucial for UI: first, to get UI, an individual must meet the
minimum employment or contribution period, typically about six months within the last year. In addition, the individual must
be without work and currently be seeking work (eligibility for the labor market). UI is usually a percentage of the last income
where an upper limit applies 83.
According to the required criteria, definition problems and monitoring problems arise with respect to informal work and the
eligibility for the formal labor market. The applicant’s or the family’s wealth, however, are of no importance for UI which saves
on administrative costs. UI is financed by contributions (payroll taxes) of employees or employers or both. It might also be
complemented by general public funds.
Compared to other instruments, UI has several advantages 84: first, because the number of members is high, UI provides
sufficient coverage. This also holds true for periods of serious and long-term economic shocks. Secondly, people pay money
in periods of high employment (and hence highly aggregated demand) and receive money in periods of low employment
(and hence low aggregated demand), so that total demand is smoothed by UI systems. Thirdly, in contrast to severance
payments (see below), unemployment insurance leads to no distortions in the incentives to substitute workers already
working for the company by new workers. Finally, people might be more willing to accept jobs with high employment risks,
which is a desired side effect, presupposing that high-risk jobs have higher positive externalities 85.
However, UI is restricted to smooth the income situation of formerly employed people, but not appropriate to protect the poor
in a society since getting UI requires that applicants have been part of the formal labor market for a sufficiently long period of
time. In fact, empirical studies show that the poor are disproportionately less unemployed than the rich in many developing
countries 86. The reason is that the poor simply cannot afford being unemployed for a long period of time and hence accept
any kind of low wage job.
83
The degree of UI can be captured by the Generosity Index measuring the product of the replacement rate (defined as the average percentage from the
last net income) and the recipiency rate (defined as the percentage of unemployed people getting UI). For instance, the Generosity Index is almost the
same in the United Kingdom as in the US, but the replacement rate is much higher in the USA whereas the recipiency rate is lower. Vroman and Brusentsev
(2005:63).
84
Vodopivec (2005a:3).
85
Acemoglu and Shimer (2000).
86
De Ferranti et al. (2000) for Mexico and Uruguay; Esguerra et al. (2001) for the Philippines, and Edwards and Manning (2001) for an overview.
52
5 • Unemployment Compensation
Given these caveats of unemployment insurance, it is straightforward that it exists in most highly developed countries, but
plays only a minor role in emerging and developing countries. With respect to the implementation in different types of
countries, basic facts are as follows 87:
OECD countries: Almost all countries have UI, the self-employed are usually excluded, replacement rates are between 40
percent and 75 percent of recent average earnings, with maximum rates of 80 percent in Sweden and 90 percent in Denmark.
The minimum requirement is usually a period of 6 months of work in the last year. Eligibility for the labor market applies, and
the system is usually funded by contributions from both employees and employers 88.
East Europe and transition countries in Central Asia: UI also exists in most of these countries, and requirements and
replacement rates come close to those in OECD countries. However, whereas employers pay for UI in all countries, employee
contributions exist in only 9 out of 21 countries considered.
Latin America and the Caribbean: UI exists in several countries 89. Replacement rates differ, but they are usually considerably
lower than in OECD countries. In some countries (Argentina, Brazil and Uruguay), replacement rates are linked to minimum
wages. In most countries the system is financed by both employers and employees.
Asia: UI exists only in a few countries, namely Bangladesh, China, Iran, South Korea and Taiwan 90, but only South Korea
and Taiwan offer coverage for workers of all firms (firms with less than 5 workers are excluded in Taiwan, however). Conversely,
Iran excludes persons who are privately insured, and China restricts coverage to workers in public sector enterprises and
some collective enterprises. Iran and Taiwan have replacement rates of about 50 percent, whereas China and South Korea
have flat rates of about 10 percent below minimum wages. Other requirements come close to those in OECD countries, and
systems are again financed by payroll taxes of employers and/or employees 91. In practice, UI plays only a minor role even
in those countries where it formally exists since the recipiency rate is much lower than in OECD countries. For instance, in
South Korea, temporary, daily and part-time workers are excluded, and the high percentage of people working in the informal
part of the labor market is clearly never covered. Empirical studies for South Korea estimate that only about 15 percent of all
spending on social protection for the unemployed is paid through unemployment insurance 92.
Africa: UI exists only in few countries, namely Algeria, Egypt, South Africa and Tunisia. South Africa and Egypt have
replacement rates of about 50 percent whereas Algeria and Tunisia pay part of minimum wages. Again, contribution systems
differ between payroll taxes by employees and/or employers 93.
5.1.2. Unemployment Assistance (UA)

Like UI, access to UA requires that applicants be without work and currently searching for work. In countries with both UI and
UA, UA sets in after a certain period, and amounts are lower than under UI. UA is usually not calculated as a percentage of
the last income but as a flat rate 95. Conversely to UI, UA requires not only that people have been employed for a sufficiently
long period and that they are eligible for the labor market, but also that the applicant’s and/or the family’s wealth is below a
critical threshold.
In the European Union, unemployment insurance, unemployment assistance and social assistance (see below) all exist
simultaneously. This implies that UA needs to be higher than social assistance since distinguishing between the two systems
would otherwise not make sense. Comparing the financing side to UI, general public funds usually play a higher role whereas
contributions by employers and employees play a lower role.
87
See Vodopivec and Raju (2002: 19-31) for a more comprehensive overview on the implementation of the different instruments discussed in section 5.2.
88
Exceptions are the USA, Iceland and Ireland where only employers pay, and Luxembourg where the system is funded by employees’ payments only.
89
These are Argentina, Barbados, Brazil, Chile, Ecuador, Mexico, Uruguay and Venezuela.
90
91
In China, the system is financed by employers and state subsidies, but employees do not pay.
92
Vroman and Brusentsev (2005:139).
93
In Tunisia, the system is completely state financed.
94
See the overview below.
53
Principal facts around the world are as follows:
OECD countries: About half of all countries have UA, and almost all of them adopt UA complementary to UI. Exceptions are
Australia, Hong Kong and New Zealand who provide unemployment assistance, but no unemployment insurance. Benefits
are flat-rated over time, and are usually not related to the former income (means-tested minimum income). Exceptions
include Australia, Austria and Germany where payments are calculated as replacement rates (e.g. around 50 percent in
Germany) and hence depend on the former income. In countries where UI, UA and social assistance all exist simultaneously,
payments are always below UI but above social assistance. Eligibility conditions vary considerably among countries and
depend on whether UA comes closer to UI or closer to social assistance. In countries where UA comes close to UI and differs
simply in that replacement rates are lower, the eligibility conditions are the same as under UA, and UI sets in after a certain
period of time when UI terminates (for instance after 12 months in Germany in general). Additionally, it is often required that
the income of the household and total assets are sufficiently small which brings UA close to social assistance. Conversely to
UI, UA is usually not financed by payroll taxes of employees and employers but through general tax revenues.
East Europe and transition countries in Central Asia: In countries where UA is provided 95, it is almost never (Estonia
having the only stand-alone system of UA) adopted complementary to UI. Systems come close to those in OECD countries.
Other countries: Outside the OECD countries and Eastern Europe, UA hardly exists. Exceptions are Hong Kong, Tunisia
and Mauritius where benefits are means-tested minimum income flat rates.
5.1.3. Social Assistance (SA)

In contrast to UI and UA, the eligibility for social assistance is not generally restricted to those who are available for the labor
market. However, individuals who are healthy and young enough to work are often required to search for jobs and to do
some mandatory minimum low paid work. SA thus comes close to UA but the amount is often lower and equal to the amount
that is granted for those who are not or no longer available for the labor market. Definition problems and administrative costs
are thus similar to those for UA. Depending on the program, the income from low paid work may be kept additionally for
incentive reasons. Section 5.3. will only briefly touch on this subject; more details are given in chapter 6 which specifically
deals with social assistance.
5.1.4 Individual Saving Accounts (ISA)

As under unemployment insurance, eligibility to individual saving accounts requires that the applicant is laid-off and has
worked in the formal sector for a certain period of time. As UI, ISA are financed through payments by employers and/or by
employees. Contributions are mandatory as for unemployment insurance.
However, there is an important difference since individual saving accounts are personal based funds, which are available
only to the individual under consideration. Thus, ISA are in fact mandatory savings that can be used in cases of unemployment.
Eligibility thus requires additionally to UI that sufficient funds are available in the individual account. If funds are not used up
during periods of unemployment or if there are no unemployment periods at all, the remaining part of the fund can be taken
after retirement. The idea of individual saving accounts can be combined with the general principle of unemployment insurance
such that workers make mandatory contributions not only to their own individual saving accounts but also to a Solidarity
Fund used for unemployed people who do not have enough funds in their ISA yet. To the best of our knowledge, such a
system has been only implemented recently in Chile 96, but it is widely discussed in theory in literature (see below). Elements
of this system also exist in Brazil where eight percent of wages are deposited into an individual account 97, in Singapore and
in the Philippines 98.
95
These are Bulgaria, Czech Republic, Estonia, Hungary, Poland, Romania, Russia, Slovak Republic and Slovenia.
96
Conerly (2002), Vroman and Brusentsev (2005:157f.), Sehnbruch (2006) and Acevedo et al. (2006).
97
Vodopivec and Raju (2002:70).
98
Choon and Tsui (2003) for Singapore and Vodopivec (2005a:14) for the Philippines.
54
5.1.5 Severance Pay

Severance pay are lump-sum payments from employers to their laid-off employees. Thus, only (former) employees and their
employers themselves are involved, and there is no redistribution among different firms and different workers. Another
difference to UI is that dismissed workers do not need to be eligible for the labor market anymore.
Severance pay can either be mandatory or agreed upon in private labor contracts. They usually increase with the number of
years an employee is with a company, a typical formula being that one month’s salary is paid for each year. One problem of
severance payments is that they are not always credible since firms may be bankrupt. This problem can be mitigated by
introducing mandatory severance payment saving accounts as in Columbia (see below).
Principal facts around the world are as follows:
OECD countries: Severance pay exists in all European OECD countries as well as in Australia, Canada, USA, Japan and
New Zealand. They are not always mandatory by law; but if not, they are part of collective bargaining agreements (as in
Germany and the Netherlands, for instance). Severance payments apply to dismissals both for reasons of economic
redundancy and for personal reasons, but they are restricted to no-fault dismissals. They are financed by employers themselves,
in many cases accompanied by state assistance. Payments vary among countries, a typical payment scheme being that one
month’s salary is paid per year of service. Austria is the only OECD country that has established mandatory saving accounts
for severance pay ensuring that firms can in fact fulfill their obligations in cases of dismissal.
East Europe and transition countries in Central Asia: Severance payments exist only in the Czech Republic, in Hungary
and in Poland. Conditions come close to those in OECD countries.
Latin America and the Caribbean 99: Severance pay plays an important role and exists in most countries. Payments are
similar to those in OECD countries, but other conditions apply. First, state assistance is not provided. Secondly, dismissals
for economic causes are prohibited in all Latin American countries, except Argentina and Chile. Often, severance pay is
offered for voluntary quits. To overcome the problem described above, i.e. that severance pay often applies to firms who will
not be able to fulfill their liability obligations, Columbia was the first country to establish a system where employers need to
collect funds on mandatory saving accounts. This does not only improve the coverage of the system, but it also reduces real
wages since workers anticipate that severance pay will in fact be paid in case of economic redundancy 100. Mandatory saving
accounts for severance pay exist also in Argentina, Brazil, Chile, Ecuador, Panama, Uruguay and Venezuela 101.
Asia: Severance pay exists in Bangladesh, China India, Indonesia, Malaysia, Pakistan, the Philippines, Singapore, Sri
Lanka and Thailand. Again, state assistance for employers is never granted, and the conditions come close to those in
OECD countries. Severance payments in Indonesia are far above average conditions in Asia, and even high compared to
OECD countries (see below).
Africa: Finally, severance pay plays a minor role in Africa and exists only in Botswana, Libya and Tanzania 102.
5.1.6. Layoff Taxes

Layoff taxes are paid by firms in cases of workers’ dismissals, so that the financing side of layoff taxes resembles severance
pay. However, layoff taxes do not go to the workers dismissed themselves, but to general funds used to finance unemployment
insurance or unemployment assistance. Thus, with respect to eligibility, layoff taxes can be designed as either unemployment
insurance and/or unemployment assistance. In other words, the eligibility side of layoff taxes resembles unemployment
insurance systems, whereas the financing side resembles severance pay.
99
Ferrer and Riddell (2005) for an overview.
100
Kugler (2005) for an analysis of the economic benefits of the system.
101
Gill and Ilahi (2000).
102
Minimum period of service varies considerably (3 months in Tanzania compared to 60 months in Botswana).
55
In reality, layoff taxes as such do not exist yet, but the experience rating system in the USA includes elements of layoff taxes.
Experience rating means that the contributions to the general unemployment insurance fund depend partly on the number of
workers dismissed in the past, so that dismissal becomes more costly for firms who are not likely to be bankrupt 103. The
practical impact is usually considered to be weak, however.
Finally, job security regulations, which make it more expensive to dismiss workers, can be interpreted as substitutes for
experience rating. This however, is beyond the scope of this analysis 104.
5.2. Incentive Effects of Different Instruments of Unemployment Compensation

All of the instruments described above can be analyzed under two different objectives: First, the suitability to fulfill the alleged
aims, namely income smoothing and protection of the poor. Secondly, the instruments have negative incentive effects both
for employers and employees, and these negative incentive effects that may increase unemployment, decrease the money
available for active labor market policy, and weaken the economy as a whole. The suitability to mitigate income shocks and
poverty has already been briefly sketched above. In the following, incentive problems for workers as well as for firms will be
discussed. Of course, the distinction between incentives for workers and firms is somewhat artificial as, for instance, lower
effort incentives for workers will also negatively influence firms’ decisions to hire and fire workers for any real wage given as
they can expect lower productivity. Hence, when analyzing the impact of different instruments on firms’ decisions, it is
implicitly assumed that productivity levels are given.
5.2.1. Unemployment Insurance, Unemployment Assistance and Social Assistance

Distortions for (Potential) Workers’ Incentives
The basic incentive problems of unemployment insurance and unemployment assistance as well as of social assistance are
obvious: Since people get money when they are unemployed and pay money when they are employed, the incentive to
avoid unemployment is diminished 105. More precisely, the following incentive effects can be distinguished:
Effort incentives: Being insured reduces the incentive to avoid dismissal when being employed, and hence has negative
effort effects.
Incentives to invest in education: The incentive to invest in education may also be diminished, since the probability of
being unemployed is decreasing with higher education. The incentive is reduced when the income difference with and
without employment decreases.
Job search incentives: The incentive to search for a new job is also reduced for similar reasons. Theoretically, the total
impact of unemployment insurance on search efforts is ambiguous due to the countervailing effect that more resources
enable a more effective job search. Empirically, however, studies confirm that the negative effect dominates such that total
incentives for job search are diminished 106. Search efforts provide an important argument why benefits should decrease
over time 107: Since a long period of unemployment indicates that finding a well paid job is unlikely, compensation schemes
need to set higher incentives to accept low paid work when the period of unemployment increases, and this requires a
decreasing compensation rate 108.
103
See: e.g. O’Leary (2000) and Rösch (2007) for a description of the US system.
104
See: Heckmann and Pages (2000) for an empirical investigation of the impact of job security regulation on unemployment in Latin America.
105
Of course, this is true for all insurance contracts, so that the only question is how can these unwarranted incentive effects be mitigated.
106
Calmfors and Holmlund (2000).
107
As pointed out in section 2, such a decreasing compensation scheme is implemented in almost all OECD countries by the joint use of unemployment
insurance and unemployment assistance.
108
See: e.g. Wang and Williamson (1996) for a calibration using US data and Karni (1999) for an overview.
56
Impacts on wage formation: Both UI and UA increase the bargaining power of potential employees because they lead to
higher exit options - individuals who can rely on social protection can demand higher wages. This may increase real wages
especially for low income levels, and may hence lead to higher unemployment rates not only because of the three incentive
problems just described, but also because of higher wages for any productivity level given. It has often been pointed out that
the higher bargaining power for workers or labor unions may also make it much more difficult to recover from economic
shocks: Recovering from shocks requires reductions in real wages, and this may often be very difficult if unemployment
benefits can not be shifted downwards in the same period. The interaction between shocks and high unemployment benefits
has in fact been used to explain the persistence of high unemployment rates in many European countries 109.
For all of these negative incentive effects, it does not play an important role whether unemployment insurance or unemployment
assistance applies - what matters is the difference between the income situation with and without employment. However, as
described in section 2, unemployment assistance is often only granted when the family’s income is below a critical threshold,
and unemployment assistance then comes close to social assistance. This has two opposite incentive effects: On the one
hand, for those whose family income will be above the threshold, the negative incentive effects are restricted to UI since they
will not be eligible for UA anyway. This is especially important for countries that are less concerned about income smoothing
and mainly interested in protecting the poor. It is then better to exclude former workers with high family incomes. However,
it may also set unwarranted incentives for other family members who may reduce their working load or search effort to make
the applicant eligible for unemployment assistance.
It has already been emphasized in the sectio above that Unemployment Assistance (UA) combines elements of both
Unemployment Insurance (UI) and Social Assistance (SA): As under UI, eligibility requires that applicants have worked
before for a certain period; and as under SA, applicants need to show proof that the family’s income is below a certain
threshold. Since combining these two requirements is likely to maximize administrative costs (and potentially also the incentive
problems on the applicant’s side) we see no good reason for implementing an unemployment assistance system as this
would presumably maximize administrative costs: Eligibility to unemployment insurance requires proving suitability for the
labor market, but is granted independently of the financial situation of the family. Social assistance, on the other hand, is
granted regardless of the former and the future employment situation, and thus requires only evidence that the family’s
wealth is sufficiently low. Unemployment assistance, however, requires to control both the availability for the labor market (as
under unemployment insurance) and to monitor the wealth and income of the applicant and the family (as under social
assistance) and thereby maximizes administrative costs. Another weakness of unemployment assistance compared to social
assistance is that individuals have an incentive to pretend being unemployed even though they do not intend to work at all,
which also leads to a waste of resources.
In summary, there seem to be no good reasons why distinguishing between unemployment assistance and social assistance
is necessary; whenever unemployment insurance expires, it should be enough to ensure a minimum standard of living.
Removing unemployment assistance and restricting instruments to unemployment insurance and social assistance is also
consequent with respect to the two objectives of unemployment compensation, namely smoothing income streams and
protecting the poor. Following the insurance principle, the family’s wealth should be of no importance with respect to the first
objective. By contrast, the former employment situation should have no impact when the objective is protecting the poor.
However, the arguments against unemployment assistance do not contradict an important insight derived above, namely
that unemployment insurance payments should decrease over time to set correct incentives in searching for jobs and for
accepting low paid work. Admittedly, the difference between UI and UA in many countries can be interpreted as a device to
implement such a rate, which is decreasing over time (the high unemployment insurance is only paid for some time, and
afterwards the lower unemployment assistance rate sets in). However, it is straightforward that this objective can also be
reached with two instruments only - unemployment insurance and social assistance 110. Furthermore, keeping in mind that an
109
Blanchard and Wolfers (2000). However, there are also other authors including Solow (2000) who emphasize that other factors as restrictive monetary
policy and highly regulated product markets may be more important.
110
Vodopivec (2004b:5).
57
58
unemployment assistance system is not funded by contributions of employers and employees (this would make it an
unemployment insurance system), it is likely that benefits above minimum standards of living would be fiscally unsustainable
in developing countries anyway. Hence, payments should either depend on former employment and on the availability for the
job market or on the applicant’s wealth, but not on both of them.
Whereas these arguments vote for not adopting unemployment assistance at all in developing countries, comparing
unemployment insurance to social assistance shows that the former should also play only a minor role in developing countries.
Protecting the poor is obviously more important compared to smoothing income streams, and official unemployment and
poverty are not closely related in developing countries 111. However, it is worth being emphasized that this does not necessarily
mean that other kinds of income support targeted especially for those who have formerly been employed should not play an
important role in developing countries; see the assessment of severance payments, saving accounts and layoff taxes below.
Distortions for the Incentives of Firms

First of all, it is worth noting that it does not make a difference for the incentives of firms whether the money is spent for
unemployment insurance, unemployment assistance or social assistance; all that matters is how much money is paid by the
firms (or their employees) to finance the required funds, and under which circumstances the money is paid 112. In all systems
described in section 5.2., firms contribute to the funds proportional to the number of workers they employ, but they do not
contribute to funds in case of dismissals (see the experience rating system in the USA, however). It follows immediately that
the incentive to hire workers is diminished due to higher costs, and that the incentive to dismiss workers is much higher than
the socially optimal incentive: from the point of view of economic welfare, a worker should only be dismissed if his expected
productivity in other jobs (and hence the economic opportunity costs) is higher than his productivity in the current job (in case
of unemployment, his future productivity is even zero). By contrast, firms compare the worker’s productivity with his real
wage (including payroll taxes for unemployment insurance), and whenever the real wage is higher than the expected productivity
in future jobs (which is usually the case in layoffs), they have an inefficiently high incentive to dismiss workers. This can be
considered as the main economic problem of unemployment insurance.
Apart from this fundamental inefficiency, firms may also have an incentive to create wrong jobs and hire wrong workers.
Consider two jobs, one of them being very risky and promising high rewards when successful, the other one being more
conservative in the sense that a dismissal is very unlikely, but that rewards are moderate. Even in cases where the expected
reward is higher in the second case such that the second job dominates with respect to the average return and with respect
to lower risk, firms may nevertheless prefer creating jobs of the first kind since only the bright side is internalized whereas the
dark side (the dismissal risk) is partly externalized to all insurance payers and to the state 113. However, this argument is not
as straightforward as the incentive problem described in the previous paragraph because there are also countervailing
effects: In recent contributions, it is also argued that the higher incentive for risk taking caused by the externalization of costs
may just offset the effect that people are risk averse, and that high risk jobs tend to provide higher positive externalities 114.
Empirical Results 115

Most empirical studies confirm that higher replacement rates also increase the time of unemployment, i.e. they in fact
decrease the incentives to search for new jobs or to accept low paid work. This seems to be almost uncontroversial for US
data 116. For Europe, however, results are more mixed, and some studies find no impact at all 117. Analyzing only the impact
of the duration where unemployment benefits are granted (and not the generosity index as described in section 5.2.), most
111
Vodopivec (2004a: 9) and Rösner (2007:32f.).
112
In fact, empirical studies (see e.g. De Ferreanti et al. 2000) show that it does not make much of a difference whether employers or workers contribute to
unemployment insurance.
113
Karni (1999:24).
114
Acemoglu and Shimer (2000).
115
An extensive and systematic overview on empirical studies up to 2001 is provided in Vodopivec and Raju (2002:55-59, Table 4.4).
116
Card and Levine (2000), Hotz et al. (2002) and Juradja and Tannery (2003).
117
Roed and Zhang (2003).
59
studies also find a positive impact on unemployment 118. In particular, it is observed that the escape rate (i.e. the probability
of finding a new job) increases dramatically when unemployment insurance benefits expire 119.
Of course, it needs to be taken into account that empirical studies relying on international comparisons should be interpreted
carefully since there are many impact factors that can hardly be controlled for (for instance, the enforcement of programs is
different, other policy measures both inside and outside the labor market play important roles, and replacement rates can not
always be made completely comparable) 120. Unemployment rates have been heavily reduced in Denmark, Ireland and the
Netherlands even though there are strong unions and extensive systems of social protection, and the positive developments
for the three countries are usually attributed to other factors as a moderate wage growth, social dialogues, an active labor
market policy, part-time arrangements (for the Netherlands), and high foreign direct investments (for Ireland). However, all
three countries have also streamlined their unemployment compensation systems: they have reduced the duration, have
tightened the eligibility criteria, have improved their monitoring systems, have implemented stricter work search requirements,
and have adopted a more flexible definition of suitable work 121.
Summing up, it seems fair to say that the overwhelming majority of empirical studies confirm the intuitive incentive effects
pointed out in the theoretical considerations above.
5.2.2. Individual Saving Accounts (ISA)

Any system of unemployment protection faces the same fundamental trade-off between risk sharing and incentives: The
higher the degree of risk sharing, the higher the incentive problems. A full system of individual saving accounts may completely
eliminate these incentive problems, albeit at the expense of also completely eliminating risk-sharing among different potentially
unemployed people. This does not mean that the same result can be reached by simply eliminating unemployment income
support at all since such a system would still need to rely on social assistance to prevent an extreme form of poverty. The
main advantage of individual saving accounts is that it in fact implements a system of mandatory saving, and hence makes
self-protection obligatory at least to some degree.
The positive incentive effects of ISA compared to the traditional forms of unemployment protection (i.e. UI and UA) are
obvious: Since employees will get the money saved on their accounts even if they never become unemployed, the difference
between the income situation with and without unemployment is much higher than with traditional unemployment insurance
or unemployment assistance. The mitigating effect on disincentives refers to all problems discussed above as all of these
problems depend on the difference between the income with and without employment.
Of course, restricting unemployment income completely to ISA would leave those individuals, who have insufficient ISA
coverage, without (or with low) income. Hence, ISA should to be accompanied by solidarity funds where payments are not
used individually but are paid to a general fund available to all unemployed individuals who contribute to ISA, and hence also
to the fund. Such a fund obviously suffers from the same incentive problems as UI and UA, so that the combination of ISA
and solidarity funds combines obligatory self-insurance with elements of (also obligatory) traditional unemployment benefits.
Keeping both, the theoretical considerations and the empirical results on the inefficiencies of traditional insurance systems
in mind, it seems very likely that combining ISA and solidarity funds is an important step in the right direction 122. Note that the
traditional insurance system can simply be interpreted as the extreme case of ISA and solidarity funds, where payments to
solidarity funds amount to 100 percent while the principle of self-insurance is completely neglected. Thus, from the point of
view of economic theory, our argument is very simple: Combining ISA with solidarity funds allows for an “interior” solution,
and there is no reason why a corner solution with only unemployment benefits or mandatory individual saving should be
optimal.
118
Toomet (2005) and Biewen and Wilke (2005) who find a positive correlation form in Germany.
119
Meyer (1990).
120
121
See: e.g. the overview in Vroman and Brusentsev (2005:96ff.). Accordingly, Boone et al. (2000) find that optimal sanctions for abuse of unemployment
insurance payments are much higher than those usually observed.
120
See also: Orszag et al. (1999), Yun (2001) and Hopenhayn and Hatchondo (2002).
60
We strongly believe that the principle of mandatory self-insurance implemented via individual unemployment saving accounts
should play an important role especially in developing countries where protection of the poor is clearly a more important
issue than smoothing income streams. Relying only on social assistance and ignoring any insurance element in unemployment
income support would sacrifice the opportunity to ensure that those who have steady and sufficiently highly paid jobs for at
least some time will never ever become poor again. This reduces the funds needed for social assistance, and can hence
help to make social benefit systems sustainable. We wish to emphasize this point as it explains why we vote in favor of an
unemployment insurance system (designed as a combination of ISA and solidarity funds as it is the case in Chile) even as
part of an instrument mix that is mainly concerned about protecting the poor.

ISA are mainly designed to change the incentives of potentially unemployed individuals, but they also have high impacts on
firms’ incentives to hire new workers. Under ISA, income support is basically provided by self-insurance so that firms do not
have to pay for this part of support which reduces payroll taxes and hence the cost of labor. Of course, this argument is
based on a comparative analysis of different instruments, but this is the relevant case when designing an optimally integrated
system of unemployment compensation.
Note that, since the money on ISA is paid back to the employees who pay for it in any case, ISA does also not increase
wages by shifting the labor supply curve upwards. This does not hold for the (expected) part of the money going to solidarity
funds as workers with a relatively low unemployment risk anticipate that this part of the money is in fact an (expected)
subsidy for other workers. For these amounts paid to solidarity funds, it is not likely that it makes a high difference whether
the money is paid by firms or workers. Hence, the positive effect of ISA on firms’ incentives to hire is again based on the self-
insurance principle of ISA, whereas the amounts paid for solidarity funds will have similar effects to an unemployment
insurance system funded by payroll taxes for either employers or employees (or both of them).
Empirical Results
To the best of our knowledge, mandatory individual saving accounts have only been recently implemented in Chile, and
there is no empirical evidence on whether the positive theoretical predictions will indeed be observed in reality. Moreover,
neither unemployment insurance nor unemployment assistance existed in Chile when individual saving accounts were
implemented, so that empirical studies can not measure the different impacts of the systems, but only the impact of ISA itself
(which is obviously not the relevant question).
Because of this lack of empirical background evidence, the simulation results in Brown et al. (2006) deserve attention. They
find that moving from a combined system of UI and UA as existing in many European countries to a full ISA-system might
reduce the unemployment rate in countries as France and Germany by almost 50 percent. As expected, the reduction
potential is increasing in the generosity index explained in section 2 as the incentive problems become more severe when
the degree of coverage increases. Brown et al. (2006) emphasize that their results have nothing to do with cutting social
security, as total benefits remain the same, so that the decline in unemployment can be attributed to positive incentive
effects. Other simulation studies also confirm large benefits of moving from a traditional UI/UA-system to ISA 123. Even
though these simulations do clearly not apply directly to developing countries, they strongly support our conclusions drawn
earlier that mandatory individual saving accounts are a step towards the right direction.
5.2.3. Severance Pay and Layoff Taxes

An important aspect of severance pay described in section 5.2. (see also Annex 5) is that they are paid as lump-sums such
that they are sunk after a worker has been dismissed (ex post perspective). Hence, they do not create adverse incentives for
investing in education and in searching for new jobs 124. However, they may still negatively affect effort incentives on the job
123
Kling (2006) for simulation results referring to the USA.
124
There may be small income effects due to the higher overall wealth, but these are neglected here as they are not likely to be important for developing
countries.
61
up to the point where the requirements for no-fault dismissal are still fulfilled, as severance payments are not sunk at this
point in time (ex ante perspective).
These considerations are important as they show that the relative impacts of severance pay on workers’ incentives are
highly different from those of unemployment insurance. With respect to job search incentives and education incentives,
severance payments are obviously superior - while unemployment insurance sets highly negative incentives, there are
(almost) no incentive problems with severance pay. With respect to the incentives on the job, however, the relative incentive
effects are theoretically ambiguous due to countervailing effects. First, if payroll taxes paid by employees decrease because
severance payments are used as a substitute to unemployment insurance, then net wages ceteris paribus increase, and
since this also increases the difference between the situation with and without a job, it also leads to higher effort incentives.
However, there is the countervailing effect described above, since effort incentives are reduced by the possibility to collect
the severance pay. Hence, relative effort incentives are unclear.
Let us now consider layoff taxes where the payments of firms in case of dismissals do (at least partly) not go to dismissed
workers themselves, but to general funds, which are used as a substitute for unemployment insurance. In other words, the
only difference to unemployment insurance is the financing side, whereas the eligibility conditions and the generosity index
can be designed in similar ways. Concerning the incentives of firms (see below), layoff taxes come close to severance pay
as firms pay in the very moment where employees are dismissed; and it does not make a difference for firms whether
payments go directly to dismissed workers or to general funds. For workers, however, the incentive effects of layoff taxes are
not identical to severance pay but to those of unemployment insurance and unemployment assistance since benefits from
layoff taxes are not supposed to be sunk, but are supposed to be paid out according to similar criteria as for unemployment
insurance 125. As long as the difference between the income with and without work is the same, and as long as the eligibility
criteria are also the same, incentive effects are similar to unemployment insurance. For workers’ incentives under layoff
taxes, we thus simply refer to subsection 2.1.1 above.

First of all, it must be emphasized here that severance pay and layoff taxes set similar incentives for firms since firms do not
care whether their payments in cases of dismissals go directly to former employees or whether they are used to finance the
unemployment insurance system. Hence, there is no differentiation between severance pay and layoff taxes in this subsection.
There are typically two incentive effects: the incentives to hire and the incentives to fire, and let us compare layoff taxes to an
unemployment insurance system where funds are financed by payroll taxes (see the overview above).
The incentives to hire depend on total expected labor costs, and these costs in turn consist of two parts: the wages paid
during employment, and the expected payments after terminating an employment contract. Hence, payroll taxes and layoff
taxes are substitutes (i.e. they have effects going in the same direction) with respect to hiring incentives since both of them
increase the expected cost of hiring a worker: Payroll taxes by higher payments during employment and layoff taxes by
higher payments after employment. Since both are substitutes and since both increase expected costs, the negative
employment effects depend on the magnitudes of the two effects.
However, besides this fundamental result, four additional insights arise with respect to hiring incentives:
First, it is always possible to find a layoff tax that leaves the expected cost of labor unchanged compared to a status quo
where unemployment insurance is financed by payroll taxes. In other words, the fact that both instruments are substitutes
does not offer arguments against layoff taxes.
Secondly, layoff taxes are more important for firms hiring high-risk workers (i.e. workers with a high unemployment risk)
since payments arise only after it turns out that it is better to dismiss the worker. Thus, the impacts on different firms are not
the same: more conservative firms would prefer layoff taxes, whereas high risk firms would prefer traditional unemployment
insurance financed by payroll taxes.
125
Blanchard and Tirole (2006).
62
Thirdly, as with any job security regulation, layoff taxes set high incentives to agree upon short-term contracts, which prevent
from layoff taxes in cases of dismissal. In other words, employers and employees might have a joint incentive to evade layoff
taxes in a creative, and socially unwarranted way (to avoid the word corruption). It follows that, as under other job security
regulations, an intelligent system must be implemented that prohibits short-term contracts, which are only signed to circumvent
layoff taxes. This seems to be a major practical problem with layoff taxes.
Fourthly, layoff taxes can not be enforced in case of bankruptcy, and if mandatory saving accounts for layoff taxes are
implemented (similar to the mandatory saving accounts for severance pay in Chile), it must be ensured that these accounts
do not have the same effects as payroll taxes, i.e. firms must be sure to get the money back after some time. Summing up,
such a system may be difficult to design, and lots of restrictions and indirect effects need to be taken into account.
After having discussed firms’ incentives to hire, we now turn to the incentives to fire. The result is completely different when
comparing layoff taxes and traditional unemployment insurance financed by payroll taxes: since the effects go in opposite
directions, the two instruments are no longer substitutes, but complements with respect to firing incentives: payroll taxes increase
wages during employment, and hence increase the incentive to terminate the relationship. By contrast, severance payments
and layoff taxes have no (direct) impact on wages, but increase costs of termination, and hence enhance the incentive for stable
employment relationships. The main advantage of layoff taxes compared to the traditional financing of unemployment insurance
is thus that the private and economic incentives to fire workers are aligned: since unemployment insurance is internalized by
firms through layoff taxes, they will fire workers if - and only if - this is economically appropriate 126.
However, severance pay and layoff taxes do not only affect the incentives to hire and fire, but they may also influence the
decision whether to substitute one worker for another. This is straightforward for severance payments: Suppose that a
company wants to hold the number of employees constant, and that the productivity of an applicant is 100 whereas the
productivity of a long-term employee is only 80. If severance payments amount to 30, the company has an incentive to keep
the long-term employee which leads to inefficiently low fluctuations and hence to efficiency losses for the whole economy.
With layoff taxes, these disincentives can theoretically be avoided by decoupling payments from single dismissals and by
making them dependent on the total number of employees. In practice, however, this will be difficult since the total number
of employees depends on many factors that need to be controlled for. Summing up, severance payments (and layoff taxes)
reduce fluctuation rates since the decision between workers already in the company and new applicants is not cost neutral.
This is an important drawback compared to traditional unemployment insurance that needs to be taken into consideration,
as with traditional unemployment insurance, contributions are the same for all employees.
Empirical Results
As pointed out above, severance pay and layoff taxes can be interpreted as measures that increase the firing costs, and
thereby also the expected costs of hiring. Ceteris paribus means that severance pay, as well as any other job security
regulation, decreases hiring incentives as long as they do not substitute for other instruments. To the best of our knowledge,
the problem with all empirical studies is that they do not consider job security regulations and severance pay as substitutes
for other instruments (especially unemployment insurance), but analyze the impacts of severance pay under the assumption
that everything else remains the same. The most famous study is still the one by Heckmann and Pages (2000) who consider
a composite index for job security regulation (JSR) and compare Latin America, the Caribbean and OECD countries. In
particular, three results are worth mentioning 127:
126
This point is used in Blanchard and Tirole (2006) to argue that layoff taxes should play the dominant role in unemployment compensation systems.
127
See also Besley and Burgess (2004) for India and Haltiwanger et al. (2003) for OECD countries and transition economies.
63
1. The problem is lowest in the Caribbean because JSR plays a relatively unimportant role. The reason is that the layoff
rules are based on the common law system, which means that parties are relatively free to bargain the terms of the
contract.
2. JSR itself is highest in Latin America, because financial distress of firms is not accepted as reason for layoffs, so that
severance pay is usually serious.
3. However, the impact of JSR on unemployment is much worse in richer OECD countries. This is (convincingly) explained
by the fact that there are no additional unemployment benefits in Latin America, so that incentives to search for new jobs
are extremely high.
As mentioned, the problem with this study and with other empirical investigations is the ceteris paribus-condition. It may well
be the case that severance payments reduce unemployment if they serve as substitutes for other parts of labor costs.
Consistent with this argument, Garibaldi and Violante (2005) use labor union power as a proxy for wage flexibility and find
that severance payments reduce unemployment if wages are flexible.
The importance of the indirect effects of severance pay on wages is best captured in the influential study by Kugler (2005)
about the mandatory saving accounts for severance pay in Colombia. She finds that mandatory saving accounts have
reduced real wages by about 60 percent to 80 percent of total severance payment contributions, which can be attributed to
the fact that employees can now be sure that they will in fact get the legally obligatory or contractually agreed upon severance
payments. Without mandatory saving, employees must calculate the risk that firms are bankrupt at the time of dismissal, so
that effects on real wages are diminished 128. Summing up, it seems fair to say that there are no empirical studies contradicting
our theoretical conclusion that severance payments (financed by mandatory saving) and layoff taxes could also play an
important part of an integrated unemployment compensation system, even in developing countries.
5.3. The Current Situation in Indonesia

5.3.1. Social Security in Indonesia – Some Brief Remarks from the Perspective of
Unemployment Compensation
Out of the instruments discussed in this paper, unemployment insurance, unemployment assistance, individual saving accounts
for unemployment benefits, mandatory saving accounts for severance payments and layoff taxes do not exist in Indonesia,
and social assistance plays only a relatively minor role. Severance pay, however, is quite important (see below). Furthermore,
there are several features of social security programs that are both important and informative in our context: On the one
hand, there are programs where funds can, under certain circumstances, also be used in case of unemployment. On the
other hand, programs aimed at reducing poverty after retirement are informative for the (potential) design of unemployment
compensation both with respect to their incentive effects and their effects on the labor market as well as with respect to their
coverage.
For these reasons, those aspects of the social security regulation in Indonesia, which provide important insights for the
design of unemployment compensation, are summarized in section 5.3.2. After some brief remarks on social assistance in
section 5.3.3., recent developments with respect to severance pay are discussed in section 5.3.4. Conclusions are provided
in section 5.4. based on both, the analysis in section 5.2. as well as the current situation in Indonesia.
128
MacIsaac and Rama (2000) find that in developing countries only about 50 percent of legally justified severance payments are in fact granted.
64
5.3.2. Some Brief Remarks on Social Assistance

JAMSOSTEK
The most important legal act on social security is JAMSOSTEK, enacted in 1992, which provides support for work injuries,
death benefits, health care and a provident fund for workers in companies in the formal labor market with more than ten
employees.
Even though members are allowed to withdraw their accrued balances in case of unemployment under certain circumstances 129,
JAMSOSTEK does not include unemployment compensation. Except for the provident fund, contributions are exclusively paid
by employers (see Table 8).
Since JAMSOSTEK refers only to the formal labor market and to companies having at least ten employees, it follows that
about 80 percent of workers are excluded 130.
Table 9: Contributions/Premiums for the JAMSOSTEK Program (percentage of wages)
Program Employers Workers Total
Workplace Accident Benefits Program (JKK) 0.24-1.74 (5 classes) - 0.24-1.74
Death Benefits Program (JK) 0.3 - 0.3
Retirement Benefits Program (JHT) 3.7 2 5.7
Healthcare Benefits Program (JPK) 3-6 - 3-6
Total 7.24-11.74 2 9.24-13.74
Source Arifianto (2006: 60).
But even for companies in the formal sector with more than ten employees (hence for the companies the legal act is targeted
to), the participation rate is relatively low due to the following reasons 131:
First of all, companies can avoid the obligation to participate in JAMSOSTEK if they show proof that they have implemented
a superior social protection system for their employees as a substitute. And since JAMSOSTEK provides only very little
returns on investment 132, private contracts can be beneficial to both employers and employees simply because they are
more efficient.
Secondly, companies have invented various evasion strategies to circumvent their obligations. For instance, they hire workers
on contract status even for a long time instead of making them permanent employees. This does not remove the legal
obligation to contribute to JAMSOSTEK, but workers do not always seem to be aware of this fact. Another example is
subcontracting between different employees as common in the cigarette industry since these subcontracting employees are
(at least partly) exempted from JAMSOSTEK contributions.
129
Tambunan and Purwoko (2002:34).
130
Comparable programs to JAMSOSTEK are TASPEN for civil servants and ASKES and ASABRI for the armed forces personnel.
131
Tambunan and Purwoko (2002:46) for similar arguments.
132
It has been estimated that JAMSOSTEK provides an average return 38 percent below the inflation rate, and hence a considerably negative real return
on investment (ILO 2003:94).
65
Thirdly, the success of these evasion strategies might indicate that JAMSOSTEK is not well accepted, and that employees
may be afraid that forcing their employers to participate in the program may increase the probability of becoming unemployed
in exchange for only a relatively small expected benefit. And finally, it is estimated that only about half of the employers who
are required to participate actually comply even in cases where their legal obligation is clear 133.
Table 10: Employment and JAMSOSTEK ‘s Members
Description 1994 1995 1996 1997 1998 1999*
Number of wage employees (million) 25.1 25.7 26.3 27.1 26.3 26.1
Number of formal employers (thousand) 182 182 183 183 183 183
Membership of JAMSOSTEK
Wage employees (million)** 7.6 9.1 11.3 13.4 14.95 16.0
Employers (thousand) 51.9 60.1 77.7 68.7 82.5 86.6
Potential members (million)
Employees (1-3a) 17.5 16.6 15.0 13.7 11.4 10.1
Employers (2-3b) 130.4 121.9 114.3 105.3 100.5 96.4

* Until September 1999.
** Salaried workers are not included.
Source:Tambunan and Purwoko (2002: 45).
Administrative costs of JAMSOSTEK are estimated at around 10 percent of total contributions where the main part is spent
on the old-age program 134. Compared to other countries, these administrative costs are not excessively high, and this can at
least partially be attributed to the fact that the contributions are collected from employers and restricted to firms with more
than ten employees. This needs to be taken into account as it is sometimes recommended in the literature that employees
should bear a higher percentage of the contributions 135. From an economic point of view, however, transaction costs seem
to be most important as the impact of whether employers or employees pay on net wages and on hiring incentives is known
to be low 136.
JAMSOSNAS
As pointed out above, the coverage of JAMSOSTEK is extremely limited due to the fact that neither self-employed people
nor workers in the informal sector of the labor market are included. To overcome this problem, JAMSOSNAS (Law No. 40/
2004) was endorsed in 2004. JAMSOSNAS goes far beyond JAMSOSTEK, and mandates that social security coverage be
extended to the informal sector of the labor market, to the unemployed and to the poor. However, the law itself is very short
and the details necessary for implementing it have not been designed yet.
Since, in stark contrast to JAMSOSTEK, JAMSOSNAS could eventually cover all Indonesian citizens, it is argued in the
literature that it might overshoot and weaken the Indonesian economy 137. In the light of the arguments in section 5.2., we
tend to share this assessment: JAMSOSNAS would mainly be financed through contributions by employers and employees
in the formal sector of the labor market whereas workers in the informal sector are supposed to participate with fixed
133
ILO (2003).
134
Angelini and Hirose (2004: 49).
135
Tambunan and Purwoko (2002: 57).
136
However, following the arguments of Yves Guérard, the above mentioned 10% ratio might be misleading: if expenses are expressed in percentage of
benefits, i.e. the product which is delivered under the program, expense ratios are around 30 percent even before taxes and dividends.
137
Arifianto (2006).
66
payments. In practice, this would most probably mean that most of the money would be collected in the formal labor market:
First, the fixed payments would be highly below the contribution rates in the formal sector. Secondly, since workers in the
informal sector tend to change their employers very quickly it is very difficult to collect fixed payments in the informal market
without prohibitively high transaction costs 138. In fact, many examples show that it has always been very difficult to collect
social security contributions from the informal sector.
Table 11: Estimated JAMSOSNAS Contributions/Premiums for Formal Sector Workers (percent of wages)
Program Total Current Total Known Total JAMSOSNAS

Contribution Rates JAMSOSNAS Contribution Rates
of JAMSOSTEK Contribution Rates (Estimated)
Workplace Accident 0.24-1.74 0.24-1.75 0.24-1.75

Benefits Program (5 classes)
Death Benefits Program 0.3 Unknown 0.3

(Paid in full by employers)
Retirement Benefit Programs 5.7 Unknown 16

(equal contributions for (10 percent for old-age
old-age pension and pension and 6 percent
old-age savings schemes) old-age savings schemes
Healthcare Benefits Program 3-6 6 6
Total 9.24-13.74 6.24-7.75 22.54-23.75
Source: Arifianto (2006: 63).
As mentioned, details have not yet been agreed upon, but estimations show that the burden in the formal sector would go far
beyond the one under JAMSOSTEK 139, and it is very unlikely that such sharp increases in labor costs would not lead to
seriously higher unemployment rates.
Besides the fact that the program may simply be too ambitious, incentive effects seem to be underestimated. First, individual
saving accounts are not considered as a promising alternative under JAMSOSNAS, and all the problems associated with
payroll taxes are likely to reduce the program’s efficiency. Second, there is an additional problem that plays a higher role
compared to industrialized countries: as the dividing line between the formal and the informal labor market is less clear in
developing countries, the incentive to evade contributions by escaping in the informal market is higher, and might be a
profitable alternative if contributions indeed rise to the numbers listed in the section below. Examples for such behavior
include the case of Argentina where more than 50 percent of the workers in the formal labor market evaded their social
security contributions when the pension scheme was reformed in a way that led to much higher contribution rates. These
evading strategies have considerably contributed to the budget deficits in Argentina. In short, JAMSOSNAS might be
reconsidered both with respect to the generosity and with respect to the instruments used to finance the system.
5.3.3. Some Brief Remarks on Severance Pay

Besides the two programs just described, there are a variety of welfare programs targeted to provide a safety net to at least
some members of the informal sector 140. Based on these programs, social assistance is not provided on a regular and
steady basis but is rather targeted to support the poor in critical situations. After the crisis in 1998, new social safety net
138
For a general discussion of extending social protection coverage to the informal sector workers see chapter 7.
139
These estimations take the expected increase in the average age of the population already into account.
140
Tambunan and Purwoko (2002:37) and Lindenthal (2004:20).
67
programs (JPS) have been designed and total expenditures are estimated at around 1.25 percent of GDP in 2000 141.
Important parts include food security programs (cheap rice distribution) and a small scheme to provide payments to old
persons without family or community support. It needs to be emphasized, however, that these payments are only granted on
a temporary basis and that they cannot substitute for family or community support in the long run. For a more detailed
treatment of social assistance see chapter 6.
In the literature, it is often emphasized that the social assistance programs suffer from (1) an inequality between different
regions, (2) the fact that they are poorly targeted with high rates of leakages to the non-poor, and (3) from high administrative
costs 142. For instance, it is estimated that about 30 percent of the total budget for the subsidized rice program are administrative
costs.
5.3.4. Severance Pay

Severance payments have a long tradition in Indonesia, and they are currently regulated under the Manpower Act 13/2003.
The literature agrees that severance rates have increased by about 200 percent since 1986, and that long-term employees
have benefited most with a 50 percent increase since 2000 143. The increase in severance payments is also due to the fact
that minimum wages have increased, and since severance payments are defined as multipliers of wages, higher minimum
wages also lead to higher severance payments. An unusual feature of the current regulation is that severance payments are
also granted for voluntary quits and for employees who are dismissed for major transgressions such as stealing or violent
behavior 144.
Depending on the time of employment, the figure below shows the development of severance pay since 1986 145.
Figure 6: Severance and Long Service Pay for Lay-Offs and Plant Closings
Multipliers of Monthly Wages
Years of Employment
Source: World Bank (2006c: 5).
Annex 7 shows that, expressed as a multiplier of monthly wages, Indonesia is (almost) leading in the world with respect to
the generosity of severance payments. However, this is somewhat misleading at least when it comes to a comparison with
industrialized countries as these countries mainly protect unemployed people through unemployment insurance and
141
WORLD BANK (2000a).
142
Lindenthal (2004:40f).
143
WORLD BANK (2006c:4).
144
Manning (2003:80f).
145
Annex 6 shows how current severance payments differ depending on the cause of dismissal.
68
unemployment assistance such that severance payments are only complementary. The immense increase in the international
ranking, especially compared to other developing countries, is nevertheless striking. From the economic analysis in section
5.2., it is already clear that we prefer severance payments to unemployment insurance, but the generosity obviously deserves
further discussion.
5.4. Preliminary Recommendations and Open Questions

Based on our evaluation in section 2 and considering the current situation in Indonesia as described in section 3, we now
provide some very preliminary recommendations about how a future system of unemployment compensation could be
designed. We do not go into details, neither with respect to administrative aspects and magnitudes (measured by a generosity
index, for instance) nor with respect to a timetable. Nevertheless, we hope that the following insights may prove helpful for
further discussion:
1. Mandatory individual saving accounts could be an important part of an integrated system of unemployment
compensation: compared to unemployment insurance and unemployment assistance, individual saving accounts considerably
reduce incentive problems on the worker’s side. As in Chile, they can be combined with contributions to solidarity funds in
order to provide insurance for those who are not sufficiently covered. Of course, these contributions to Solidarity Funds
increase labor costs in the same way as payroll taxes for unemployment insurance and unemployment assistance do, and
they hence need to be kept small in all developing countries. But even in developing countries, implementing mandatory
individual saving accounts seems reasonable because those who have formal labor contracts for at least some time will then
be kept out of a general social benefit system. This seems to be very important as a system providing a minimum standard
of living for the poor could otherwise not be sustainable in a reasonable time horizon.
A problem of mandatory saving accounts (as of any kind of unemployment benefits, however) is that they can hardly be extended
to the informal sector of the labor market. It is hence worthwhile to think about subsidies for voluntary saving accounts for
informal workers. Although subsidies are costly by definition, such a system may reduce payments for social assistance as they
set incentives for informal workers to self-insure against periods of unemployment and underemployment 146.
2. As has been pointed out in subsection 5.3.4, severance pay plays already an important role in Indonesia. In the light of the
theoretical discussion in section 5.3., we believe that severance pay is indeed superior to unemployment insurance.
However, there are at least two critical aspects we wish to emphasize: on the one hand, the generosity of severance
payments has sharply increased in the last years, an aspect which needs to be reconsidered. On the other hand, however,
the number of firms actually participating in the program is quite limited. Thus, lower generosity together with higher participation
rate ensured by stricter law enforcement could probably increase the program’s efficiency.
A particular problem with respect to the participation rate is bankruptcy. We thus recommend mandatory saving accounts
for severance pay as a second main pillar of an integrated system. Empirical studies about Colombia have demonstrated
that mandatory saving accounts for severance pay can considerably reduce real wages, and can thereby diminish the total
increase in expected labor costs for firms caused by severance payments. This is a very important result which backs up the
argument that negative incentive effects of severance payments are smaller than those associated with traditional financing
by payroll taxes. To the best of our knowledge, introducing mandatory insurance for severance pay (which comes close to
mandatory saving accounts) is currently being discussed in Indonesia 147.
In the same way as for individual saving accounts, these accounts for severance pay could be complemented by payments
to solidarity funds, i.e. by layoff taxes, which do not go to the dismissed workers themselves, but to general funds. Of course,
these payments will have lower effects on real wages (i.e. the reduction rate will be smaller), and may hence negatively
affect labor demand, but this is inevitable for all insurance systems. Hence, the basic idea is the same for individual saving
accounts and for severance pay; the main difference being that mandatory individual saving accounts are financed by
146
Suprobo et al. (2007), subsection 5.2 for a similar suggestion.
147
The Jakarta Post (2006).
69
employees whereas the severance pay accounts are

financed by employers. Both funds can be accompanied
by Solidarity Funds.
3. General systems of either unemployment

insurance or unemployment assistance financed by
payroll taxes are not recommended 148. Their incentive
effects on both sides of the labor market (i.e. for workers
and for firms) are severely negative, and administrative
costs 149 and monitoring costs are likely to be high in
developing countries 150. This holds especially for
unemployment assistance if payments are made
dependent on requirements, former employment and
sufficiently low wealth. Motivation impacts are adverse
compared to individual saving accounts, and negative
spillover effects for other family members also need to
be taken into account 151. However, we wish to emphasize
that traditional unemployment insurance in fact has one
important advantage compared to severance payments
and layoff taxes: It has no adverse effects on the relative
costs of workers already in the firm and new applicants.
4. When designing a combined system of mandatory

individual saving accounts and mandatory savings
for severance pay, both of them complemented by
solidarity funds, the following aspects seem to be
important:
Most obviously, it needs to be carefully thought about how the two parts of the system work together. For instance, are
individual saving accounts really compatible with layoff taxes? How do incentives overlap, and do they reinforce each other
or are they at least partially crowded out? To see the problem, assume that the individual saving account of an employee
under consideration has sufficient coverage. Will he then still at least have partial access to payments funded by layoff
taxes? And, even without layoff taxes and only severance pay instead, should the severance pay be deducted from the
amount on the individual saving account the individual has access to?
To avoid the problem that severance payments cannot be paid in case of bankruptcy, mandatory saving accounts (self-
insurance) or mandatory insurance can be established. One carefully needs to think about which of the two options is
superior, or whether firms should be allowed to choose between the two alternatives. At the moment, we tend to prefer
mandatory saving accounts, i.e. self-insurance.
In developing countries, it is notoriously difficult to define who is actually unemployed since the boundaries between the formal
and the informal labor market are blurred. Even though this has usually been discussed with respect to traditional unemployment
insurance, it also applies to the solidarity funds from individual saving accounts and layoff taxes 152. Hence, benefit eligibility may
148
This contradicts the position in Tambuan and Purwoko (2002:55f), for instance, who recommend unemployment assistance on a flat-rate basis.
149
In practice, one needs to implement systems allowing for synergies, usually referred to as “one-stop registrations”. See Vroman and Brusentsev
(2005:190f.) for a discussion of the topic.
150
See: in similar vein, Vodopivec (2004c:6) and Vodopivec (2005a:19).
151
Vodopivec (2004b:6).
152
70
hardly be enforceable in developing countries, especially when the possibilities to work in the informal market are taken into
account 153. Nevertheless, this should not be taken as an argument against the components suggested in this paper - the
problems apply to any kind of insurance system, and are not specific for saving accounts and severance pay.
Private transfers that play an important role in developing countries, and which can make all kinds of solidarity funds more
sustainable since less money is needed, should not be crowded out.
The same holds for informal employment protection arrangements, which are also predominant in developing countries.
Households in all developing countries have invented strategies to smooth consumption in cases of income shocks, and we
partially agree that one need to be aware that these traditional strategies should not be crowded out. However, it must also
be taken into account that these strategies may involve higher costs than the incentive costs associated with severance pay
and saving accounts (and even with traditional insurance principles): Empirical research suggests that the consumption
fluctuations caused by income shocks in Indonesia are not much higher than in the USA, and one might conclude that
families in Indonesia have adequate self-insurance against income shocks. However, the same study shows that the behavioral
distortions leading to consumption smoothing are severe: many households reduce their spending on education during
shocks, and other household members are much more likely to work when the head of the household is unemployed 154.
All of the four points described certainly deserve further research, and designing a system taking these (and other) points
seriously is difficult. However, it should have become clear that the advantages of mandatory individual saving accounts for
formal workers, voluntary (subsidized) saving accounts for informal workers, and mandatory saving accounts for severance
pay, combined with solidarity funds compared to traditional insurance are high, so that thinking about appropriate designs is
worthwhile.
Finally, we need to emphasize that we have neglected public works on the one hand and minimum wages on the other hand.
With respect to public work, this is only due to the restricted perspective of our analysis, but we certainly do believe that
public work is a powerful instrument to reduce problems of people who do not find work in the labor market 155. An interesting
result of empirical studies that goes beyond the scope of our paper is that active labor market policy should mainly count on
services in cities and on agriculture in rural areas, as the multipliers (i.e. the number of jobs created divided by the amounts
spent) are then highest 156.
With respect to minimum wages, however, we are much more skeptical. Minimum wages can only be beneficial if they lead
to redistributions from firms to workers, and not to higher unemployment, which requires first that profits of firms are sufficiently
high to avoid bankruptcy, and second that firms do not have access to other labor markets without minimum wages. Empirical
studies for Indonesia, however, show that minimum wages lead to higher unemployment in regions with extremely poor
people, and hence seem to support only those workers who live above the poverty threshold anyway 157.
153
154
Chetty and Looney (2005).
155
Vodopivec (2005a: 17) for a convincing summary of the benefits of public work.
156
Suryadarma et al. (2007).
157
Suryahardi (2001).
71
72
6. Social Assistance:
Reforming Non-
Contributory Social
Protection
- Social Assistance comprises the non-contributory provision of public aid to targeted poor and or certain groups
considered as vulnerable. Modern social assistance involves three dimensions: income support, family benefits and
social services.
- Social assistance measures in Indonesia display a high degree of fragmentation. Providing effective social assistance
requires horizontal and vertical integration. Therefore, as a first step in the direction of further harmonization, the
complex web of inherited social assistance programs should be consolidated and simplified and administration
delegated to a newly founded institution operating under the umbrella of the Ministry of Social Affairs.
- In contrast to social security, which is financed via individual contributions of participants, taxes provide the most
important financing source for social assistance. In terms of financing social assistance, the success of reform efforts
crucially depends on complementing tax system reforms to stabilize the required revenue base.
73
6.1. Social Assistance – Conceptual Framework

Social assistance comprises the non-contributory provision of public aid (monetary or in-kind transfers) to targeted
poor or certain groups considered as vulnerable. Its role and design with regard to targeted poor differs with respect to
the perception of behavior expected from citizens to obtain their way of living either from the labor market or from welfare
benefits. While in some countries like Britain and the USA, the accent has been on differentiation between “non-deserving”
or “deserving” poor and a more punitive approach towards the former, in other countries, like in France and the Scandinavian
countries, it is considered a society’s obligation to avoid social exclusion by providing individuals with the necessary means
for integration. Modern “activating” social assistance policy tries to provide the lifetime poor with the comprehensive social
protection they need and the transient and working poor with sufficient incentives and opportunities for their full integration
into work.
Historically, social assistance is the oldest form of social protection and dates back in Europe to the 13th and 14th century,
when townships and churches set up the first welfare provisions to aid poor or sick members of their communities. The
principle that anyone who falls into need and is unable to help himself could count on support from community was deeply
rooted then in religious or privately motivated benevolence, charity and group solidarity but still not entrenched in law. This
situation dramatically changed when in the wake of the Industrial Revolution during the 18th and 19th century mass poverty
grew in unprecedented scale so that in order to avoid public revolt some more institutionalized response to the so-called
“Social Question” had to be found. While in some countries existing Poor Laws were adapted to the new needs, the introduction
of Social Insurance through the Bismarck Legislation of 1881 certainly was a milestone concerning social protection of the
working class (see chapter 3). Replacement income during periods of work interruption, either temporary (sickness or
unemployment) or definitive (permanent work incapacity or retirement) was as such quite effective in alleviating poverty.
However, as the social insurance principle was based on the ability to work and on contributory formal employment, all those
who were temporarily or for lifetime unable to work were systematically left out. In other words: contrary to the claim of
universality and comprehensiveness of the Beveridge approach, the Social State Model of social protection was
not poverty-resistant.
This unsolved question led to the necessity of additional social assistance schemes as a non-contributory provision of
economic or non-economic public aid to targeted poor and needy individuals or groups, financed by tax-revenues and
usually subject to means-testing. Unlike the universal flat-rate income approach, means-testing is considered
indispensable as the principle of subsidiarity requires striking a balance between assistance from others and self
responsibility. The controversy between the guaranteed income approach on the one side and means-tested social assistance
on the other is further spurred on criticism of traditional social assistance as focusing too much on reducing current poverty
and emergency relief and too little on issues of long-term and structural causes of poverty. Furthermore, the modern
notion of social assistance as a fundamental human right to enable recipients to lead a dignified life in the sense of
avoiding social exclusion has influenced conceptualization in many countries so that the strict former separation
between social assistance and social insurance has developed more towards a continuum 158. One of the major
reasons for this lies in the fact that employment patterns in the Western World have substantially changed from standardized
full-time unlimited work contracts in the age of industrialization towards part-time work and often unstable “atypical”
employment conditions. Especially, the new phenomenon of “working poor”, i.e. employees who need social assistance to
complement their income from work, has led to new considerations regarding the integration of insurance and assistance
benefits towards redressing social inequalities in a more generalized form. A further argument for integration consists in the
disincentive effects for those caught in unemployment or (better) underemployment traps where the loss of social benefits,
in conjunction with the impact of taxation and rising mandatory contributions to social security, can wipe out any advantage
expected from extending earned income. Thus, strictly separating social assistance from social insurance could be a severe
hindrance to the transformation of informal into formal employment.
158
This has been especially the case in the Scandinavian countries and in France, where the term social exclusion was defined as a rupture of relationship
between individual and society due to failure of societal institutions to integrate. Berger-Schmitt (2000:4).
74
6 • Social Assistance: Reforming Non-Contributory Social Protection
In order to promote full economic and social integration, modern social assistance schemes have to provide the categories
of people who cannot work with income security and social protection and those who are able to work with sufficient incentives
and opportunities to integrate themselves as fast and as far as possible into the workforce. This differentiation between the
“lifetime poor” and the “transient poor” has led to a threefold dimension of modern social assistance: income support,
family benefits and social services.
Income support involves monetary transfers to low income households. Income support aims to provide those financial
resources needed for upholding a certain basic living standard. The current discussion on cash transfers in low and middle
income countries added a further dynamic dimension to the impacts of income support on poverty regarding them rather as
an investment: In the long-term, cash transfers are intended to break the intergenerational cycle of poverty by fostering
investments in human capital (nutrition, education, health). Any form of income support inevitably raises the question of
adequacy and targeting. National income lines as at-risk-of-poverty thresholds not only have to be country-specific but
also have to consider deviations in the cost and standard of living in different provinces and for different strata of society.
Furthermore, mere headcount figures on poverty do not answer questions on the intensity of individual poverty (severity of
risk) or on the density of population just above the national income poverty line and thus vulnerable to fall back (exposure to
risk) 159. It is important in this situation on the one hand to keep individuals and households in a sustainable distance from
imminent exposure to the risk of poverty, and on the other, to avoid the poverty trap where a potential increase in earnings
from work fails to translate into a net income increase that is felt by the individual to be a sufficient return for the additional
effort.
Targeting versus universalism is another controversial issue as it is often argued that targeting leads to high administrative
expenses and tends to stigmatize recipients. On the other hand, while targeting is necessary for minimizing both errors of
inclusion (benefits go to the non-poor) and errors of exclusion (the poor not getting benefits), the only universalistic scheme
would be a guaranteed basic income for every citizen as was intended in the Beveridge Report but which not even in Britain
159
Indonesia, for instance, has an extremely high share of “near-poor”. Data collected shortly after the 1998 crisis show that while “only” 22 percent of the
population was officially poor, about 45 percent were vulnerable to poverty. Data from more recent years show a very high degree of households moving in
and out of poverty. For example, more than38 percent of poor households in 2004 were not poor in 2003. World Bank (2006b:28).
75
was put into practice and today hardly exists. Another counter argument relates to the intrinsically self-selective character of
social assistance in that potential beneficiaries have to face a pre-defined actual risk and are eligible only to the corresponding
benefits. Finally, it can be argued that the universality criterion refers to the human right to be protected from poverty, but not
to the eligibility for the benefits themselves.
Family benefits pose another threat in the context of adequacy as especially large families could accumulate social transfers
for health, food, housing, and education for each of their members so that their income from social assistance related
measures might exceed potential income from work. The setting of upper income ceilings to qualify for benefit in consideration
of family allowances is therefore necessary. This can be done reducing the amount of transfers when family income exceeds
a certain sum or means-testing, mindful coordination of different allowances is also helpful. Apart from these dangers of
over-accomplishment, family benefits often have strategic importance for poverty alleviation of single, e.g. widowed,
parents and especially for women. In general women not only receive less attendance and lower income but also have
less opportunity to better their living situation than men usually have concerning education and social status. In many
countries, single mothers constitute by far the largest group among poor households and non-existent or affordable access
to childcare is for many the greatest hindrance for gainful employment.
Social services, or the provision of access to social services, have become an integral component of social assistance
when poverty is no longer considered as a single dimension but as a complex and multifaceted problem of social exclusion
involving the accumulation and combination of several types of deprivation: lack of education, deteriorating health conditions,
homelessness, loss of family support, non-participation in the regular life of society, and lack of job opportunities. Social
exclusion goes beyond poverty and as each type of deprivation increases the other a vicious circle results, leading from
long-term unemployment to the break of family ties, and eventually to marginality and delinquency. While direct income
support remains important, social services help individuals not only to live in human dignity, but also to (re)gain
employability and become an active and integrated member of society. In many low and middle income countries
strong efforts are being made to secure access specifically to health and education services for poor people.
However, the problem of stigmatization of social assistance recipients needs to be considered, too. On one side,
stigmatization aggravates social exclusion of low income people. As social exclusion is one dimension of poverty, social
assistance undermines its own objective in this
regard, i.e. mitigating poverty, if it actually
fosters stigmatization. One important aspect to
influence stigmatization is the degree of
visibility of social assistance reception: The
easier social assistance recipients are identified
as such, the more likely is their societal
stigmatization. On the other side, a certain
prevalence of stigmatization reduces the danger
of creating a dependency-mentality. Thus, the
task is to select instruments that minimize but
not necessarily completely abolish sensitivity
to potential stigmatization 160.
160
For example, with regard to targeting and stigmatization any public proclaiment of eligible persons/households should be avoided. Another example
refers to equal access of health services and stigmatization: in this context a closer linkage between social insurance and social assistance contributes to
a reduction in possible stigmatization of social assistance recipients: if people insured under contributory social insurance hold the same ID cards as
people financed by public subsidies providers are not able to distinguish between both groups and thus poor people face a lower probability being
stigmatized on behalf of their income status. However, this option presupposes that all population groups including the poor households are entitled to the
same benefit package of health services. Although in many country contexts this might not be immediately achievable, it should nevertheless be the
ultimate objective.
76
6.2. The Current Situation

Prior to the Asia Crisis of 1997/98, the Indonesian government mainly trusted in the trickle down effects of economic growth
for poverty alleviation. This neglect of direct intervention changed when in the aftermath of the crisis previous downtrends in
poverty levels dramatically reversed especially in the considerable near poor fraction of the population. As an immediate
response and following recommendations of the World Bank and ADB, several Social Safety Net programs (JPS) to help
especially poor people to cope with their steeply declining purchasing power were hastily implemented by a wide array of
government agencies providing health services (Kartu Sehat), subsidized rice (Raskin) and fuel subsidies and scholarships
for poor children 161 as well as a multitude of small social assistance programs for specific target groups (abandoned
children, disabled persons, poor elderly, etc.) operated by the Ministry of Social Welfare 162. Early assessments of the JPS ad
hoc measures show low coverage, poor targeting and, in general, little coordination 163.
Since then, the newly elected government has undertaken considerable efforts to better the performance of existing programs,
reduce or terminate others and introduce new ones. One of the main government concerns was to establish free access to
health services for the poor as had been stipulated in Law No. 40/2004. Providing financial access to health services for the
poor is an important step towards breaking the vicious cycle between poverty and illness: for the poor, the risk of severe
illness and earlier death from disease is considerably higher than for those who are financially better off. Poor people are
more exposed to health risks as they often experience ill working and living conditions, with limited access to clean water or
adequate sanitary facilities. In addition, the poor are the most vulnerable as they are less able to recover from the direct and
indirect costs of illness: apart from forgone earnings, the poor often have to cut down expenditures on necessities, such as
food and clothing, or are forced to take their children out of school because they can no longer afford the school fees.
In 2002 the Ministry of Health launched programs to provide for financial assistance for health care for poor people. The
Social Safety Net for Health (JPS-BK) aimed at basic health and maternal services as well as nutritional improvement, and
the Alleviation of the Health Impacts of Oil Subsidy Reduction (PDPSE-BK) program providing free basic health services to
the poor via health cards. This was followed by ASKESKIN a public healthcare scheme for the poor providing free-of-charge
primary healthcare services at public health centers (PUSKESMAS) and inpatient treatment in third-class hospital wards,
predominantly for those poor possessing health insurance cards 164. The scheme is run by the Ministry of Health. Whereas
PUSKESMAS are granted block grants for operational requirements, equipment and medicine, funds for inpatient care are
channeled via PT ASKES for inpatient treatment. Funding is provided in form of block grants from the government. Although
this new program has formally entitled around 60 million households to free access to healthcare, utilization by the poor has
not substantially increased probably due to several aspects, possibly also including stigmatizing effects of health cards or
non-acceptance by practitioners and hospitals. Furthermore, funding PUSKESMAS through block grants does not stimulate
incentives for PUSKESMAS to improve access and service quality (see last chapter in this report for more information on the
ASKESKIN scheme).
ASKESKIN is a component of a wider program called PKBS-BBM (Program Kompensasi Pengurangan Subsidi – Bahan
Bakar Minyak). Education assistance is provided via the PKBS-BBM in the form of either an operational aid program to
primary and secondary schools (BOS) aimed at canceling school fees, as well as providing scholarships for poor senior
secondary school students (BKM) 165.
Another program originally covered by PKBS-BBM is the unconditional cash transfer (UCT) program or subsidi langsung
tunai (SLT), which was introduced in Indonesia in 2005. The UCT was carried out between October 2005 and December
2006 providing 19.2 million poor and near poor with this form of income support. The UCT was replaced by a Conditional
161
Sumarto et al (2004:3).
162
Weber (2006:107).
163
Although the intended beneficiaries of these programs are the poorest quintile, they have access to only 31 percent of health cards, 39 percent of
scholarships and 29 percent of RASKIN benefits (Lindenthal 2004:21). Tambunan and Purwoko (2002) present similar conclusions.
164
Weber (2006:107).
165
Knoess (2007).
77
78
Cash Transfer (CCT) program in 2007 166. In contrast to UCTs, conditional cash transfers provide income support to poor
families conditional upon investments in human capital (school attendance, health care, and/or nutrition). They do not impose
any restrictions on the usage of transfers as long as certain defined health and education related conditionalities are fulfilled.
For instance, education grants are tied to school enrollment or school attendance, whereas health grants are contingent
upon health check-ups or immunization of infants. The Indonesian CCT program includes both an education as well as a
health component. Education-related conditionalities refer to the attendance and enrollment in primary and secondary school.
Health-related conditionalities refer to health checks and the care of pregnant women and children between 0-5 years. The
cash transfer is channeled directly to the recipients through post offices in order to reduce transaction costs and handed out
to the female head of the household every three months. The target groups of the CCT program in Indonesia are very poor
households with children between 0 and 15 years and/or a pregnant mother at the time of registration. Each household will
receive funding for up to six years. A pilot project started in 2007 targeting 500,000 households. If successful, the number of
covered households will gradually expand to cover a maximum of 6.5 million households. The entire implementation period
is scheduled for 2007-2015. The CCT is only implemented in areas where the supply of health and education services is
already adequate in terms of quantity and quality.
BAPPENAS is the responsible ministry for the conceptual development of the CCT program. The implementation and
coordination is carried out by the Department of Social Affairs. For this purpose, a Program-Operating Unit (POU) has been
designated in the Directorate of Social Assistance within the Department of Social Affairs. The POU will report to a Ministerial
Level Steering Committee and operate under the technical supervision of a Technical Committee. The Steering Committee
is composed of members from several ministries. An important role of the steering committee is to ensure the participation
and commitment of all relevant ministries. At the provincial level a Technical Coordination Committee with representatives of
the Ministries of the Steering Committee at the national level helps to coordinate program activities between national and
provincial levels, to ensure cooperation of relevant stakeholders at the provincial level and to socialize the program.
Further, the Ministry of Social Affairs runs social welfare programs (BKSP) which are targeted at specifically vulnerable
groups, which include neglected elderly, mentally disabled persons, and persons suffering from chronic illnesses by providing
income support. A temporary component is targeted at the victims of natural disasters. The funds for this program are
channeled to the recipients via specific service provider organizations 167. In addition, it provides social services for certain
groups like poor families living in slum areas, abused children and adults, and drug addicts 168.
At local level, several social assistance programs are in place, too. Examples include health insurances with pro-poor
components in Jembrana and Kupang (starting 2007). The local government in Jembrana also provides subsidies and
scholarships to students. In Makassar, poor people have received free healthcare at PUSKESMAS since 2006 and free
education in selected areas, starting in 2007 169. In East Java forms of income support are provided to poor households or for
instance in the Surabaya Municipality the local government subsidizes the “Movement of Mother Loving Purse” helping poor
mothers to improve prenatal care 170. In addition to the above mentioned programs, several districts had already implemented
their own district health insurance programs under an earlier program (JPK/Gakin scheme).
Besides public programs for social assistance at central and local level, a broad variety of informal and more traditional
arrangements for mutual solidarity and self-help are active, for example, neighborhood and community assistance schemes
in the form of rice-sharing (Jimpitan), funeral attendance (Layatan) and numerous other mostly custom-based forms of risk-
sharing or risk-coping strategies 171. As valuable as these voluntary arrangements are as an expression of social capital and
166
For a comprehensive overview on cash transfers in Indonesia with a specific emphasis on conditional cash transfers, see Knoess (2007).
167
In addition, the Ministry of Social Affairs is in charge of the so-called KUBE/LKM program, which is providing grants to microfinance institutions (KUBE)
respectively, channeling grants from donors through an APEX-institution to microfinance institutions (LKE). Although this program does not directly provide
social assistance to poor households, it promotes productive self-help as it helps to relax liquidity constraints faced by poor households, thereby fostering
income generation by strengthening access to credit for poor people.
168
ASKESOS is not mentioned here as it is a contributory program and thus not qualifying as social assistance.
169
Bahagijo (2007).
170
ADB (2006:90).
171
Ibid., pp. 114.
79
social cohesion inherent in society, they should not be overestimated in their potential outreach and financial sustainability in
case of severe crisis.
Tab. 12 summarizes the main social assistance programs and categorizes them according to the different types of social
assistance programs, i.e. income support, social services, and family benefits 172.
Table 12: Types of Social Assistance in Indonesia (2007)
Income Support Family Benefits Social Services
- CCT program (very poor - Social services for poor families

households) living in slum areas,
- BKSP (neglected elderly, abused children and
mentally disabled persons, None adults, and drug addicts
and persons suffering from - ASKESKIN (health)
chronic illnesses) - BOS/BKM (Education)
- Local level initiatives - Various health and education
(East Java) initiatives at local level
The Indonesian performance is impressive concerning the range of social assistance programs provided. However, a
coordinated and long-term oriented approach to social assistance is still lacking. Prevailing problems particularly
seem to relate to either the organizational set-up (coordination at central level and between local and central level) or the
financial sustainability. These issues will be taken up in the following chapter.
6.3. Reform Options

6.3.1. Organization of Social Assistance
The general set-up of social assistance within the framework of separation was discussed in chapter 4.2. This chapter
focuses on the issues of decentralization and coordination. Within the context of the organization of social assistance the
question of how much to decentralize social assistance is of utmost importance. Decentralization has the potential to improve
social assistance delivery by bringing services closer to clients and strengthening the capacity of local governments and
providers to plan and deliver services and to monitor outcomes. Thus, subsidiary decision-making in the provision of social
assistance is desirable for better targeting as local knowledge can be used to limit leakages in the form of errors of unwanted
inclusion or exclusion 173. It also helps ensure that municipalities plan their budgets and expenditures carefully, and that they
do not end up lobbying the central government for large transfers. However, decentralization also entails certain risks as the
absence of sufficient attention to local resources and capacity can lead to an erosion of services and even deepen geographic
inequalities.
Empirical evidence on the degree of centralization and its impact on performance related indicators is scarce. An innovative
study on the relationship between the degree of decentralization and distributive outcomes of social assistance arrangements
in 15 European countries came to the conclusion that the degree of centralization resp. decentralization is an important
determining factor for the performance of social assistance programs as it affects both distributive effectiveness as well as
efficiency of measures. Regarding distributive effectiveness of social assistance, i.e. when the percentage reduction
of initial poverty is the relevant outcome, highly centralized systems achieve the highest figures, while more
decentralized systems exhibit lower effectiveness. With respect to distributive efficiency, i.e. percentage reduction
172
As information on local level programs is rare, no claim for completeness is made.
173
Targeting is especially difficult in Indonesia because of the significant size of the informal sector, which complicates measurement of income for accessing
benefit eligibility since beneficiaries may be reluctant to report income if they are not paying taxes or social security contributions.
80
of initial poverty set in relation to the share of social assistance expenditures in GDP, the results seem to suggest
that the systems with a medium degree of decentralization do better than either extremely centralized or decentralized
systems 174. A medium degree of decentralization seems also to be preferable when looking at another indicator for
distributive efficiency, i.e. the percentage reduction of initial poverty, where similar results are presented 175.
Concerning the present situation in Indonesia, political and administrative decentralization under the policy of “local autonomy”
since 2001 has led to more responsibilities at the provincial and district government level but often without corresponding
financial resources and administrative know-how for implementation (see chapter 2.2.). Thus, any decentralization of social
assistance needs to be considered within these current constraints.
In addition, vertical coordination of existent programs is necessary. The case of social health protection for the poor illustrates
the necessity: The national program ASKESKIN is “accompanied” by various local initiatives. These duplications hinder
efficiency. Further, the example of social health protection also demonstrates the constraints levied upon future system
development due to path dependency: The regional health insurances developed under JPK (Gakin scheme) and ASKESKIN
are not only parallel initiatives but also in conflict 176. Thus, any delay in vertical harmonization of social assistance might lead
to an acceleration of problems in the future.
Furthermore, there is also a lack of vertical coordination insofar as at the community level there are no specific administering
bodies, such as public welfare centers and social workers, to which measures from the central government could be directed
for implementation. On the macro-level of central government, there is again a deficit concerning horizontal coordination
between the many different programs and implementing institutions.
In the context of vertical and horizontal coordination, the design of the CCT program is a very good example as it involves a
mechanism for horizontal as well as vertical coordination. However, it is still only one instrument. Therefore, as a first step in
the direction of further harmonization, the complex web of inherited social assistance programs should be consolidated
and simplified and administration delegated to a newly founded institution operating under the umbrella of the
Ministry of Social Affairs.
Assessment of foreign experience, especially from Latin America, could also be helpful in this context, as many of these
countries share the same experience concerning persistence of poverty and extension of the informal sector. Recent social
programs in Latin America and the Caribbean concentrate on strengthening human capital offering productive opportunities
and improving family and community environment. In the past few years, conditional cash transfer programs have been
successfully used to generate synergies between human capital formation and poverty reduction strategies. A recent report
summarizes main experiences from Latin America (see ECLAC/United Nations 2006).
6.3.2. Financing Social Assistance

In contrast to social security, which is financed via individual contributions of participants, taxes provide the most important
financing source for social assistance. Considering the unfortunate relation between the number of poor and near poor
persons in Indonesia and available tax resources, the potential outreach of social assistance is limited and any ad hoc
measures should be avoided. Basically, the decision is either to provide a relatively broad range of social assistance measures
to very limited groups of persons or to focus on a more limited number of measures (such as health services or modest forms
of income support via cash transfers) provided to a wider share of the poor population.
In any case, strengthening the outreach of social assistance programs involves the need to increase available tax revenues
or to reallocate expenditures among sectors. Increasing tax revenues does not necessarily involve an increase in tax rates.
Strengthening the institutional basis of the existent tax system will lead to an increase in the total amount of taxes received.
Several potential options exist:
174
Hölsch and Kraus (2003).
175
Hölsch and Kraus (2002).
176
The regional initiatives legally challenged PT ASKES. The Supreme Court ruled that the central government’s decision to implement the program through
PT Askes violated the decentralization laws and PT Askes has to coordinate with all local governments (World Bank 2006b:32).
81
Ensuring compliance: The table below shows a continuous upward trend in domestic revenues. However, the Indonesian
tax system, which mainly relies on income taxes, is still of limited effectiveness. The number of registered taxpayers
amounts to 3.5 million only, out of a total of 40 million households 177. Even if one takes into consideration that among the
total number of households a significant proportion is not able to pay taxes the rate of non-compliance will still be quite
high. Considering the large informal sector this result is hardly surprising. Therefore, any strategy of providing social
assistance to poor households needs to be complemented by appropriate tax system reforms. Steps to improve
the system are currently being undertaken. The Directorate General of Taxes, the Indonesian agency responsible for
tax administration is pursuing three administrative goals: to raise taxpayer compliance, to raise public trust in the tax
administration and to improve the productivity of tax officers.
Table 13: Actual Government Revenues (in Billion Rp)
2005 2006 2007
Domestic Revenue 494148 635926 720389

Tax Revenue 346834 409020 509462
Income Tax 175380 208834 261698

VAT 101295 123028 161044
Other taxes 70159 77158 86720
Non Tax Revenue 147314 226906 210927
Source: Ministry of Finance/BPS-Statistics Indonesia
Changing taxation structure: Legal tax reforms for changing the taxation structure constitute another option to increase
revenues. In general, there are two options: using direct taxes, i.e. taxing income or assets, which are tied to individual
productivity (e.g. income taxes) or indirect/consumption taxes, which are tied to economic transaction, i.e. use of
income/assets (e.g. value added tax –VAT). In the light of the current problems of the Indonesian tax system,
using indirect taxes to increase the tax base for financing social assistance seems to be an option in the short
to mid-term worth consideration as it is not easy to evade those taxes. Consumption-related taxes are specifically
an option if informal sector activity is significant. But the design of indirect taxes is crucial in order to avoid a regressive
character of those taxes (specifically VAT), thus hurting the poor more than the well-off. Differentiation of VAT is a
solution to avoid regression. It involves taxing basic necessities (certain kinds of food, such as rice) at a lower rate
than ‘luxurious’ goods, which are mostly consumed by the better off (e.g. electronic goods, such as DVD recorders).
However, the impact of variations in the tax structure on economic activities needs to be carefully considered,
too.
Within a given budget intersectorial restructuring of government expenditures as well as a reduction in transaction costs
provide an option for increasing available resources for social assistance.
Intersectorial restructuring of government expenditures: The reduction in oil subsidies since 2005 created
178
significant additional fiscal resources of up to 10 billion USD a year . However, albeit declining, total subsidies
are still high. In 2006 total subsidies were at Rp. 107,463 billion while total expenditures were at Rp. 669,880 billion
resp. Rp. 443,509 billion at central level. Thus, subsidies make up 16 percent of total expenditure resp 24.23 percent of
total expenditure at central level 179. Continued reductions in subsidies enable more fiscal space for social assistance
programs.
177
“Tax Reform: Time tTo Shift From Income- tTo Consumption-Based Tax”, The Jakarta Post (December 14, 2006).
178
ADB (2007)
179
According to the Ministry of Finance/BPS-Statistics Indonesia.
82
Reducing transaction costs: Within given budgets a reduction in transaction costs increases the share of resources
available for the poor. The preceding section highlighted the current institutional fragmentation. Improved coordination
between agents involved reduces recurrent costs per targeted household/person.
Last but not least, sustained economic growth combined with a clear pro-poor focus is clearly the eminent factor. It
increases tax revenues, while simultaneously decreasing the number of potential beneficiaries.
In addition, any decentralization of social assistance programs without a well-functioning intergovernmental fiscal
system that ensures adequate, sustainable and equitable financing of benefits and programs across municipalities is risky.
Firstly, there is the risk that without earmarking financing sources designated for social assistance programs, money will be
diverted elsewhere. Secondly, there is the risk that the poorest municipalities with the greatest need will have the most
limited resources, leading to a vicious circle in which funds are not available to those who need them the most. Thus,
premature and ill-prepared decentralization of financing responsibility for social assistance benefits can undermine
their provision, as institutional and financing arrangements are unable to cope with the requirements of administering
cash benefits. In the extreme, decentralized financing mechanisms can result in the perverse outcome of widening
inter-regional disparities in coverage, so that in poorer localities poverty levels may actually stay as they were or even
rise. Findings from studies on social assistance in Central Europe and the Baltic States support this suggestion 180.
Therefore, to ensure that a reasonable compromise between efficiency and effectiveness is found, some centralized
redistribution mechanism on macro level is necessary to control and equalize sharing the burden of poverty between richer
and poorer provinces or regions.
180
World Bank (2007c: 37).
83
84
7. Extending Social
Security Coverage to
the Informal Sector
- In contrast to any other population group in Indonesia, the non-poor informal sector workers are among the segment
of the population suffering the highest degree of exclusion at present. Whereas standard contributory social security
is designed for formal sector employees and the poor are targeted by tax-financed and means-tested social assistance
mechanisms, there are only very few programs directed at the non-poor in the informal sector.
- Considering the enormous size of the informal sector in Indonesia, the integration of the informal into the formal
economy should be the ultimate long-term objective (formalization approach). In this case no specific social protection
programs for the informal sector are necessary.
- If an exclusion of the non-poor in the informal sector from public social protection coverage is not deemed tolerable
in the short- to mid-term, specific social security programs have to be designed and implemented as standard social
insurance is not appropriate for reaching out to the informal sector (direct inclusion approach).
- However, if the Indonesian government decides to opt for universal tax-financed benefits with regard to specific risks,
the non-poor in the informal sector will automatically be included and no specific program will have to be initiated.
85
7.1. The Informal Sector: Conceptual Framework

7.1.1. Defining the Informal Sector
Even though the concept of the informal sector (IS) was already introduced in 1972 181, today, there is still no consensus on
how to theoretically define and empirically measure the informal sector. The definitions and conceptualizations of this term
vary over time, space, field of application, and organization. Around some thirty terms including survival sector, non-structure
sector, non-observed economy etc. have been and/or are currently used to describe the IS, making it a somewhat fuzzy
concept. The same is true for its potential causes, relevance and persistence. Over time, major changes in conceptualizing
and understanding the informal sector have occurred (for a summary, please see table 13).
However, the informal sector has been mainly defined in relation to interrelated defining characteristics which can be
classified into two broad categories: a differentiation distinguishing informality with respect to a) certain formal characteristics
of the employment status, the production unit and the economic activity on the one hand or a focus on b) the relation
between the economic actors/units and the state rules and regulations on the other hand.
Definition of IS related to Formal Characteristics of the Employment Status and the Production Unit
The definition of the informal sector with respect to the specific formal characteristics of the economic production
units focuses on the size of the economic unit, its level of organization, the firm-level resource endowment, the technology
applied and the location/physical place of operation. As such, informal sector production units are considered to be of small
size, mainly comprising small and micro-enterprises (productive enterprises with up to/not more than around five employees),
not constituted as separate legal entities independent of their owners (which are either private households or single private
persons) and where the employment of immediate family members, casual workers and apprentices is the norm 182. The
production in the informal sector is said to be mainly labor intensive, requiring only low-level (and often informally acquired)
skills and education. The technology employed is assumed to be simple and minimal. Often, due to a lack of access to
financial markets, financial capital investment tends to be low, but varies significantly by type and scale of informal activity.
Hence, informal sector entrepreneurs have to rely on their own personal savings or equity as their primary source of start-up
capital 183.
The following main four categories of physical work place/location are said to be prominent in the informal sector: a)
home-based workers, b) street peddlers and sidewalk vendors, c) temporary or seasonal workers at the proximity of ongoing
projects of street constructions and d) those that work between homes and streets, such as scavengers.
Sectors and economic activities typically found in the informal sector range from shoe shine services and street vendors,
which contribute rather little to output, to those activities that require more investment in capital and skills such as construction,
transport and manufacturing. Although in the past the term IS has been largely understood to be mainly an urban phenomenon,
more recent approaches, as documented in the characteristics discussed above, clearly point out that informality is not
restricted to urban regions, but is also evident in rural areas, i.e. spanning across regions/geography.
With respect to the status of employment, all types may be found. Employers, the self-employed, and the ones working on
cuenta probista in general, as well as members of producer cooperatives form part of non-wage employment; whereas wage
employment (employees) in the informal sector comprise regular and casual employees, which may include informalized
formal employment such as sub-contract workers 184 and home-based workers 185. Furthermore, informal sector employment
also includes jobs that do not provide a salary in companies that produce income, such as contributing family members.
181
This is the first time the term informal sector/economy has been officially used within an ILO study about Kenya/Uganda; however, different origins dating
back to earlier years may be found. See for example Blunch et al. (2001:5/6).
182
LACF (2007:27)
183
LACF (2007:27)
184
For a more detailed discussion on whether sub-contracted work should be included in the definition of IS, please see later discussion on informal vs.
informalized work.
185
Blunch et al. (2001:3)
86
7 • Extending Social Security Coverage to the Informal Sector
What this kind of definition provides is less a consistent and clear differentiation between the formal and the
informal sector but rather a relevant insight into its heterogeneity. Nonetheless, referring to seemingly cliché-like definitions
is obviously not appropriate for deriving policy implications.
Legalistic Definition of IS
Another defining characteristic of the informal sector is that it is said to operate outside the legal and wider regulatory
(including social protection) environment, e.g. that it is in a sense illegal. This refers to both the IS being a) “un- or under-
regulated” meaning (partially) outside the scope of laws and regulations and informality being related to b) non-compliance
with applicable law and/or regulations. Hence, legal status of an enterprise or individual is not restricted by a single criterion,
but by a multiplicity of regulations, such as tax laws, registration procedures, labor laws among others.
With the rise of new forms of employment (such as just-in-time employees) and subcontracting (in regular home work, in
registered sweatshops or in export processing zones), being associated with greater flexibility in production at the expense
of job stability and social security benefits, the so called “social protection” definition of informality has specifically emphasized
the workers´ legal status (the coverage of workers by mandated labor protection) but also the non-compliance with other
types of state regulations, as a distinguishing characteristic of informality, which is less concerned with the nature than the
quality of employment. It focuses on non-compliance by either workers or enterprises with all or some of the national
or local legislation, especially related to commercial and/or labor regulation. Informal sector working relations in
this vein are characterized by little or no appreciation of industrial relation norms and workers rights and a lack of
adequate social protection means.
This new focus expands the definition of informal sector related to the size and other formal characteristics of the economic
unit to include informal contractual arrangements among otherwise formal entities, so called informalized formal work.
Hence, this definition highlights the need for a more nuanced debate of issues related to the informal sector with a greater
emphasis on informal employment then employment in the informal sector 186, e.g. the division between informal salaried
workers in any size firm (whether formal or informal with respect to the size and other characteristics of the economic unit)
and the informal self-employed 187.
Furthermore, some enterprises or workers may operate in a grey zone by only partially complying with government
regulations or only complying with some applicable laws and regulations (i.e. such as tax law), and not or not fully with others
(for example social protection, labor law, underreporting of sales etc). Hence, a continuum of informality exists with
respect to legality, with complete non-compliance or non-regulation on the one end of the spectrum and complete compliance
and regulatory coverage on the other. As this implies that one can define the borders of the informal sector arbitrarily, as
narrowly or broadly as one wishes, it is also not precise enough to clearly distinguish the informal sector from the formal one.
7.1.2. Proposed Working Definition of the Informal Sector

As both definition approaches of the IS outlined above
exhibit different kinds of weaknesses and are open to
substantial critique if considered separately, a definition
of the informal sector seems appropriate which combines
the major elements of the two different informal sector
definitions, i.e. the production unit and legalistic view,
as used in the definition of the ILO, which can be found
in the table below. The rows represent a differentiation of
the informal sector by type of production unit (formal sector
enterprise, informal sector enterprise, household), while a
definition linking the type of employment status to the workers´
legal status is given in the columns.
186
Hussmans (2004)
187
Blunch et al. (2001)
87
Table 14: ILO Conceptual Framework: Informal employment
Source: LACF (2006:27)
Whereas formal sector enterprises may comprise big enterprises as well as dynamic and competitive small and medium ones,
informal sector enterprises consist of micro-enterprises (with no more than five employees) as well as family businesses (operated
by non-professional own account workers with or without contributing family members). Households, on the other hand, refer to
households that employ paid domestic workers and those that produce goods exclusively for their own final use. From the table
above it can be seen that informal employment refers to all numbered cells, with the exception of cell 7, whereas informal
employment outside the informal sector is given in cells 1, 2, 9 and 10.
7.1.3. Explaining Informality or Why Individuals Operate Informally

Whereas earlier approaches attached a negative connotation to informality (as defined before in a legalistic sense), nowadays
informality, viewed from a broader perspective, is considered to have the potential to contribute positively to poverty reduction,
employment, economic growth and social balance. Hence, the informal sector, if appropriately targeted and framed, may
also be conceptualized as a way of leading to development. However, in order to design such policies for the informal sector
directed at unleashing its positive developmental potential it is necessary to understand the reasons for informality, e.g. to
understand why individuals do not comply with state regulations, and hence operate informally in this sense.
Three different views prevail: the first perspective argues that individuals take part in the informal sector by chance or by
default, e.g. as a survival strategy to earn their livelihoods because they are excluded from critical state benefits along with
the circuits of the modern economy and because of restrictive constraints in the form of prohibitively costly state regulations,
which therefore need to be overcome to reduce informality. This is the basis of the more traditional view that the informal
sector is a labor market segment consisting of displaced workers queuing for better jobs in the rationed formal sector 188.
But not all enterprises and workers operate informally because they cannot afford the high formalization costs, but, on the
contrary, due to voluntary choice. Some individuals can more or less freely decide upon their optimal levels of engagement
with regulatory and legal institutions of the state, that is they choose whether to cross the relevant margin into informality
or not based on individual cost-benefit calculations, weighing the benefits associated with formality against the
related costs. Potential motives (benefits of informality) underlying this calculation might be related to increased labor
market flexibility, profitable market niches and opportunities as well as profit maximization via evasion of otherwise affordable
state-regulation 189.
188
Henley et al. (2006:6).
189
Henley et al. (2006:6).
88
A third set of informal sector explanations (which presents a somewhat middle ground perspective of the IS between the first
two discussed approaches) focuses on the subjectively perceived or existing de facto irrelevance of state norms and regulations
for specific kinds of economic transactions. This view conceptualizes the informal sector as a set of informal institutions and
personalized relations that may be either seen as substitutes in the absence of formal/state institutions or as competing
with state institutions when these are perceived as being less relevant if not irrelevant at all. In this perspective, the
informal sector coexists beside the formal sector, following its own logic in terms of norms, rules, and types with interactions
taking place that are more generally based on trust, reciprocity and other elements of social capital 190. An often-cited example
for this perspective can be found in Geertz´s seminal “Peddlers and Princes”, in which the evolution of the Indonesian bazaar
economy to that of the more rational, modern firm is described. According to Geertz, the Indonesian bazaar economy
encompasses “firms” that are relatively unregulated and hence unconstrained by state institutions, but are characterized by
a web of informal institutions related to the management of credit, risk, and irrigation 191.
Although these explanations exhibit quite different foci of attention, potential reasons and functioning of informality, they
might most adequately be seen not as contradictory, but complementary. Thus, informality in a given country should be
viewed as being caused by a confluence of factors that may most properly vary in likelihood and relevance for
different segments of the IS as defined above.
Overview: Evolution of Conceptualizations, Characteristics and Views on the Informal Sector

Over Time
The following table summarizes the main points discussed above to assist with policy development.
Table 15: Conceptualizations, Characteristics and Views on the Informal Sector
Aspects, along
which perspectives Traditional View Modern View
differ/vary over time
Legality IS individuals are characterized by IS individuals are characterized as operating on a

operating completely illegally continuum between legality and illegality
complying with some rules but not with others
Reasons for operating IS participation by chance, due to IS participation by chance, by choice and due to
informally prohibitively high internal, external, irrelevance. Motives include labor market
and inter-firm constraints flexibility, existence of profitable opportunities,
non-compliance with regulations although
possible to comply
Employment status Mainly small and micro entrepreneurs IS covers all types of employment
(employers, self-employed and family status, including non-wage and wage
labor) that manage unregulated or illegal employment as well as employment in
and unlisted enterprises informal sector and informal
employment in formal sector.
Short-or long term IS viewed as traditional and IS viewed as part of an economy,

phenomenon & linkages transitional sector, separated from formal persistent and even expanding phenomenon,
with the formal sector economy; a phenomenon of existing side by side and even differently
underdevelopment interlinked (subordinated or superordinated)
with the formal sector.
190
LACF (2007).
191
Geertz (1963).
89
Economic activity and IS mainly consists of street vendors and IS consists of a variety of economic activities,
sectors included small-scale producers; a purely ranging from casual workers in the construction
urban phenomenon sector and agriculture to home-based
sub-contracting workers in high-tech
industries; urban as well as
rural phenomenon.
Technological base IS production is more or less static; Considers that some segments of the
usage of traditional technology and IS may be highly dynamic and
copycat strategy innovative (processes, inputs, and outputs);
able to adapt to new market niches.
Earnings & poverty Earnings tend to be very low in Although poverty is often quite substantial,
the informal sector; poverty is a key, earnings vary considerably and systematically
if not defining characteristic of the IS by occupational status, with the
self-employed/micro-enterprise employers
having the highest earnings, followed by
regular wage earners, causal wage
earners, and apprentices.
Economic, societal and IS viewed as marginal and residual IS is fundamental (with significant potential of),
political contributions sector in terms of place in and contributing to productivity, employment, and
contribution to overall economy; output, poverty reduction, and the legitimacy
characterized especially by as well as functioning of economic, distributive
low productivity and political institutions, implying social balance
and inner political stability.
7.2. The Current Situation

Statistical Sources Used, Their Characteristics, Quality, and Compliance/Consistency with GTZ
Definition of Informal Sector and Implications for Data Interpretation
As the aforementioned highlights, some caution with empirical data on measuring and characterizing the informal sector in
a given country is generally warranted 192. To understand how far the statistics and other data used within the following
characterization of the informal sector in Indonesia actually comply with the proposed definition of the IS outlined in the
preceding subsection, and whether and how far comparability of the different data may be matched, a brief overview and
characterization of the data used is provided in the following 193.
In Indonesia, the IS is defined by the bureau of statistics (BPS) according to a combination of the employment status and
the type of main job. Since 2001, BPS distinguishes seven types of employment status: 1) own account, 2) self-employed
with assistance from family members, 3) employer with permanent worker /waged worker, 4) employee/staff/worker, 5)
casual worker in agriculture, 6) casual worker not in agriculture, 7) unpaid worker. Formerly the final two were embedded in
employee and own-account worker. Taking the type of main job into account, informal sector workers are a) self-employed
assisted by family workers or temporary agricultural workers and b) the self-employed without assistance, agricultural and
192
Describing and analysing the informal economy in a specific country is a rather difficult and inevitable/inherently imprecise task. International comparability
is even further constrained, if not impossible due to several reasons, for example the usage of different definitions of the informal sector by different
organisations and /or variability in defining the IS over time by the same type of organisation, a high unreliability of data due to a lack of willingness to
truthfully reveal the status of and reasons for informality in case of surveys, the sensitivity of results to a) definitions used and b) statistical extraction and
processing techniques employed and a collection of data on an ad hoc rather than on a regular basis.
193
As the existing data is often not consistent with the working definition of the IS used above, categorizing and disaggregating the IS as done in the matrix
in this chapter will be severely constrained and the description of the IS inevitably incomplete, but, at the same time, revealing potential data needs.
Furthermore, caution is warranted in compiling, comparing, and interpreting the different data. Thus, where applicable, differences in the definitions that
severely diverge from the definition of the IS in this paper will be highlighted and possible implications discussed.
90
non-agricultural freelance worker whose occupation is other than technical, professional, and related workers; administrative
and managerial worker as well as clerical and related workers (as shown below):
Table 16: Categorization of the Indonesian Labor Force
Main Occupation
Employment
Pro- Manage Clerical Sales Service Agricul- Produc- Opera- Laborer Other
Status fessional rial worker worker provider ture tion tor
worker worker
Own account F F F INF INF INF INF INF INF INF
worker
Self employed F F F F F INF F F F INF
assisted by
family member/
temp help
Employer F F F F F F F F F F
Regular F F F F F F F F F F
employee/Worker
Casual worker in F F F INF INF INF INF INF INF INF
Agriculture
Casual worker in F F F INF INF INF INF INF INF INF
non-agriculture
Unpaid worker INF INF INF INF INF INF INF INF INF INF
Source: The Situation of Indonesia Labor Force 2006, BPS, Jakarta
At least three sources can be identified that can be used to derive characteristics of the Infomal Sector in Indonesia:
the National Labor Force Survey (NLFS)

the Economic Census
the Integrated Survey of Small (ISSME) - Scale and Micro Enterprises
Not only do these sources differ with respect to the data extraction techniques used, but also with respect to how the informal
sector is defined. The NLFS distinguishes formality and informality along the line of employment status, considering all types
of employment status including employers with permanent workers and employers, as formal employment; and the rest,
including types of agricultural work, as informal. The economic census and ISSME define informality in terms of compliance
with legal norms. In this a distinction is made between directory (DE) and non-directory establishments (NDE) with the NDEs
being the basis for defining informal establishments. DEs are all entities that have legal status in any of the following
meanings: a) whether unincorporated or incorporated, b) establishments granted permission by local authority to operate
(government) and c) those registered with some ministries in charge of the sector development.
This has implications not only in terms of limited comparability of the data but also related to the IS definition proposed
above. The Economic Census and ISSME would call establishments formal that would be considered informal by the GTZ/
ILO definition. Because others only partially comply, they would be listed as informal under our definition, but also formal by
the EC and ISSE. Informality in the sense of the working definition would also comprise part of agricultural workers that are
completely excluded in the Economic Census and ISSME. Hence, figures compiled out of these data will most likely lead to
a picture of the size of the IS in Indonesia that underestimates the true figure. As both the NFLS as well as Economic
Census/ISSME exclude persons that are absent for more than six months, overseas migrant workers, who form part of the
IS in the above defined sense, will also not be included. Moreover, the NFLS defines employment as “workforce aged 15 and
above who work for pay or assist others in obtaining pay or profit for duration of at least one hour during survey week”;
hence, informal child labor is excluded and unemployment rates are expected to be quite low.
91
Another important empirical source on which this compilation of the IS in Indonesia is based, are two surveys undertaken
by the ILO in 2001 and 2003, explicitly designed to assess social security needs and capabilities of the IS in Indonesia.
These approaches, although slightly biased in terms of provinces and sectors surveyed, make use of the same kind of IS
definition as proposed in this paper.
7.2.1. Identification and Characterization of Excluded Individuals: Socio-economic

Characteristics of the Informal Sector in Indonesia
Size and Heterogeneity of the Indonesian Informal Sector
One overarching characteristic of the IS in Indonesia is its heterogeneity; particularly related to differences between
provinces; whereas some provinces are more rural, others are mainly urban; some are conflict ridden (Aceh/Nias), others
struck by natural catastrophes etc. Most of the establishments without legal entities are situated in East and Central Java
(around 21 percent), West Java (17 percent) and Sumatra (15 percent) in 2004 194.
Indonesia has faced and is still facing large scale shifts from formal to informal employment (there has been a steady
increase in the number of establishments without legal entities since 1998, which amount to 17,145 thousand in 2004),
with the IS becoming the mainstream by its sheer numbers, which is reflected by the fact that the IS comprises around 70
percent of the national labor force with the total number of workers in IS amounting to 67,866,398 195. According to Blunch et
al. (2001) around 77 percent of non-agricultural employment in Indonesia is considered to be informal.
This highlights that the IS in Indonesia is not only quite substantial but also growing. The increasing trend/growth of
informality but also the provincial variations can be traced back to a number of reasons, the most important being: a) Asian
financial crisis, b) inner political upheavals, c) terrorist attacks, as well as d) natural catastrophes such as earthquakes and
tsunami.
Sectors Represented in the Informal Sector in Indonesia

According to national statistics, the IS sector in Indonesia, excluding agriculture, is relatively service oriented. According to
Economic Census/ISSME, the IS in 2004 was predominantly represented by the following sectors: a) wholesale and retail
trade, restaurants and accommodation services (61 percent), b) manufacturing industry (16 percent) followed by c) transport,
storage and communication (13 percent), d) financial institutions, real estate, rental services and others (8 percent), and d)
mining and quarrying; electricity, gas and water supply, construction (2 percent). If one accounts for the number of employees
of establishments without legal entity, 58 percent of employees work in the wholesale and retail trade, restaurants
and accommodation services sector, around 21 percent in the manufacturing industry, only 2 percent in mining etc.
and around only 9 percent each in the remaining two sectors.
The Indonesian rural sector employs around 44 percent of the actual national labor force 196, if non-agricultural
workers are included. However, quite a substantial part of the rural economy consists of subsistence farmers, receiving
only part of their income as cash, but mainly in form of crop shares of which a large portion are used for local consumption.
This may indicate that the agricultural sector should be highly represented in the IS, too.
Other important sectors of informal employment are related to home-based work and sub-contracting, which is primarily
associated with female work. These are mainly: pottery, batik, and rattan. Whereas batik and pottery are produced for the
local or regional domestic markets, rattan is also exported 197. Generally, these activities are not based on a written contract,
which might be closely related to the fact that their contractors are generally neighbors or relatives.
194
Economic Census/ISSME, BPS (2004).
195
NFLS (2007).
196
BPS: NLFS.
197
Mehrotra and Biggeri (2002).
92
Industrial structure and employment status in the informal sector

The Indonesian economy as such is characterized by a “missing middle” in its industrial structure of which a rather small number
consists of large firms, whereas small domestic market oriented business can be found at the other extreme. Micro-enterprises
(according to the Indonesian Bureau of Statistics, comprising not more than 4 employees) account for 90.85 percent
by number of firms and for 38.10 percent by employment 198. Hence, livelihoods quite heavily depend on them.
According to NFLS, non-wage labor in the form of self-employed without and with assistance by family members together form
the majority of employment status in the IS in 2007 (accounting for 58 percent), followed by unpaid workers with 26 percent,
whereas casual wage employment accounts for 16 percent. Surveys conducted by ILO in 2001 and 2003 199 draw a somewhat
similar picture, but point out geographic differences. Whereas in urban areas two-thirds are self-employed with 36 percent of
the self-employed not employing any other person, and the remaining 64 percent employing more than one employee (with a
ratio of 5:1 employees/employers), in the urban regions, around 41 percent of the rural workforce represent casual and unpaid
workers, with only 3.3 percent receiving regular income or being regularly employed. Nonetheless, most of them (44.6 percent)
are self-employed. Around half of all respondents own their means of income such as a boat or land.
Personal Details of Individuals and Households in the Informal Sector

According to an ILO study, about 70 percent of those surveyed in urban areas were married. The average number of
dependents for married people was 2.79 per family, the number of dependents in each family ranged from 1 or 2 up to 12
(with the majority having 2 to 4 dependents). On the other hand, the number of the individuals being married in the rural
regions was higher and amounted to 93 percent, with 3.17 dependents per family. Nonetheless, more than 600 families
(which make up 27.7 percent) exist in the rural regions with more than three dependents. Generally, the age dependency
rate has decreased in the last decade to almost 50 percent, showing that the population is becoming older 200.
Poverty Characteristics of the Informal Sector in Indonesia

Even if all poor individuals in Indonesia (around 38.000.000 in 2002) would form part of the IS (which amount to roughly
63.000.000), a quite significant number of around 60 percent of individuals in the IS would have to be considered as
non-poor. In addition, according to an ILO study, although the majority of individuals in the IS sector earn less than the
average income, around 17 percent of those surveyed earned around twice the national average 201.
Hence, although there is a tendency to associate informal sector activities with poverty, it cannot be presumed that earnings
or wages are generally lower than in the formal sector. Nonetheless, a large diversity of income levels exist (Tambunan
2004), especially in urban areas or between urban and rural areas respectively, with individuals in the rural areas earning
generally substantially less than in the urban regions. Furthermore, income in these areas seems to be highly uncertain
(Wiebe 1996). Nearly all individuals in rural areas do not receive regular income and more than half of them had some
months without any income at all.
Furthermore, according to an ILO study (2001, 2003), the type of income received differs sharply from that in the urban
regions: only 19 percent receive income in cash with the remainder receiving only varying percentages of their income in
cash, but in goods or in kind. Half of the respondents received only up to 60 percent of their income as cash. Much of the
income from produce comes from sale and only less than a fifth is bartered or used for self-consumption. Females show
disproportionate shares of adverse incomes: more females receive lower incomes: around 44.6 percent of women
earned less than Rp. 200,000 per month, with the comparable figure for males being 22.4 percent.
With respect to other poverty dimensions, it needs to be mentioned that not only the maternal mortality rate can be considered
to be quite high, even much higher than other comparable countries in the region (307 per 100,000 live births), with maternal
deaths accounting for around 21 percent of all female deaths of reproductive age) but also the malnutrition rate 202.
198
The World Bank (2006d).
199
200
World Bank (2006c).
201
202
World Bank (2006c:33).
93
Living Conditions in the Rural and Urban Informal Sectors

Although Indonesia still depends on a rather rural and agriculture economy, around half of the IS live and work (often at the
same place) in urban areas, showing an increasing trend towards rural-urban migration 203, and most likely related to this, an
increase of slums and informal settlement areas. Access to safe drinking water differs significantly between rural and urban
areas. Whereas 90 percent of the urban population has access to safe water, this amounts to only 69 percent in rural regions 204.
However, this relation is somehow reversed with respect to the provision of sanitation, which is highly inadequate, being the
worst in the East and South East Asian region. Less than 1 percent of all Indonesians are provided with piped sewerage
services. According to SUSENAS data, 59 percent of the urban poor have no access to adequate sanitation, compared to 80
percent in rural areas (World Bank 2006c).
Gender Characteristics of the Informal Sector in Indonesia

Women form a major part of the informal sector in Indonesia and are over-represented in unpaid and low-paid jobs.
According to Gallaway and Bernasek (2001), up to 65 percent of economically active women in Indonesia are found in the
informal sector. Furthermore, women form the biggest share of the increasing number of overseas migrant workers, which
heavily contribute to rural income via providing remittances. As such, they work as domestic helpers, a job often characterized
by a lack of employment relationships and labor protections, often facing severe violations of worker rights, such as rape,
harassment, cuts in wages and dire working conditions.
But not only are women victims of violence and crime in case of overseas migrant work, but also in regions of civic strife and
military conflict: Aceh, Maluku, Poso, Papua, and Central Kalimantan are among the areas where community and military
violence has seriously affected the lives of women and children with women left widowed and/or stigmatized if abandoned
pregnant and child bearing. In addition, in March 2001, 1.1 million people were internally displaced, most of them women,
children and the elderly.
Women form the majority of home-based workers, amounting to around 85 percent to 89 percent of the total women
working in pottery, rattan, and batik sectors. Whereas home-based working women have mainly received primary and junior
high school levels of education, women working in the same sectors but in formal employment generally show higher levels
of education 205.
Furthermore, studies have shown that the majority of women in the IS have at least two jobs, one at home and one
outside the home, receiving less education and financial reward. From this follows that they have less free time and
means to decide for themselves and often lack basic knowledge on health risks and safe working practices. Due to this, the
following diseases are predominantly found among working women in the IS: malnutrition, anemia, infectious diseases,
malaria, visual, hearing loss, pesticide poisonings, and respiratory problems. Moreover, indications of neuropathy after more
than 20 years of pesticide spraying, allergic contract dermatitis associated with use of detergent in dish washing, back injury
after selling groceries, or seamstresses suffering from asthma 206.
7.2.2. Problem Statement

As shown above, the informal sector in Indonesia comprises a substantial and growing part of the working population, i.e.
more than 69 percent the total labor work force according to the NLFS (2007). However, the analysis in the preceding
sections also revealed that belonging to the informal sector does not necessarily imply a low income status: Around 60
percent of individuals in the IS can be considered as non-poor and the survey by Angelini and Hirose (2004) showed that 17
percent of those surveyed earned around twice the average national income 207.
203
McCulloch (2007).
204
World Bank (2006b).
205
Mehrotra and Biggeri (2002:43).
206
Markkanen (2004:19-20)
207
Again it needs to be stressed that some caution needs to be applied when interpreting those data as at present no representative survey regularly
assesses the income status of informal sector workers specifically. However, even if not fully precise, the data hint at the fact that the non-poor part of the
informal sector is non negligible.
94
The following analysis deals with how to extend coverage to the non–poor in the informal sector: Extending social
protection coverage to the poor or near poor will be achieved by adequate social assistance measures as analyzed in
chapter 6. However, the non-poor in the informal sector workers are excluded from any larger-scale social protection
program at present. They are neither eligible for measures targeted at poor households nor reached by contributory social
security as Government Regulation No.14/1993 limits the compulsory coverage to the formal sector enterprises with 10 or
more workers and a monthly payroll of Rp. 1 million.
Voluntary social security programs targeted at this group are still very limited in scope. For example, at present, JAMSOSTEK
as well as the Ministry of Social Affairs are running programs targeted at the informal sector. JAMSOSTEK is trying to extend
its insurance coverage to the informal sectors whereas the Ministry of Social Affairs is offering a saving scheme to informal
sector workers (ASKESOS). JAMSOSTEK employee’s social security scheme is a pilot project for self employed workers,
providing benefits in the case of death and funeral expenses on the basis of a regular income-related contribution. ASKESOS,
implemented by the Ministry of Social Affairs (DEPSOS) is a social welfare insurance specifically targeted at the self-employed
and informal sector workers, e.g. small traders, service sellers and agriculture laborers and fishermen, but who have an
income below Rp. 200,000 per month. It provides protection against the risk of a loss in social welfare in the event that the
main breadwinner suffers death, sickness or accident. In addition, provincial/district authorities are appointing the local
ASKESOS implementing institution, which have to be a legal entity. Commercial insurances are only an option for the better-
off segment in the informal sector and micro-insurance coverage is still limited in Indonesia.
Thus, in contrast to any other population group in Indonesia the non-poor informal sector workers are among the
segment of the population suffering the highest degree of exclusion at present.
7.3. Reform Options

7.3.1. General Reform Options
As the objective of the current reform agenda is to extend coverage to every Indonesian citizen, strategic options of how to
include the non-poor informal sector workers need to be considered. Law No. 40/2004 provides the legal basis. With regard
to beneficiaries it distinguishes between three population groups: employees (those earning a wage), participants receiving
contribution assistance and participants paying contributions (but receiving no regular wage). Particularly the latter category
comprises the non-poor in the informal sector. According to Law No. 40/2004 all benefits except pensions, i.e. health, work
accident, death and old age, shall be extended to “all participants paying contributions”. Considering the heterogeneity of the
non-poor informal sector workers is not an easy task. As will be shown in the succeeding sections it is not achieved by
simply removing the restriction mentioned above, as standard formal social security programs are not suitable for
reaching out to the entire informal sector.
With regard to each risk (i.e. health, old age, accident, and disability), three broad strategic options exist:
Formalization approach: Within this approach there would be no specific measures to include informal sector workers in
social protection schemes. Rather, efforts would be directed at the improvement of the social security system for formal
sector employees, thus increasing the incentives for informal sector workers to enter the formal economy. However, this
approach should be embedded in a comprehensive national strategy aimed at removing existing barriers for moving from
the informal into the formal sector (e.g. restrictive labor laws and regulations, combating tax evasion).
Direct inclusion approach: Direct inclusion aims at integrating the informal sector into the national social security system.
If the Indonesian government opts for universal tax-financed benefits with regard to specific risks where less path dependencies
yet exist, the entire population is automatically included and no specific programs are necessary. However, for risks covered
by social insurance mechanisms, including the non-poor in the informal sector involves the design and implementation of a
specific national program targeted at the informal sector as an integral component of the national social security system.
Informal sector workers would pay a contribution and then be entitled to the benefits offered by the national social security
system.
95
Indirect inclusion approach: Indirect inclusion would put a stronger emphasis on private activities. Informal sector workers
would not be included in the national social insurance system as such. Rather, the public sector would enable and strengthen
private approaches to social protection, e.g. by offering capacity building measures or start-up subsidies to organized groups
(large cooperatives or umbrella organizations of informal sector workers) in order to enable those groups to provide micro-
insurances to their members.
These strategies are not excluding each other. Rather complementing combinations are possible. In any case the
integration of the informal sector into the formal economy should be the overall and foremost objective. In this
context, the formalization approach tackles the essential underlying problem (i.e. reducing the informal sector), hence being
the most sustainable approach in the long run. However, in the time before the achievement of this objective the informal
sector would have to primarily rely on private initiatives. There are numerous examples worldwide where “third sector”
organizations 208 have provided social services for their members. Although third sector provision of social security might
possibly be more flexible concerning implementation and adaptation to changing needs of members, severe drawbacks
exist with regard to risk pooling, financial viability and outreach limitation. Therefore, if an exclusion of informal sector
workers in the short and mid-term is considered as non-tolerable, strategies aimed at the integration of the informal sector
should be complemented by either the direct or indirect inclusion approach. Both approaches involve high transaction costs
(e.g. developing a legal and regulatory framework, forming appropriate delivery structures, capacity building). But as the
micro-insurance sector in Indonesia is still rather small, in terms of potential outreach the direct inclusion approach offers the
better perspective. Thus, a combination of the priority objective of formalization together with the direct inclusion
approach is recommended and will be analyzed in the following sections.
The analysis of the direct inclusion approach will focus on contributory programs, as universal tax-financed benefits
need no specific separate institutional structures to include the non-poor in the informal sector.
The International Conference on Extending Social Health Insurance to Informal Economy Workers in Manila 2006 offered an
interesting forum for the exchange of existing experiences regarding the inclusion of informal sector workers. Although a
primary focus was on social health protection, the presented experiences transcended the realm of health and touched on
social protection of other risks 209. The country examples presented here will draw on the experience presented at that
conference. However, it needs to be stressed that the evidence-base for successful examples of how to integrate the
informal sector into social security programs is still weak. Thus, much of the following discussion will be at the
conceptual level. Nevertheless, the lack of best practices should not discourage policy-makers: Being among the pioneers
who implement coherent and encompassing social security programs for the informal sector offers the chance of providing
the best practice model for other countries.
7.3.2. Integration of the Informal Sector: the Formalization Approach

Constraints Faced By Informal Enterprises and Laborers
As pointed out above, the formalization of the informal sector should be the ultimate objective. This renders it necessary to
analyze the constraints informal sector workers are facing that form a barrier to enter the formal market. In addition, existing
incentives, which induce informal sector participants to stay voluntarily within the informal sector, have to be taken into
account too.
Although there are differences between rural and urban enterprises as well as big compared to small and micro-enterprises,
the following summarizes some of the most important constraints faced by all types of enterprises: One of the major
constraints faced by enterprises in Indonesia arise from the demand side in form of a lack of sufficient (domestic) demand
for products, which may be strongly related to a lack or low level of income of the Indonesian citizens. According to several
private investment climate surveys conducted by World Bank and ADB respectively 210, among the supply side factors are:
208
The “third sector“ economy is composed of all self-help organizations which neither belong to the public nor to the private sector and are characterized
by their not-for-profit orientation and social ethics.
209
http://www.shiconferencemanila.info/.
210
World Bank ADB Private Investment Climate Survey (2003/4), World Bank (2006d).
96
a) policy uncertainty, b) corruption by local and national governments, c) high regulatory burden 211, aggravated by d)
a lack of coordination and confusion about the roles and responsibilities of national and sub-national governments
with respect to business registration and licensing, e) restrictive labor legislations 212 as well as a f) lack of access to
adequate financial services 213. In addition, an g) inefficient Tax-VAT administration, which leads to tax refunds being
either rarely received, or with quite a high time lag and h) a lack of necessary working skills 214 are also obstacles to
improve the growth of enterprises. Furthermore, several g) external shocks, such as the terrorist attacks in Bali, but also
earthquakes and Tsunami have had detrimental effects to small and micro-enterprises.
One of the major constraints faced by informal laborers is related to the high level of underemployment. Although the
level of unemployment according to national statistics amounts to around 10 percent 215, there are more than 30 million
people that are underemployed (working less than 35 hours per week), with young people (between age of 15 to 34) and
women being more severely hit. Of the underemployed, 44 percent are involuntarily underemployed with females
being more voluntarily under-employed than males. This is also supported by research done by Gallaway and Bernaseck
(2001) on examining the determinants of labor force participation for women and men in Indonesia, suggesting important
gender differences, with the decision to participate in the labor market in case of females being negatively correlated with the
existence of children in the household. In addition, the high number of tertiary educated in the urban informal sector market
is significant and may be interpreted as an indicator for formal labor market rigidities in the form of a limited availability of
skilled jobs and the entrepreneurial nature of the informal sectors. On the contrary, the education level in the rural areas is
quite different from that of the urban regions with a significant number of people with no or only primary education and low
transition rates from primary to secondary education 216. Furthermore, skills upgrading by private investors and enterprises is
rather low with only 23 percent of firms offering formal training to their employees compared to 42 in Malaysia and 69 in
China 217. All this indicates that the quality of both private and public education may still be lagging behind what is necessary,
thereby severely limiting not only SMEs but also informal laborers. Hence, the labor market in Indonesia exhibits three
major problems that severely limit the formalization of informal sector laborers: unemployment, under-employment
and low productivity due to low level of appropriate business training 218. A review of occupational safety and health
in the informal sector in Indonesia conducted by DK3N/Joedoatmodjo in 1999 219 found that among the most cited work
hazards were noise, vibration, heat stress, poor lightening, unsafe electrical wiring, exposure to dust and chemicals, and
poor ergonomics. Workers in the informal economy in Indonesia suffer from malnutrition, parasitic diseases, asthma, skin
allergies and cancers, chemical poisoning, musculoskeletal disorders, food poisoning, respiratory track problems and blood
diseases.
Strategic Problems Incorporated in Formalization Strategies

In addition to the above-mentioned constraints faced by informally operating enterprises and laborers that direct attention to
reforming the business regulatory and labor market environment, efforts to formalize informally operating enterprises are
also plagued by strategic issues. This is so, because the decision to operate formally or informally is a strategic
211
Regulatory burdens limiting SME growth and sustainability mainly include practice of different local governments imposing unnecessary regulations. As
several Kabupaten impose charges and taxes on vehicles entering or leaving their district (which are often against national law), they severely restrict
potential gains from trade between the different regions, which could be highly beneficial considering the substantial regional differences within the country.
(World Bank, 2006c:xiii).
212
For example labor law that limits sub-contracting to only work that is auxiliary to the firms operations (thereby severely limiting business opportunities and
also competitiveness of SMEs).
213
Of specific importance are the subsidies credit schemes, which have damaged the access of SMEs to credit, as they lead to a deterioration in repayment
culture and undercut commercial loans, thereby reducing the incentives for bank and non-bank (microfinance) financial organisations to develop this sector
as a profitable market niche. Furthermore micro-credit banks in rural areas are facing fierce competition, and are lacking coordination by a central bank,
which further reduces their sustainability (World Bank, 2006d:xii).
214
A survey among leading business individuals in Indonesia has shown that there is a profound lack of working and business knowledge of graduated
students from local schools due to a lack of professional and business education at high school levels but also training institutions (FIAS, 2005).
215
It needs to be considered that being employed is quite broadly defined by national Indonesian statistics as individuals having at least worked one hour
in the respective survey week.
216
Angelini, Hirose (2004).
217
World Bank-ADB (2003/4).
218
Hasoloan 2006; World Bank (2005b).
219
Hasoloan 2006; World Bank (2005b).
97
decision: whether or not individuals decide to operate formally depends, among others, on the actions taken and behavior
realized of other relevant actors or players, particularly of competitors both in the formal and informal sector, as well as the
behavior and credibility of state actors.
First of all, the decision to operate informally may be related to a lack of coordination between competitors, e.g. it may not
make any sense to comply with government regulations unless competitors do the same, as otherwise one not only has to
pay the costs of compliance but also the costs in form of a competitive disadvantage. Another dimension concerns credibility
issues, e.g. the decision to operate formally or informally also depends on whether the governmental actors at different
levels uphold their promises or not, as, once registered, enterprises are easier targets for predatory or legitimate state
interventions. A last obstacle to formalization is often related to a lack of information about already enacted programs and
policies. As indicated by a recently conducted survey 220, a lack of information of IS individuals about specific social protection
programs is still very high.
Hence, among the strategic issues that are to be taken into account in the effort to induce and increase compliance with state
regulations coordination, credibility, and information problems are of utmost importance and need to be addressed in
addition to the aforementioned reforms related to the business environment and labor market policies and regulations.
Therefore, viewing formalization as an inherently strategic problem directs attention to the question of how to achieve an
outcome in which both formal and informal enterprises comply with government regulations and in which governmental actors
enforce laws and regulations consistently and fairly. Among the most important solutions to realize such an outcome are those
that increase the costs of non-cooperation, the benefits of cooperation as well as the likelihood of detecting and sanctioning non-
cooperative behavior of the different formal and informal, private and state actors involved 221.
Recommendations for Increasing Formalization of Informally Operating Enterprises and Laborers

Mechanisms to induce formalization of informally operating enterprises of various kinds and reasons that not only take
into account the before mentioned constraints but also strategic problems of the aforementioned kind are the following:
Simplify regulatory framework: stop general subsidized credit programs and improve micro-finance markets, reduce
local government regulations that are not in line with national government laws and regulations and that inhibit inter-
regional domestic trade. In addition, reform the tax refund systems to reduce time for tax refund. This also includes
increasing labor market flexibility by reducing restrictions on outsourcing and the costs of firing a worker. This needs to
be complemented and balanced by social protection and other labor laws directed at protection of the work force (see
below).
Connect the costs and benefits of compliance with state regulations via linking the compliance decision to access to
markets for finance, skills and/or social insurance benefits.
As formalizing totally unorganized informally operating enterprises/individuals is rather difficult if not impossible, it is
advisable to work through existing intermediaries such as business associations by making benefits of membership
contingent on regulatory compliance. Where non-existent, government may even play a role in encouraging the
establishment of such business or self-employed associations. Formal enterprises (domestic as well as
international), although in a quite limited way, can also serve this objective; for example, by screening their potential
suppliers via employing criteria that are based on the compliance with certain state laws and regulations (such as
social, environmental and sustainable standards).
Apart from setting incentives mainly directed at increasing the benefits of compliance, increasing the costs of non-
compliance in the form of improving detection and sanctioning via enhanced coordination between different
state agencies and/or different levels of government is indispensable. In this vein, the lately created one-stop-
shops (OSS) are an appropriate starting point. Nonetheless, as work by IFC-PENSA (2004) shows, it is still possible to
improve performance of these one stop shops by more transparent and standard procedures of licensing in terms of
220
The Jakarta Post (December 4, 2006).
221
Kenyon (2007).
98
cost, time, and application procedures, encouraging commitment of the head of regencies to the development of OSS
in his/her region and focusing on capacity building of OSS staff with respect to macroeconomic and institutional training,
as well as exchanges among OSS staff to conduct comparisons, potentially leading to benchmarks.
Last but not least, certain types of public sector management reforms may also be advisable to increase credibility
of governmental actors. This particularly concerns direct anti-corruption measures as well as reducing the discretionary
space of governmental officials by simplifying regulatory procedures. Of particular importance in Indonesia is to streamline
the location of regulatory authority between national and sub-national government and the responsibilities of different
regulatory governmental agencies. In addition, the national government should direct attention to and eliminate anti-
competitive behavior by local governments, which often favor local economic interests by setting up barriers to entry in
particular sectors or locations 222.
Furthermore, it is indispensable that potential synergy effects between business development and employment policies
are considered, as they represent two sides of the same coin: On the one hand business development policies, particularly
those directed at SMEs and micro-enterprises, if realized in a way not only to increase profits but also to induce investments
and stimulate enterprise growth, may lead to an increase in employment. Hence, an integrated, systemic policy approach
is warranted that not only avoids conflicting relations but ensures consistency and complementarities between the
distinct policies, including measures directed at poverty reduction.
To tackle the problem of un- and especially underemployment as well as the lack of sufficient protection not only of informal
but also formal laborers, the following employment and labor market policies are considered to be of importance:
Invest in and Upgrade Workers´ Skills and Education (human capital development): Labor market and other
employment policies that improve the supply and demand side conditions on the first, second and third labor
market need to be considered as they may unfold positive implications for the development and growth of SMEs
and micro-enterprises. Especially occupational retraining and enhanced vocational training for laborers in the
second and third labor markets may be highly interlinked with the support of SMEs and micro-enterprises 223.
Hence, it may be advisable to improve and/or establish re-training, vocational education as well as general
secondary education via setting public and/or enabling investment in private business training centers 224.
Policies directly targeted at employment creation: As one of the biggest risk factors of the poor and near poor
are shocks to their incomes, a workfare program in the form of cash for work or a national employment
guarantee program more generally 225, might be an appropriate complementary strategy. The work could include
building or improving local infrastructure, including roads, water and sanitation, as well as health centers or school
buildings. However, the success of such a program very much hinges on a low wage level imposed. Experience
from other countries such as India show that a wage below a certain minimum wage is mandatory to ensure that
the scheme does not displace labor from the formal economic activities. In addition, government could increase
the number of temporary migrant workers through official channels, while at the same time improving (not
only but also social) protection of formal overseas workers and measures to prevent their exploitation abroad. Of
particular importance are such measures and livelihood programs for crises-affected areas, such as Aceh, North
Sumatra, and Eastern Indonesia.
Improve labor protection and labor rights: one major focus should be stopping exploitation at work, especially
the worst forms of child labor and exploitation of migrant domestic workers. In addition, improved occupational
health and safety regulations should be implemented and enforced, together with principles and rights at work,
which might be complemented by a bipartite cooperation between employers and unions in case of deciding upon
labor market flexibility and job security 226.
222
World Bank (2006d:xiii).
223
For a more detailed description how to approach employment and business development services more specifically, see Qualmann and Meyer Stamer
(2000).
224
Sugiyarto (2005).
225
BAPPENAS, ILO (2005).
226
ILO, Indonesian Government (2007).
99
In addition, the initiated strategies directed at reducing poverty in Indonesia more generally, could further
complement the before mentioned strategies.
In the aforementioned sense, the policy measure to improve the investment climate in Indonesia announced in a
presidential instruction (INPRES) No. 3/2006 by the Indonesian government and the newly initiated Decent Work
Program (2006-2010) initiated by the Indonesian government and the ILO are important first measures to realize the
aspects discussed.
7.3.3. Extending Social Security Coverage to the Informal Sector: the Direct Inclusion
Approach
As already mentioned the analysis of introducing public social protection mechanisms for the non-poor in the informal sector
is only devoted to the analysis of contributory social security programs. Further, the adjective ‘non-poor’ will not be
mentioned anymore in the following sections. However, it should be kept in mind that it is this specific part of the
informal sector the entire sub-section is devoted to.
7.3.3.1. Specific Characteristics of Social Security Programs for the Informal Economy
In case the direct inclusion approach is chosen, it should be avoided to simply copy programs aimed at the formal sector:
Targeting the informal sector entails peculiar problems and requires different approaches due to the specific
characteristics of the informal sector.
Formal social security makes use of mandatory membership as well as income-related contributions, which are directly
deducted from the pay-roll. Regarding the informal sector it is hardly possible to rely on mandatory pay-roll
contributions: Firstly, income flows of informal sector workers are often highly volatile. Secondly, the “informality” of the
informal sector implies the non-existence of any effective mechanism assessing the actual income of an informal sector
worker. In this context, the establishment of a monthly salary is very difficult and unreliable. Therefore, countries aiming at
including the informal sector in social insurance systems in general have to rely on flat-rate contributions. Flat-rate contributions
in turn inhibit the use of mandatory membership: Incomes in the informal sector are not only highly volatile but the levels of
income display a high degree of variation, too. As a flat-rate does not involve a redistribution of incomes from high to lower
levels, joining should be voluntary in nature. In addition, promoting mandatory employment would require the explicit
identification of the eligible target group, which again due to the “informality” of the respective sector is hardly possible.
Table 17: Key Differences between the Informal and Formal Sector
FORMAL SECTOR INFORMAL SECTOR
Membership Mandatory Voluntary
Contributions Income related contributions Flat rate contributions
Thus, voluntary membership and flat-rate contributions are the two key constituent features of informal sector programs. In
this regard, publicly provided social security measures for the informal sector are very similar to micro-insurance approaches.
However, a limited potential to include some segments of the target group in standard social security exist. This
issue will be taken up further in the section below.
100
7.3.3.2. Specific Risks of Social Security Programs for the Informal Economy
Demand Side
The voluntary nature of membership renders it necessary to analyze factors promoting or limiting demand, i.e. willingness-
to-pay (WTP) of informal sector workers, for social security measures. The WTP is the monetary equivalent of the specific
subjective individual preferences. In fact, a recent survey conducted among 2,500 informal sector workers in five provinces
showed both male and female informal sector workers were reluctant to join any social security scheme 227. Another survey
however pointed to the fact that quite a significant proportion of the informal economy workers in the urban region signaled
their willingness to contribute to an adequate social security system (around 41.4 percent), whereas this is only true for 16
percent in the rural regions, which account for around 60 percent if calculated together 228. A thorough assessment of the
demand for social security is therefore essential. Otherwise, the risk of reaching only low enrollment rates is high.
The following factors play a role in determining the effective demand for social security:
Risks faced by informal sector workers: The benefits offered should be in tune with the risks actually faced by
informal sector workers reflecting their preferences. The (non-representative) survey conducted by Angelini and Hirose
(2004) found that in rural areas old age security was regarded as the priority social security need (25.92 percent all male
respondents and 28.13 of all female respondents) while in urban areas health security ranked highest (40.00 percent of
all male respondents and 49 percent of all female respondents). The second priority in rural areas among male and
female respondents was health security (23.9 percent/25.95) and education among male respondents (23.9 percent).
In urban areas male referred to work injury (23.00 percent) and females to old age security (19 percent) as their second
priority 229. McCord, Ramm and McGuiness who conducted focus group interviews to assess the potential demand for
micro-insurance identified the following priority risks: (1) serious illness, (2) education of children, (3) poor harvest, (4)
death of a relative and (5) social obligations 230.
Income Level: Obviously, the individual income level influences the WTP. The higher (lower) the individual income
level, the higher (lower) the individual WTP. In addition, the Ability-To-Pay (ATP) reflects the budget constraint faced by
informal sector workers. In addition to the WTP, the ATP needs to be considered as the level of contributions is not
income-related: Informal sector workers may be willing to pay, but are not able to pay for the provided service due to
their low income. Obviously, the ATP is determined by the individual income level. The more evenly income is distributed
among informal sector workers, the smaller the variance of the aggregated WTP and ATP. The survey conducted by the
ILO indicated the ATP for social security per month in urban areas ranged from 35 percent, being adamant that they
could not contribute, which correlates with the income levels where 31 percent earned less than Rp. 200,000 per month
or half the average minimum salary and a further 27 percent earned Rp. 2,000,000-Rp. 4,000,000 per month which is
somewhere between half and the average minimum salary. Around 17.5 percent indicated that they could afford to pay
Rp. 45,000 per month, which equals the average contribution to JAMSOSTEK (for both employer and employee).
About 34 percent of respondents could contribute about Rp. 20,000 per month. The majority of respondents in the rural
areas, around 43.5 percent, stated that they are not prepared to contribute to a scheme (whether voluntary or compulsory).
A third indicated that they could contribute less than Rp 10,000 per month and around 22.3 percent could contribute
more than Rp. 10,000 per month 231.
Familiarity with the concept of insurance: It is necessary to determine how familiar the target group is with the risk-
pooling mechanism and whether they are open to this approach. As the concept of insurance is highly complex, potential
227
“Social Security Yet To Reach The Informal Sector”, in: The Jakarta Post, December 4, 2006.
228
229
The preference with regard to old age instead of health in rural areas is surprising. However this result should be treated with caution. Firstly, Angelini and
Hirose themselves mention that specifically the sample of interviewed persons in rural areas contained a relatively high percentage of older persons leading
to bias. Secondly, as the survey was aimed at interviewing informal sector workers in general and not particularly targeted at the non-poor it is possible that
a high share of interviewed persons were already covered by the health program preceding the ASKESKIN scheme and therefore considered an additional
coverage of health as unimportant.
230
McCord, Ramm and McGuiness (2005:4).
231
232
McCord, Ramm and McGuiness (2005:55-56).
101
clients might not understand the benefits offered by insurance if there were no prior experiences with the so-called
“insurance culture”. Specifically, the fact that contributions must be paid on a regular basis (i.e. non- state-contingent),
whereas the realization of a (monetary) benefit in a given period is not certain (i.e. state-contingent) might be considered
as unfair. A lack of understanding would inhibit demand. The better the educational background of people, the more
likely they are to be familiar with the insurance concept. The educational levels within the informal sector differ widely,
specifically and not surprisingly between rural and urban areas. McCord, Ramm and McGuiness (2005) provide anecdotal
evidence on problems related to a limited understanding of the insurance concept as such and the social security
program offered by JAMSOSTEK to formal workers 232. A general lack of awareness of social security is also reported
by the findings of Angelini and Hirose.
Cultural or religious norms: Cultural or religious norms may also hinder the development of insurance. Insurance
deals with potential losses, which may occur in the future. But in some societies talking about future negative events or
taking precautionary measures to prevent them, is considered as a taboo, because doing this would lead to the actual
occurrence of these events. Again, McCord, Ramm and McGuiness (2005) report some evidence on these kinds of
hindering social norms: Ganesha Microfinance had offered micro-health insurance for a brief time. A majority of the
target group thought that by offering such a product, Ganesha was encouraging them to become sick and die leading to
the perception that the MFI was wishing them bad luck (McCord, Ramm and McGuiness 2005: 57). However, above all,
the compliance with Sharia Law seems to be essential in the Indonesian context. The existing SSABs up to now
operate as for-profit organizations. Although the law 4/2004 explicitly states the mandate to transform the SSABs into
non-profit organizations, prevailing misperceptions about the instrument offered could prohibit enrollment.
Information: Lack of information is another factor leading to a low realized demand: Demand may be inhibited simply
by the fact that potential clients do not know about the availability of an insurance service.
Trust in government institutions: If people do not have confidence in the national social security system, they might
be hesitant to join any program. Angelini and Hirose (2004) found evidence on a common distrust in government
institutions.
The quoted examples show that all potential problems inhibiting demand are present within the Indonesian context. However,
as only specific cases are known, the actual relevance of each category for program design is still an issue for further
investigation.
Supply Side
Also at the supply side social security programs for the informal sector face certain risks, which are not relevant for formal
social security schemes.
Low contribution recovery rate: As informal sector programs cannot rely on automatic pay-roll deduction, contributions
need to be regularly collected individually. If no specific information and enforcement system is in place, individual
members face an incentive to withhold their payment as they fear no consequences. Further, informal sector workers
often face highly irregular income flows. Due to this aspect they might not be able to pay regular monthly contributions.
Both factors contribute to the risk of a low contribution recovery rate, i.e. the relative share of contributions received in
comparison to total expected contributions.
Adverse selection: The problem of adverse selection refers to the risk that only persons with a high risk profile join
whereas low risk persons do not join. The presence of adverse selection leads to a higher claim ratio thus resulting in
higher costs. The presence of adverse selection, as well as difficulties in dealing with adverse selection, depends on the
specific risk in question.
Free-rider behavior: Voluntary membership allows people to join the scheme when they are in need and drop out
afterwards. This behavior decreases the efficiency of risk pooling and threatens financial stability. The presence of free-
rider behavior, as well as difficulties in dealing with free-rider behavior, depends on the specific risk in question.
232
102
Limited size of the pool: Insurance works by sharing risk across a large population. The smaller the risk pool the less
efficient any risk pooling is and the higher the volatility of claims and related costs.
Specifically with regard to adverse selection and free-rider behavior the extent of those insurance-related risks depends on
the type of risk covered (e.g. both are more prominent or difficult to cope with in regard to health risks than old age).
The following table summarizes the specific risks faced by programs targeted at the non-poor in the informal sector.
Table 18: Specific Risks of a Social Security Program for the Non-Poor in the Informal Sector
Risks
Demand Side Low enrollment due to

* Limited demand for social security (WTP)
* Low ATP
Supply Side Financial instability due to

* Low contribution recovery rate
* Presence of adverse selection
* Presence of free-rider-behavior
* Limited size of the risk pool
7.3.3.3. Program Development

The design of a contributory social security program for informal sector workers is a combination of social insurance with
some technical elements of private commercial insurances. It also resembles the design of micro-insurances in
several ways, as they are a combination of both types of insurances too.
Considering the heterogeneity of the informal sector, also in terms of regional disparities, it should be stressed from the
beginning that a “one-size-fits-all” program most likely will not provide the most suitable solution. Program components
should rather be adapted to regional characteristics. This, however, should happen under the umbrella of a concerted
centrally coordinated program.
Moreover, it has been shown above that informal sector programs exhibit certain risks that endanger stability. It should be
kept in mind that the implementation of effective mechanisms is achievable but it is not easy.
Potential for Voluntary Inclusion in a Standard Social Security Scheme

Standard social security might be open up to professional independent workers, i.e. persons with a “quasi-formal” employment
status (e.g. lawyers, doctors). However, the availability of income sheets or other information sources should be a pre-
condition in order not to create incentives leading to a further solidification of the informal sector. In any case, this is more an
option for the higher income segment of the informal sector, as there is no contribution on behalf of the employer and thus,
the contribution to be paid will be relatively high.
Preparatory Studies
Naturally, the efficiency of any program design is dependent on the amount of available information. In this context, the
following preparatory studies are recommended:
1. Identification of target group in terms of size, income levels and regional distribution: This study would
allow for a more comprehensive assessment of the relevant target group, i.e. the non-poor in the informal sector.
2. Analysis of WTP of the target group: This study would estimate the WTP of the target group, test for the
relevance of the above-mentioned factors influencing the WTP and probably identify further relevant factors. The
103
assessment of the average WTP should be combined with the analysis of its distribution, as in general the informal
sector is very heterogeneous in many dimensions, therefore the individual WTPs might naturally differ within the
target group. As people typically face the incentive to quote lower values of their WTP than their actual values
when asked directly for it, assessing the WTP requires the application of specific adequate methods.
3. Analysis of the ATP of the target group: The ATP is based on individual with respect to household budget
constraints. For an assessment of the ATP, information on the individual and - household incomes and household
based expenditure patterns respectively - needs to be obtained. The former aspect can be assessed with the help
of information derived from studies No. 1 and No. 2. Similar to the analysis of the WTP the average ATP as well as
its distribution should be assessed. The entire assessment of the ATP can also be incorporated into the study
design of study No. 2.
4. Identification of target group in terms of risk characteristics/profile: Calculating the expected total costs of
the program, the required contribution per member as well as information on the expected costs of claims per
member is needed. To assess these costs specific information on the target group is needed (e.g. utilization rates
of health services, cases and average costs of cases for health insurance or e.g. age structure, life expectancy
and mortality rates in the case of pensions). If these data are not available it is very difficult to derive reliable
estimates of expected costs.
5. Identification of existing local-level contributory programs: Not enough is known on initiatives being
implemented at the local level and pursuing the same aim. A study should be compiled listing those programs and
exploiting ways how to link up to them.
Studies No. 1-4 should be random in nature to enable a representative sample. Although more difficult and probably
more expensive to accomplish, analytical results from random studies are more reliable as they are less biased
with respect to statistical methods that deal with existing bias. They provide a more reliable basis for any planning
reducing the risk of reform failures, thus saving costs during the implementation phase.
Designing and Financing a Benefit Package for Informal Sector Workers

The design and the financing of a benefit package are ultimately interrelated: The more (less) comprehensive the benefit
package, the higher (lower) the costs and the resulting financing requirements. The lower (higher) the financing capacities,
the more restricted (less restricted) the benefit package. Thus, both aspects will be dealt with in one section.
Basically, the three major steps in designing a benefit package are as follows: (1) Pre-defining and costing a benefit
package, (2) Calculating financial capacity, (3) Adjusting benefit package to financial capacity.
The following factors influence the design of the benefit package:
Expressed preferences of the target group: As outlined above the benefit package should reflect the priority
risk(s) identified by informal sector workers.
Existing benefit package for the formal sector: As already stressed above, the formalization of the informal
sector, i.e. the integration into the formal economy, is the ultimate objective. In this context, it seems recommendable
to offer a benefit package that is less comprehensive in comparison to that offered to the formal economy. Otherwise,
any social security program targeted at the informal sector will suffer from the unintended side effect that it decreases
individual incentives for moving from the informal to the formal sector of the economy.
Financing restrictions: Basically, a scheme for the informal sector can rely on two financing sources: Individual
contributions and public co-subsidies. Obviously, the design of a social security program for the informal economy
faces a dilemma: Although a contributory capacity on behalf of the target group exists, the average ATP might be
relatively low. At the same time when considering the size of the informal sector the public capacity for co-subsidies
104
is rather low, too. Thus, a benefit package for informal sector workers, which is based on a sustainable financing
strategy, will necessarily be limited 233. In this context, finding the appropriate balance between self-financing and
public co-subsidies is ultimately a political decision.
In case the top priority social security needs differ among rural and urban regions, a differentiated benefit package according
to the specific needs of urban and rural areas could be offered.
Further, if the WTP and ATP vary strongly among the non-poor in the informal sector it is further possible to offer a
flexible benefit package consisting of different individually combinable modules (either in regard to different risks
or to differentiations within each risk category) which allows individuals to choose.
Financial Management
Given limited financing capacities a sound financial management of the entire program increases the potential
outreach by reducing and stabilizing costs.
Informal sector programs face the risk of a low contribution recovery rate, which may stem from (a) irregular incomes of
informal sector workers and (b) enforcement problems of contribution collection. The first aspect can be addressed by
offering flexible payment procedures. Members can be given the choice to pay their contributions either on a monthly,
quarterly or annual basis. The timing of contribution payment further can be set in accordance with the timing of their income
flows, e.g. workers from the agricultural sectors receive a higher income during harvest seasons. Regarding the latter
aspect, the introduction of an exclusion mechanism can strengthen enforcement, e.g. a temporary exclusion of members
after a certain threshold of outstanding contributions for a certain period or until outstanding contributions are paid.
In order to reduce the risk of adverse selection, membership should be based on the household and not on the
individual, i.e. making enrollment of dependants compulsory. Further, introducing group enrollment reduces adverse
selection even more. The free-rider problem can be prevented by introducing a minimum waiting period before any
claims can be made.
As was pointed out, the relatively small size of the risk pool may lead to a high volatility of claims. Further, it is difficult to
derive reliable estimates of expected costs ex ante. Thus, an adequate reserve rate is important. Given a fixed financing
capacity however social security programs face a further trade-off: Setting the reserve rate at a realistic level imposes a
further limiting constraint on the comprehensiveness of the benefit package (with given financial capacities) but promotes
financial stability. Setting the reserve rate at a lower level loosens this constraint but endangers financial stability.
It goes without saying that an effective financial management depends on functioning information systems.
Marketing and Information

As pointed out above factors like a lack of information on the availability of social security, a lack of understanding of
insurance principles, potential conflicting cultural or religious norms or a general distrust in government institutions might
prevent people from joining a social security program. A sensitive, extensive and continuing information and marketing
component, which addresses all of these issues is therefore essential for the success of any program targeted at
the non-poor in the informal sector. The WTP-study will offer information on the relevance of these three aspects in the
specific Indonesian context. This information should be adjusted with regard to specific regional/local backgrounds.
The information and marketing campaign should already begin a certain time before the actual operation of the
social security program starts. However, the time in between should not be too long as any positive effects of the market
campaign might dwindle again as people simply tend to forget.
233
Further, in contrast to social security programs for the formal sector, informal sector programs do not involve any interpersonal redistributions of income.
If the income distribution within the informal sector is very uneven there is another trade-off between self-financing and exclusion of persons in the informal
sector.
105
In addition, the marketing strategy should fit the profile of the target group 234. This includes taking into consideration
the diversity of the informal sector with regard to the educational background. Persons with a lower educational background
need different strategies than persons with a higher educational background.
Institutional and Organizational Set-up

Before any further actions regarding the inclusion of the informal sector can be pursued it is necessary to put this
topic on the political agenda as several important policy decisions still have to be made, e.g. allocation of responsibilities
at the ministerial level and with regard to the implementing authority, i.e. SSAB, the type of program to be adopted.
As the foremost objective regarding the informal sector is its formalization, i.e. integration into the formal economy, once this
has been achieved, a specific program for the informal sector is dispensable and participants will be integrated into the social
security for the formal economy or non-contribution-based social assistance. Thus, all social security programs for the
informal sector should be considered as temporary programs and will cease in the mid- to long-term.
This aspect is of crucial importance for the institutional and organizational development even if the transition stage is rather
long. Removing institutions and organizations once they are established is a difficult process involving the balancing of
multiple interests of stakeholders involved. Therefore, the transitional character should be considered when discussing
the institutional and organizational set-up.
Regarding the legal and regulatory framework Law No. 40/2004 provides the legal foundations for any coordinated integration
of the informal sector in national social security programs. However, it is still not precise on the exact definition of the target
group as well as on concrete mechanisms how to achieve the inclusion of the informal sector and should be improved with
regard to those aspects once the decisions at the policy level have been made. Separate regulations for informal sector
program need to be developed, too. Laws and regulations should explicitly state the temporary character and point at
the intended objective of formalization.
With respect to the decision-making authority the responsibilities at the ministerial level are still a matter for discussion.
The Ministry of Labor seems to be one of the potential candidates. On behalf of the implementing authority the question of
how to organize the administration arises similarly to social security for the formal sector (see chapter 4.2). As any program
targeted at the informal sector involves high transaction costs, only one SSAB should be responsible for implementing all
measures targeted at the informal sector in order to avoid any doubling of efforts. Further, as the mid- to long-term objective
is the formalization of the informal sector, any social security program specifically targeted at the informal sector is a temporary
measure. The implementing authority will not function any longer once the informal sector is to a large extent integrated into
the formal economy. Therefore, also the activities of the SSAB are of a temporary nature.
In the light of these aspects JAMSOSTEK seems to be a suitable SSAB: With increasing formalization workers insured
under the informal sector program would simply move to the formal sector program. However, the allocation of responsibility
to a SSAB also depends on the type of strategy actually chosen for overall social security administration, i.e. Harmonization,
Separation or Integration (see chapter 4.2). As for other social security programs alike, the supervising authority should be
allocated to the Directorate of Insurance within the Ministry of Finance.
One last important question refers to the allocation of responsibilities between the local and the central level. With
regard to the regional disparities of the informal sector, a “one-size-fits-all” program will most probably not provide the most
suitable solution. Rather program components should be adapted to the prevailing specific regional characteristics under the
umbrella of a coordinated and concerted program anchored at the central level.
The analysis of the IS in Indonesia has shown that the hazards faced, but also the ability to contribute to social security is
quite different for rural and urban regions but differs also with respect to the employment status and the sectors. Furthermore,
the constraints faced by the IS vary across but also within provincial lines. Taking these differences along several lines into
234
McCord, Ramm and McGuiness point out some examples of inappropriate marketing strategies for the low-income market, e.g. posters that are filled
with complex text or television advertising when most of the target group do not have a television (McCord, Ramm and McGuiness 2005:70).
106
account, it seems to be necessary that each program should have a regional component, e.g. should take account of
regional differences. Among others, this would imply to take advantage of Indonesia´s decentralized administration,
which allows for different strategies to be adopted in different regions. This, in turn, makes it indispensable to
enhance and streamline decentralization efforts even further than already initiated. However, learning from past
experiences, a coordination mechanism should be in place to avoid a overlapping responsibilities and programs.
The role of centrally and locally financed co-subsidies also needs to be discussed. Any (partial) reliance on local finances
should be accompanied by appropriate fiscal equalization measures as the poorest regions, which the lowest revenue
base, will most likely have the highest demand for social security and thus will be confronted with the highest expenditures,
too.
Administrative Capacity
Administrative capacities at central and local levels play a crucial role in extending coverage to the informal sector specifically
as targeting the informal sector poses specific problems with regard to registration, compliance, collection of contributions,
and record keeping.
Further, to limit the supplementary operational costs, which add up to the individual level of contributions, efficient administrative
procedures, rules and supervision mechanisms are essential.
7.3.3.4. Contracting with the Private Sector to Target the Informal Economy
Contracting key agents or groups from the private sector and assigning them an intermediary role helps to mitigate several
of the specific risks faced by social security programs for the informal economy. These intermediary groups encompass a
wide array of affiliations, for example, cooperatives, umbrella organizations within the informal sector, religious groups or
non-commercial microfinance organizations. These different kinds of groups can be captured by the term community-based
organization (CBO): A community-based organization (CBO) is a non-commercial organization with the primary objective
being to provide a collective good for the members of the organization. A CBO is based on a pre-existing community, which
in turn can be based upon geographic linkages, professional affiliations, religion or some other kind of joint activity. Often,
social relations within a CBO are – at least to certain extent – personalized 235.
235
Working with CBOs could help to increase the effectiveness of informal sector approaches in several ways. Advantages of CBOs include the following:
(a) Existing group affiliations: CBOs provide already pre-existing group affiliations, (b) Availability of information: CBOs possess more information on local
conditions than outsiders do, (c) Trust: As members of CBOs often share a similar social-economic background in one way or another (e.g. geographic
linkages, type of employment, religion or some other kind of joint activity) and the relations of group members are (to a certain degree) personalized, the
level of trust among them is higher than the level of trust towards “outsiders”, e.g. public agents. Specifically, if a certain distrust in public activities is
existent, CBOs could take the place of a substituting trust building device, and (d) Transparency: If the group in question is not too large in size and is further
characterized by personalized and permanent relations, members of CBOs might easily monitor each others actions, leading to a high degree of transparency
within those groups. However, the extent of those potential advantages should be taken as naturally given, but rather need to be assessed ex ante and
monitored ex post.
107
However, even when working with CBOs, membership on an individual or preferably household basis should still be possible
as otherwise non-organized persons within the informal sector would still be excluded from any arrangement.
More specifically, CBOs could assume the following functions:
Promoting higher coverage via group enrollment: Working with CBOs allows targeting entire groups instead of
individuals or households. Introducing group enrollment reduces the risk of adverse selection and improves risk pooling
via an increased risk pool thereby reducing the size and the volatility of claims. In addition, it contributes to lower costs
per member due to fixed cost degression. Depending on the financial situation of the scheme, a part of these risk
smoothing and cost reducing effects could be transferred to members and organized groups (OGs) by offering premium
discounts, which in turn might induce an increase in demand.
Information and Marketing: Key agents of CBOs could act as facilitators in promoting enrollment by providing information
on the scheme to their group members. Due to their informational advantages those agents are more likely to effectively
address informal sector workers as they know how to talk to them and which objections might exist. In addition, due to
the higher level of trust within CBOs, informal sector workers are more likely to believe any information provided by
those key agents rather than information provided by outsiders.
Collecting contributions: As pointed out previously, the collection of contributions is a further challenge. Within this
context, contracting key agents of CBOs as premium collection agents would allow for a reduction in transaction costs.
They possess more information on the actual income situation and thus on the ATP than outsiders. In addition, due to
a higher level of trust within CBOs people might be more willing to submit payments to their leaders than to outsiders.
In addition, CBOs often require a certain regular payment from their members. Therefore, channels for receiving payments
are already in existence.
By carrying out these functions, contracting with CBOs could result in a higher effectiveness of social security programs
for the informal sector in the following ways:
Increased enrollment rates, thus higher effectiveness in terms of coverage,
Higher premium recovery rates, thus higher effectiveness in terms of higher revenues and increased financial stability,
and
Reduced size and volatility of claims, thus higher effectiveness in terms of lower and more predictable costs and
increased financial stability.
However, working with CBOs is not an approach that is suitable under all conditions. Rather, its applicability depends
on the fulfillment of certain conditions, i.e. certain prerequisites have to be met:
A basic pre-condition relates to the organizational degree of the informal sector: A high share of informal sector
workers needs to be organized via CBOs. Further, eligible CBOs must not be too small in size.
If CBOs assume a more active role key agents need to receive appropriate trainings.
As the workload associated with a public-private cooperative venture is not insignificant, appropriate incentives
should be offered to key agents in order to induce them to actually fulfill their tasks. In terms of monetary
incentives, performance-based payments (e.g. payments depending on either the number of new members attracted
or total contributions collected) are a viable option. However, when working with religious leaders most probably
other instruments need to be employed (e.g. donations). If the activity results in an increased reputation of the
respective key agent this might serve as an indirect supplementary incentive.
In addition, this approach presupposes the existence of an adequate administrative capacity of contracted
groups, which includes the fulfillment of organizational, administrative standards and financial standards (e.g.
stability, bookkeeping) that need to be ensured.
108
Table 19: Contracting with CBOs
Role of CBOs Impact Prerequisites
Group Enrollment * Increase financial stability * CBOs exceed critical size

* Reducing and stabilizing total costs in terms of membership
(reducing adverse selection, improving * CBOs might need incentives for group
efficiency of risk pooling) enrollment (e.g. premium discount)
* Reducing costs per person
(fixed cost degression)
Information * Increasing membership (better information * CBOs exceed a critical size in terms of
and Marketing on local conditions and mutual trust) membership
* CBOs possess adequate administrative
structures and capacities
* Key agents receive training with regard to
the product they are supposed to promote
* Key agents are offered appropriate
incentives
Collection of * Increase in financial stability * CBOs exceed a critical size in terms of

Contributions * Increasing premium recovery rate (better membership
information on local conditions, mutual * CBOs possess adequate administrative
trust and higher transparency) structures and capacities
* Key agents receive training with regard to
the product they are supposed to promote
* Key agents are offered appropriate
incentives
* Monitoring mechanisms exist (also
internal monitoring (book keeping etc.)
The relevance of each condition depends on the specific functions carried out by CBOs. Simply using CBOs for targeting
groups instead of individuals is the least demanding function whereby CBOs as such assume a passive role only. Contracting
CBOs with respect to key agents as information and marketing or contribution collection agents transfers more active roles
to them, which requires stronger conditions to be met.
The table above summarizes the potential roles of CBOs, the associated impact on performance and the necessary
prerequisites to be met.
With respect to the implementation, contracting with the private sector should not be considered a uniform approach across
Indonesia. Rather, again it is necessary to adapt this general approach to prevailing regional and local conditions
with regard to the above-mentioned prerequisites. For each region it should be considered if it is reasonable to involve
CBOs at all, and, if yes, which functions they should carry out. Specifically, when considering the limited global experience
up to now, initiating differing pilot programs provides a basis for fruitful insights. A necessary first step is a compilation
and assessment of existing CBOs to be able to decide on the way ahead.
As a general characteristic of the informal sector in Indonesia, membership in cooperatives and more formal
associations is rather low in both urban and rural areas. In urban areas, a survey conducted by ILO (2001 and 2003)
showed that membership in cooperatives or other associations makes up only around 6 percent, whereas around 12.6
percent of respondents in rural areas are members of rural cooperatives (KUD) and 84.7 percent of rural individuals surveyed
were not members of any association or group. A survey conducted by UNICEF confirms these tendencies, showing that
109
none of the surveyed home-based female workers is a member of a workers’ organization, or has even heard of the existence
of such an organization.
However, alternative organizations exist, for example:
Existing Networks of Microfinance Institutions: Indonesia possesses a well-established network of microfinance

providers. Contracting with relevant microfinance umbrella organizations would not only ensure access to large groups
of potential clients but those types of organizations might also be most likely to provide the professional and organizational
capacity to support the functioning of an informal sector social security scheme.
Religious leaders and organizations: Considering the influence of Muslim religious leaders within the Indonesian
context, ensuring the support of religious organizations, kyais (Islamic scholars) or ulamas (Islamic scholar-religious
leaders) for promoting a social security program would be a very important asset. Ensuring the agreement of the takaful
insurance board is another important aspect. Religious leaders and groups (depending on their capacities) could contribute
to all three functions: They could promote group enrollment as they reach large groups. They are extremely influential
and thus very much qualified for informational and marketing activities. As they are highly respected, they could also act
as contribution collection agents.
Arisans: Indonesian arisans are non-registered community-based organizations that provide a limited range of social
protection services to their members, for example coverage of health expenses, festivity expenses, educational expenses
and business set-up costs. Each member is required to pay monthly contributions, which are based on the financial
capacity of each member. Basically, three types of arisans exist: (1) Family or clan arisans, with an approximate
membership of 100 to 600 persons, (2) Ketua RT – Batak Kampung arisans, which is comprised of all Batak men and
their families living in the same neighborhood (or other ethnic groups), and (3) national level women’s (catholic) church
arisans, which amounts to approximately 225,000 members. Arisans are widespread in many Indonesian regions, but
originally stem from communities whose origins are rooted in the Sundanese/Javanese population or from the Batak
236
community .
7.3.3.5. Country Examples

Considering the income status of Indonesia, the size of the informal sector is surprisingly large. Thus, relevant country
examples should come from countries where the informal sector is of similar size. However, as already stressed, there is no
solid knowledge available on “good practices”. As the Philippines and Vietnam are quite advanced in implementing
programs directed at the informal sector, the following describes some of the central design features of the respective
programs and deduces lessons learnt so far 237.
The Philippines
Overview
The Individual Paying Program (IPP) in the Philippines was established in 1999 and is a component of the National
Social Health Insurance Program (NSHIP). It is targeted at the informal sector and free-lance professionals (lawyers, doctors
etc.). The formal sector and poor households are covered by separate components of the NSHIP. The IPP offers the same
benefit package as other components of the NSHIP (e.g. formal sector), but relies on flat-rate contributions, which are
payable quarterly, semi-annually or annually in order to take into account the irregular incomes of informal sector workers.
The contributions are combined in one pool comprising all participants of the National Social Health Insurance Program.
236
237
Schmidt et al. (2005) and the websites of the International Conference on Social Health Insurance in Developing Countries in Berlin 2005 (www.shi-
conference.de) and of the Social Health Insurance Conference in Manila 2006 (www.shiconferencemanila.info) provide more extensive information.
110
In 2003, problems, such as low coverage, low premium recovery rates and signs of adverse selection led to a
redesign of the program: A pilot program, called POGI, was started which worked with organized groups (OG). These OGs
work as marketing and premium collection agents while certain incentives are offered to the OGs in exchange for their
cooperation (e.g. cooperatives receive a certain amount for each new member). This measure should aid in increasing and
stabilizing membership and improving premium recovery rates. However, no measures were taken to reduce adverse selection
or the presence of free-riding behavior (e.g. there was no penalty for people joining the scheme whenever in need and opting
out afterwards). In 2005 the situation was still problematic: Membership still was low with a coverage rate of 18.8 percent
of the total target group only. In addition, outstanding premiums (only about 75 percent of registered members were actively
paying), a still continuing increase in utilization rates and average claim values as well as a high percentage of catastrophic
claims (26 percent in comparison to 14 percent for other programs) contributed to an estimated overall loss of 121 percent
in 2005.
Thus, further changes were introduced and a re-designed program launched. A major new aspect is that the program
(now called KaSAPI) targets larger organized groups and makes group enrollment compulsory (at least 70 percent must be
enrolled into the program), which should contribute to cost stabilization and cost reduction by reducing adverse selection as
well as the fixed costs per member (e.g. administrative costs).
This cost saving effect enables PhilHealth to offer two different types of discounts. The first discount is a premium discount,
which is shared between the OGs and the individual members. The size of the discount depends on the size of the respective
OG, i.e. number of eligible members for IPP, and the percentage of enrollment and varies between 3.5 percent and 7
percent. A second discount is related to the frequency of payment and rewards advance payment: The highest discount is
offered if the total premium is paid annually (3 percent), the lowest if it is paid quarterly (1.5 percent). Again, this discount is
shared between OGs and individual members 237. The maximum combined total discount is 10 percent, the minimum total
discount is 5.5 percent.
The incentive structure offered by the present set-up of the IPP is envisaged to create a “triple-win” situation: PhilHealth
should benefit from an increased financial stability through an increased membership rate, reduced adverse selection and
an improved premium recovery rate. The OGs receive an additional income, which could help to cover the additional
administrative costs they incur. For the individual members of the OGs the discounts might offer an additional incentive to
enroll.
Further changes include measures to prevent free-rider behavior by introducing a waiting time: A minimum number of
payments now are required before any availments are made. Further, a premium differential between professionals and
informal sector workers has been introduced.
Lessons Learnt
The example from the Philippines shows first of all that targeting the informal sector is an incremental learning process that
involves a step-wise adoption of program design from experiences derived to implementation. Secondly, it shows the relevance
of the problems mentioned in the preceding sections when analyzing the specific risks of social security programs for the
informal sector and thus the necessity to include appropriate measures, e.g. against free-riding behavior and adverse selection
right from the beginning. PhilHealth included such types of measures only since the 3rd phase of the program. Thirdly, the
PhilHealth experience illustrates that working with private sector organizations is not an ultimate solution and that the
characteristics of the respective organizations matter. Therefore, PhilHealth changed the requirements for eligible OGs in
the 3rd phase now targeting larger groups only. Despite the issue of size, the lack of individual incentives for OG group
leaders to actually fulfill their obligations could also be responsible for the rather poor performance in terms of membership
and premium recovery rates. Fourthly, as described above, PhilHealth offers the same benefit package in its IPP program as
237
An interesting feature of this discount is that is up to the OGs how to organize the advance payment internally. It therefore allows for a situation, in which
the individual members of the OGs, who in general prefer to pay small amounts regularly, still can continue that way and the OGs pay to PhilHealth an
annual advance premium and collect the money from their members in installments.
111
in its other components. It is still an open question whether this solution – although for sure desirable from an equity perspective
– is a financially feasible solution. Within the 3rd phase PhilHealth tried to tackle this problem by introducing premium differentials
between professionals and informal sector workers, thus introducing some kind of income-related redistribution. However, it
is still questionable if this is sufficient to finance a full benefit package.
Vietnam
Overview
Vietnam is implementing a program to include the informal sector in the national social health insurance scheme238.
Three time phases can be distinguished. From 1992-2002 small scale pilots of voluntary rural health insurance were
implemented in order to raise awareness and introduce the concepts of social health insurance. There was no regulatory
framework. Coverage fluctuated around only 200,000 members during this period. In 2003, Circularr 77 was introduced
which harmonizes the benefit package with the compulsory scheme and introduced further regulations. The Vietnam Social
Security (VSS) agency is responsible for the fund administration. Membership is voluntary and on an individual basis. To
prevent free-rider behavior a waiting period applies. The benefit package included inpatient and outpatient care. Members
were charged a flat-rate but co-payments of 20 percent applied. Furthermore, exclusion and benefits caps applied, as well,
an enrollment threshold minimum of at least 40 percent of the total uninsured from communities/mass associations was
introduced to counter adverse selection.
In 2005 several changes were introduced. No co-payments are required anymore and the enrollment thresholds are
reduced to 10 percent of total uninsured households in each community, and 30 percent of the membership in mass
organizations. For the revenue collection specifically trained “health insurance agents” are employed who receive a 6-7
percent of total revenues collected as a performance incentive. Up to 2006, 8,000 health insurance agents have been
trained. Local authorities also receive 1-2 percent of all revenues as an incentive for cooperation.
Membership shows a very impressive upwards development (although the total coverage rate is still low). Covering
27,944 informal sector workers in 2003 membership increased to 1,394,485 in 2006. However, at the same time the
financial situation of the fund is rather instable and worsening: In 2004 actual expenditures were 126 percent of the
available treatment fund, which increased tremendously to 727 percent in 2006 239. This situation is to a certain extent
induced by adverse selection and a low premium recovery rate leading to funding deficits. To deal with these problems the
following amendments are planned but not yet implemented: (a) Moving from individual coverage to family coverage to
reduce adverse selection, and (b) Increase contribution levels but at the same time encourage more people (near poor) to
join with a government subsidy for at least 30 percent of premium.
Lessons Learnt
Again, the Vietnamese example shows an incremental learning process. Secondly, the program can be considered a success
in terms of the dynamic upward trend in membership, although the coverage rate is still low. This might be due among others
to the fact that Vietnam explicitly included information and awareness-raising measures in its program design. Thirdly,
utilization rates increased indicating another success in terms of access to health services, which is the ultimate objective of
extending access. However, this is only one side of the story: The increase in utilization rates is most likely not only due to
unmet demand, but also to inefficient moral hazard behavior on behalf of providers. Fourthly, in terms of financial performance
however, the results are disappointing. The planned amendments might mitigate this problem. Nevertheless, the question
remains again, if a comprehensive benefit package is financially feasible.
238
Besides the informal sector, i.e. farmers, self-employed, the scheme is also open to students, school children, and dependents of the compulsory health
insurance scheme. However, the analysis deals only with the groups of the farmer and self-employed. This is possible as separate data exist for the
mentioned target groups.
239
Source: Social Security Vietnam (only farmers and self-employed).
112
7.4. Summary
I. The high developmental potential of the IS in Indonesia necessitates extending governmental efforts to target
policies explicitly at IS and to include all IS segments.
From the mere size it can be clearly seen that the IS is significantly contributing to employment in Indonesia, thereby
mitigating structural inequalities by absorbing workers in the formal sector who would otherwise form part of the unemployed.
However, the contribution of IS to GDP is still rather low. According to Schneider (2002) the informal economy in percent of
GDP in 1999/2000 in Indonesia was 19.4 percent which amounted to USD 276,8 billion in 2000, but remittances by informal
overseas workers are an important source of foreign exchange and a growing income source for poor rural households, as
are remittances from the urban to rural households. In addition, the IS in Indonesia is also a major provider of basic services
and needs, such as food, clothing, and shelter, contributing significantly to the livelihood and survival of the urban but also
rural population.
This indicates that the IS in Indonesia has quite a substantial potential to contribute to poverty reduction, but also economic
growth and development more generally, and highlights that the efforts of the Government of Indonesia to direct programs at
the informal economy, especially the informal poor, are not only leading in the right direction but also need to be dramatically
intensified and extended so as to cover all IS segments.
II. Regarding social protection the integration of the informal sector into the formal economy and thus, a stepwise
integration of former informal sector workers into “standard” social insurance should be the ultimate objective
(formalization approach). However, prohibitively costly constraints are major obstacles for unfolding the
developmental potential of the IS. Thus it is necessary to target the underlying causes of informality and to take a
systemic/integrated approach to policy design.
The reason for the still rather slow process of unfolding the above-mentioned developmental potential could mainly be
traced back to several prohibitively costly constraints. The analysis of constraints faced by both informal sector enterprises
and labor suggests that informality in Indonesia is less due to choice but more to chance.
This shows that there is not only a high need for governmental policies targeted directly at the IS, but that there might also
exist a quite significant demand for such policies.
Policies are needed that tackle (reduce, if possible eliminate) the more profound and underlying causes of IS constraints, to
enable the IS to unfold its developmental potential. This, in turn, warrants attention to include differences in constraints
between IS workers and enterprises but also gender issues in policy design.
Furthermore, it highlights that efforts merely directed at enforcing compliance with existing state regulations might not lead
to the expected and needed results, quite the contrary. Mainly focusing on enforcing governmental regulations of such highly
constrained enterprises without considering their specific needs and possibilities would imply that most of these enterprises
would have to close down.
Furthermore, besides reducing regulatory burdens and labor market rigidities, strategic issues and synergy effects between
different kinds of employment, labor market, business development as well as poverty reduction and growth policies need to
be explicitly considered.
III. Whereas the poor in the informal sector can be reached by social assistance schemes, the non-poor are still
excluded. Enabling an inclusion of the non-poor in the informal sector into social protection strategies in the short-
to mid-term calls for a separate social insurance scheme for the informal sector as standard social security is not
appropriate (direct inclusion).
Although the formalization of the informal economy should be the ultimate objective, during the time before achieving this
objective, the informal sector would have to primarily rely on private social protection initiatives, which are rather limited in
outreach. Thus a combination of the formalization and the direct inclusion approach is recommended.
113
Providing social insurance for the non-poor in the informal sector is different to standard social security. Basic differences
relate to the type of contribution (flat-rate) and type of membership (voluntary). Due to these differences informal sector
schemes exhibit a variety of risks that endanger their effectiveness in terms of coverage and financial sustainability. Reliable
data for planning, a careful design and skilled administrative personnel for its implementation are therefore essential.
If the Indonesian government opts for universal tax-financed benefits with regard to certain risks, no specific scheme is
necessary for those risks.
IV. The high heterogeneity of IS in Indonesia, especially in terms of provincial and geographic differences, not only
necessitates appropriately identifying and categorizing IS segments with respect to SP related idiosyncrasies but
also necessitates a regionally aligned approach.
The informal sector in Indonesia displays a high degree of heterogeneity in terms of socio-economic characteristics with
widely varying regional disparities.
A “one-size-fits-all” program most likely will therefore not provide the most suitable solution. Rather, program components
should be adapted to regional characteristics. Making use of Indonesia’s decentralized administrative structures is thus
advisable. This, however, should happen under the umbrella of a concerted centrally steered and coordinated program.
114
8. The Role and Financing
of Puskesmas in the
Indonesian Health
System
- The Government of Indonesia has taken a major step forward towards universal coverage by reforming the social
security law and implementing a social health insurance scheme with the objective of universal coverage.
- Primary health care represents the basis of the health care pyramid in Indonesia and is formed by over 7000 community
health centers, Puskesmas.
- Puskesmas play a very important role due to their proximity to the population, thus increasing the possibility of
delivering healthcare across the archipelago.
- Contractual arrangements, roles and responsibilities between the insurance provider and health service providers
are still in the process of being finalized.
- Since 2005, two modes of financing Puskesmas have been applied: insurance-based funding via an insurance
carrier (PT Askes) and direct funding of Puskesmas using block grants from the central and local government –
both offering different incentive structures and involving different requirements
115
8.1. Puskesmas: Service Delivery and Financing

Recent surveys have indicated that service delivery at a Puskesmas is perceived to have improved over the past two years
and utilization has increased.
Access
Four out five households reported have access to a Public Health Center (Puskesmas), and 59 percent have access to a
general hospital. The average distance to a Puskesmas is 2.9 kilometers (1.5 kilometers in urban vs. 4.2 km in rural areas),
while the travel time (using various modes of transportation) is 17 minutes (10 minutes in urban vs. 23 minutes in rural
areas), according to the GDS 1+ 2004 240 .
Financing and Fees

Puskesmas are funded to a large extent by the central (APBN) and local government (APBD) budget as well as contributions
from patients. Fees from patients, with a rate determined at approximately USD 0.25/visit (about Rp. 2,500), make up only 2
percent of Puskesmas revenues in Indonesia 241 . Under the regional autonomy law, financing of public health services is the
responsibility of the district governments. The local government receives block grants from the central government (DAU –
dana alokasi umum) from the central government. In addition to block grants, the local government receives additional
income from local taxes, portions of natural resources and some earmarked central government spending in the health
sector (DAK – dana alokasi khusus). Government funding is channeled to Puskesmas through the local health offices (Dinas
Kesehatan), depending on the allocations for health determined by the local government. These allocations vary significantly,
some districts allocate very little for health, while others might even provide free services at the Puskesmas. Unfortunately,
in most cases these variations are not based on health needs, but rather due to district revenues; the higher the district
revenue, the higher the health expenditures. There is hardly any variation in the share of district spending on health, despite
significant regional variations in health outcomes, thus leading to inequality across districts.
The fees for consultation at a Puskesmas vary by regions and are determined by the local government according to different
Perda (Peraturan Daerah). In line with the government’s aim to provide affordable health care in public facilities, the fees are
significantly below the actual unit costs, regardless of the patient’s income. This uniform approach for all income groups
further increases inequality because patients with normal to higher income are in a position to pay higher fees and benefit
disproportionably from the government subsidies for Puskesmas. The fees range from approximately Rp. 2,000 to 5,000,
and in general comprise consultation with a doctor or nurse and the provision of medication, which are included in the List of
Essential Drugs by the MOH (Daftar Obat). The patients often receive three different supplements: a vitamin supplement
(often supplied as a vaccination), a painkiller and one further disease-related medicine 242. In the event that a patient can
provide a card that states that he or she is poor, they can request a reduction or an exemption of fees.
Medication not listed in the List of Essential Drugs and services, such as dental fillings, are subject to additional fees. These
additional charges can often total three to five times the official fees. Nevertheless, the prices are often below the fees
charged in public hospitals. In general, it can be stated that although the official fees in Puskesmas are lower than in other
facilities, there are various means for the individual Puskesmas to increase the contributions, for example, by offering
additional services, prescribing drugs not listed in the List of Essential Drugs, having different charges depending on whether
the consultation takes place with a physician or a nurse, or requesting informal fees in order to compensate for the very low
fees that are considerably below the actual unit costs. Some districts are forced to rely on contributions by the public in order
to run the Puskesmas services more effectively, and call upon the community to provide cars or motorcycles and other items,
as well as act as supporting staff, which is particularly requested of women 243. According to the GDS survey in 2004, costs
240
Based on findings of GDS1+ 2004 and Susenas (2006), GDS 2 (2007) is currently under preparation and preliminary findings are included here as well.
241
World Bank (2006b).
242
According to Mende and Tydecks (2003), this is because Indonesians often believe that a variety of supplements will rather cure them than a single
treatment.
243
Mende and Tydecks (2003).
116
8 • The Role and Financing of Puskesmas in the Indonesian Health System
do not seem to be an obstacle in acquiring health services for most households. In the event of illness, only eight percent of
households mentioned costs as a deterrent to seeking Puskesmas medical services, whereas over a quarter thought this to
be the case for private services (i.e. private physicians). In the case of Puskesmas, prohibitive costs were more of an issue
in rural areas where approximately ten percent of respondents raised this issue. Unsurprisingly, costs were more of an issue
for the poor, although still only for a minority (e.g. 13 percent using income measures). In the case of a serious illness, 1 out
of 3 respondents preferred consulting with a private doctor rather than going to the Puskesmas. Predictably, in case of
serious illness poor households (31 percent) are more likely to go to Puskesmas than non-poor ones (17 percent) (GDS 1+
2004).
Quality of Services
The overall quality of health care services in Indonesia is low, with a low availability of medication, inadequate infrastructure
and often an insufficient supply of health care personnel. Low quality facilities, a lack of clean water and low living standards
appear to provide little incentive for health workers to remain at their assigned posts, especially in remote areas 244. Puskesmas
indicators from the GDS1+ further indicate that the average Puskesmas only has between 75 and 80 percent of the basic
drugs and medication that such facilities should have, and there are also shortages in term of essential vaccines. Most
Puskesmas receive medication from the local health office (Dinas Kesehatan). If the local government does not supply the
Puskesmas adequately, they are unable to compensate for shortages because their own budgets tend to be insufficient.
Therefore, Puskesmas are often associated with a low quality of services and treatment, especially among the non-poor who
have the opportunity to seek alternative health care 245. In the last two years, a growing number of the population perceives
that the quality of services in Puskesmas has improved. This applies to the condition of facilities, attentiveness of staff,
service fees, availability of medicines and waiting time (GDS 2 2007). Nevertheless, more than 25 percent of clients demand
further improvement in medicines and vaccines and 22 percent expect service fees to be more affordable in the future.
8.2. Sustainable Financing for Puskesmas and Social Health Insurance

Participation in health insurance remains low in Indonesia. Between 2003 and 2005, participation in health insurance schemes
decreased slightly from 21.3 percent of the total population to 19.8 percent, leaving around 80 percent of the population
uninsured. In both years, the health card represented the largest share of insurance participation. Askes insurance decreased
slightly in 2005, as well as the self-insured category. Little inequality exists in access to health insurance. The pro-poor
distribution of the health card has decreased inequality regarding access to other types of insurance, such as private insurance,
JAMSOSTEK, and ASKES, but it does question the sustainability of a social health insurance scheme that is based on
contributions by its members and government subsidies for the poor that are only meant to be temporary 246.
In the current set-up, the richer quintiles benefit from subsidies to the Puskesmas just as much as the poor and they utilize
these services almost as frequently. Finding the optimal level of contributions for the richer quintiles and being able to better
target the poor, who might also be in a position to contribute a small amount or be totally exempted, are both key to increasing
the resources available at the Puskesmas and to improving quality of services. These issues need to be taken into consideration
when consolidating and further implementing the current social health insurance scheme. A coverage of standardized minimum
service packages by the social health insurance and additional services, which are charged higher for the non-poor and at
affordable lower prices for the poor (upon presentation of the ASKESKIN card), might be an option to consider further and to
already initiate minimal payments by the poor, who are now also paying for health services and do not see fees as an
obstacle to utilizing Puskesmas.
244
Berber et.al. (2007).
245
World Bank (2003) and Adang (2007).
246
The government subsidy/contribution for the poor is meant to be temporary assuming that their income will eventually rise above the poverty line.
117
GDS1+ data collected from over 120 Puskesmas showed that three-fourths of Puskesmas heads also run private practices
of their own, which has implications on the time they are able to devote to their work in public health centers. Thus, it is also
important to take this aspect into consideration when subsidizing health care or designing a sustainable Social Health
Insurance Scheme. Supporting private sector provision of health care will also be important for the poor. Indonesia’s approach
to private providers expanded the opportunities for private doctors and nurses to practice and added to the availability of
health services to both poor and non-poor households. In the 1980s, when the low salaries of government health workers
made it difficult for them to keep practicing their profession, the government – rather than restricting levels of employment
and raising salaries – allowed its staff to maintain private practices outside of their normal working hours 247. Today, the
majority of healthcare professionals in Indonesia engage in the delivery of both public and private services. While this dual
position of public health providers created perverse incentives, lowered the quality of services in the public health system
(mainly due to the reduced number of hours these doctors put into public practices, as well as the usage of drugs and other
public facilities in their private practices), it also allowed the private provision of services to develop and the average number
of hours served by trained physicians and paramedics to increase. Arguably, one of the reasons why Indonesia is probably
not doing too badly in most health outcomes, despite low levels of public spending, is that private provision services have
filled the service provision gap in areas where public provision has been inadequate in supply or quality. In this situation,
private providers are very much part of health service delivery in Indonesia, and training, contracting and monitoring of
services need to be an integral part of government health policy.
Spending can improve healthcare outcomes, however it is equally important to improve the quality of health policy-making
and health institutions. Important steps to improve Indonesia’s health outcomes include:
Improving the quality of basic healthcare;
Increasing the utilization of secondary care by the entire population, especially the poor;
Strengthening preventative care and intensifying programs and national campaigns that tackle communicable diseases,
particularly in remote and less developed areas of Indonesia.
Two of the three above-mentioned targets are closely linked with Puskesmas – as they are the government institutions
responsible for providing basic health care, strengthening preventive care, as well as promoting and implementing national
campaigns to tackle communicable diseases – especially in remote areas in which Puskesmas are often the only accessible
public health facility.
Increases in the utilization of health facilities accompanied by improvements in the quality of services are a major strategy to
improve public health outcomes. Increasing government spending and income-based patient contributions for Puskesmas
are, therefore, an important mechanism to achieving health for all, while at the same time strengthening a mixed provision of
health care facilities in the public and the private sector.
8.3. Conclusions regarding the Financing of Puskesmas

The increase in the utilization of Puskesmas indicates that the expansion of SHI in Indonesia has increased access and
coverage to basis health services. Unfortunately, the rapid assessment of the Askeskin program has also revealed that the
government contributions to the social health insurance scheme have not been targeted effectively, and benefit the non-poor
rather than the poor. This does not comply with the underlying principles of social health insurance, which only foresees the
role of government in 1) being responsible for formulating implementing regulations for the Social Security System; 2)
ensuring the fair administration of the system and making health facilities available to participants; and 3) registering and
helping to pay contributions for poor people.
247
World Bank (2003).
118
The brief experience of Puskesmas receiving funding from a third-party health insurance company, such PT Askes, as well
as an overall desk-review on Puskesmas financing and utilization has identified issues that need to be addressed in the
future if the Government of Indonesia decides to re-channel funding for Puskesmas services via PT Askes or another
institution as a service provider for health:
Capacity Building and Empowerment of Involved Stakeholders

In general, little has been adjusted to prepare Puskesmas for new modes of funding through a national social health
insurance scheme beyond the transfer of per-capitation payment of Rp. 1000 per person per month based on the
targeting outcome for poor in the catchment area of the Puskesmas. Administrators, providers and the target beneficiaries
are not sufficiently informed about the social health insurance components and procedures, thus leading to the changes
in the program set-up and modes of payment for Puskesmas.9
Regulations need to be clarified and socialized adequately to Puskesmas, District Health Offices and the population in
order to clarify benefits and procedures.
Heads of Puskesmas should be empowered to develop business plans and calculate their costs more efficiently based
on actual unit costs.
Local health offices should support Heads of Puskesmas in advocating for increased budgets from the local government.
Funding and Allocation of Funds

Although the preliminary findings of the GDS 2 survey might lead to the assumption that Puskesmas revenues (see
table with Puskesmas revenue which indicated a significant increase in the non-routine income from Askes compared
to routine income from Askes) have been increased due to the Askeskin program, the assessment of Askeskin also
shows that coverage of the poor is low and that the utilization of the card among the rich was higher than among the
poor.
119
Allocation of Funds for the Askeskin component which was based mainly on the number of poor was not appropriate. It
needs to consider local conditions, i.e., health status of the poor, geographical situation and poverty index and local capacity
of the area.
In general, channeling and absorption of funds were delayed introducing the possibility of poor absorption and leakages. A
funding information system using an effective accountability system should be established. To avoid leakage, the budget
utilization review should be conducted (verification and retrieval) in relation to the services utilization.
Program Management
Program monitoring and evaluation are not yet effective for health facilities contracted by PT Askes, as well as for the
direct funding to Puskesmas. Monitoring of funds should be improved with verification instruments, utilization review
and retrieval system.
The complaint resolution systems are not established in many areas and not functioning in the others. The structure is
to be established considering the need to improve quality and satisfaction level of health care services.
The establishment of the institutional structure is not yet consolidated. There still is loose interaction and coordination
among stakeholders. There is little accommodation for participation of the private sector, including private health facilities.
A clear job description is needed to avoid duplication and gaps in providing health care for the poor.
8.4. Options for the Future Inclusion of Puskesmas in a SHI Scheme

The Indonesian government has reformed the National Social Security Law. On 19 October 2004, the government passed a
bill providing for the foundation of a National Health Insurance Program (NHIP) and decided to implement a social health
insurance with universal coverage. Civil servants and private sector employers make contributions and the government is
making contributions on behalf of the poor. However, contributions by the non-poor remain low, thus endangering the
sustainability of a social health insurance scheme.
In 2005, all contributions from civil servants, private employers and the government contributions on behalf of the poor were
administered by PT Askes and channeled to contracted service providers, including Puskesmas. This arrangement was
changed in the second half of 2005 to direct government (tax-based) funding to Puskesmas and only hospital (in-patient
care) via PT Askes (social health insurance). In 2006, this was changed once again to overall channeling of funds via PT
Askes. The system has since been revised yet again to the current status of insurance contributions to PT Askes for in-
patient care and direct provision of government funding to the Puskesmas.
Reviewing the situation (desk study only), there are two options regarding the financing of Puskesmas:
The first is to opt for a social insurance mechanism which would mean that everyone would contribute to the insurance
funds and that the contributions of the poor are made by the government and that funds are channeled via PT Askes to
all service providers for both out-patient and in-patient care, thus including Puskesmas.
The second would be to continue with the current model and cover in-patient care only by the social health insurance
and continue to make tax-based government contributions to Puskesmas for out-patient care.
Regarding the first option, it is clear from reviewing the above listed issues that the current subsidies from the government
are disproportionately benefiting the non-poor, which is not in line with the principles of social health insurance and universal
coverage defined as access to key preventive, curative and rehabilitative health interventions for all at an affordable cost,
thereby achieving equity in access. Thus, contracting Puskesmas via PT Askes and making insurance-based payments
that are contributed by the members would increase a balanced financing of these health facilities from all members rather
than only funding Puskesmas from government revenues. This would reflect the fact that the utilization of Puskesmas is
equally distributed among all quintiles and should therefore also be funded by all.
120
The inclusion of Puskesmas in the service provider network of PT Askes could stimulate more utilization and outcome based
funding of health services depending on the needs of those insured. It would further encourage Puskesmas to make budget
plans according to actual costs and adjust prices for services accordingly. The contracting of Puskesmas for individual health
services could/should nevertheless ensure that a clear distinction is made between individual services and public health
services, such as health campaigns and spraying mosquitoes. To this end, it would definitely require continued and in some
areas, even increased, provision of block grants from the central and local government to Puskesmas for operational costs
and public health services, taking regional disparities and health outcomes into consideration.
In order to gradually prepare Puskesmas to meet requirements of a social health insurance scheme and related administrative
needs without facing difficulties, the following aspects may be considered:
Contributions
An objective of health-financing policy is equity in financing: households contribute to the health system on the basis
of ability to pay. The current fees at the Puskesmas are very low and it has often been stated that the fees have not been
an obstacle for a utilization of Puskesmas, even for the poor. All members, including the near poor and poor, should
therefore contribute to the Social Health Insurance funds, thus also suggesting that the poor contribute a very small
amount that is supplemented by the government. Only the poorest of the poor should be totally exempted from
contributions.
This would also create a more conducive environment for the development of a sustainable social health insurance with
increasingly universal coverage, because the poor would also contribute according to what they can afford and enhance
the concept of solidarity among all income groups.
Minimal payments would also help dismantle the prevailing idea of ‘free health services’ that the current government
contributions on behalf of the poor for the Askeskin program have provoked. Small co-payments would increase awareness
that it is rather meant as an addition to their contributions as long as they are not in a position to pay the full costs.
A major barrier in accessing public health facilities are transport costs. It would make sense to develop modalities to
reimburse the poor for transportation costs and use part of the government funding for Askeskin for this purpose.
Quality of Services
Increase quality of services including availability of medicine.
Develop Minimum Service Standards for Puskesmas.
Increase incentives for Puskesmas staff, especially in remote areas, including a clearer division of tasks between
physicians and paramedical staff.
Ensure availability of health staff during hours of operation (to counter the high rate of absenteeism of health staff in
Puskesmas).
Encourage a mix of service providers (public and private) with uniform standards and fees. This will increase
competition, ensure that patients use facilities that are most suitable for specific illness and also encourage Puskesmas
to focus on the services they can provide most efficiently.
Develop standardized costing for services and outputs that are based on actual unit costs. This would imply an
absolute change from the current highly subsidized tariffs that are considerably below the actual unit costs.
Continued technical support and stewardship during transition to universal coverage by providing support in management,
oversight and training to all relevant stakeholders, as well as continued ‘socialization” of the concept of social health insurance.
This is currently still not adequately implemented. Heads of Puskesmas have complained that they are not sufficiently
informed about the procedures.
121
A further factor is the availability of skilled administrative personnel to facilitate the effective administration of a
nationwide system. This would require an empowerment of Puskesmas staff to successfully communicate and coordinate
with PT Askes and specifically entail the following:
Capacity building to develop their financial management skills to develop a business plan;
Enhanced negotiation skills for finalizing contracts with the insurance provider;
Development of an output based monitoring and evaluation system.
The second option would be to continue subsidizing Puskesmas directly with block grants for operational costs from the
central and local government. This would not allow for the clear distinction of individual and public health services although
it does offer a solution until certain preparations are met to gradually prepare Puskesmas for future contracting in the
national Social Health Insurance. A few issues could nevertheless be adjusted in this current set-up to make Puskesmas
service delivery more efficient and demand oriented:
Local Government should allocate money for health according to actual needs of the public health facilities, based on
actual unit costs, utilization, geographic conditions, and health outcomes.
Encourage health facility management to develop improved monitoring and evaluation systems and develop business
plans and address the needs of their community to their local government and request funding based on regional
needs.
Define more clearly the division of tasks for Puskesmas: what should they be offering and how should they charge for
these services? This should be reviewed and analyzed thoroughly to improve government budget allocation.
Adjust regulations for retribution to the new situation and allow for the development of alternative models that might be
developed regionally, because Puskesmas have complained that they are receiving less income from the public, but
are still having to pay the same amount of retribution to local government.
The decision regarding which model of financing is most suitable still needs further review and analysis, however, it can
already be stated at this stage that the incentives for the Puskesmas to offer quality services would be, of course, higher if
they were to be contracted under a health insurance scheme, rather than funded by block grants, as is currently the case.
The block grants are not performance based and thus do not offer any additional motivation for Puskesmas management
and staff to increase resource effectiveness and service quality.
122
9. Conclusions
There has been remarkable progress in Indonesia’s social protection system from the crisis to date. Prior to 2005, Indonesia’s
social protection system was characterized primarily by (i) post-crisis-era safety net programs, and (ii) large commodity price
subsidies and transfers, provided primarily by fuel products, and (iii) a rudimentary social insurance system covering only a
minority of the population. The implementation of Law No. 40/2004 on the National Social Security System is currently
introducing a comprehensive reform of the existing system. Law No. 40/2004 provides the legal basis for the above-stated
objective of social protection coverage for all. A task force, headed by the Coordinating Ministry of People’s Welfare, is
working on the introduction of new regulations, as well as the mapping and harmonization of existing regulations relevant in
the context of Law No. 40/2004. It has made considerable achievements since its appointment: finalization of the mapping
and extensive review of existing laws, as well as a comprehensive design of a harmonized legal and regulatory framework.
BAPPENAS, the Indonesian National Planning Authority, is involved in the ongoing reform process and committed to further
develop strategies on how to optimize the process and possibly introduce alternative options based on international experience.
These policy recommendations are specifically relevant in the preparation of the five-year plan 2010-2015 for the further
development of a system of social protection that will be launched at the end of the current legislative period in 2009.
It is evident that all future decisions and reforms should take the ongoing reforms into account and need to be closely
discussed with all stakeholders involved in the reform process. The current reform status also determines how much space
is left for proposing or introducing new systemic options concerning the social protection of diverse risks.
This approach to combine the Social State Model with welfare elements seems to be the most suitable in Indonesia because
it is not easy to decide whether Indonesia should turn to the Welfare State or to the Social State approach for Social
Protection. A composition of public social insurance and social assistance schemes together with provisions of the private
sector in three overlapping structures is recommended.
Regarding the importance of reforming the administration of Social Security, the report reviews possible options and concludes
that harmonization and rationalization of regulations between existing administrative bodies is the preferred option at present
in Indonesia. It is acknowledged that harmonization represents an important first step. However, given limited technical
benefits of harmonization in comparison to the other two options, the country may consider proceeding further in the mid-
term. Options for later reforms are: (a) institutional integration of risk-related activities under one specific administrative body
for each risk in standard social insurance, and (b) institutional separation of activities for the poor and other vulnerable
groups, i.e. social assistance. In this context, responsibility for social assistance should then be centered on a newly founded
institution, perhaps under the umbrella of the Coordinating Ministry of Social Affairs.
123
With regards to social health protection, the systemic question, i.e. which instruments to apply, appears to be answered by
combining social insurance and tax-financed means-tested social assistance to the poor. Many country examples, such as
South Korea or Costa Rica, show that this method is a feasible option for achieving universal coverage. Questions concerning
social health protection reform, therefore, refer primarily to technical issues within an already defined system. The ratification
of Law No. 40/2004 on the NSSS established the basic principles of universal social security and a time frame for the work
necessary for the new system to be implemented throughout Indonesia. Within the boundaries set by the new NSSS law and
the reform and implementation steps taken since then, however, many decisions still need to be made, for example, with
regard to the choice on how to pool funds, i.e. single vs. multiple funds.
The question as to whether the Government contributions for health services for the poor in community health facilities
(PUSKESMAS) should be paid via the regular health insurance carrier or directly to the Puskesmas using block grants from
the central and local government is reviewed separately. Considering the specific incentive structures and different requirements
involved, insurance-based financing seems to be the more sustainable and effective option in the long term.
With regard to pensions, a finding is that, in absence of comprehensive social insurance providing life-long old age benefits,
the reform agenda in Indonesia is wide open for innovative solutions, as there are only a few limiting path dependencies to
respect. One conclusion is that as prevention against demographic risks, funding elements could gradually complement
pension insurance. Then, when the new national pension scheme is fully established, the share of pay-as-you-go financing
could be gradually reduced in favor of the funding principle. Low-income earners could be supported to a certain extent by
tax-financed instruments. A tax-financed universal basic pension scheme could be introduced with pensions based on a
fixed percentage of local minimum wages. However, there are certain drawbacks to consider. First, like all universal schemes,
it could be very costly to guarantee such a basic pension considering the high percentage of the poor and near-poor population.
Second, the option for a non-contributory pension could create negative incentives for participation in contributory social
insurance, so this solution could endanger moving from informal to formal employment. Another approach to address the
problem of old age poverty could be seen in the future role of social assistance, which would be subject to means-testing and
targeted to all persons with income below the national poverty line. This option, however, is not a substitute to the previous
options, but rather a complementary one.
In stark contrast to social health protection, the issue of unemployment compensation or which instruments to choose for
unemployment compensation is still under strong debate at present. Preliminary recommendations from this report propose
a combined system of mandatory individual saving accounts and mandatory savings for severance pay, both of them
complemented by solidarity funds. General systems of either unemployment insurance or unemployment assistance financed
by payroll taxes are not recommended.
In terms of providing social assistance, horizontal and vertical coordination of existent fragmented programs, as well as of
those to be initiated in the future – from an external perspective preferably under the umbrella of the Ministry of Social Affairs
– is essential. For financing social assistance, several options exist to enhance the tax base without increasing taxes.
However, sustained economic growth is central for extending social assistance as it increases tax revenues, while
simultaneously decreases the number of potential beneficiaries.
How to incorporate the non-poor in Indonesia’s large informal sector and how to provide and finance at least basic social
assistance to the high numbers of poor or near-poor are two of the most crucial challenges on the way to universal coverage.
Concerning an extension of social protection coverage to the informal sector, it was deliberately emphasized that the reduction
of the informal sector and its stepwise integration into the formal economy should be the ultimate and foremost objective. An
inclusion of informal sector workers into social insurance needs to rely on the design and implementation of specific programs
adopted to informal sector characteristics as standard social insurance is not appropriate for reaching out to the informal
sector. Due to these differences informal sector schemes exhibit a variety of risks that endanger their effectiveness in
coverage and financial sustainability. Reliable data for planning, a careful design and skilled administrative personnel for
implementation are, therefore, necessary preconditions.
124
9 • Conclusions
Reviewing the current situation and analyzing the options for the various social protection schemes for citizens revealed that
the efficacy of ongoing and future reforms will depend highly on developments and progress in other sectors. Most prominently,
it will depend on private sector development and the expansion of the formal sector, and, respectively, reduction of the
informal sector, tax system reforms, and sustained efforts in decentralization reforms. As economic growth clearly impacts
the feasibility and outreach of publicly provided social protection, continuing efforts to promote an enabling environment for
economic growth are essential. This includes, among other things, the creation of a conducive and investment-friendly
business environment.
Moreover, reform options also need to be carefully designed in order to prevent adverse impact on the labor market. Another
important aspect is that although social protection in the long run has the potential of enhancing social stability and social
cohesion by achieving equity, it demands solidarity and ownership from the entire population. Communication and marketing
of reform efforts can aid in achieving support. However, in the end, people will only support a social protection system if they
receive benefits and if it operates in a credible way. Most likely, the society, specifically the economically stable, will expect
solid evidence that they are also benefiting from the system and not just primarily financing it.
All in all, institutions need to be strengthened, and technical support and capacity development need to be available at all
stages of the reform.
Further, reform efforts should be realistic and consider prevailing constraints. For example, considering the unfortunate
relationship between the number of poor and near-poor persons in Indonesia and the available tax resources, the potential
outreach of social assistance is limited and any ad hoc measures should be avoided. Basically, the decision is either to
provide a relatively broad range of social assistance measures to a very limited group of persons or to focus on a more
limited number of measures (such as health services or modest forms of income support via cash transfers) provided to a
wider share of the poor population.
The questions of sequencing and timing of reform steps are beyond the scope of this paper. However, a continuous fine-
tuning and further development of the Road Map for Social Protection Reform in Indonesia are, naturally, very important
issues.
Last but not least, it needs to be emphasized that achieving universal coverage, as intended by the Government of Indonesia,
is a lengthy and incremental process, especially in a vast and heterogeneous population like that of Indonesia. With its long-
term political commitment to achieving social security for all, the Government of Indonesia has met an essential pre-condition
for success.
125
126
8 • Conclusions
Annex
127
128
Annex 1: Characteristics of Social Security prior and according to Law 40/2004
Social Security System prior to Law 40/2004 Law 40/2004 on Social Security Reform
GENERAL GENERAL
Administration Administration (Art.2-5) and Management of Social Security Funds

Social Security Administering Bodies (SSABs): governmental profit-oriented (Art.47-51)
limited companies, Perseros (PT) • Social Security Administering Bodies (SSABs): existing SSABs (Art.5
1. PT TASPEN- The government civil servants’ social insurance and 47)
Provides: Lump-sum cash benefit at retirement age, death benefit, o JAMBOSTEK, TASPEN, ASABRI, ASKES
cash value before retirement, invalidity pension o Additional SSABs may be established
2. PT ASKES- The health insurance for government civil servants, • Government allowed to take specific measures to safeguard
retirees, veteran and their families financial soundness (Art.48)
Provides: health care • No cross-subsidies (Art.49,2)
3. PT ASABRI- The armed forces social insurance • Technical reserves required [liabilities of SSABs arising in the
Provides: protection against the loss of income due to old age, context of meeting future liabilities to participants] (Art.50)
termination of employment, death, pension • Supervision carried out by an authorized institution (Art.51)
4. PT JAMSOSTEK – The employeesí social security • Participants have the right to obtain and SSABs are required to
Provides: accident insurance, provident fund, death insurance, health provide at any time about accumulation of c and return and
care benefits of old age pension, public pension, life insurance scheme
(Art.49,3&4)
• Supervision and monitoring carried out by authorized institutions.
• The Indonesian National Social Security Association (INSSA) is an
association, which linksup the SSABs.
Participation and contributions Participation and cont ributions (Art. 13-17)

• Entitlement/compulsory for: employees of the civil service, members of the • Compulsory, but adoption of this principle contingent on economic capacity
armed forces and employees of private companies with a certain staff size of people and government; initial phase: compulsory for formal sector
(10 employees and more) and total payroll (Rp. 1 million and more). employees; voluntary for informal sector workers and inclusion of poor
According to the law all workers are entitled to be voluntary members of the people
Jamsostek scheme • Income-related or fixed amount
• Contributions: Contribution from employers, employees or combination of • Contribution assistance: Contribution paid by government on behalf of poor
contribution from both and destitute people (in the 1st stage of reform for health benefit only)
HEALTH HEALTH
ASKES (Govt Regulation No. 6/1992): Health Benefit (Art.19-28)

• Principle: Social Insurance • Principle: Social insurance Coverage: universal (intened)
• Membership: • Membership (Art. 21, 28):
o Government employees (1968) o including family (up to 5 members); inclusion of additional
o Including family members(spouses and up to two dependent family members possible by payment of additional contributions
children under 21 years of age) o after termination of work membership continues for six months
o Veterans and fighters for national independence without paying contributions; in case of unemployment/
• Benefits: comprehensive package, but differentiated in different classes permanent disability and subsequent inability to pay,
regarding IP. Included: primary health care, secondary health care, membership is financed by government (taxes)
inpatient care; normal and abnormal delivery; simple laboratory and X-ray • Benefit package (Art.22): ‘health promotion, preventive and
procedure; pharmaceutical; supplement, eye glasses, dental prosthesis curativetreatment, and rehabilitation incl. medicines and medical
• Financing (Art.22, 27): supplies used as necessary’; [medical services and treatment,
o Contribution: employees: 2 percent; government: 0,25 percent immunisation, contraception services, OP and IP services, emergency
(2004), ca. 1 percent (at present), 2 percent (by 2007) services, other medical procedures incl. blood transfusions and heart
o Provider payment: capitation fee, per case, per diem surgery], IP: ‘standard health services’; exclusion to be defined by
o Level of payment is determined by MoH and MoIA to ensure SSAB (Art.26)
ASKES’ solvency • Financing (Art.22, 27):
o Contributions formal sector: income related (with limit), jointly
JAMSOSTEK (Law No.3/1992):
employer-employee
• Principle: Social insurance
o Contributions for persons not receiving a wage: fixed amount
• Membership:
o Contribution assistance (tax-based): contributions paid by
o SHI for private employees (1992)
government on behalf of poor and destitute people
o Obligatory for all employers having more 10 or more employees;
o Co-payments: possible ‘for types of services that may induce
includes family up to 3 children; employ a workforce of 10 (ten)
misuse’ Provider (Art.23,24):
or more or pay wages of at least Rp 1,000,000 (one million
o Contracted government-owned or private health facilities o
rupiah) a month (Government Regulation No. 14/1992)
Exceptions in case of emergencies
• Benefit: Ambulatory care, hospitalisation and maternity
Annex
129
130
• Financing: o Compensations for regions where no adequate health facilities are
o Contribution: employers pay 3 percent (6 percent) of wage; wage available
ceiling: Rp 1ml per month (constant since 1993) o Payment mechanisms:(a) Amount of payment to health facilities for
each region to be determined between SSAB and association of
health facilities in the region concerned; (b) Payment within 15days
of receipt of a request for payment [budget or fixed amount per
capita]
o Drugs and medical supplies: standard list and ceiling prices
• Healthcare delivery system, quality assurance program and healthcare
financing system to be developed by SSAB (Art.24,3)
Work Accident Benefit (Art.29-34)

• Principle: Social insurance (Art.29)
• Coverage: Participants paying contributions
• Benefit (Art.31): oproviding health benefits and cash compensation (unique
lump-sum depending on degree of disability) in case of permanent total
disability or death following a work accident or work-related illness
• Financing (Art.34, 31):
o Contributions: income related; employers only (varying to risk
environment of work)
o Contributions for persons not receiving a wage: fixed amount
o Co-payments: for certain services/accidents; employers only
Providers (Art.32):
o See above
OLD AGE/ DEATH/DISABILITY OLD AGE/ DEATH/DISABILITY
Old Age Benefit Old Age Benefit (Art.35-38)

PT Taspen (Govt Reg No.25/1981): • Principle: social insurance [joint contributions] or mandatory savings
• Benefit: Lump sum payment on retirement [benefits accrue from accumulated contributions] (Art.35)
• Financing: • Coverage: Participants paying contributions Benefit (Art.37):
o Contribution: Employee: 3.25 percent of salary
PT Jamsostek (Law No.3/1992) o Lump-sums paid to participants when reaching the age of
• Benefit: Lump sum payment contribution + interest (JAMSOSTEK JHT) retirement, after death, or suffering permanent total disability
• Financing o Amount is based on total accumulated contributions plus return
o Contribution: employee:2.00 percent, employer: 3.7 percent o Partial payment possible after 10 years
PT Asabri (Govt Reg No.67/1991): • Financing (Art.38):
• Benefit: Lump sum payment on retirement o Contributions formal sector: income related, jointly employer-employee
• Financing: o Contributions for persons not receiving a wage: fixed amount
o Contribution: Employee: 3.25 percent of salary,
Pension Benefit Pension Benefit (Art.39-42)

Principle: Social Insurance • Principle: social insurance or mandatory savings [basically social insurance
PT Taspen (Law No.11/1969) : mechanisms, but providing the opportunity for employees who are entering
• Benefit: Annuity benefit for life (is equal to 16.5 months of salary) retirement age but have not fulfilled min contribution period to have their
• Financing: contributions be treated as compulsory savings] (Art.39)
o Contribution: employee: 4.75 percent of salary, government: state • Coverage: employees paying contributions
budget • Benefit (Art.41):
o Monthly cash amount (in case of old age, disability, widow/widower, child,
PT Jamsostek: ---
parents of unmarried participants) after having paid contributions for at least
PT Asabri (Law N. 6/1966): 15 years that ensures a decent standard of living, i.e. basic needs (Art.39)
• Benefit: Annuity benefit for life o In case of reaching retirement before p. is entitled to reach the
• Financing: accumulated amount plus return
o Contribution: employee: 4.75 percent of salary, government: state • Financing (Art.42):
budget o Contributions formal sector: income related or fixed, jointly
employer-employee
Death Benefit Death Benefit (Art.43-46)

PT Taspen: -- • Principle: social insurance or mandatory savings (Art.43)
• Coverage: participants paying contributions
PT Jamsostek (Law No.3/1992):
• Benefit (Art.45): fixed amount
• Benefit: Lump sum death benefit and funeral
• Financing (Art.46):
• Financing:
o Contribution formal sector: income-related, employers only
o Contribution: Employer – 0.3 percent of gross wages (ILO 2004: 22)
o Contributions for persons not receiving a wage: fixed amount by
PT Asabri: --- participant
Annex
131
Annex 2: Various Funding Sources of Puskesmas according to routine and non-

routine income
Summary for variables: um1_4 by categories of: um1_2 (routine income of Puskesmas)
um1_2 min mean max

------------------------------------------------------------------------------------------------------
Patient 15,000 1.47e+07 2.99e+08
APBN 33,050 8.89e+07 9.79e+08
ASKES 74,000 1.01e+07 1.20e+08
APBD 129,499 1.92e+07 1.59e+08
Other 20,948 3.98e+07 4.35e+08
------------------------------------------------------------------------------------------------------
TOTAL 15,000 5.97e+07 9.79e+08
Summary for variables: um2_4 by categories of: um2_2 (Non-routine income of Puskesmas)
um2_2 min mean max

------------------------------------------------------------------------------------------------------
Patient 25,000 1.87e+07 1.17e+08
APBN 20,000 5.14e+07 8.51e+08
ASKES 1,685,000 6.63e+07 6.06e+08
APBD 600,000 1.35e+07
Other 23,016 7.18+07
------------------------------------------------------------------------------------------------------
TOTAL 20,000 4.92e+07 8.51e+08
Source: GDS 2, presented in 2007
132
Annex
Annex 3: Social Health Protection in Latin America
Brazil248
Contrasting the general trend towards market-oriented systems on the Latin American subcontinent in the 1980es and
1990es, the largest country in the region set up a mainly tax-funded national health system. The constitution establishes
health care as a civil right and a public responsibility. Against this legal background, Brazil integrated the formerly existing
social insurance schemes into the new Unified Health System (Sistema Unitário de Saúde, SUS) that is organized according
to general characteristics of the Beveridge model. Through the SUS, all citizens are entitled to appropriate medical care, and
the idea behind was to overcome the great disparities in provision among different social groups and between urban and
rural communities. The core elements of the Brazilian health sector reform were universality, equity of access, participation
and integrated provision. In the country’s federative structure, decentralization is an important principle: the federal states
have a major influence on all health policy decisions, while the municipalities are responsible for health care provision and
contribute 15 percent of overall funding.
The remarkable feature of the Brazilian system is the combination of public financing, which at least theoretically guarantees
universal coverage, with heterogeneous provision of health services including private provision on a significant scale. Whereas
three-quarters of outpatient care takes place in public health centers and outpatient clinics, the majority of inpatient treatment
under the SUS is provided in private hospitals and clinics. The separation of financing and service provision facilitates
competition between service providers, which in principle can be expected to increase the efficiency of the system. But the
great fragmentation of providers, and above all the predominance of a fee-for-services system within the payment structure,
creates harmful incentives and cost-inflating induction of demand.
Moreover, coexistence with private health insurance schemes, which sell additional insurance to the majority of the upper
and middle classes, gives rise to considerable fairness and efficiency problems. Thus, in spite of the tax-financed state
system, medicine remains a multi-tier system, particularly as SUS provision outside the major cities is extremely sketchy and
inadequate. At the same time, service providers, which are often closely connected with insurance providers, tend to implicitly
privilege private health insurance schemes by charging expensive services to the state scheme even if patients are insured
privately, on the basis that all Brazilians are legally entitled to use the SUS.
Another relevant challenge faced by the new system appears to be the trade-off between decentralization and coordination.
Recently, inter-municipal consortia have been implemented where neighboring municipalities form a partnership to supply
high-complexity health care. Despite potential gains of scale, scope and coordination associated to such partnerships, the
study shows that there are free rider incentives to be dealt with in order to ensure stability to the consortium. The free rider
problem arises from the fact that one of the members of the partnership may benefit from the gains of joint provision of health
services without participating in its financing. This problem is aggravated by a constitutional clause that states that health
care funded by public institutions cannot segregate, so that every citizen should have access to any public health care
establishment when needed.
248
Frenk and Londoño (1997); Teixeira et al. (2006).
133
Chile 249
Since its market-oriented social sector reform in 1981, Chile is generally considered a typical example of market-driven
health sector reforms. The intention was to re-organize the widely state-run system in a way that allowed for an increasing
role of private insurers and providers. Health protection is compulsory for all citizens who were initially mandated to spend 4
percent of their taxable income. The low wage-related rate, however, soon turned out to be insufficient for covering health
expenditure, and contributions were stepwise incremented until reaching 7 percent. In theory, all citizens have the freedom
to either enroll with the public social health insurance scheme or with one of about 15 private insurance companies. The
latter, however, can select their enrollees according to their ability to pay. The entrepreneurial logic forces for-profit insurance
companies to make sure that the expected expenditure for services does not exceed the income from premiums.
However, even in times of robust economic growth and relative welfare in the late nineties, affiliation to private health
insurance companies never exceeded one third of the population. The coexistence of a solidarity-driven public sector and a
for-profit private sector operating with risk-adjusted premiums has led to a two-tier health insurance system. Risk selection
applied by private insurance companies causes serious equity and fairness problems. Unpredictable, often existentially
threatening co-payments have become a serious problem for the users of the Chilean health care system, and coverage-
lacks have become a major menace for patients.
More than 25 years after the wide reaching sector reform, the results are relatively far away from the initial intentions. The
pretended extension of private health care and financing has not been achieved, and more than two thirds of the citizens of
the South American country still depend on the public services 250. Efficiency gains are to be observed rather in the National
Health Fund (FONASA) than in the private health market. Since the end of the military regime in 1990, democratic governments
have invested heavily in public service. At the same time, FONASA underwent a series of internal reforms and a re-structuring
of its functions.
Achieving universal coverage is one of the major challenges in most developing countries. Whilst most analysts are focusing
upon the effects of privatization and competition in health financing, another fundamental lesson learned from Chile is
generally under-represented in the current debate. Today, however, it is also one of the very few countries in Latin America
that provide practically universal coverage in health. This has become possible due to the combination of the Bismarck- and
the Beveridge-system. The formal economy and parts of the informal sector are counting for a contribution-based insurance
system. The poor are protected by a tax-financed welfare system administered by the same public social health insurance
FONASA. Both public sub-systems are solidarity-driven and their combination guarantees for progressive financing and
effective re-distribution in the public health care sector (Bitrán/Giedion 2003, p. 62). A set of waivers and exemptions within
the public system is diminishing the negative social effects and the discrimination produced by out-of-pocket payments.
Altogether, under the roof of FONASA an effective linkage of contributing and non-contributing members has been implemented
and continuously managed.
249
Bitrán and Giedion (2003: 60-66), Holst et al. (2004).
250
About 10 percent of the population rely on the autonomous insurance schemes run by the armed forces including the police, the large universities and
some public enterprises as the national copper industry CODELCO, and others (Holst 2001: 19,79).
134
Annex
Colombia 251
Prior to the Colombian health sector reform of 1993, employment-based social health insurance schemes covered about
one third of the population. Formal sector workers were compulsorily enrolled to closed pools and paid mandatory contributions.
While the largest fund, the Instituto de Seguridad Social (ISS), covered private-sector workers, public-sector workers were
enrolled in funds exclusive for state-owned enterprises, public universities or government units. Besides two large public-
sector funds (Cajanal for central government employees and Caprecom where workers in the state-owned telecommunications,
television and postal enterprises were enrolled), the overall risk pool was divided in more than 1,000 small-scale insurance
schemes.
The 1993 reform was inspired by the fundamental changes Chile had applied to its health system 12years earlier. However,
Colombians tried to avoid the obvious undesired effects that became more and more visible in Chile. By creating a contributory
system for all formal sector workers and employees, the previously existing monopolistic funds were exposed to a sort of
managed competition, since new insurance companies were allowed to enter the market and citizens had the right to choose
the insurer according to their own preference. Financing is based on mandatory payroll-contributions of 12 percent of the
salary (employer 8, employee 4 percent).
In order to avoid market failure and inefficient competitive behaviour by health insurers (most of them were private, and
some for profit), Colombians tried to implemented effective regulation. In order to prevent competition on prices, contributions
within the contributory scheme are pooled in a single equalization fund, which then allocates insurers a risk-adjusted community-
rated contribution for each enrollee defined by age, gender and geographic location (urban/rural). Thus, enrollees pay
according to their ability to pay, but insurers receive contributions according to the individual risk of the insured persons.
Furthermore, a standardized benefit package defining services covered and excluded, qualifying periods for full coverage
and co-payment rates has been established, insurers are mandated to contract all applicants, enrolment is compulsory for
all registered workers, and full portability of benefits is guaranteed.
With regard to the challenge of extending social health protection beyond formal sector employees and workers, the creation
of a subsidized system besides the contributory scheme is especially worth mentioning. Citizens who are economically
unable to pay contributions are required to register in an Administrator of the Subsidized System ARS). However, the number
of poor and indigent people enrolled in the subsidized scheme still remains below expectations.
For targeting the needy and the poor, Colombia has developed a Beneficiary Identification System (SISBEN) based on
comprehensive questionnaires and individual contacts by social workers. Entitlement to the subsidised system is conditioned
by an adequate SISBEN qualification. Financing of the subsidised scheme relies on tax revenue and on the solidarity
contribution of the better-off who pay 13 percent instead of 12 percent for health insurance, and the additional 1 percent
contribution is designated to subsidize the poor. The benefit package covered by the solidarity scheme is more limited
compared to the comprehensive one of the contributory system, and the main advantage for beneficiaries are co-payment
exemptions at public health care providers.
Altogether, despite the impressive first place Colombia occupied in the 2000 World Health Report regarding the financial
equity, the 1994 reform has failed to overcome the pre-existing equity problems of social security. The health system is still
exclusive, and compared to the population share covered previously by the public health care system the coverage rate has
not effectively increased. Health has remained an individual and not a social right, and Colombia is still far away from
providing universal coverage without discrimination. The country experience shows rather that even applying complex
regulations the concept to universalize health seen as an individual consumption good does not ensure to achieve universal
coverage.
251
Hernández (2002) Castaño and Zambrano (2005).
135
Costa Rica 252

Costa Rica can be considered as one shining example for the successful implementation of universal social health insurance.
Today, the health sector in the small Central American country is mainly funded by the Costa Rican Social Security Fund
(CCSS), with some health promotion, disease prevention and environmental health provided by the Ministry of Health
(MoH). Besides other intrinsic factors of the Costa Rican society, effective social protection has certainly played an important
role for improving the health status of the population. Today, Costa Rica’s health indicators resemble those of industrialized
countries, rather than those generally exhibited by countries with a similar per capita income (€1300). The progress in
health is based on the Social Security principles of solidarity, universality and equality.
The CCSS was created in the early 1940s to administer the social security insurance system and has been very successful
in stepwise expanding social protection to the whole population. Like practically all social health insurance schemes around
the world, the CCSS started with formal sector workers, and already in 1945 their family members were also covered. The
next relevant steps of extension were the inclusion of pensioners enrolled in the contributory pension scheme in 1959 and of
the elderly without mandatory old-age insurance in 1974. On year later, voluntary insurance was implemented for self
employed for whom affiliation became compulsory in 1984. Lastly, social protection was extended to informal sector workers
in 2000 when the ”Law for the Protection of Workers” was passed. Due to high commitment, good governance, and social
dialogue, social protection and access to health services are now more or less universal.
In 2004, the CCSS spent 69.8 per cent of total health expenditure for providing curative and rehabilitative care, individual
preventive services such as immunization, and some educational services. In addition, the National Insurance Institute (INS)
covers the treatment, rehabilitation and compensation of policyholders for occupational illnesses, work accidents and
automobile-related injuries. The government is responsible for public health services and spends additional resources for
subsidizing the health and the pension schemes.
Since the 1994 health sector reform, all public health services have been transferred to the CCSS that currently is running
and operating all the country’s 29 hospitals, providing 95 per cent of hospital services and around 70 per cent of all consultations.
Except for three private clinics, virtually all health facilities are operated by the CCSS, the MOH or the INS, and form part of
the national health system. While the quality of care for inpatient services is considered to be quite high, dissatisfaction with
the quality of care provided in ambulatory settings is perceived to increase. Health-care providers in the MOH and CCSS are
mainly paid in salary, although the CCSS has been experimenting with other options, including the Company Medicine
Scheme (where a company pays the physician’s salary and provides a clinic for employees) and capitation payments to
physicians or cooperative clinics.
252
Sáenz and Holst (2007).
136
Annex
Mexico 253
The history of social health protection in Mexico started in 1943 when the Mexican Social Security Institute (Instituto Mexicano
del Seguro Social, IMSS) for formal, private sector, salaried workers and their families were established. In 1959, the Institute
of Social Services and Security for Civil Servants (Instituto de Seguridad y Servicios Sociales de los Trabajadores del
Estado, ISSSTE) began to cover government employees and their dependents. From the beginning, the Bismarck-borne
Mexican system is characterized by its segmentation between formal workers entitled to social security benefits, and self-
employed, unemployed, non-salaried and informal-sector workers whose health care needs were in principle attended by
the Ministry of Health.
Lack of universal social protection in Mexico is intertwined with the existing health system segmentation between the formal
sector, with access to social security, and the rest of the population that is excluded from this type of protection. This situation
has contributed to the prevailing health inequalities and unfair health financing. While more than 60 percent of the wealthiest
quintile of the population was insured, the figure was approximately 10 percent for the poorest quintile. Further, in the
poorest states only one-fifth of households were insured. Each year, between two and four million households either spent
30 percent or more of disposable income (total income less spending on food) on health or crossed the poverty line due to
health spending. Further, 85 percent of these households were uninsured and the majority was from the poorest deciles.
For improving social health protection, Mexico has recently devised a structural reform in order to extend social protection in
health to families not covered by conventional employment-based social insurance. The reform of the Mexican health care
system is designed according to the concept of ”managed competition” to gradually correct major financial disequilibria and
comprises an operational program called Popular Health Insurance, which offers uninsured Mexicans voluntary access to
publicly subsidized coverage for a comprehensive set of services at the primary and secondary levels, as well as a gradually
expanding set of costly high-specialty interventions. It is expected that by 2010, at the end of a seven-year transition period,
universal coverage will be reached.
However, some doubts exist whether the cornerstone of the 2004 reform, the voluntary Popular Health Insurance (PHI), will
resolve the problems of the public health care system. The next two or three years will be critical for the viability of PHI: both
families and states will face increasing difficulties in paying the insurance contributions; health infrastructure and staff might
turn out to be insufficient to guarantee the health package services; and the contracting of private providers will further strain
state health ministries’ ability to strengthen service supply. Moreover, redistribution of federal health expenditure favoring
PHI at the cost of the IMSS might further endanger public health care delivery.
253
Knaul and Frenk (2005), Frenk et al. (2007), Laurell (2007).
137
Annex 4: Protecting the Elderly Poor
The following pages present four countries’ (Germany, United Kingdom, Switzerland and Sweden) approaches to include
protection of old age income for elderly poor into their pension systems. They reflect the three models of first tier pension
provision: basic pensions, resource/means-testing or minimum pensions: (1) Basic pensions are flat rate benefits independent
of an individual’s earnings history but with the requirement of minimum contribution periods. (2) Resource-tested benefits
vary in the amounts paid out to individuals in accordance with other sources of income available to the beneficiary. The rules
applicable to retirees may be identical to those of the population in general. (3) Minimum pensions are paid independent of
the beneficiary’s possible other sources of income.
A Germany
A.1 Main Features of the Old Age Protection System
Germany’s statutory public pension system has traditionally served to provide income replacement for employees
when they go into retirement. It is a social insurance-type, contributory and earnings-related pay-as-you-go scheme.
Benefits are calculated on the basis of average lifetime earnings using a points system that sets individual earnings
in relation to national average earnings. Past earnings and pensions in payment used to be indexed in line with
the growth of average earnings. However, this has recently been changed and a dampening factor is now applied
that takes into account the effects of changes in the system dependency ratio, the relationship between the
number of beneficiaries and the number of contributors. The system maintains a strong link between individual
contributions and benefits. It is not intended to be redistributive between high and low earners. The contribution
rate of, currently, 19.9 percent of salary is paid in equal shares by employee and employer. Contributions are
capped at about 2.1 times the average relevant earnings 254.
While an estimated 63 percent of employees are covered by additional voluntary occupational schemes or maintain
individual private old age provision 255, the social insurance scheme, which covers close to 90 percent of the
population and on average provides two thirds of overall old age income 256, continues to be the main factor of old
age security. As this is certain to change in the future, participation in additional old age provision has recently
been rising considerably.
A.2 Protecting the Elderly Poor

In line with the intended principle of maintaining a close link between contributions and benefits in the German
social insurance pension system, there is no basic pension or minimum pension included in the system as such.
Recent reforms to the pension system have reduced the gross individual replacement rate for an average earner
from 48.7 percent to 39.9 percent 257. However, more importantly, for workers earning just 50 percent of the
national average salary during their entire working life, the reform effect was that their net pension benefit was
reduced from 39.7 percent to 32.6 percent of national average earnings 258.
These changes, combined with high unemployment and increasingly fragmented employment histories have led
to an increasing number of smaller pensions – a trend that is likely to continue.
254
OECD (2007:127).
255
OECD (2007:78).
256
Nationaler Strategiebericht Alterssicherung 2005, Deutscher Bundestag, Drucksache 15/5571, Berlin 27.05.2005, p. 8.
257
OECD (2007:66).
258
OECD (2007:67).
138
Annex
Elderly poor can apply for social assistance, i.e. a means-tested benefit to top-up their individual old age incomes
to the politically defined minimum income level. However, in the past, a significant number of elderly poor is
known to have been ashamed to claim social assistance. To remedy this situation, the introduction of a targeted
old age basic protection scheme (Grundsicherung im Alter) in 2003 introduced a new logic: social assistance
rules were eased for applicants above age 65, and the benefit no longer had to be applied for at social assistance
offices but through the pension insurance administration.
The most important change has been that social assistance offices no longer take recourse to the income (or
assets) of an elderly applicants’ relatives. The fear of causing financial hardship for their children had caused
many retirees who would have been entitled to social assistance to refrain from an application – even in cases
where recourse to family members’ resources would have been highly unlikely. In order to reach as many entitled
retirees as possible, the pension insurance institutions are obliged to contact pensioners with pension amounts
below a certain threshold in order to inform them about the possibility of receiving this means-tested top-up.
The basic old age protection scheme is financed from general tax revenue.
On average, the social assistance payments will lift beneficiaries to an income level of about 19 percent of
national average earnings (with regional variations in line with differences in housing costs) 259. About 1.8 percent
of the population above the age of 65 receive payments from the social assistance scheme for the elderly.
According to the OECD’s standardized calculations 260, the resource-tested old age benefits paid to pensioners in
Germany amount to 1.1 percent of all payments to retirees from the public pension system. The entire remaining
percentage is the share of the earnings-related social insurance scheme.
B United Kingdom (UK)

B.1 Main Features of the Old Age Protection System
The UK’s state pension system comprises two tiers. The first tier, a flat-rate basic pension, has traditionally
functioned to provide minimum protection in old age. A second tier, an earnings-related additional pension, was
introduced in 1975. While this meant a departure from the philosophy to have the state provide only a universal
minimum, the pensions payable from the system were too small to constitute an actual income replacement. The
earnings-related state pension has since been made more redistributive and, as a result of a major reorientation
of the country’s pension policies over the past three years, is now being scaled down and re-oriented to provide
an additional benefit not only in recognition of employment but also for caring responsibilities.
This combines with the country’s strong tradition of voluntary defined benefit occupational pension schemes and
various forms of individual private provision. In order to protect this culture of private provision, the introduction of
the state earnings-related pension was coupled with a “contracting-out” mechanism that allowed individuals to
have a part of their contributions to the state scheme transferred to an occupational or individual pension plan
that fulfilled certain criteria.
Benefits from the state system are dependent on contributions or credits. 44 (39) years for men (women) are
required for the full basic pension. Contributions are not for pensions in particular but are general for all branches
of national insurance (health care, maternity benefits, unemployment etc). They are split between employee and
employer and capped at about 1.15 times national average earnings. Non-contributory credits are assigned to
people caring for children or disabled relatives, and also to low-income earners.
259
OECD (2007:127).
260
To facilitate international comparison of pension systems, the OECD calculated the theoretical weight of the different tiers in the public pension systems
of its member countries. Using each country’s distribution of earnings and assuming full-career employment, the results show the relative importance of the
different components in a country’s pension system as in force in 2004. For lack of suitable data it has not been possible to consider the growing number
of interrupted and fragile employment histories. The results will therefore tend to underestimate the absolute importance of the first tier components that are
aimed at protecting the elderly poor. (OECD 2007:50-51).
139
B.2 Protecting the Elderly Poor

The UK’s state pension system contains a number of elements to particularly help low income earners obtain
sufficient pension entitlements and payment amounts. Employees earning above the lower earnings limit (of
around 15 percent of national average earnings) but below a primary threshold (of around 17.5 percent of national
average earnings) are exempt from contributions but accrue entitlements to the state basic pension.
In 1980 the UK changed the indexation rule for the state basic pension. While in the past it had been up-rated
annually in line with the growth of national average earnings, this was replaced by the retail price index. By the
end of the 1990s it had become clear that this had led to an erosion of the value of the state basic pension that for
an increasing part of the pensioner population meant a life in poverty.
As an initial response, since 2000 the price indexation of the basic state pension has been enhanced by discretionary
increases to maintain real value of the basic pension. In a first round of reforms, a strictly means-tested minimum
income guarantee was introduced (1999) and regularly up-rated in line with the growth of average earnings. As
there were fears that the means-testing would discourage private provision, this was changed in 2003 to a system
(Pensions Credit at about 20 percent of average earnings) where income from voluntary private old age provision
was only partially included in the means-test. The same set of reforms introduced changes to the additional state
pension. Under the new rules, the benefit from the state additional pension for those with earnings below a low
earnings threshold of about 42 percent of national average earnings was determined as if they had contributed at
the level of the threshold.
According to OECD calculations 261, these reforms left the individual replacement rate for an average earner at
30.8 percent but improved the net pension benefit for low earners from 29.4 percent to 36.0 percent of national
average earnings.
A second round of major changes to the UK’s pension system, which is in the process of being introduced 262,
contains further elements to protect the poor: the minimum contribution period for the state basic pension will be
reduced to 30 years and the indexation of pensions in payment will return to be with the growth of average
earnings. It is intended that the state basic pension, which today replaces about 15 percent of average earnings,
will return to a level of 19 percent or 20 percent.
B.3 Relative Importance

At present, the resource-tested component (Pension Credit) of old age income in the UK is of major importance
for pensioners. The first round of reforms mentioned above, already reflected in the OECD’s standardized
calculations 263, was intended to change this in the medium-term. Accordingly, the basic state pension will (again)
constitute the main share of public old age provision. Just over half of public transfers to pensioners (50.8 percent)
will be based on this component of the pension system. Just 0.5 percent will be from resource-tested benefits,
while 33.8 percent will relate to the credits apportioned to earnings below the contributions threshold or to the
minimum accrual rates in the earnings-related second tier. The earnings-related component of the public pension
system is of minor relevance as it provides only 15 percent of the total payments.
Legislation enacted in 2007 and further changes already announced are likely to shift these weights yet again.
261
OECD (2007:67).
262
Department for Work and Pensions: Security in Retirement – Towards a New Pension System, London, May 2006.
263
Cf. footnote \h 7 of this annex.
140
Annex
C Switzerland
C.1 Main Features of the Old Age Protection System
Switzerland’s pension system 264 combines a contributions-financed, earnings-related social insurance scheme
with near universal coverage, and mandatory defined benefit occupational pension schemes.
The social insurance scheme has a rather strongly redistributive character as contributions are not capped, but
benefits are at twice the minimum benefit or about 36 percent of earnings. Contributions are 8.4 percent of
pensionable earnings – split between employee and employer. Pensions in payment are indexed in line with the
average of earnings growth and price inflation.
For the occupational schemes, the government sets contribution rates (which vary by age and sex), minimum
rate of return and a minimum annuity rate. However, the occupational schemes are free to apply more generous
rules than legally expected of them, and many do so. Low earners are exempt from participation in the occupational
schemes. Reductions to the minimum rate of return were prescribed by the government when rates of return on
the financial markets plummeted at the beginning of this decade. Nevertheless, the system has been very stable
and the high predictability of its payouts brings it very close to a defined benefit system.
C.2 Protecting the Elderly Poor

Protection for the elderly poor is to a large extent built right into the Swiss social insurance pension system. The
minimum pension at 18 percent of national average earnings compares to the social assistance level found in
other countries. Targeted supplementary benefits are paid to those having to rely exclusively on the minimum
pension. They will lift overall minimum old age income to about 24 percent of national average earnings. The
supplementary benefits are tax-financed.
According to the OECD’s standardized calculations 265, Switzerland’s resource-tested supplementary benefits are
of little importance. This shows the relatively strong redistributive element in the country’s second tier social
insurance pension system which itself constitutes 68.4 percent of transfers to people in old age. The mandatory
occupational defined benefit plans add the remaining 31.5 percent.
D Sweden
D.1 Main Features of the Old Age Protection System
In 1999, Sweden changed its social insurance pension system to notional defined contribution (NDC) rules. In an
NDC system, an individual’s contributions are recorded in an individual account and annually credited with a
notional return. In Sweden that rate of return is calculated on the basis of the growth rate of national average
earnings. As the amounts accumulating in the individual accounts are notional capital that is not backed by
assets, the system continues to operate on a pay-as-you-go basis. Contributions are payable at 16 percent of
pensionable earnings, split evenly between employee and employer. The system has a redistributive component
as contributions are not capped but benefits are, at about 1.3 times average earnings 266.
Sweden also maintains a buffer fund, a funded reserve that is part of the pay-as-you-go notional accounts system.
On retirement the notional capital is converted into an annuity payment. The level of this lifelong pension payment
is dependent on the average remaining life expectancy at the age of retirement. Pensions in payment are increased
annually using a formula linked to the growth of average earnings and to a dampening mechanism based on the
development of the system’s assets (notional and real) and liabilities.
264
Queisser, Monika / Vittas, Dimitri: The Swiss Multi-Pillar Pension System. Triumph or common sense?, World Bank, Development Research Group,
Washington, D.C., August 2000.
265
266
OECD (2007:184).
141
In addition to the social insurance pension, there is the so called premium pension, financed through a mandatory
contribution of 2.5 percent of pensionable earnings to be paid into private defined contribution pension funds,
which in the future are expected to provide about half of the average old age income.
Additional voluntary occupational schemes are also common.
D.2 Protecting the Elderly Poor

Sweden’s mechanism to protect the elderly poor is the so called guarantee pension, a pensions-income-tested
top-up to the pension paid from the notional accounts. Other sources of income are not included in the means-
test. The guarantee pension is set at about one third of national average earnings. In order to protect the incentive
to participate in formal employment and pay pension contributions, the guarantee pension has two withdrawal
rates. For very small pensions up to about 20 percent of average earnings it is 100 percent. Beyond that level the
withdrawal rate reduces to 48 percent. Accordingly, the pension payment is at least lifted to the guaranteed
minimum, but decreasing top-ups apply to NDC pensions as high as almost half of average earnings 267. The
guarantee level is price indexed. The guarantee pension is financed through transfers from the general budget 268.
The minimum guarantee certainly has its role in assuring that, even after the reforms introduced by the NDC
system that had aimed at making old age provision financially sustainable, the individual net replacement rate for
low earners is still 42.8 percent (reduced from 44.7 percent) 269.
According to the OECD’s standardized calculations 270, Sweden’s minimum pensions constitute 4.7 percent of the
country’s wealth transfers to the elderly. The earnings-related public systems provide just under half of all transfers
(49 percent), while the private mandatory or quasi-mandatory defined benefit and defined contribution plans have
shares of 26.4 percent and 19.9 percent respectively.
267
OECD (2007).
268
Palmer, Edward: The Swedisch Pension Reform Model. Framework and Issues, World Bank Social Protection Discussion Paper Series No. 0012,
Washington, D.C., June 2002, p. 2.
269
OECD (2007:67).
270
142
Annex
Annex 5: Overview of Different Systems for Unemployment Compensation
Program Basis Duration Eligibility Financing
Past contributions Employees

required, no-fault and/or
Perc entage of
dismissal, ability employers,
Unemployment Insurance (UI) past wage; Limited
and willingness general
often declining
to work needs to public funds
be proven (tax revenues)
Similar to
Limitation Requirementsof
UI, or by
Total family depends on both UI and SB
Unemployment Assistance (UA) general
income existence of need to be
public funds
SB fulfilled
only
Family
Total family General
Social assistance (SA) Unlimited income below
income public funds
critical threshold
Employees
Percentage of
Individual and/or employers
Individual Saving Accounts (ISA) past wage; Limited
funds available (mandatory
often declining
savings)
Employer in
Lump-sum case of no-fault
payment based dismissal, may
One-time No-fault
Severance pay on years of be by
payment dismissal
employment and mandatory
last wage savings of
employers
Payments to
finance UI;
Layoff taxes Differs from UI
potentially Similar to UI Similar to
(Radical version of only in
based on years or UA severance pay
experience rating) financing method
of employment
and last wage
Source: Based on different sources, see e.g. Vodopivec and Raju (2002), 15f.
143
Annex 6: Months of Severance and Long Service Pay, according the Cause of
Separation, Indonesia 2003
Basic Cause of Separation
Econo- Bank- Retirement Change of Dismissal Voluntary

mic ruptcy Death/ Firm Quit
Cause Illness*
Status
Dis- Quits Min. Maj.

missed Offense Offense
Rates Months of Pay
Serverance Pay
Less than one year service 1 mth. 2 2 2 2 1 1 0 0
Three years service 4 mth. 8 8 8 8 4 4 0 0
Five years service 6 mth. 12 12 12 12 6 6 0 0
Ten years service 9 mth. 18 18 18 18 9 9 0 0
Twenty years service 9 mth. 18 18 18 18 9 9 0 0
Maximum 9 mth. 18 18 18 18 9 9 0 0
Long Service Leave

Less than one year service 0 0 0 0 0 0 0 0 0
Three years service 2 mth. 2 2 2 2 2 2 0 0
Five years service 2 mth. 2 2 2 2 2 2 0 0
Ten years service 4 mth. 4 4 4 4 4 4 0 0
Twenty years service 7 mth. 7 7 7 7 7 7 0 0
Maximum 10 mth. 10 10 10 10 10 10 0 0
Total Severance
Less than one year service 2 2 2 2 1 1 0 0
Three years service 10 10 10 10 6 6 0 0
Five years service 14 14 14 14 8 8 0 0
Ten years service 22 22 22 22 13 13 0 0
Twenty years service 25 25 25 25 16 16 0 0
Maximum 28 28 28 28 19 19 0 0
* In lieu of a company financed pension scheme. Prolonged illness for 12 months or more.
Source: Manning (2004), 27.
144
Annex
Annex 7: Severance Pay as Tax on Monthly Wages
Country Tax in monthly wages Tax as percent of annual wages Rank
Bolivia 4.756 39.6 1

Portugal 4.166 34.7 2
Indonesia (2003) 409 34.1 3
Ecuador 4.035 33.6 4
Turkey 3.973 33.1 5
Peru 3.796 31.6 6
Honduras 3.53 29.4 7
Colombia 3.493 29.1 8
Indonesia (2000 to 2002) 3.420 28.5 (8)
Chile 3.380 28.2 9
Spain 3.156 26.3 10
ElSalvador 3.134 26.1 11
Mexico 3.126 26.1 12
Costa Rica 3.121 26 13
Indonesia (1998 and 1999) 3.041 25.3 (13)
Argentina 2.977 24.8 14
Venezuela 2.955 24.6 15
Dominican Republic 2.814 23.5 16
Panama 2.718 22.7 17
Nicaragua 2.563 21.4 18
Trinidad & Tobago 2.548 21.2 19
Indonesia (average 1996/97) 2.502 20.8 (19)
Uruguay 2.232 18.6 20
Paraguay 2.168 18.1 21
Jamaica 1.920 16.0 22
Guyana 1.890 15.8 23
Greece 1.804 15.0 24
Brazil 1.785 14.9 25
Austria 1.784 14.9 26
Belgium 1.729 14.4 27
United Kingdom 1.451 12.1 28
Switzerland 1.247 10.4 29
Poland 1.219 10.2 30
France 1.143 9.5 31
Germany 1.140 9.5 32
Norway 0.912 7.6 33
Canada 0.553 4.6 34
Australia 0.443 3.7 35
New Zealand 0.221 1.8 36
United States 0 0 37
Simple averageall Countries* 2.4 19.8
* Excludes figures for Indonesia.

Source: Heckmann and Pages (2000) for all Latin American, Caribbean and OECD countries. Hiring tax for these countries in 1999. World Bank
(2006c) for Indonesia.
145
Bibliography
Acemoglu, D.; Shimer, R. [2000]. Productivity Gains from Unemployment Insurance. European Economic Review. 1195-
1224.
Acevedo, G.; Eskenazi, P., and Pages, C. [2006]. Unemployment Insurance in Chile: A New Model of Income Support for
Unemployed Workers. Social Protection Discussion Paper 0612. World Bank. Washington, DC.
Adang, Setiana [2007]. Social Health Insurance Development as an Integral Part of National Policy: Recent Reform in the
Indonesian Health Insurance System. GTZ-ILO-WHO.
ADB [2006a]. Country Strategy and Program. Indonesia 2006-2009. Jakarta.
ADB [2006b]. Indonesia, in: ADB Country Outlook 2007. p. 207-212. Jakarta.
ADB [2006c]. Social Protection Index: Indonesia. Manila
ADB [2007]. Indonesia Public Expenditure Review, Jakarta
Angelini, J., and Hirose, K. [2004]. Extension of Social Security Coverage for the Informal Economy in Indonesia. ILO
Working Paper 11. Manila.
Arifianto, A. [2006]. The New Indonesian Social Security Law: A Blessing or Curse for Indonesians? ASEAN Economic
Bulletin, 23, 57-74.
Arifianto, Alex et al. (The SMERU Research Institute) and Tan, Ellen (World Bank Indonesia) [2005]. Making Services Work
for the Poor in Indonesia: A Report on Health Financing Mechanisms in Kabupaten East Sumba, East Nusa Tenggara. A
Case Study. Report Prepared for the World Bank Indonesia Office.
Bahagijo, S. [2007]. A Few Good People and Local Governments, in: The Jakarta Post, 19th June 2007.
BAPPENAS, ILO [2005]. A National Employment Guarantee Programme for Indonesia: an Approach, Jakarta.
BAPPENAS [2006]. Laporan Akhir Hasil Asesmen Cepat Program PKPS-BBM 2005 Bidang Kesehatan (Final Report on the
Findings of the Quick Assessment of the PKPS-BBM 2005 Health Program). Paper.
Barber et al. [2007]. Differences in Access to High-Quality Outpatient Care in Indonesia. Jakarta.
Barr, N.; Diamond, P. [2006]. The Economics of Pensions. Oxford Review of Economic Policy, 22/2006.
Berger-Schmitt, R. [2000]. Social Cohesion as an Aspect of the Quality of Societies: Concept and Measurement. Centre for
Survey Research and Methodology (ZUMA), EU Reporting Working Paper 14/2000.
Besley, T., and Burgess, R. [2004]. Can Labor Regulation Hinder Economic Performance? Evidence from India. The Quarterly
Journal of Economics, 19, 91-134.
Biewen, M., and Wilke, R. [2005]. Unemployment Duration in West Germany: Do the IAB Employment Subsample and the
German Socio-Economic Panel Yield the Same Results? Journal of the German Statistical Society, 88, 209-236.
Bitrán, Ricardo; Giedion, Ursula [2003]. Waivers and Exemptions for Health Services in Developing Countries. Worldbank –
Social Safety Net Primer Series, Washington (http://wbln0018.worldbank.org/HDNet/hddocs.nsf/
65538a343139acab85256cb70055e6ed/2327dc75151b9f1385256cf0005e323c/$FILE/0308.pdf
Biro Pusat Statistik Indonesia / BPS Statistics Indonesia www.bps.go.id/sector/pubfin
Biro Pusat Statistik (BPS) [2007a]. Population and Type of Activity 2004, 2005, 2006 and 2007. http://www.bps.go.id/sector/
employ/table1.shtml
Biro Pusat Statistik (BPS) [2007b]. Population 15 Years of Age and Over Who Worked by Main Industry 2004, 2005, 2006
and 2007. http://www.bps.go.id/sector/employ/table2.shtml
Biro Pusat Statistik (BPS) [2007c]. Population 15 Years of Age and Over By Main Employment Status 2004, 2005, 2006 and
2007. http://www.bps.go.id/sector/employ/table3.shtml
Biro Pusat Statistik (BPS) [2007d]. Unemployment by Educational Attainment 2004, 2005, 2006 and 2007. http://www.bps.go.id/
sector/employ/table4.shtml
146
Bibliography
Blanchard, O., and Tirole, J. [2006]. The Joint Design of Unemployment Insurance and Employment Protection: A First Pass.
IDEI Working Paper 258. Toulouse.
Blanchard, O., and Wolfers, J. [2000]. The Role of Shocks and Institutions in the Rise of European Unemployment: the
Aggregate Evidence. The Economic Journal, 110, 1-33.
Blunch et al. [2001]. The Informal Sector Revisisted: A Synthesis across Space and Time. The World Bank, Social Protection
Discussion Paper Series, No. 0119, July 2001.
Boone, P.; Fredriksson, P.; Holmlund, B., and Van Ours, J. [2001]. Optimal Unemployment Insurance with Monitoring and
Sanctions. IZR Discussion Paper 401.
BPS, BAPPENAS, UNDP [2004]. Indonesia National Human Development Report 2004. http://www.undp.or.id/pubs/ihdr2004/
ihdr2004_full.pdf
Briggs, A. [1961]. The Welfare State in Historical Perspective. Archives Européennes de Sociologie, II, 2.
Brodjonegoro, Bambang; Department of Economics, University of Indonesia [2004]. Three years of Fiscal Decentralization
in Indonesia: Its Impacts on Regional Economic Development and Fiscal Sustainability. http://www.econ.hit-u.ac.jp/
~kokyo/APPPsympo04/Indonesia(Bambang).pdf
Brotowasisto, Asisto et al. [1988]. Health Care Financing in Indonesia. Health Policy and Planning; 3(2):131-140.
Brown, A.; Orszag, J.M.; Snower, D.J. [2006]. Unemployment Accounts and Employment Incentives. Working Paper 1274.
Institut für Weltwirtschaft, Kiel.
Calmfors, L.; Holmund, B. [2000]. Unemployment and Economic Growth: A Partial Survey. Swedish Economic Policy Review,
7, 109-153.
Card, D.; Levine, P. [2000]. Extended Benefits and the Duration of UI Spells: Evidence from the New Jersey Extended
Benefit Program. Journal of Public Economics, 78, 107-138.
Carunia, M. [2006]. Social Security Yet to Reach Informal Sector, in: The Jakarta Post, 12th May 2006.
Castaño, Ramón Abel; Zambrano, Andrés [2005]. Biased selection within the social health insurance market in Colombia.
Paper presented at CERDI – 2nd International Conference: Health Financing in Developing Countries, 1-2 Dec. 2005,
Clermont-Ferrand.
Centre of Population and Policy Studies. Gadjah Mada University. [2007]. Rapid Assessment Report. Presentation.
Chetty, R., and Looney, A. [2005]. Income Risk and the Benefits of Social Insurance: Evidence from Indonesia and the
United States. NBER Working Paper 11708, Cambridge.
Chetty, Raj; Looney, Adam [2005]. Income Risk and the Benefits of Social Insurance: Evidence from Indonesia and the
United States. Cambridge: Working Paper w11078.
Choon, C.N., and Tsui, A. [2003]. Life Annuities of Compulsory Savings and Income Adequacy of the Elderly in Singapore.
Journal of Pension Economics and Finance, 2, 41-65.
Conerly, W.B. [2002], Chile Leads the Way with Individual Unemployment Accounts. National Center for Policy Analysis,
Brief Analysis No. 424.
Consultative Group for Indonesia (CGI) [2006]. Policy Note on Decentralization for the 15th CGI Meeting in Jakarta. http://
siteresources.worldbank.org/INTINDONESIA/Resources/CGI03/277309-1150258924822/2654372-1150259412106/
2654374-1151377030050/DecentrlPolicyNote.pdf
De Ferranti, D.; Perry, G. E.; Gill, I.S., and Serven, L. [2000]. Securing our Future in a Global Economy. World Bank Latin and
Caribbean Studies, Washington, DC.
Department for Work and Pensions, United Kingdom [2006]. Security in Retirement – towards a New Pension System.
London.
Deutscher Bundestag [2005]. Nationaler Strategiebericht Alterssicherung 2005. Drucksache 15/5571, 27.05.2005. Berlin.
Djaelani, F. [2002]. Development of Health Insurance Products in Indonesia. Paper at the Asia-Pacific Summit on Health
Insurance, 22.-24.5.2002. Jakarta.
147
Döring, D. [1997]. Is the German Welfare State Sustainable? In: Koslowski, P.; Follesdal, A. (eds.): Restructuring the Welfare
State. Theory and Reform of Social Policy. Berlin, Heidelberg, New York, pp. 351.
ECLAC/United Nations [2006]. Shaping the Future of Social Protection: Access, Financing and Solidarity. Montevideo, Uruguay.
Edwards, A., C., and Manning, C. [2001]. The Economics of Employment Protection, and Unemployment Insurance Schemes:
Reflections on Policy Options for Thailand, Malaysia, the Philippines and Indonesia, in: Betcherman, G., and R. Islam
(eds.): East Asian Labor Markets and the Economic Crisis: Impact, Responses and Lessons.
Esguerra, J.; Ogawa, M., and Vodopivec, M. [2001]. Options of Public Income Support for the Unemployed in the Philippines.
The Philippine Review of Economics, 38, 37-66.
Esping-Andersen, G. [1990]. Three Worlds of Welfare Capitalism, Princeton.
Ferrer, A., and Riddel, C.W. [2005]. Unemployment Insurance Saving Accounts in Latin America: Overview and Assessment.
World Bank, Washington, DC.
Fias [2005]. Indonesia – Aceh/Nias: Mini-Diagnostic Analysis of Investment Climate, September 2005.
Firdausy, C. [2006]. Social Security yet to Reach the Informal Sector, in: The Jakarta Post, 4th December 2006.
Frenk, Julio; Knaul, Felicia; González-Pier, Eduardo; Barraza-Lloréns, Mariana [2007]. Poverty, Health and Ssocial Protection.
iIn: GTZ-ILO-WHO 2007, pp. 22-31.
Gallaway, J.H.; Bernasek, A. [2002]. Gender and Informal Sector Employment in Indonesia, in: Journal of Economic Issues,
Vil. 36, No.2, June 2002, pp. 313-320.
Garibaldi, P.; Violante, L. [2005]. The Employment Effects of Severance Payments with Wage Rigidities. Economic Journal,
115, 799.
Geertz, C. [1963]. Peddlers and Princes: Social Development and Economic Change in Two Indonesian Towns. Chicago.
Gill, I.S.; Ilahi, N. [2000]. Economic Insecurity, Individual Behaviour and Social Policy. Paper prepared for the Regional Study
Managing Economic Insecurity in Latin America and the Caribbean. World Bank, Washington, DC.
GTZ, Ministry of Home Affairs [2003]. Decentralization in Indonesia since 1999 – An Overview. http://www.gtzsfdm.or.id/
dec_in_ind.htm
GTZ-SHI, Dinas Kesehatan Propinsi Jawa Tengah [2007]. Pemetaan dan Analisa Pembiayaan dan Pelayanan Kesehatan di
Daerah. Jakarta.
GTZ-SHI, Coordinating Ministry for People’s Welfare [2006a]. Pemetaan dan Analisa Metoda Pembiayaan di Rumah Sakit.
Jakarta.
GTZ-SHI, Coordinating Ministry for People’s Welfare [2006b]. Pemetaan dan Analisa Peraturan Perundangan Terkait Jaminan
Sosial. Jakarta.
GTZ-SHI, Coordinating Ministry for People’s Welfare [2006c]. Pemetaan dan Analisa Sektor Informal. Jakarta,
GTZ-SHI, Coordinating Ministry for People’s Welfare [2006d]. Pemetaan Proses Perumusan UU SJSN dari Tahun 2000 –
2004. Jakarta.
GTZ-SHI, Coordinating Ministry for People’s Welfare. [2006e]. Social Security System Reform in Indonesia. Jakarta.
GTZ-SHI, Ministry of Health [2007]. Pemetaan dan Analisa Peraturan Perundangan di Bidang Kesehatan. Jakarta.
Haltiwanger, J.; Scarpetta, S.; Vodopivec, M. [2003]. How Institutions Affect Labor Market Outcomes: Evidence from Transition
Countries. Paper presented at the World Bank Economist Forum.
Hasoloan, M.A. [2006]. Country Report: The Indonesian Labor Market. Paper presented at the OECD Forum on the Restated
OECD Jobs Strategy.
Heckmann, J.; Pagés, C. [2000]. The Cost of Job Security Regulation: Evidence from Latin American Labor Markets. Working
Paper 7773, Cambridge.
Henley, A. et al. [2006]. On Defining and Measuring the Informal Sector. IZA DP No. 2473.
Hernández, Mario [2002]. Health Reform, Equity, and the Right to Health in Colombia. Cad Saúde Públ 18 (4), 991-1001
http://www.scielo.br/pdf/csp/v18n4/10181.pdf
148
Bibliography
Hölsch, Katja; Kraus, Margit [2003]. Poverty Alleviation and the Degree of Centralisation in European Schemes of Social
Assistance, in: ZEW Discussion Paper No. 03-16.
Holst, Jens; Laaser, Ulrich; Hohmann, Jürgen [2004]. Chilean Health Insurance System: a Source of Inequity and Selective
Social Insecurity. J Publ Health 12, 4, 271-282, Springer-Verlag Heidelberg. http://dx.doi.org/10.1007/s10389-004-0026-5.
Hopenhayn, H.; Hatchando, J.C. [2002]. The Welfare Consequences of Alternative Designs of Unemployment Insurance
Saving Accounts. University of Rochester.
Hotz, V.J.; Mullin, C.H.; Scholz, J.K. [2002]. Welfare, Employment, and Income: Evidence on the Effects on Benefit Reductions
from California. American Economic Review, 92, 380-384.
Hussmans, R. [2004]. Measuring the Informal Economy: From Employment in the Informal Sector to Informal Employment.
Buerau of Statistics, ILO, Policy Integration Department, Working Paper No. 53. Geneva.
IFLS. [1997]. Organization of the Indonesian Health Sector (Appendix C). Survey.
IFC-PENSA [2004]. Review of One Stop Shops in Indonesia. Jakarta.
ILO [date not specified]. Indonesia Working out of Poverty: Social Protection for All, K6 from “Overview of Literature”. Jakarta.
ILO [1999]. Restructuring of the Social Security System in Indonesia. Geneva.
ILO [2003]. Social Security and Coverage for All - Restructuring the Social Security Scheme in Indonesia: Issues and
Options. Jakarta.
ILO, Indonesian Government [2007].: Indonesia: Decent Work Country Programme 2006-2010. Jakarta.
Indonesian Government [1945]. UU Dasar Negara Republik Indonesia Tahun. Jakarta.
Indonesian Government [1979]. UU No. 4/1979 Tentang Kesejahteraan Anak. Jakarta.
Indonesian Government [1981]. Peraturan Pemerintah No. 41/1981 Tentang Pelayanan Sosial Bagi Fakir Miskin. Jakarta.
Indonesian Government [1992]. UU No. 23/1992 Tentang Kesehatan. Jakarta.
Indonesian Government [1999]. UU No. 39/1999 Tentang HAM. Jakarta.
Indonesian Government [2002]. UU No. 23/2002 Tentang Perlindungan Anak. Jakarta.
Indonesian Government [2004a]. UU No. 10/2004 Tentang Pembentukan Peraturan Perundang-Undangan. Jakarta
Indonesian Government [2004b]. UU No. 32/2004 Tentang Pemerintahan Daerah. Jakarta.
Indonesian Government [2004c]. UU Nomor 34/2004 Tentang Tentara Nasional Indonesia. Jakarta.
Indonesian Government [2004d]. UU No. 40/2004 Tentang SJSN. Jakarta.
Indonesian Government [2006]. Himpunan Peraturan Perundang-Undangan RI. Jakarta: PT Ichtiar Baru Van Hoeve.
Indonesia Rapid Decentralization Appraisal (IRDA), Third Report [date not specified]. http://unpan1.un.org/intradoc/
groups/public/documents/APCITY/UNPAN017807.pdf
International Conference on Social Health Insurance in Developing Countries, Berlin 2005. www.shi-conference.de
Ishengoma, E.; Kappel, R. [2006]. Economic Growth and Poverty: Does Formalisation of Informal Enterprises Matter?
GIGA-Working Paper No. 20/2006.
Jordan, Rolf [2005]. Soziale Sicherheit in Indonesien. Anmerkungen zur Armut, Gesundheit und Bildung. Fokus Asien Nr. 25.
Jurajda; Tannery, F.J. [2003]. Unemployment Durations and Extended Unemployment Benefits in Local Labor Markets.
Industrial and Labor Relations Review, 56, 324-348.
Karni, E. [1999]. Optimal Unemployment Insurance: A Guide to the Literature. Social Protection Discussion Paper 9906.
World Bank, Washington, DC.
Kenyon, T. [2007]. A Framework for Thinking about Enterprise Formalization Policies in Developing Countries. World Bank
Policy Research Working Paper No. 4235.
Knaul, Felicia; Frenk, Julio [2005]. Health Insurance in Mexico: Achieving Universal Coverage through Structural Reform.
Health Affairs 24 (6), 1467-1476. http://content.healthaffairs.org/cgi/reprint/24/6/1467
149
Knoess, Johanna [2003]. Introduction of Conditional Cash Transfer Schemes in Indonesia: aBrief Summary, Study
commissioned by the Social Protection Section, GTZ.
Kompas [2007]. Askes belum bayar klaim Rp. 11 miliar ke RS Soedono Madium. 20th July 2007.
Koslowski, P. [1997]. The Social State in the Post-Modern, in: Koslowski, P.; Follesdal, A. (eds.): Restructuring the Welfare
State. Theory and Reform of Social Policy. Berlin, Heidelberg, New York, 351.
Kugler, A. [2005]. Wage-Shifting Effects of Severance Payments Savings Accounts in Colombia. Journal of Public Economics,
89, 487-500.
Kutzin, Joseph [2000]. Goal Framework for Social Health Insurance Development. World Health Organization, Regional
Office for Europe, Copenhagen.
LACF [2007]. The Informal Sector: What Is It, Why Do We Care, and How Do We Measure It?, LACF_021-042, 5/8/07.
Lampert, H.; Althammer, J. [2001]. Lehrbuch der Sozialpolitik. München.
Lanjouw, Peter et al. [2001]. Poverty, Education and Health in Indonesia: Who Benefits from Public Spending? Paper.
Laurell, Asa Cristina [2007]. Health System Reform in Mexico: A Critical Review. Int J Health Serv 37 (3), 515 – 535.
Law No.22/1999 on Regional Governance. http://www.ipd.ph/logolinksea/resources/Law22_99_n_eluc.pdf
Lindenthal, R. [2004]. The Challenge of Social Protection for All: Policy Options for Indonesia. UNSFIR-ILO Working Paper,
Jakarta.
Londoño, José-Luis; Frenk, Julio [1997]. Structured Pluralism: towards an innovative model for health system reform in Latin
America. Health Pol 41 (1), 1-36
Macisaac, D.; Rama, M. [2000]. Mandatory Severance Pay in Peru: An Assessment of its Coverage and Effects Using Panel
Data. World Bank, Washington, DC.
Manning, C. [2003]. Labour Policy and Employment Creation: An Emerging Crisis? Technical Report prepared for Bappenas,
Jakarta.
Manning, C. [2004]. Legislating for Labour Protection: Betting on the Weak or the Strong? Research School of Pacific and
Asian Studies, Working Paper 2004/08.
Markkanen, P.K. [2004]. Occupational Safety and Health in Indonesia. ILO Subregional Office for South-East Asia and the
Pacific, Manila, Philippines, Working Paper No. 9.
Martineau, J-N.; Guérard, Y. [2005]. Implementing an Indonesian National Social Security System (SJSN Law No. 40/2004),
Phase I Report, Jakarta.
Marzolf, James. [2002]. The Indonesian Private Health Sector: Opportunities for Reform. An Analysis of Obstacles and
Constraints to Growth. Discussion Paper. World Bank Indonesia.
McCord, M. et al. [2005]. Microinsurance. Demand and Market Prospects Indonesia, Microinsurance Center.
McCulloch, N. [2007]. Pathways Out of Poverty During an Economic Crisis: An Empirical Assessment of Rural Indonesia.
World Bank Policy Research Working Paper 4173.
McLeod, D. [2003]. Community-Based Social Services: Practical Advice Based upon Lessons from Outside the World Bank.
Social Protection Discussion Paper Series, 0327/2003.
Mehrotra, S.; Biggeri, M. [2002]. Social Protection in the Informal Economy: Home-Based Women Workers and Outsourced
Manufacturing in Asia. UNICEF Innocenti Research Centre, Innocenti Working Paper No. 97.
Mende, Kathrin; Tydecks, Debora [2003]. Basisgesundheitsversorgung in Indonesien: Die Bedeutung der Puskesmas für
die nationale Gesundheitsversorgung. Southeast Asian Studies Working Paper No. 19. University of Bonn.
Meyer, B. D. [1990]. Unemployment Insurance and Unemployment Spells. Econometrica, 58, 757-782.
Ministry of Health [2004]. Keputusan Menteri Kesehatan Nomor 1241/Menkes/SK/XI/2004 Tentang Penugasan PT.
ASKES Dalam Pengelolaan Program Askeskin. Jakarta.
Ministry of Health [2005]. Makalah Diskusi Kelompok Penetapan Paket Manfaat Jaminan Kesehatan. Jakarta.
150
Bibliography
Ministry of Health [2006]. Himpunan Peraturan Perundangan Bidang Kesehatan. Jakarta.

Ministry of Health [2007]. Pedoman Pelaksanaan Jaminan Pemeliharaan Kesehatan Mayarakat Miskin 2005, 2006, 2007.
Jakarta.
O’ Leary, C. [2000]. U.S. Unemployment Insurance: Progress and Prospects. Upjohn Institute for Employment Research.
Organization for Economic Co-operation and Development (OECD) [2007]. Pensions at a Glance. Public Policies across
OECD Countries. Paris.
Orszag, J. M.; P. R. Orszag;, D. J. Snower; Stiglitz, J.E. [1999]. The Impact of Individual Accounts: Piecemeal vs.
Comprehensive Approaches. Paper presented at the Annual Bank Conference on Development Economics, World
Bank, Washington, DC.
Qualman, R.; Meyer-Stamer, J. [2000]. Wirtschafts- und Beschäftigungsförderung in systemischer Perspektive. GTZ, Abteilung 41.
Palmer, Edward [2002]. The Swedish Pension Reform Model: Framework and Issues. World Bank Social Protection Discussion
Paper Series No. 0012, Washington D.C.
Pattinasarany, Daan [2007]. Decentralization and Public Service Delivery. Preliminary Results from the Governance and
Decentralization Survey 2006 (GDS 2).
PT Askes [2006]. Himpunan Peraturan Perundang-Undangan yang Berkaitan dengan PT Askes jilid 1 dan 2. Jakarta.
Queisser, Monika; Vittas, Dimitri [2000]. The Swiss Multi-Pillar Pension System. Triumph or common sense? World Bank
Development Research Group. Washington D.C.
Roed, K.; Zhang, T. [2003]. Does Unemployment Compensation Affect Unemployment Duration? The Economic Journal,
113, 190-206.
Rösch, M. [2007]. The Unemployment Insurance in the USA. http://tiss.zdv.uni-tuebingen.de/webroot/sp/spsba01_W98_1/
usa6d.htm
Rösner, H. J. [2007]. Social Protection in Western Europe. GTZ, Eschborn.
Rösner, H.J. [1995]. Fundamentals of Social Policy Consultation in Development Cooperation, in: W. Lachmann / H.J.
Rösner (eds.): Social Security in Social Market Economy. Conceptual Foundations and Procedural Principles. Konrad-
Adenauer-Foundation, Sankt Augustin, 234-256.
Rösner, H.J. [1995]. Social Market Economy as a Normative Model of Governmental Social Policy, in: W. Lachmann / H.J.
Rösner (eds.): Social Security in Social Market Economy. Conceptual Foundations and Procedural Principles. Konrad-
Adenauer-Foundation, Sankt Augustin, 34-53.
Sáenz-Pacheco, Alberto; Holst, Jens [2007]. Equity and Solidarity in Social Health Insurance; Chances and Risks of the
Costa Rican Way towards Universal Coverage. in: GTZ/ILO/WHO 2007, 5-51.
Satriyo, Hana A. et al. [2003]. Indonesia Rapid Decentralization Appraisal (IRDA). The Asia Foundation. Jakarta.
Scheil-Adlung, Xenia [2004]. Sharpening the Focus on the Poor: Policy Options for Advancing Social Health Protection in
Indonesia. ILO, Extension of Social Security Papers, 19/2004, 9.
Schmidt, J.O. et al. [2005]. Linking up Social Protection Systems in Developing Countries. BMZ/GTZ, Bonn.
Schulenburg, J.-M.; Greiner, W. [2007]. Gesundheitsökonomik, Heidelberg.
Sehnbruch, K. [2006]. Unemployment Insurance or Individual Savings Accounts: Can Chile’s New Scheme Serve as a
Model for other Developing Countries? International Social Security View 59, 27–48.
Sigit, H.; Surbakti, S. [2006]. Chapter 3, Indonesia, in: Asian Development Bank (ed.): Social Protection Index for Committed
Poverty Reduction, 2006, 91.
Simarmata, D. [2006], Tax Reform: Time to Shift from Income to Consumption-based Tax, in: The Jakarta Post. 14th December
2006.
Social Health Insurance Conference, Manila 2006. www.shiconferencemanila.info
Social Insurance and Allied Services. [1942]. Report by Sir William Beveridge. American Edition, New York.
151
Solow, R. [2000]. Unemployment in the United States and in Europe: A Contrast and the Reasons. ifo Studien, 1, 1-12.
Strauss, John et al. [2004]. Indonesian Living Standards Before and After the Financial Crisis. Evidence from the Indonesia
Family Life Survey. RAND Corporation Monograph Series. Singapore.
Sugiyarto, G. [2005]. Creating Better and More Jobs in Indonesia: A Blueprint for Policy Action. ERD Policy Brief No. 43.
Sumarto, S.; Suryahadi, A.; Widyanti, W. [2004]. Assessing the Impact of Indonesian Social Safety Net Programs on Household
Welfare and Poverty Dynamics. SMERU Working Paper.
Suprobo, T.; I. U. Tarigan; D. Weiss [2007]. The Informal Sector in Indonesia and Social Security, Jakarta.
Suryadarma, D.; A. Suryahadi; S. Sumarto [2007]. Reducing Unemployment in Indonesia: Results from a Growth-Employment
Elasticity Model. SMERU Working Paper.
Suryahardi, A. [2001]. Wage and Employment Effects and Minimum Wage Policy in the Indonesian Urban Labor Market. A
Paper from the SMERU Research Institute.
Suryahardi, A. [2001]. Wage and Employment Effects and Minimum Wage Policy in the Indonesian Urban Labor Market.
SMERU Working Paper.
Suwando, Kutut; INFID Annual Lobby [2002]. Decentralization in Indonesia. Jakarta.
Suwandono, Agus et al. [no date specified]. WHO Case Study: Indonesia.
Tambunan, T. H.; Purwoko, B. [2002]. Social Protection in Indonesia, in: Adam, E., M. von Hauff, and J. Marei (eds.), Social
Protection in Southeast Asia & East Asia, 21-73.
Tambunan, T. [2004]. Urban Poverty, Informal Sector and Poverty Alleviation Policies in Indonesia. EGDI and UNU-WIDER
Conference “Unlocking Human Potential: Linking the Informal and Formal Sectors, 17-18 September 2004, Helsinki,
Finland, Parallel Session 4.1.
Teixeira, Luciana; Bugarin, Mauricio; Dourado, Maria Cristina [2006]. Intermunicipal Health Care Consortia in Brazil: Strategic
Behavior, Incentives and Sustainability. Int J Health Plann Mgmt 21 (4). 275–296. http://www3.interscience.wiley.com/
cgi-bin/fulltext/113394617/PDFSTART.
Thabrany, Hasbullah et al. [2003]. Social Health Insurance in Indonesia: Current Status and the Plan for National Health
Insurance. Paper pPresented in WHO SEARO Workshop on Social Health Insurance. New Delhi.
The Jakarta Post [2006]. Govt Wants Severance Pay to be Covered by Insurance. 13th December 2006.
Toomet, O. [2005]. Does an Increase in Unemployment Income Lead to Longer Unemployment Spells? Evidence Using
Danish Unemployment Assistance Data. Aarhus University. Working Paper 2005-07.
UNESCAP [2003]. Country Reports on Local Government Systems: Indonesia. http://www.unescap.org/huset/lgstudy/new-
countrypaper/Indonesia/Indonesia.pdf
UNICEF / BAPPENAS [2004]. Report on MDGs. Jakarta.
Unni, J.; Rani, U. [2002]. Social Protection for Informal Workers: Insecurities, Instruments, and Insittutional Mechanisms.
ILO, September 2002.
Unni, J.; Rani, U. [2003]. Regional Overview of Social Protection of Informal Workers in Asia: Insecurities, Instruments and
Instiutional Arrangements. Economic and Social Commission for Asia and the Pacific, Gender & Development Discussion
Paper Series No. 14, December 2003.
Usman, Syaikhu; SMERU [2007]. Indonesia’s Decentralization Policy: Initial Experiences and Emerging Problems. http://
www.smeru.or.id/report/workpaper/euroseasdecentral/euroseasexperience.pdf
Vaughan, E.; Vaughan, T. [2003], Fundamentals of Risk and Insurance, New York.
Vodopivec, M. [2004a]. Introducing Unemployment Insurance to Developing Countries. World Bank. Washington, DC.
Vodopivec, M. [2004b]. Comparing Unemployment Insurance and Unemployment Assistance. World Bank Employment
Policy Primer, 3. Washington, DC.
Vodopivec, M. [2004c]. Unemployment Insurance: Efficiency Effects and Lessons for Developing Countries. World Bank
Employment Policy Primer, 5. Washington, DC.
152
Bibliography
Vodopivec, M. [2005a]. Choosing the System of Unemployment Income Support. World Bank, Washington, DC.
Vodopivec, M.; Raju, D. [2002]. Income Support Systems for the Unemployed: Issues and Options. Social Protection Discussion
Paper 0214, World Bank, Washington, DC.
Volbracht, Renate [2005]. Bringt Dezentralisierung Autonomie in die Regionen? http://home.snafu.de/watchin/II_1_05/
Dezentralisierung_Autonomie.htm
Vroman, W.; Brusentsev, V. [2005]. Unemployment Compensation throughout the World: a Comparative Analysis. W.E.
Upjohn Institute for Employment Research, Michigan.
Wang, C.; Williamson, S. [1996]. Unemployment Insurance with Moral Hazard in a Dynamic Economy. Carnegie-Rochester
Conference Series on Public Policy, 44, 1-41.
Weber, A. (ed.) [2006]. Social Protection Index for Committed Poverty Reduction. Asian Development Bank (ADB), Manila.
WHO [2002]. Trade in Health Services: Global, Regional and Country Perspectives. Washington DC: Pan American Health
Organization, Program on Public Policy and Health, Division of Health and Human Development.
WHO [2004]. Paper for the Executive Board on Social Health Insurance.
Widiatmoko, Dono; Gani, Ascobat [no date specified]. International Relations with Indonesia’s Hospital Sector. Report Prepared
for the World Health Organization.
Wiebe, F. [1996]. Income Insecurity and Underemployment in Indonesia´s Informal Sector. World Bank, East Asia Pacific
Country Department III, Policy Research Working Paper No. 1639.
World Bank [2000a]. East Asia. Recovery and Beyond, Washington DC.
World Bank [2000b]. Health Strategy in a Post-Crisis, Decentralizing Indonesia. Report No. 21318.
World Bank [2005a]. Decentralization, Service Delivery and Governance in Indonesia. Findings from the Governance and
Decentralization Survey (GDS) 1+/2004. 2005.
World Bank [2005b]. Indonesia: Ideas for the Future: Creating Jobs, in: Indonesia Policy Briefs, January 2005.
World Bank [2006a]. Indonesia Public Expenditure Review 2007. Jakarta.
World Bank [2006b]. Making the New Indonesia Work for the Poor. Jakarta.
World Bank [2006c]. The Impact of Formal and Informal Labor Regulations on Business in Serang District. Jakarta.
World Bank [2006d]. Revitalizing the Rural Economy: An Assessment of the Investment Climate Faced by Non-Farm
Enterprises at the District Level. Consultative Draft, 38777 v.2.
World Bank [2007a]. Indonesia: Economic and Social Update. Jakarta.
World Bank [2007b]. Indonesia Public Expenditure Review 2007. Jakarta.
World Bank [2007c]. Social Assistance in Central Europe and the Baltic States. Washington, D.C.
World Bank’s Decentralization Thematic Team: What is Decentralization?
http://web.worldbank.org/WBSITE/EXTERNAL/TOPICS/EXTPUBLICSECTORANDGOVERNANCE/EXTDSRE/
0,,contentMDK:20246045~menuPK:509216~pagePK:210058~piPK:210062~theSitePK:390243,00.html” http://
web.worldbank.org/WBSITE/EXTERNAL/TOPICS/EXTPUBLICSECTORANDGOVERNANCE/EXTDSRE/
0,,contentMDK:20246045~menuPK:509216~pagePK:210058~piPK:210062~theSitePK:390243,00.html
Yun, J. [2001]. On the Integration of Unemployment Insurance with Pension through Individual Saving Accounts. World
Bank. Washington, DC.
153
GTZ Jakarta
Deutsche Bank Building 20th Floor
Jalan Imam Bonjol No. 80
Jakarta Pusat 10310 Indonesia
Tel. : +62 21 319 24007
Fax. : +62 21 319 24070
Email : gtz-indonesia@gtz.de
Web : www.gtz.de/indonesia

Option For Social Protection Reform in Indoensia - 2008

Uploaded by

Document Information

Original Title

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

Option For Social Protection Reform in Indoensia - 2008

Uploaded by

Copyright:

Available Formats

Options for Social Protection Reform

Deutsche Gesellschaft für Technische Zusammenarbeit (GTZ) GmbH

GTZ Office Jakarta

Design & Production

All Rights Reserved/Printed in Indonesia

Abbreviations and Acronyms

KUD Rural Cooperatives (Koperasi Unit Desa)

3. The Social State Model: a Model for Indonesia? 13

4. Social Insurance: Reforming Contributory Social Security 23

5.1.4. Individual Saving Accounts (ISA) 54

6. Social Assistance: Reforming Non-Contributory Social Protection 73

7. Extending Social Security to the Informal Sector 85

Figure 1 Administrative Divisions in Indonesia 12

• Analyzing institutional options for administering a comprehensive system of social protection

The report is divided into eight sections:

A summary of conclusions and open issues is presented in chapter nine.

Table 1: Number and Percentage of Population below the Poverty Line

Area / Year Poverty Line (Rp / Capita / Month) Number of Percentage of

Food Non Food Total

This led to the introduction of new local taxes, which in turn

In order to solve the above-mentioned problems and

The government has also begun to draft a ”Grand

Figure 1: Administrative Divisions in Indonesia

Province (Propinsi), 33,

Village (Desa and Kelurahan)

3.1. Models of Social Protection

Table 2: Main Principles of the Bismarck and Beveridge Model

Bismarck Model Beveridge Model

Objectives Status-oriented Subsistence-oriented

Benefits Income-related National minimum

3.1.1. The Bismarck Model

3.1.2. The Beveridge Model

3.1.3. Comparison of Bismarck and Beveridge

3.2. Guiding Principles of the Social State Model

3.3. The Social State Model within the Indonesian context

3.3.1. Present Situation of Social Protection Policy

3.3.2. The Vision: Indonesia - a Social State with Welfare Elements

Figure 2: Core Elements of the Social State Model

1. The type and degree of uncertainty caused by the risk

2. The relative size of the loss and

3. Whether the risk is idiosyncratic or covariate

Table 3: Characterizing Risks

Illness Old Age Unemployment

Characteristics of insured risk

However, two aspects need to be stressed:

4.1.1. Organizational Principles of Social Insurance

4.1.2. Financing Social Insurance

* Who is paying contributions: Employee, employer, or shared between both.

4.2. The Administration of Social Security

4.2.1. The Current Situation

4.2.2. Reform Options

Table 4: Administrative Options

Integration Risk pooling (size and composition Social insurance

Separation Economies of scope Social assistance

Harmonization Avoiding political conflicts ‘1st step’ only

4.3. The National Social Health Insurance Program: Work in Progress

Figure 3: Timeline of Health Sector Reform in Indonesia 72

2004 2007 2009

4.3.2. The Current Status of the System

7. The benefit packages offered by health insurance programs vary greatly.

State-provided health insurance program for

Civil Military and War veterans

Employers / 2% (gradual) National govern-