See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/269599946

Evaluating Jakarta's Flood Defence Governance: The Impact of Political and

Institutional Reforms

Article  in  Water Policy · June 2012

DOI: 10.2166/wp.2012.119

CITATIONS READS

20 6,660

4 authors, including:

Niki Frantzeskaki B. Enserink

Utrecht University Delft University of Technology
193 PUBLICATIONS   10,360 CITATIONS    78 PUBLICATIONS   1,393 CITATIONS   

SEE PROFILE SEE PROFILE

Wim Ravesteijn
Delft University of Technology
74 PUBLICATIONS   522 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Connecting Nature: Co-producing nature-based solutions to achieve environmental, social and economic resilience in cities View project

Connecting Nature View project

All content following this page was uploaded by Niki Frantzeskaki on 18 February 2016.

The user has requested enhancement of the downloaded file.

 Provided for non-commercial research and educational use only.
Not for reproduction or distribution or commercial use.

This article was originally published by IWA Publishing. IWA Publishing recognizes
the retention of the right by the author(s) to photocopy or make single electronic
copies of the paper for their own personal use, including for their own classroom use,
or the personal use of colleagues, provided the copies are not offered for sale and
are not distributed in a systematic way outside of their employing institution.

Please note that you are not permitted to post the IWA Publishing PDF version of
your paper on your own website or your institution’s website or repository.

Please direct any queries regarding use or permissions to waterpolicy@iwap.co.uk

 Water Policy 14 (2012) 561–580

Evaluating Jakarta’s flood defence governance: the impact

of political and institutional reforms
Imelda Simanjuntaka, Niki Frantzeskakib, Bert Enserinka
and Wim Ravesteijnc
a
Delft University of Technology, Faculty of Technology, Policy and Management, Policy Analysis Section, Jaffalaan 5,
2628BX, Delft, The Netherlands
b
Corresponding author. Erasmus University Rotterdam, Faculty of Social Sciences, Dutch Research Institute for Transitions,
Burgemeester Oudlaan 50, 3000DR, Rotterdam, The Netherlands. E-mail: frantzeskaki@fsw.eur.nl
c
Delft University of Technology, Faculty of Technology, Policy and Management, Technology Dynamics Section, Delft,
The Netherlands

Abstract

Flood defence management in Jakarta is a critical governmental activity, since Jakarta is a low-lying delta
metropolis. Its protection against flooding is crucial for the continuing of economic activities. Despite the
urgent need for action, the implementation of flood defences like the Eastern Flood Canal has been a lengthy
and strenuous process. The first plans to construct the Canal date back to the early 1970s and gained support
early during the policy design process but the actual construction only started in 2003 after institutional and pol-
itical changes. Earlier evaluations point to a lack of financial resources as a cause of delayed implementation. We
have explored the causes beyond budgetary reasons, and have used Kingdon’s streams model to structure the
policy design and implementation process, and to analyze the co-evolution of politics (the political stream), pol-
icies (the solutions stream) and social-ecological context (the problem stream). We have paid due attention to the
effects of decentralization and the realized political reforms. Our research has revealed three causes for the
implementation delay: (a) the disconnection of political and institutional developments for problems and solutions
before 1998; (b) institutional misalignment between municipal and ministerial authorities; and (c) an absence of
critical actors in the policy process despite their control of financial and administrative resources. Implementation
was only feasible when political and institutional developments coincided.

Keywords: Decentralization; Infrastructures; Jakarta; Kingdon’s streams model; Policy implementation;

Politics; Water management

doi: 10.2166/wp.2012.119

© IWA Publishing 2012

562 I. Simanjuntak et al. / Water Policy 14 (2012) 561–580

1. Introduction

Jakarta is a city located in a low-lying delta area and is annually exposed to flooding. Of the
65,000 ha area of Jakarta, about half is situated in flood-prone areas. According to current data from
the Jakarta Public Works Agency, about 78 out of 92 sub-districts in Jakarta are prone to flooding
(Kementerian Pekerjaan Umum, 2011). Flood risk remains a threat to Jakarta and the magnitude of
floods has been growing over the last decade (e.g. flood events in 1996, 2002, 2007 and 2010). In
an interview with the Jakarta Globe in early 2010, Sutopo Purwo Nugroho, head of disaster-flood miti-
gation at the Agency for the Assessment and Application of Technology (BPPT), stated: ‘Jakarta will
never be completely free from floods’ (Primanita, 2010). Flooding impacts widespread areas, inducing
huge economic and social damage for the city (NEDECO, 2002; Bappenas, 2007). According to earlier
studies, one factor contributing to the worsening of floods in Jakarta is inadequate flood defence infra-
structures (Steinberg, 2007). Although flood defence infrastructure programs have been on the
government agenda since 1970, most of these policies have never been implemented due to political
and public resistance.
One of the flood control infrastructure programs for which implementation has been delayed for more
than 30 years was the Eastern Flood Canal (EFC). The EFC project plan was first presented in 1973, in
the Master Plan of Drainage System and Flood Control for Jakarta. The 1973 Master Plan was con-
structed by the Ministry of Public Works (MPW) in collaboration with the Dutch government. Even
though the EFC design was finished in 1973 and its construction was planned to be completed
before 1980 (NEDECO, 1973), implementation of the project was only begun in 2003. Infact, work
has not been fully completed yet, and even if it had, would only have been able to reduce flooding
in East Jakarta by about 30%, according to Sutopo Purwo Nugroho in the same Jakarta Globe interview
(Primanita, 2010).
There are many contradictory opinions regarding the reasons for the delay of the EFC implemen-
tation. In the public media, the Indonesian government claimed that the land acquisition process was
holding up implementation. Research documents charge Jakarta’s authorities with having had
a significant role in the delay (Steinberg, 2007). According to Caljouw & Nas Pratiwo (2004),
budget allocations which were not used to their purpose was the main factor contributing to non-
implementation of the EFC in the past. Steinberg (2007) claimed that the delay was due to the difficulty
of attracting investors, resulting in an inadequate budget for its implementation.
Our research has revealed that reasons for the delay to the EFC lay beyond the budget limitation. In
fact, many Indonesian infrastructure projects which required high budgets did get implemented in the
past. Given that, this paper intends to answer the following question: what were the decisive causes
of the implementation delays and the opposition faced by the Jakarta city government regarding the
EFC?
The causes of delay and opposition have been sought in two areas: (i) the institutional structure
in Indonesia, and (ii) the way in which the decision-making process of the EFC was organized.
We pay attention to the transition from centralization to decentralization, which led to a new
distribution of responsibilities and thus created new opportunities. We claim that the institutional
structure, its changes and the impacts of these changes on the policy-making process explain the
issues causing the delay and resistance, as well as the eventual start of EFC implementation
after 2003.
 I. Simanjuntak et al. / Water Policy 14 (2012) 561–580 563

2. Research methodology

We conducted desk research and in-person interviews for data collection. The desk research focused
on the history of the EFC and the institutional development in Indonesia regarding water management
(see Appendix A, Table A.1, available online at http://www.iwaponline.com/wp/014/119.pdf). Eighteen
semi-structured in-person interviews were conducted in Jakarta, Indonesia in April–May 2010. The
interviewees were selected based on their role and position during the implementation process of the
EFC and included government officials, non-governmental organization (NGO) spokespersons and citi-
zens who lived near the canal (see Appendix A, Table A.2, available online at http://www.iwaponline.
com/wp/014/119.pdf). The interviews with citizens focused on the causes (and motivation) of their
opposition; officials and other stakeholders were questioned to understand the formal and informal insti-
tutional structure and the interactions at district, provincial and national levels.

2.1. The streams model

For structuring and analyzing the data, we applied Kingdon’s (1985) streams model. The streams
model offers the opportunity to take into account politics along with problems and policies, whereas
other models of policy processes tend to neutralize politics, such as the rounds model (Teisman,
2000) and the garbage can model (Cohen et al., 1972). The streams model: (a) has an explicit focus
on the politics of the policy process; (b) captures institutional arrangements; (c) suits the description
of lengthy and complex policy- and decision-making processes; and (d) assumes that external events
create windows of opportunity, which seemed to play an important role too in the policy process con-
cerning the EFC. In effect, the streams model turns the traditional rational model of policy making
upside-down and gives scope to problems looking for solutions, as well as to solutions looking for suit-
able problems (Enserink et al., 2010).

2.2. Streams and their confluence

Kingdon (1985) identified three streams in policy making: a problem stream, a solutions stream and the
stream of political events; their confluence is the precondition for getting an issue solved. The problem
stream contains the issues that need to be solved – in our case the recurring floods in Jakarta. The policy
stream contains (a multitude of) deliberate answers to tackle problems; in our case, amongst others, the var-
ious plans for an EFC. The political stream is a combination of national mood, pressure groups and turnover
in office. For the latter, however, we prefer to use Zahariadis’ (2003) redefinition of Kingdon’s political
stream in terms of governing party ideology, which fits better with the Indonesian democratic and insti-
tutional landscape. According to Kingdon (1985: 84), issues get on the agenda when ‘a problem is
recognized, a solution is available, and the political climate makes the time right for change’. The confluence
of these three streams at critical junctures, or ‘policy windows’ which are open only for a short time when
condition are right, is the precondition for getting a matter on the agenda (Mucciaroni, 1992).
Kingdon describes two more categories of windows: problem windows and political windows.
Problem windows stem from problems that arise exogenously to the political stream and demand
a policy response; they are typically more or less rational responses to prior problems. They
564 I. Simanjuntak et al. / Water Policy 14 (2012) 561–580

occur when the problem stream and policy stream meet but no connection to the political stream is
present; if the latter remains absent, the opportunity for problem solving passes by. Political win-
dows, by contrast, are associated with inherently preferred solutions chasing problems which can
serve as rationales for a policy (Zahariadis, 2003). The latter mostly occur due to political
events: changes of government, shifts in the ‘national mood’, and the rise and fall of political for-
tunes (Lipson, 2007).
Open policy windows do not, though, automatically bring about policy change. There must be
deliberate efforts to seize the opportunity and push the problem onto the agenda before the favorable
conditions change. This is done by so-called ‘policy entrepreneurs’. Policy entrepreneurs play a criti-
cal role in ‘coupling’ problems and solutions during such windows of opportunity (Mucciaroni,
1992). Policy entrepreneurs are typically the actors who recognize problem or political windows
and work to open and expand them. They do so by linking solutions to problems and by working
to get the resulting policy packages accepted by decision makers (Huitema & Meijerink, 2010).
According to Kingdon, policy entrepreneurs often have a good reputation, good networking skills
and are willing to invest resources. As we will show, the opportunities for policy entrepreneurs in
flood management policy making in Suharto’s regime in Indonesia (1965–1998) were limited.
Only after the regime changed could policy entrepreneurs effectively exploit newly occurring
policy windows.

3. Institutional context of water management in Jakarta

The territory of Indonesia is divided into autonomous provinces, districts (kabupaten) and
municipalities (kota), with districts and municipalities being at the same level of government
though set apart by the one being rural (districts) and the other being urban (municipalities).
Within districts and municipalities there are sub-districts (kecamatan) which are smaller administra-
tive government units. Each sub-district is further divided into villages. Villages in rural areas
are called desa, while in urban areas they are referred to as kelurahan. Figure 1 shows how the
different levels are organized and how the power and responsibilities of the central government
have been transferred to lower levels of government by co-administration, decentralization and
deconcentration.
Decentralization was enforced by Law 22, which was passed by Parliament in 1999. Beginning in
2000, central government transferred powers and responsibilities to the regional (provincial and district)
level in almost all sectors, including water and land policy. After decentralization, the bupati (district
head) and walikota (municipal head), as the head of the autonomous local government, became directly
responsible to the local assembly, the Dewan Perwakilan Rakyat Daerah (DPRD). Starting in 2000, the
national sectoral agencies were abolished and the civil servants of these agencies were placed in regional
governments (Usman, 2001). The transition from centralization to decentralization has continued since
then.

3.1. Jakarta’s Eastern Flood Canal

Jakarta is one of Indonesia’s provinces and is headed by a governor. It is made up of five

municipalities (Jakarta Utara, Jakarta Timur, Jakarta Barat, Jakarta Selatan and Jakarta Pusat) and
 I. Simanjuntak et al. / Water Policy 14 (2012) 561–580 565

Fig. 1. Government levels in Indonesia.

each of them is headed by a mayor. The EFC is located in the Jakarta Region, with the Java Sea to its
north, Tangerang District (Banten Province) to the west, Bekasi District (West Java Province) to its east
side, and with Bogor District (West Java Province) to its south. According to government decree ‘DKI
Jakarta1 No. 3504/2003’, the construction of the EFC trajectory starts from the Cipinang River and ends
in the Java Sea. The land required for the canal development is 405.2 ha, of which 147.9 ha are in North
Jakarta and 257.3 ha in East Jakarta (Wawan, 2007).
At different points, the EFC cuts through three sub-districts in the East Jakarta district (Jatinegara,
Duren Sawit and Cakun), and one sub-district (Cilincing) in the North Jakarta district. In total, 11
villages border the canal lying in East Jakarta, whilst two more, Rorotan and Marunda, lie in the
North Jakarta district. More details of the sub-districts and villages can be seen in Table 1.
Over the last decade, the eastern and northern area of 15,400 ha of DKI Jakarta has developed into
industrial and warehousing areas and settlements, which have complicated construction plans (see
Table 2). The areas in Eastern Jakarta that had to be acquired for the construction of the canal were
70% dominated by settlement areas. Northern Jakarta, which is downstream, was dominated by agricul-
tural activities with a smaller proportion of settlements.

1
The Special Capital Territory of Jakarta (Indonesian: Daerah Khusus Ibu Kota Jakarta).
566 I. Simanjuntak et al. / Water Policy 14 (2012) 561–580

Table 1. Description of areas of Eastern Flood Canal.

No. Villages Length (m) District
1 Cipinang Besar Selatan 770 East Jakarta
2 Cipinang Muara 758 East Jakarta
3 Pondok Bambu 2,072 East Jakarta
4 Duren Sawit 1,705 East Jakarta
5 Pondok Kelapa 193 East Jakarta
6 Malaka Jaya 433 East Jakarta
7 Malaka Sari 717 East Jakarta
8 Pondok Kopi 1,816 East Jakarta
9 Pulo Gebang 3,137 East Jakarta
10 Ujung Menteng 2,884 East Jakarta
11 Cakung Timur 2,019 East Jakarta
12 Rorotan 3,055 North Jakarta
13 Marunda 3,615 North Jakarta
Source: Bappeda, 2003.

Table 2. Percentage of land use by village in the area of Eastern Flood Canal in the year 2001. (Data taken from Bappeda, 2003.)
Percentage of land use by village in the area of the Eastern Flood Canal in
2001 (Penggunaan Lahan) (%)
Village Settlement Industry Other
Cipinang Besar Selatan 64.99 2.50 32.51
Cipinang Muara 88.44 0.35 11.21
Pondok Bambu 87.21 0.00 12.79
Duren Sawit 77.56 0.00 22.44
Pondok Kelapa 67.81 1.20 30.99
Malaka Sari 94.20 0.00 5.80
Malaka Jaya 71.96 0.00 28.04
Pondok Kopi 79.17 0.10 20.73
Pulo Gebang 84.00 3.00 13.00
Cakung Timur 64.19 11.50 24.31
Ujung Menteng 20.47 18.79 60.74
Rorotan 41.45 4.69 53.86
Marunda 21.05 10.03 68.92
Total 66.35 4.01 29.64

4. Historical context

The policy processes of the EFC cannot be separated from the preceding development of Master Plans
for the Flood Control and Drainage System in Jakarta (see Table 1). In this section we sketch the his-
torical development of the EFC.
The idea to construct the EFC was first launched by the Dutch engineer Van der Beer in 1918. His
idea was to control the water volume discharged on Jakarta through 13 rivers by diverting the runoff of
these rivers through canals at the east and west side of the city towards the sea.
 I. Simanjuntak et al. / Water Policy 14 (2012) 561–580 567

After Independence Day (17 August 1945), flood management in Jakarta received attention by the cen-
tral government due to Jakarta’s strategic importance as an economic and governance center. In 1965,
central government established the Commando project of Flood Control (‘Kopro Banjir’), which focused
on infrastructure development and was fully funded and implemented by central government.

4.1. The first master plan

Given Jakarta’s vulnerability to flooding, flood prevention was prioritized in the initial stage of devel-
opment of the flood Control Master Plan. In the 1970s, Van der Beer’s original idea was adopted by
NEDECO, a Dutch engineering consultant which collaborated with Indonesia’s government, and the
idea was presented in the new ‘Master Plan of Flood Control and Drainage System for Jakarta’ (Kop,
2008: 374–375). In this first Master Plan, the EFC was recommended to protect the flood-prone area in
the eastern part of Jakarta. But, even though the EFC’s design was finished in 1973, it was not directly
adopted by the Indonesian government or, as Kop (2008: 375) notes, it was ‘eternally postponed’. We
found that a lack of political commitment to allocate resources (financial and administrative) resulted in
the non-direct adoption and non-implementation of Master Plan I. Given the political conditions during
Suharto’s regime, the government did not perceive the construction of the EFC as urgent. The fact that
the EFC would provide flood protection to 20% of the areas of Jakarta (WJEMP, 2002), where only
low-income households were settled (NGO interviewee), made it hard to gain political support.

4.2. Thirty years of standstill

After the development of this first flood control Master Plan, many more studies were executed (until
2003) to formulate long-term solutions for Jakarta’s flood control (see Table 3). These studies were con-
ducted by central government in collaboration with several donor communities (especially from The
Netherlands and Japan). These studies produced several refinements of previous master plans (Master
Plan II (1991), III (1997) and IV (2002)). In every new master plan, the EFC was again present as a
solution to control flooding in Eastern Jakarta. During the development of subsequent master plans,
the design of the EFC went through several revisions and refinements.
As presented in Table 3, the policy process of the EFC lasted more than 30 years before a decision taken
was actually implemented. Two distinct phases can be distinguished in the macro-political context of the
EFC policy process: first, the EFC Master Plans were constructed during Suharto’s regime, when govern-
ment practice in Indonesia was highly centralized and autocratic and the public infrastructure development
funds relied on external loans; second, at some stage in the process of the EFC design, the transition from
centralization to decentralization of administration systems in Indonesia took place, in around 1999.
Figure 2 illustrates that the agreement to implement the EFC and start construction was reached only
after the institutional shift. In the next section we elaborate on the impacts of this change.

5. Institutional change

During the development of Master Plans I, II and III, the government system in Indonesia was cen-
tralized, and water resources management was the responsibility of central government (see Table 1). As
described in Sections 2 and 4 above, the administrative arrangements related to the water sector and
568 I. Simanjuntak et al. / Water Policy 14 (2012) 561–580

Table 3. Master plan development of Eastern Flood Canal, Jakarta, Indonesia.

Year National level events East canal events Description
1965 Establishment of Jakarta flood Plan and construction of flood
prevention project by the central infrastructure under the
government (‘Kopro Banjir’) authority of the Ministry of
Public Works (MPW)
1969–1998 General Suharto’s New Order Water resources and other
Government in power governmental functions
consolidated to central
government
1973 Master Plan I (Master Plan of Flood Master Plan I produced by the
Control and Drainage System for Department of Public Works
Jakarta) (NEDECO, 1973) and Electricity (central) in
collaboration with the
Netherlands Engineering
consultant, NEDECO
1974 Water Law No. 11 According to this law, the flood
control infrastructure project
was fully funded and
implemented by the central
government (MPW)
1991–1993 Master Plan II The collaboration between the
Department of Public Works
and the Japan International
Consultant produced a new
master plan (revision of
previous master plan)
1996 Severe flood inundated almost New study about Jakarta flood was The ‘Jakarta Flood Control
all Jakarta’s land area, conducted by central government Advisory Mission (JFCAM)
especially in North and East in collaboration with The by NEDECO’ (1996)
Jakarta Netherlands government produced a new study about
floods in Jakarta and an
alternative design of the
Eastern Flood Canal
1997 Master Plan III The collaboration between
Department of Public Works
and Japan International
Consultant produced a new
master plan
1998 Fall of Suharto’s regime The transformation from
centralization to
decentralization is initiated
1999 Decentralization Law 22/1999 Eastern Flood Canal plan is Change of institutional
and Fiscal Equalization Law integrated with the spatial arrangement in water sectors.
25/1999 planning of Jakarta 2010 Provincial government has
autonomous power in
regulation and policy related
to water and flood
management
(continued)
 I. Simanjuntak et al. / Water Policy 14 (2012) 561–580 569

Table 3. (continued)

Year National level events East canal events Description

2002 Severe flood hits Jakarta. It was
recorded that disastrous floods
affected 10,000 ha of Jakarta’s
built-up area
2003
2004 New Water Law No. 7. First
direct election
2007 Another devastating flood hits
Jakarta; one of the worst
floods ever experienced
a
MoU, Memorandum of Understanding.
Master Plan IV
Kick-off of flood canal construction
by President Megawati Soekarno
Putri
More stakeholders involved
in developing the Master Plan
IV
MoUa between central
government and provincial
government to implement
Eastern Flood Canal signed
Central government increased its
commitment to accelerate the
completion of the Eastern
Flood Canal by allocating
more budget than before

Fig. 2. Macro context of the Eastern Flood Canal policy process.

flood risk management can be characterized as top–down, leaving the actors at the bottom level (respon-
sible for implementation) with limited discretion and flexibility. This section describes the institutional
and political changes during and after 1998.

5.1. Central government in control

Water control and water management were regulated based on Law No. 11/1974. According to this
law, the power to implement policies for flood control management resided with the central government
and, more specifically, with the MPW (which had the authority and responsibility to coordinate macro
570 I. Simanjuntak et al. / Water Policy 14 (2012) 561–580

planning, technical planning, supervision and the implementation power related to water resource man-
agement). Related to flood control infrastructure development, the MPW had supervisory control of
implementing institutions appointed by them. Consequently, all flood control management activities
were centrally controlled and had to be approved by the MPW (see Figure 3).
To implement the legislation of Law No. 11/l974, Regulation No. 22/1982 was issued eight years
later. According to this regulation, the Water Resources Development Plan was to be provided by
the MPW and, more specifically, by the Deputy of Natural Resources and Environment of the National
Development Planning Agency (Bappenas). Moreover, the MPW had the authority to appoint and estab-
lish organizations at the national or regional level to perform certain tasks related to water management,
including flood control infrastructure development and management. The regional authorities (provin-
cial and lower level) were responsible only for performing specific administrative tasks delegated by the
central government.

5.2. Decentralization: provinces in control

After the fall of Suharto’s regime in May 1998, the structure of the relationship between central gov-
ernment and the provinces or lower level administration changed from being typified by centralization
and co-administration to a more decentralized governmental system, through the establishment of Law

Fig. 3. Administrative bodies of the water sector after decentralization. (Bappenas: Deputy of the Natural Resources and
Environment of the National Development Planning Agency at national level; Bappeda: Deputy of Natural Resources and
Environment at provincial level.)
 I. Simanjuntak et al. / Water Policy 14 (2012) 561–580 571

No. 22/1999. Regional governments now had the authority to enact their own regulations, formulate
their own plans, programs and fiscal policies, raise and retain revenue and exploit natural resources
including in the water sector (Bhat & Mollinga, 2009).
These new autonomy and decentralization laws put pressure on the provinces in many aspects of
flood control and management. The institutional changes implied that the provinces were now ultimately
responsible for flood management and for addressing flood damage in Jakarta, where before responsi-
bility had been handled by central government (see Figure 3). Additionally, river basins lying inside a
province and not nationally strategic were to be managed by the province concerned (Bhaat et al., 2005).
Related to flood control developments, the Public Work Agency had authority to give permission and
approval for the development of public infrastructure services, including to the flood control infrastruc-
ture in Jakarta (Capacity Building of Drainage Management of Jakarta (WJEMP, 2002)).
The provinces also gained responsibility for the operation and management of flood protection infra-
structure and for the Flood Forecasting and Warning System (FFWS). Provinces had the responsibility
to conduct activities of all relevant agencies providing information concerning water-related and flood
risk issues. Agreements were made between the provinces and their districts regarding how to manage
floodwaters. In summary, decentralization after 1999 gave power to the provinces to direct flood defence
policy and become responsible for dealing with flood issues.

6. Explaining the delay

The above description of the institutional and political changes in Indonesia indicates that it was
events in the political stream that were crucial for the implementation of the Jakarta flood infrastructure,
and not the lack of infrastructure solutions and policy. In this section, we discuss three decisive causes
for the implementation delay of the EFC: (i) institutional misalignment between the regional and
national levels; (ii) the absence of critical actors in the policy process; and (iii) the disconnect between
political and institutional developments, further evinced by the fact that actual implementation was feas-
ible only when political and institutional developments coincided. Hence, we will differentiate between
two time periods, pre- and post-1998, and put our focus on the latter post-Suharto period.

6.1. Institutional misalignment

During the period of centralization, power and responsibility for planning and policy making relating
to flood control infrastructure resided with central government (Bhat et al., 2005). Thus, implementation
of the master plans relied on the commitment of central government to provide budgetary and legislative
support (such as laws, regulations and procedures) to the organizations responsible for the
implementation.
In practice, provinces were responsible for the implementation of the master plans but the critical
resources needed to realize their implementation were not adequately provided for. In cases where
resources were provided, they were insufficient to initiate the planned infrastructure projects. According
to our interviewees, the master plans lacked any legislative support (i.e. there was an absence of clear
procedures, budget schemes and regulations) to realize them. For instance, whilst the cost of all works
set up in the 1973 Master Plan amounted to Rp592 million, the annual budget for the project in the mid-
1970s was only approximately Rp1.5 million (Soenarno & Sasongko, 2000).
572 I. Simanjuntak et al. / Water Policy 14 (2012) 561–580

6.2. No commitment

We also observed that the local province itself had a low commitment to push and promote the
implementation of the EFC. When attending meetings, officials from the province were present only for
formal reasons and did not reveal any new facts about the flooding problems or Jakarta’s spatial planning
to central government. Consequently, the content of the master plan studies did not change and consisted
mainly of technical design revisions which neglected the changing land use (Master Plans 1973, 1993,
1997). This lack of commitment from the province may be related to the institutional structure of this
period since, prior to 2002, provincial governments saw themselves as the representatives of central gov-
ernment and not as the representatives of their citizens. Assigned administrators were easily replaceable,
making them more interested in sustaining their position than in providing better public services. In this
period the institutional structure of the political stream prevented any opening up of policy windows.

6.3. After decentralization

As we showed in Figure 2, after decentralization, the EFC construction was finally kicked off by cen-
tral government in 2003. The transition from centralization to decentralization brought about relevant
changes in the governance style of the public administration, including in the water sector, that is, dele-
gating power and responsibilities to the regional level. Moreover, the new freedom of the press, and the
mobilization of public opinion after new severe flood events influenced the decision-making process of
the EFC.

6.4. Lack of transparency

The decision to construct the canal was taken early in the transitional period by the Jakarta Province
in consultation with the national government, and carried out with limited public participation or invol-
vement of NGOs and district government. The policy process was dominated by government actors and,
consequently, the EFC Master Plan III was only accessible to central and provincial authorities without
being presented to the public (based on our field research data; see Appendix A, Table A.2, available
online at http://www.iwaponline.com/wp/014/119.pdf). Transparency and openness are considered good
governance principles belonging to practices in the decentralization style, but they were not applied in
the transition period when the decision to construct the canal was made. When the decision to
implement the EFC was published, opposition from citizens rose, given that people were allowed to
voice their concerns and to defend their interests (after decentralization). People living near to or on
the areas designated for the canals tried to block the government’s decision by holding on to their
lands, and this then blocked the construction of the canals.

6.5. Lack of communication

The lack of communication and coordination between the province and district governments was
another cause for implementation delay. It led to a vague plan about the construction and to a failure
of land acquisition, and land availability was a critical factor to accelerate the completion of the
canal. Before 2003, the province was responsible for providing land but land availability was not
well prepared or managed by the province, as other priorities existed. Even today, there are designated
 I. Simanjuntak et al. / Water Policy 14 (2012) 561–580 573

construction areas that still remain to be freed from dwellings so as to give space for flood defence
measures. The slow land acquisition and physical construction activities of the EFC are executed in
parallel.
This land acquisition problem arose because the decision to implement EFC did not involve the dis-
tricts, even though the district authorities were fully acquainted with the current condition of the EFC
areas. Instead, rather than involving the district authorities, the land acquisition process was attempted
from top–down. In 2003, the provincial Governor established a decree about the procurement of land for
canal construction (in the same year as the construction of the canal was kicked-off by central govern-
ment), a decree which had to be followed by the districts. Two problems arose here: first, a lack of
coordination between province and districts as to the status of land supply for the EFC, and second,
the Governor’s decision did not fit with the concept of district autonomy (under decentralization). In
fact, questions of land use had been left entirely to district autonomy, meaning that the governor has
no authority to override the land management plans of a district.
The land acquired to construct the EFC spans across 11 villages in the Eastern Jakarta district and two
villages in the Northern Jakarta district, and thus the cooperation of both these districts was very impor-
tant. During our interviews, we found that this raised a new problem. The designated lands had been
utilized mostly as informal settlement areas (70%) and for business, agriculture and public and social
facilities. The question of land acquisition became a very complex problem due to the lack of spatial
planning undertaken by DKI Jakarta. In the 1997 Master Plan, in which the EFC was first planned
to be constructed, it was mentioned and emphasized that land-use planning needed to be conducted
and enforced so that land use would not hinder implementation in the future. This land use plan was
not followed by the province. Real problems emerged after 1999 when land affairs were decentralized
and decisions on land-use planning were given to the district government. By the time the EFC was
being constructed, the land which should have been allocated to the canal had changed into a settlement
area. It was found that, even after construction work had started, some of the designated lands were still
under development to become residential real estate.
During the policy-making process of the EFC, the role of the province was not settled and limited.
Since the province was clearly knowledgeable about Jakarta’s flooding problem, it was expected that
during its attendance in decision-consultation meetings it could have exercised pressure to take actions
regarding flood issues in Eastern Jakarta. Instead, we found that the province had very limited influence
on the decision-making process, even during the decentralized period. For example, the Jakarta Province
initiated the establishment of the Governor’s decree in order to implement the EFC but this action could
not be followed up due to the MPW’s instruction to disrupt its operation (according to one interviewee);
this is an anomaly of the decentralization era. After decentralization, it was expected that provinces and
municipalities would enjoy the authority to take initiatives and decisions. This situation shows that
decentralization did not improve institutional performance in decision-making nor in implementation.

6.6. Involving critical actors

Several actors who held important resources were not involved throughout the policy design and
implementation process. Based on our findings, we can list several of these so-called ‘critical’ actors
(Enserink et al., 2010), who should have been involved because of their direct involvement in
implementation procedures, their control of resources, and their interests in the success of the project.
These critical actors were the Spatial Planning Agency, the Ministry of Finance, and the municipalities.
574 I. Simanjuntak et al. / Water Policy 14 (2012) 561–580

According to De Bruijn & ten Heuvelhof (2008), critical actors who are not involved during a
decision-making process may try to use their resources or power to block the decision. They may try
to redefine the decision, postpone implementation or implement the decision in a different way from
the one intended. Moreover, they may also adopt a passive attitude and not give much attention and
support during implementation. Such typical behavior was also observed (and revealed) during the
implementation phase of the EFC.
The absence of land owners in the problem framing and planning phases caused tough opposition in
the implementation phase. Since the landowners knew that the EFC could not be implemented without
the availability of land, they used this condition to secure their interest by requesting very high compen-
sation from the government. As mentioned earlier, another illustration is the absence in the planning
process of the district authorities, despite their knowledge of the actual land use. Consequently, land
that should have been designated to the construction of the EFC became settlements and industrial areas.

7. When political and institutional developments meet

If we take a look at Table 3, we see that the consensus on construction of the EFC was reached in the
aftermath of the 2002 flood. In this section, we intend to specify the mechanisms by which the flood in
2002 became a unique event that brought about government attention and an immediate response. The
decision to implement the EFC in 2003 is related to the changing contextual conditions during this
period. Drawing on Kingdon’s model, we conclude that, in the preceding years, a political window
opened as consensus was gained after the institutional and political changes in Indonesia and the pol-
itical stream coincided with the policy stream. The flood event was the catalyst that opened a policy
window where problems, solutions and politics could meet.
Using Kingdon’s streams model to analyze the EFC policy process, the evolution of the EFC can be
explained in terms of the meeting of policy windows, problem windows and politics streams. Figure 4
shows three streams as identified by Kingdon related to the EFC implementation process. The flood
events and severe floods that hit Jakarta can be positioned within the problem stream that opened a pro-
blem window. The master plans (to respond to the flood problems) constitute the solution stream. The
decentralization, democratization and the associated freedom of press, as well as the societal movement
of freedom of expression (after the fall of Suharto’s regime) give a new dynamic to the political stream.
The provinces and the media can be characterized as policy entrepreneurs who tried to seize opportu-
nities to merge the three streams. This merging resulted in the recognition of the urgency of EFC
implementation by central government. Each stream depicted in Figure 4 and the role of the policy
entrepreneurs will be further discussed in the following paragraphs.

7.1. Problem stream

The flood problem in Jakarta has been present for many years. Severe flood events in 1970, 1979, 1996
and 2002 paralyzed the city for weeks. These floods claimed numerous victims and caused huge economic
losses, accounting for an estimated US$1,000 million for the 2007 flood alone (Yamin, 2009; Tanuwidjaja
et al., 2010). The annual floods kept the problem stream alive, while the severe flood events functioned as
focusing events to open up problem windows whenever new policies and master plans were initiated. After
the political changes, growing public concern and public debate forced policy makers to put the flood
 I. Simanjuntak et al. / Water Policy 14 (2012) 561–580 575

Fig. 4. The streams model of the Eastern Flood Canal policy process, Jakarta, Indonesia.

problems on the agenda, which led to new issues emerging in the political stream. The 2002 flood was a
critical event opening up a policy window for the EFC as problem, solution and political streams met.

7.2. Policy stream

From the early 1970s, a series of master plans was developed but never adopted by the Indonesian
government. The policy stream never met the political stream and, due to the absence of these links
until 2002, the EFC proposals were turned down and thus not implemented by central government.
Only after the regime change in 1998 did a decentralization process start leading to changes in the insti-
tutional setting and a new distribution of responsibilities to district and regional authorities. The regular
flood events can be conceptualized as the problem window that allowed the EFC to appear on Jakarta’s
flood management agenda.

7.3. Political stream

The institutional shift from centralization to decentralization after the fall of Suharto was character-
ized by shared power between district, regional and national governments. The change supported the
better functioning of the democratic system in Indonesia, which is now characterized by direct free elec-
tions, freedom of press and an increasing power of civil society. In a way similar to that considered by
576 I. Simanjuntak et al. / Water Policy 14 (2012) 561–580

Zahariadis (2003), all these political and institutional changes can be conceptualized as expressions of
the changes in political ideology and determining elements in the political stream. The new division of
power and responsibilities between government bodies opened the political window within which the
problem of uncontrolled flooding was linked to the EFC alternative by the province. The changes in
the role and position of the provinces enabled Jakarta’s Province to push central government to give
(financial) support to implement the EFC.

7.4. Freedom of the press

Freedom of the press was important when linking the problem and solution streams. The discussion
of the causes and solutions for Jakarta’s flood problems between intergovernmental bodies, academics
and professionals spilled over into the media. For the first time, during the flood event in 2002, various
media (e.g. television channels, newspapers) published the reality of the flood impacts and criticized the
government’s failure in providing an adequate flood infrastructure in Jakarta.

7.5. Public interest

The increase of society’s power due to direct elections can be conceptualized as an important element
belonging to the political stream. This is demonstrated by the fact that the agreement between central
and provincial government was announced in 2002, and EFC construction was kicked off in 2003 –
close to the first direct elections in 2004. We contend that the prospect of these direct elections may
well have increased the government’s commitment towards servicing the public’s interests. Politicians
wanted and needed to show to the public that the current government has both the willingness and com-
mitment to speed up the implementation of flood control infrastructure to ensure Jakarta’s protection
against flooding in the future. This political determination was a precondition for a successful political
campaign for the future election period.

7.6. Policy entrepreneurs

The joining of issues within a policy window is accomplished by policy entrepreneurs – advocates of
particular policies who seek to get them on the agenda and accepted by decision makers (Huitema &
Meijerink, 2010). Kingdon claims that disparate actors can play the role of policy entrepreneurs, at
different times and in different decision settings. This theoretical proposition is corroborated by our find-
ings from the EFC policy process in the period from 1973 to 2002.
We have identified three policy entrepreneurship initiatives promoting the EFC as a (partial) solution to
Jakarta’s flooding problem. As described above, two of the master plan studies (in 1973 and 1996) were
produced by Indonesia’s government in collaboration with the Dutch government, and the first policy
entrepreneurs and source of initiatives were the Dutch consultants engaged. These Dutch advisers pro-
moted the EFC in the respective master plans as the best alternative to safeguard Jakarta from flooding.
However, they were not very successful, as the political conditions (with a highly centralized autocratic
regime) were not favorable; they acted in the problem window but failed to engage the decision-makers.
The second policy entrepreneur identified is the Jakarta district authority, which exploited the public
discontent after the 2002 flood, by demanding the immediate implementation of the EFC. The Jakarta
district authority became an important policy entrepreneur in the revival of the EFC. After
 I. Simanjuntak et al. / Water Policy 14 (2012) 561–580 577

decentralization, district authorities were held accountable for flood protection. Consequently, the issue
moved higher on their agenda and they started applying pressure to change the way that the provinces
and the national government managed water.
In 2002, after decades of lacking political support, the central government began to shift towards a
supportive posture and started to advocate the EFC as an alternative for Jakarta’s flood defence. This
shift came after the negotiations between central and provincial government, which were initiated after
the latest flood events and led to the establishment of the Memorandum of Understanding (MoU) on
EFC implementation. The provinces should be considered the third policy entrepreneur in this game,
opening the window for funding and implementation of the EFC by extorting a MoU from the govern-
ment. The changing attitude and position of the provinces is an expression of their new focus on
serving their constituents rather than obeying the national government.
The media could be considered to be the fourth policy entrepreneur, as they created a sense of
urgency that contributed to opening up a policy window. Even though the media did not directly pro-
mote the EFC, media pressure in broadcasting the reality of the impact of floods and information about
the failing flood management in Jakarta helped to put pressure on elected officials at all government
levels and connected the 2002 flood event to the EFC alternative.

8. Discussion and conclusion

In our introduction we stated that many explanations and contradictory opinions exist regarding the
causes for the delay of the implementation of Jakarta’s flood defence policies. Suggestions were made
that insufficient funding, lack of investors, bad governance and the difficult land acquisition process
were factors holding up implementation. By reconstructing the policy-making process for the EFC,
while applying Kingdon’s streams model, we have tried to answer the question given the decisive
causes of the implementation delays and the opposition experienced by the Jakarta city government
to flood defence policy making. In this section, we discuss some of the findings.
This case study shows that the delays in implementing the EFC cannot be contributed to a lack of
policy making; between 1973 and 2003 when implementation started, several studies were carried
out to formulate long-term solutions for Jakarta’s flood defences. The causes of delay and opposition
were mainly located in the political sphere, especially in the institutional structure of Indonesia
before decentralization in 1999. Two distinct phases can be distinguished in the macro context of the
EFC policy process, with a transitional period in between. The first phase occurred during Suharto’s
regime when several master plans for flood protection were developed. In this period, government prac-
tice in Indonesia was highly centralized and autocratic, and public infrastructure development funds
relied on external loans. The transition from centralization to decentralization started in 1999. Decentra-
lization implied that the provinces and districts gained power to direct policies related to flood defence
management, and the provinces became responsible for dealing with flood issues and for policy
implementation.
We found three decisive causes for the delays in implementation of the EFC:

i) The disconnection of political and institutional developments, and of the problems and solutions
streams before 1998;
578 I. Simanjuntak et al. / Water Policy 14 (2012) 561–580

ii) An institutional misalignment between district and national levels during the transition period (1997–
2002) and after decentralization; and
iii) The absence of critical actors in the policy process.

The institutional misalignment was most prominent when discussing the role of the provinces in
policy making during the Suharto regime and in the transition period. At first, their role was marginal
as they were not consulted and had no authority whatsoever. Implementation of the EFC was not con-
sidered urgent because most of the decisions reflected central government’s interest, and not district or
regional public interests. During the transition period, these old institutional arrangements were slowly
dissolving.
Implementation was possible only after changes in the political climate and with institutional decen-
tralization. The rearrangement of the political stream allowed district and regional authorities to take
responsibility for flood protection measures and to act as policy entrepreneurs, connecting politics
with problems and policies. Only after 2002 were provincial and Jakarta district authorities able to
push their agenda. At the same time, communication with other stakeholders was still inadequate.
The non-involvement and absence of critical actors such as the district authorities, landowners and
city dwellers led to disinformation about actual land use patterns and the resistance of landowners,
which then caused serious delays after the decision to start construction had been taken. As we have
stated above, even actors who were holding important resources were not involved in the policy
design process.
The 2002 flood events were the catalyst creating the urgency to open a policy window, and policy
entrepreneurs were supported by the new freedom of the press which allowed for criticism and
public discussion of problems, possible remedies and their consequences. The 2002 flood and the
first glimmer of the upcoming 2004 general election were opportunities seized by the policy entrepre-
neurs to push decisions onto the agenda. For these reasons, we conclude that our findings add to earlier
studies on infrastructure governance in Jakarta, bringing forward additional political and institutional
reasons to explain the lengthy decision-making and implementation processes.
Kingdon’s streams model has been used to show how the EFC finally got attention from national
policy makers and was considered to be a solution by the Indonesian government. The decision to
finally implement the EFC emerged through the active intervention of policy entrepreneurs by joining
(i) the flood events (problem stream), (ii) the master plans that proposed the EFC as an alternative to
cope with the Jakarta’s flood problem (solution stream), and (iii) the decentralization, supported by
the freedom of the press and the direct election after the fall of Suharto’s regime (political stream).
The provinces, Jakarta’s districts and the media can be characterized as successful policy entrepreneurs
which grabbed the opportunities and joined these three streams, so as to push central government to see
the urgency of EFC implementation and to provide guarantees for funding.
This case study suggests that Kingdon’s model is a powerful tool to describe actual decision-making
processes. We have been able to highlight the important role of policy entrepreneurs and to show that
policy entrepreneurs can only be successful when the institutional and political order allows them to
operate. The Dutch advisers who operated in the early centralized period clearly did not have sufficient
leeway to move their problem window to a policy window; the case study shows that a policy entrepre-
neur should seize opportunities. In the EFC case, an extreme event in the problem stream was the
catalyst for further mobilizing policy entrepreneurs to take action and push the decision forward.
 I. Simanjuntak et al. / Water Policy 14 (2012) 561–580 579

Acknowledgement

The authors would like to thank Dr Chris Tapscott for his constructive remarks and discussion of a
previous version of the current paper at the IRSPM XV 2011 Conference held in Dublin, Ireland.

References
Badan Perencanaan Daerah Propinsi Daerah Khusus Ibu Kota Jakarta (Bappeda) (2003). Penyusunan Rencana Kawasan Banjir
Kanal Timur: Laporan Final (Preparation of the East Flood Canal Zone Plan: Final Report). Lembaga Penelitian Dan Pem-
berdayaan Masyarakat (LIPI), Bandung Institute of Technology, Bandung, Indonesia.
Bappenas (2007). Direktorat Pengairan dan Irigasis – Kedeputian Sarana dan Prasana Kementerian Perencanaan Pemban-
gunan. Nasional/Badan Perencanaan Pemabangunan Nasional Penanggulan Banjir Jakarta (Directorate of Irrigation, Deputy
facilities of Infrastructures). National Planning Agency, Jakarta, Indonesia.
Bhat, A. & Mollinga, P. P. (2009). Transitions in Indonesian water policy: policy windows through crisis, response through
implementation. In: Water Policy Entrepreneurs: A Research Companion to Water Transitions Around the Globe.
Huitema, D. & Meijerink, S. (eds). Edward Elgar Publishing, Cheltenham, UK, pp. 158–174.
Bhat, A., Ramu, K. & Kemper, K. (2005). Institutional and policy analysis of river basin management: the Brantas river basin,
East Java, Indonesia. Policy Research Working Paper Series, No. WPS 3611. World Bank, Washington, DC.
de Bruijn, J. A. & ten Heuvelhof, E. F. (2008). Management in Networks: On Multi-actor Decision Making. Routledge,
New York & Canada.
Caljouw, M. & Nas Pratiwo, P. J. M. (2004). Flooding in Jakarta: towards a blue city with improved water management, Paper
presented at the 1st International Conference on Urban History, Surabaya, Indonesia, 23–25 Aug 2004.
Capacity Building of Drainage Management of Jakarta (WJEMP) (2002). Drainage Management for Jakarta: Strategic Action
Program Development (DKI 3-9). Jakarta, Indonesia.
Cohen, M. D., March, J. G. & Olsen, J. P. (1972). A garbage can model of organizational choice. Administrative Science Quar-
terly 17(1), 1–25.
Enserink, B., Hermans, L., Kwakkel, J., Thissen, W., Koppenjan, J. & Bots, P. (2010). Policy Analysis of Multi-actor Systems.
Lemma, Amsterdam.
Huitema, D. & Meijerink, S. (2010). Realizing water transitions: the role of policy entrepreneurs in water policy change. Ecol-
ogy and Society 15(2), p. 26. [online] http://www.ecologyandsociety.org/vol15/iss2/art26/, accessed 9 May 2012.
Kementerian Pekerjaan Umum (2011). See: www.pu.go.id, accessed 29 February 2012.
Kingdon, J. W. (1985). Agendas, Alternatives and Public Policies. Harper-Collins, New York.
Kop, J. (2008). Hydraulic engineering, sanitary engineering and bridge building in new Indonesia. In: For Profit and Prosper-
ity: The Contribution Made by Dutch Engineers to Public works in Indonesia (1800–2000). Ravesteijn, W. & Kop, J. (eds).
KITLV Press, Leiden, The Netherlands, pp. 346–379.
Lipson, M. (2007). A garbage can model of UN peacekeeping. Global Governance 13(1), 79–97.
Mucciaroni, G. (1992). The garbage can model and the study of policy making: a critique. Polity 24, 459–482.
NEDECO (1973). Master Plan for Flood Control and Drainage System of Jakarta. Government of Jakarta, Jakarta, Indonesia.
NEDECO (1996). Jakarta Flood Control Advisory Mission (JFCAM). Government of Jakarta, Jakarta, Indonesia.
NEDECO (2002). Final Report Quick Reconnaissance Studies Flood Jabodetabek 2002. Government of Jakarta, Jakarta, Indonesia.
Primanita, A. (2010). Jakarta will never be free of flooding, expert warns. Jakarta Globe, 19 February 2010. http://www.
thejakartaglobe.com/city/jakarta-will-never-be-free-of-flooding-expert-warns/359533, accessed 9 May 2012.
Soenarno & Sasongk, D. (2000). Participatory Planning and Management for Flood Mitigation and Preparedness in the City
of Jakarta. (ENRDD/WMRS/W/164(Phase II)). Djoko, Jakarta, Indonesia.
Steinberg, F. (2007). Jakarta: Environmental Problems and Sustainability. Asian Development Bank (ADB) Manilla, Philippines.
Tanuwidjaja, G., Widjaya, J. M. & Tallar, R. Y. (2010). Creative collaboration in urban polder in Jakarta, in the framework of
integrated water management. Green Impact Indonesia. See: http://greenimpactindo.wordpress.com/2010/03/28/creative-
collaboration-in-urban-polder-in-jakarta-in-the-framework-of-integrated-water-management/, accessed 7 December 2011.
 580 I. Simanjuntak et al. / Water Policy 14 (2012) 561–580

Teisman, G. R. (2000). Models for research into decision-making processes: on phases, streams and decision-making rounds.
Public Administration 78(4), 937–956.
Usman, S. (2001). Indonesia’s Decentralization Policy: Initial experiences and emerging problems. SMERU Research Institute
Working Paper. London, September 2001.
Wawan, F. (2007). Enterprise Resource Planning. Menyelaraskan Teknologi Informasi dengan Strategi Bisnis, Informatika,
Bandung, Indonesia.
WJEMP (2002). Drainage Management for Jakarta, Strategic Action Program Development (DKI 3–9), Jakarta, Indonesia.
Yamin, K. (2009). Will Jakarta ever see an end to its perennial flooding problem? Jakarta Globe. See: http://www.thejakartaglobe.
com/news/will-jakarta-ever-see-an-end-to-its-perennial-flooding-problem/309739#Scene_1, accessed 7 December 2011.
Zahariadis, N. (2003). Ambiguity and Choice in Public Policy: Political Decision Making in Modern Democracies. George-
town University Press, Washington, DC.

Received 11 July 2011; accepted in revised form 29 December 2011. Available online 5 April 2012

View publication stats