Fighting corruption in the World Bank’s
Kecamatan Development Program
Andrea Woodhouse
June 2002
The author is a World Bank consultant working on a World Bank-funded community
development project in Indonesia, the Kecamatan Development Program (KDP). The views
expressed in the paper are the author’s own and do not necessarily reflect the views of the
World Bank or KDP.
1 A previous, unpublished version of this paper appeared under the title The Dynamics of Rural Power in Indonesia: Fighting Corruption in a
World Bank Community Development Program
Table of Contents
Executive Summary
What works to limit corruption in a large, rural development project in a country with endemic corruption,
a weak legal system and a history of top-down political control by a powerful state bureaucracy?
The Kecamatan Development Program (KDP) is a $370 million World Bank community-driven
development project in Indonesia that funds infrastructure and small loans in over 20,000 villages
nationwide.2 Its approach to combating corruption is based on an analysis of the political economy of
corruption in Indonesian villages and is two-pronged. First, it aims to change the conditions that breed
corruption in villages by breaking existing monopolies over information, resources, and access to justice.
Second, it aims to prevent corruption in the project itself by skewing the incentives of the project structure
against corrupt behavior.
At the heart of KDP’s anti-corruption approach is the principle that villagers themselves have decisionmaking
power over planning, procurement and management of funds. Some of the concrete measures of
its approach include:
· simplifying financial formats so that they can be understood easily by villagers
· transferring funds directly into collective village bank accounts
· insisting that all financial transactions have at least three signatures and that at least three quotations are
found for the procurement of goods, to be shared publicly at village meetings
· insisting that details of all financial transactions are posted on village notice-boards
· requiring that regular village meetings are held to account for project funds at which villagers have the
right to suspend further disbursements of funds if irregularities are found
· providing village-level sources of information and channels for complaints that are independent of
local government
· intensive field-level supervision by elected village facilitators (two in each village) and sub-district level
project facilitators
· independent monitoring of the project by NGOs and local journalists
Although these measures have had some success, corruption in KDP persists. This paper examines where,
why and how this takes place. Its findings are based on a series of in-depth, ethnographic field interviews, a
review of the KDP field experience, an incentives analysis of the KDP project cycle and a survey of the
existing literature on village governance in Indonesia. The paper aims to get a sense of the anatomy of
corruption in KDP villages: of how the actors perceive their interests, what motivates them, what kinds of
constraints they face, and what kinds of steps they take to resolve their problems. The underlying aim is to
assess what kinds of anti-corruption measures are likely to succeed in a local-level project that exists in a
systemically corrupt environment and whose size and breadth (20,000 villages nationwide) makes
centralized control and monitoring of funds impossible. The paper use corruption also as a lens through
which to view snapshots of social and political change in Indonesian villages.
The paper argues that corruption is primarily a problem of incentives, and can be fought effectively only by
changing the costs and benefits attached to corrupt behavior. It claims also that local context and social
norms are key to understanding how these incentives can be changed in order to reduce corruption.
2 The follow-on project (2002-2006), KDP 2, is a $420 million project. Combined with KDP 1, this makes KDP the largest World-Bank poverty
reduction/community development project in Asia ($790 million).
The first part of the paper examines the conditions that enable corruption to flourish in Indonesian villages.
Corruption in Indonesian villages is encouraged by a combination of factors. High levels of bureaucracy
and red tape multiply the loci of rent-seeking in villages. A history of impunity for corruptors, combined
with a legacy of oppression for whistleblowers, means that it is almost never in the interest of any individual
villager to protest corruption. The Indonesian state’s administrative structure concentrates power in village
elites which, combined with a weak and corrupt judicial system, impedes poor people’s access to justice and
control over decision-making. Finally, the Suharto government’s conscious strategy of de-politicizing
villages, combined with the use of development funds as a means of patronage and control, has created an
environment where corruption is rarely resisted.
The second part of the paper examines corruption in KDP, based on a review of the KDP field experience
and an incentives analysis of the project cycle. Corruption in KDP takes several forms, including budget
mark-ups, collusion, bribes and kick-backs to local officials. The analysis reveals that the elements of the
project most effective in limiting corruption are transparency, community participation, and the provision
of independent channels for resolving complaints. Information and local control are key elements in both
preventing and fighting corruption: the most successful strategies for fighting corruption in KDP have
hinged on publicizing anti-corruption activities, garnering wide local support, and using sanctions credibly.
Project facilitators are also key to fighting corruption: they provide a channel of information to villagers that
is independent of local government and, because they are backed by the central KDP structure, they have
more protection from threats and intimidation than ordinary villagers. There is evidence also of some
governance spillovers from KDP, illustrated by examples of villagers using their experience of KDP as a
precedent for protesting against corruption in other projects.
The incentives analysis of the project cycle identifies the stages of the project cycle that are most vulnerable
to corruption. These lie at the stages of proposal preparation (formation of false borrower groups for small
loans); release of funds (collusion among bank account signatories to embezzle funds); and implementation
(collusion and corruption in procurement). The analysis highlights several ways in which corruption in
KDP could be better prevented. These include improving information dissemination; working with social
sanctions to make the incentive structure less conducive to corruption; increasing incentives for KDP staff
to fight corruption; instituting measures at specific stages of the project cycle intended to limit monopoly,
clarify discretion, and improve accountability; and supporting the capacity of project facilitators to come up
with flexible local solutions to their problems.
This paper is an evaluation of the kinds of anti-corruption project design measures that are likely to succeed
in a large, decentralized community project in an environment that is itself systemically corrupt and an
environment whose history and state structures have embedded the conditions for corruption. It makes
explicit the varying successes and failures of a project that attempts to promote village-level governance
reforms in an uncertain and diverse political environment where national reforms have not yet solidified.
In doing so, it illustrates the impacts and dynamics of political transformation, local power structures and
social change in rural communities in Indonesia.
Several people gave me help and advice at all stages of the report. It is the result of their efforts as much as
First of all, I would like to thank Scott Guggenheim, whose help and guidance provide the foundation for
the entire paper. Second, I would like to thank Bob Klitgaard (RAND), who has continually gone far
beyond the call of duty as a peer reviewer and has advised me not only on the paper itself but also on how
best to make it useful. Third, I would like to thank Sarwar Lateef, who has given considerable ongoing
support to the research and its wider aims.
The KDP team in Jakarta helped me at all stages. I would like to give special thanks to Ela Hasanah
(KDP’s project historian and a member of the team that originally designed KDP), who collaborated with
me at every stage of the research. I am grateful to Pieter Evers, who contributed an analysis of the rural
judiciary in Indonesia to this paper and whose work on village governance informed many of my initial
ideas. Ibnu Taufan, Susan Wong , Susanto Simanjuntak, Victor Bottini and staff from KDP’s “Handling
Complaints Unit” helped identify problems, arrange fieldwork and assess field findings. Milissa Day gave
me invaluable advice on fieldwork and writing.
Several people helped me with fieldwork. These include the villagers who participated in the research; my
field assistant, Liebe Ibet; and local KDP field staff: Martius, Subeyang, Idrus, John Masengi, Sentot Satria
and several kecamatan facilitators. The facilitators deserve special mention for helping me despite their
already-overstretched workload. They are the front line for anti-corruption activities in KDP and deserve
credit for their efforts to fight corruption despite the risk of threats and intimidation involved in doing so.
I am indebted to many people who read the draft and suggested improvements. These include my peer
reviewers, Susan Rose-Ackerman (Yale), Robert Klitgaard (RAND), Michael Richards (World Bank) and
Paul McCarthy (World Bank). They also include Surendra Agarwal, Vivi Alatas, Judith Edstrom, Ben
Fisher, Cyprian Fisiy, Richard Holloway (UNDP), Kai Kaiser, Menno Pradhan, Naseer Rana, Unggul
Suprayitno and Stefanie Teggemann. I would like to thank Stephen Woodhouse (Unicef) for sharing his
insights on social development in Indonesia, and would like to thank David Madden and Claire Smith for
their comments and ongoing support throughout the analysis.
I am grateful for having had the opportunity to present and discuss my findings with classes of students at
the John F. Kennedy School of Government at Harvard University and the Center for Agrarian Studies at
Yale University. I would like to thank the students themselves for their insightful comments, and would
like to thank Lant Pritchett (Harvard) and Michael Dove (Yale) for inviting Scott Guggenheim and me to
their classes. I am also grateful to Peter Eigen (Transparency International), who gave me the same
opportunity as part of his anti-corruption course at the School of Advanced International Study (SAIS) at
Johns Hopkins University.
A consensus is emerging in a large number of countries that effective poverty reduction requires a strong
governance framework: accountable institutions, participatory political systems, transparency and the rule
of law. Fighting corruption is an essential part of this. Corruption hampers growth; it diverts public
services from the poor; and it discourages foreign investment in developing countries.3
The World Bank’s anti-corruption strategy calls for action in four key areas:
· Preventing and reducing corruption in World Bank projects
· Assisting countries that ask for help in reducing corruption
· Incorporating concern for corruption into country analysis and lending
· Contributing to the international effort to fight corruption
This report is about the first area of action: preventing and reducing corruption in World Bank
projects. A first step towards this is an examination of different types of existing projects to see how
corruption takes place in them, which design features work to allow it or prevent it, and what lessons can
be drawn from their experiences so that future projects limit corruption instead of contributing to it.
This report supports that aim. It consists of an analysis of where, why and how corruption occurs in a large
decentralized community development project in Indonesia, the Kecamatan Development Program (KDP).4
KDP is one of the first World Bank projects to make anti-corruption measures an explicit element of its
design. Reviewing the analytical framework and samples of the project’s field experience gives us useful
material for improving future programs that are similar in design.
The key assumption of the report is that corruption is primarily a problem of incentives, and can only be
fought effectively by changing the incentive structures attached. The paper is divided broadly into three
· An examination of the context of community development in Indonesia
· An analysis of KDP, based on field research, to see
– what kinds of corruption there are in KDP
– where the incentives in the project cycle lie for corruption
– which project design features enable or prevent corruption
– what the best strategies have been for fighting corruption in the project
· Recommendations for improvement
3 On the effect of corruption on growth, see Mauro, Paulo. “Corruption and Growth,” Quarterly Journal of Economics, Aug. 1995, 110 (3), pp.
681-712. For the effect of corruption on public service delivery and foreign investment, see World Bank. Helping Countries Combat Corruption:
the Role of the World Bank. Washington, DC: Poverty Reduction and Economic Management (PREM) Network, 1997, or World Bank. Helping
Countries Combat Corruption: Progress at the World Bank since 1997. Washington, DC: Operational Core Services (OCS) and Poverty
Reduction and Economic Management (PREM) Network, 2000.
4 For information about KDP, see its annual report. Kecamatan Development Program. Second Annual Report 1999/2000. Kecamatan
Development Program National Secretariat: September 2000.
Why KDP?
Why study KDP? Three features of KDP make it an especially interesting subject of study when thinking
about how to fight corruption in development projects.
First, KDP is different from traditional development projects. It takes as its starting point the question
‘what do villagers want?’ rather than ‘what do villagers need?’, drawing on community development
experience that suggests both that communities themselves are best at identifying what they need and that
the results, when they do so, are more likely to be sustainable. In other words, it is demand-driven. It is
also highly decentralized. Most decisions are taken not at the center but in over 900 sub-districts
(kecamatan) and 20,000 villages across Indonesia.
KDP’s ‘different-ness’ makes it interesting for a few reasons. First, its size and disparateness make it
impossible to control from the center. This diminishes the usefulness of traditional anti-corruption
instruments, such as audits and central monitoring. Instead, KDP tries to enable villagers themselves to
become agents of anti-corruption, by trying to ensure broad participation and transparency. Second, KDP
is an experiment. The hypothesis, borne out by the findings of a growing body of research, is that this kind
of project will work better than centralized, supply-driven projects in terms of sustainability, benefiting poor
people, and—of special interest to this report—in terms of corruption.5 But we don’t really know. Such
projects are fairly new for the World Bank and for Indonesia, so the sample of experience from which to
draw on is limited. That is why it is important to investigate how KDP works and, more importantly, why
it works the way it does.
Despite its differences, KDP is still similar enough to other development projects to provide useful crossproject
comparisons. It is large, and covers an area of great geographic and cultural diversity. Like all
World Bank projects, it works through government channels. Unlike bilateral or NGO projects, the funds
are in the form of a loan to the Government of Indonesia (GOI), so KDP could not bypass the
government completely even if it wanted to do so. And like other projects in the country, KDP is in
Indonesia. The connection here is not spurious: it means that the project operates in a very specific
sociological and historical context, certain features of which—such as the administrative setting—have a
concrete effect on how the project can be run.
Second, KDP is in a country at a critical juncture in its history. Indonesia’s reform process is unstable.
Although great changes have taken place, many of the authoritarian power structures of the Suharto-led
‘New Order’ period remain in place, and corruption is entrenched.6 The proposed reforms threaten the
interests of some powerful forces in Indonesia, including those of the influential military.7 Against this
background, Indonesia is trying nevertheless to push ahead with its ambitious program of reforms. KDP is
at the front of this reform agenda. It is as much a governance program as a poverty alleviation program
5 For a discussion of the links between participatory, demand-driven community development approaches and sustainability, see Gross, Bruce;
van Wijk, Christine; and Mukherjee, Nilanjana. Linking Sustainability with Demand, Gender and Poverty. Water and Sanitation Program:
December 2000.
6 The 2001 Corruption Perception Index (CPI), released by Transparency International in June 2001, ranks Indonesia as 88th out of a survey of 91
countries, above only Nigeria and Bangladesh. The survey is a ‘poll of polls’ from various different sources and attempts to measure perceptions
of corruption across countries. Transparency International. Corruption Perception Index 2001. Transparency International: 2001. Retrieved July
1, 2001
7 Only about 25% of the military’s budget comes from government sources. The rest comes from off-budget activities, including businesses and
protection rackets in local areas. Reforms that threaten to reduce their power thus directly threaten their material interests. For a discussion of the
role of the military, see International Crisis Group. Indonesia: Keeping the Military under Control. International Crisis Group: 2000. Retrieved
August 7, 2001 .
and, where it works as it should, challenges elite control of village resources and decisions. The question is
whether, in such a political context, a program such as KDP can itself be a significant agent of change,
crystallizing nascent trends towards democracy into concrete changes in how village decisions are made.
Corruption is a lens through which to view this.
Third, KDP’s network gives us a unique window onto the state of village governance in Indonesia.
Although rapidly urbanizing, Indonesia is predominantly rural: most of its population live in villages. But,
whereas a great deal of attention is being paid to the effect of democratization at national level, very little is
known about the extent to which such changes are reflected at village level. Villages are sites of hierarchy
that, institutionally, have favored the interests of the powerful. KDP works closely with village institutions
and has a network of over 2000 facilitators working in over 20,000 villages across Indonesia. It thus affords
us a rare inroad into seeing what is happening in these villages, yielding a rich and unusual information set
about social change in Indonesia. Focusing on corruption enables us to examine issues closely related to
this change, such as accountability, participation and control of resources.
This report thus has a dual purpose. Its primary purpose is to ask how corruption can be prevented in a
large, decentralized, community-level development project. Its subsidiary purpose is to gain interesting side
snapshots of social transformation in Indonesia: how is it happening, and what does it look like?
Analytical framework
In order to ask how corruption can be prevented in a project like KDP, this report starts with explaining
what we mean by corruption and by outlining our analytical approach.
Corruption does not consist simply of bribes in grubby envelopes, but comes in many forms. It exists when
firms collude to fix prices above the market rate. It exists when civil servants demand a cut of funds ‘to
process’ public projects. It exists when officials hire their relatives for jobs that others could do better. The
forms of corruption are wide and various.
The World Bank uses a simple definition that covers many types of corruption: the abuse of public office for
private gain.8 This kind of definition is common in the corruption literature.9 It is broad enough to
encompass most types of corruption, including those that are difficult to measure, such as nepotism.
However, it is focused enough to draw parameters around corruption: to distinguish, for instance, between
bribery in the public sphere and gift-giving in the private sphere.10 Furthermore, by mentioning neither the
law nor morality, it allows two important distinctions to be made: (a) between what is corrupt and what is
illegal; (b) between what is corrupt and what is immoral. Although there may be large overlaps among
these concepts, their contours are distinct and it is important to keep them analytically separate.
8World Bank, Helping Countries Combat Corruption p. 8.
9 See Klitgaard, Robert; Maclean-Abaroa, Ronald; and Parris, H. Lindsey. Corrupt Cities. Oakland, CA: Institute for Contemporary Studies
Press, 2000, or Bardhan, Pranab. “Corruption and Development: A Review of Issues,” Journal of Economic Literature., Sept. 1998, XXXV,
pp.1320-1346 for a discussion of different definitions and types of corruption. The definition we use covers many types of corruption: officials
misuse their offices in this way when they accept bribes; when they skirt procedures in exchange for favors; and when they ignore illegal
practices for a fee, for instance.
10 This does not suggest that corruption is confined to the public sphere alone. Corruption does take place in the private sector. However, the
World Bank is mainly concerned with corruption in the public sector.
Analytical approach
Picking the right kind of analytical approach to combating corruption is instrumentally as well as
theoretically important. Many anti-corruption strategies fail. Often, this is because they do not make the
right analytical distinctions and are excessively legalistic or moralistic. The anti-corruption literature
suggests that such anti-corruption campaigns are unlikely to have lasting effects.11
Legalistic anti-corruption strategies attempt to combat corruption through strengthening laws or creating
rules. They focus on creating a regulatory environment that tightens the loopholes for corruption. The
assumption is that corruption can be fought by strengthening regulations against corrupt behavior.
Taken by itself, this assumption is flawed. Although regulations are arguably a necessary condition for
fighting corruption, they are not a sufficient condition. Almost all countries have anti-corruption laws on
their books, yet the incidence of corruption among countries varies widely, without correspondence to the
strength or number of existing rules. Anti-corruption drives that are limited to regulations may
paradoxically increase opportunities for corruption, through breeding regulations and red tape. More
fundamentally, though, they fail to distinguish clearly between rules and the motivation for following rules.
Fighting corruption demands focusing on motivations and opportunities, not the rules themselves.
Anti-corruption strategies couched in terms of morals fail for similar reasons. Such anti-corruption
strategies talk of ‘cleaning up’ and of the return of ‘honesty’ and ‘values’ in public life. They imply that
corruption would be rooted out if only corrupt officials were replaced by honest, public-spirited individuals.
The problem with this is that the proper locus of morality ultimately is people, whereas corruption lies in
systems. Combating corruption effectively demands regarding it as a problem of systemic rather than
personal failure. When an official steals money from a development budget, her motive may simply be to
send her son to a good school; and the important failure is not in her but in that her system allows her to
get away with stealing. Fighting corruption sustainably demands changing the system of rewards and
punishments so that stealing is no longer in her interest or anyone else’s.
With both legal and moral approaches, success is subject to circumstance; when they succeed they do so on
an ad-hoc basis and with largely temporary effects. This is because they fail to recognize that corruption is a
problem of incentives, not of wickedness or rules. The most important element of any sustainable anticorruption
strategy is
to change the underlying system of incentives so that agents are no longer motivated towards corrupt behavior.
This is the key premise of this report. In this kind of framework, legal actions are important, but only
insofar as their enforcement changes the cost-benefit equation.
The premise contains a hidden assumption about the way people act. It assumes that people are rational
and self-interested—that, given a choice, they follow whichever course of action benefits them most. The
assumption, however, need not be narrow or economistic. It can be made with varying degrees of strength.
A strong version of the claim might define self-interest as allegiance only to the individual self and define
benefits in material terms. We make a weaker version, which may have diminished explanatory power but
11 See Klitgaard et al, Corrupt Cities.
which we believe captures better why people act the way they do. Here, ‘self-interest’ may encompass
allegiance to one’s family or peer group as well as to oneself, so that the claim can explain human behavior
motivated by feelings of solidarity or social bonds. The benefits need not be calculated materially, but may
include social status, power or some other intangible reward. And, although this version still implies that at
some level self-interest is a universal phenomenon, it leaves room for its content to vary widely across
different cultural settings.
Indeed, one assumption of this report is that social norms have a significant effect on how people perceive
their interests. The practical challenge for projects wishing to minimize corruption is to understand how
the social norms work in the particular context of the project and to set up the incentive systems
accordingly. The way that actors in a village-level community project in Indonesia perceive their interests
will differ from the way that those in a national highways project in China perceive theirs, and the
effectiveness of different anti-corruption instruments will vary accordingly.
The difference, however, lies not simply in the social norms themselves, but in how they get expressed
through institutions. The more grassroots and local the project, the more important the social norms and
personal relationships characteristic of village institutions are likely to be. The opposite is true of a largescale
project whose participants are unknown to one another.
It is precisely for these kinds of reasons that an anti-corruption project design or strategy cannot be formed
a priori. It demands instead a knowledge both of local context and of the experience of similar types of
projects. Knowledge of local context is best formed with the involvement of local participants, who often
know best which kinds of incentives will have relative importance.
In the framework of incentives, we can see that
Corruption exists where the benefits outweigh the costs
§ The benefits may include the amount of money the agent stands to gain, the social benefits that might accrue from
this money, and the increase in power that might come from the transaction
§ The costs will include the severity of punishment, the monetary value of any fines, and any associated social costs.
§ The calculation of benefits and costs will depend on the risk of getting caught and being held accountable. This
ratio will be lower where the agent has wide discretion over the transaction—in other words, where few other
agents are able to see what is happening—and where there are few checks and balances over the agent’s actions.
The report is based on an analysis of identified corruption cases in KDP, field visits to ten villages and three
provinces, and in-depth, on-site interviews with central KDP project staff, KDP field consultants,
government officials and villagers. It also makes use of information gathered during KDP supervision
missions to provinces other than those visited for this report. The report especially makes use of the views
of KDP’s project historian and of staff from KDP’s Complaints Handling Unit, who track and follow up
corruption cases that get reported.
The report is not a survey. It is an investigation of process. Its aim is to get a sense of the anatomy of
corruption in KDP villages: of how the actors perceive their interests, what motivates them, what kinds of
constraints they face, and the steps they take to resolve their problems.
With this in mind, we attempted for the first part of the analysis to reconstruct a limited number of
corruption cases in different KDP locations. In each village, we tried to interview most of the people
involved in the corruption case, from ordinary villagers to local government officials, ‘corruptors’ included,
to see how each of the main players experienced and perceived the case. We attempted also to get the
stories of those unconnected with the case. This enabled us to get a multi-layered, ‘thick’ description of
what had happened. Sometimes, interviewees participated by drawing maps and social diagrams of the
The second major part of the research involved a review of samples of the KDP field experience and a
cost-benefit analysis of the KDP project cycle. The project is too large and complex to be susceptible as a
single entity to a cost-benefit analysis. To do it, therefore, we (a) broke the project cycle down into a series
of discrete steps, small enough to be analyzed in terms of incentives; and (b) analyzed each step of the
project cycle in terms of incentives for corruption. We did this by asking, at each stage of the project cycle:
· Who are the main actors?
· What are the main operations under their control?
· What discretion do they have over their actions?
· What interest do they have in engaging in corrupt practice?
· What sanctions apply? What is the risk of being discovered?
· What are the accountability mechanisms attached? Do the actors have to explain themselves?
· What are the social norms attached? Do these affect how the actors will perceive their interests?
The findings were drawn from an analysis of KDP corruption cases, field research, and the views of KDP
staff. In assigning weight to incentives they attempt to take into account as much as possible the views of
project participants.
The kind of information gathered from this exercise is necessarily anecdotal. With limited time and
resources, we were not able to go to more than three provinces. There was also an inherent selection bias
in picking locations: the locations visited were those where corruption had already been reported. This is
therefore neither a comprehensive nor a representative picture of village-level corruption in Indonesia.
Nevertheless, it gives us interesting discrete pictures of what corruption in KDP villages looks like, and an
idea of which elements of the project work to encourage or to limit it.
I. The context of community development in Indonesia
Political change & decentralization
Indonesia is in transition. At least two elements of this transition are likely to have a major impact on
village life in Indonesia. One is the effect of political change. The other is decentralization.
Political change
People… aren’t really scared to report problems. They’re not scared because things have changed. Before, people
used to be scared of the army, or the police. They’re not anymore.
District-level KDP consultant, Toli-Toli district, Central Sulawesi
The Indonesian political environment has changed significantly since the demise of the New Order. It has
been characterized by a trend toward democratization and the strengthening of civil society, an attempt to
redefine the position of the military in political life, and the devolution of power to local government. It
has been characterized also by a breakdown of law and order, a rise in civil unrest, separatism, and intercommunal
These changes have affected village life in Indonesia in several ways. In some places, they have served to
make latent conflict overt. The New Order government placed an emphasis on order and harmonious
social relations, and potential conflict was largely stifled.
Political changes at the center have altered this. Exacerbated by economic difficulties and enabled by a
breakdown in the will and ability of the security apparatus to impose order, they have removed the
underpinnings of an apparent stability. In its space, tensions have emerged over issues such as land, access
to resources and political power—tensions manifested often in ethnic terms. What it means for these
villages is that once-set relationships are in flux: old hierarchies are being redefined and cross-cut with new
cleavages, leading to a shift in the way villagers perceive their respective interests.
Political reform (reformasi) has also changed the landscape. The impact of reformasi at the national level,
manifested in open criticism and protest, has been significant.12 Its impact at village level is unclear.
Almost no evidence has come forward to suggest what type of impact, if any, reformasi is having on political
expression in villages. Anecdotal evidence from this report suggests that there has been some kind of
trickle-down effect, whereby villagers have some awareness of national changes and are conscious of
increased possibilities for protest.13 There thus seems to be some scope for a renewal of political
expression in villages, if limited at present to a heightened optimism in people’s minds. However, whether
12 This is so even barring those displays of political protest that are alleged not to be genuine expressions of political opinion but displays of unrest
organized from the top by groups wanting to create an impression of instability.
13 Villagers we met spoke often of reformasi and regional autonomy. Their level of political awareness was high. Mainly, this was because they
watched Jakarta news on national television.
this changes local power dynamics remains to be seen, given that many New Order institutions remain in
place at village level.
In January 2001, Indonesia started to devolve most functions of national government to its 362 districts
(rural kabupaten and urban kotamadya). If fully implemented, this will make Indonesia one of the most
decentralized countries in the world. The legal framework for decentralization makes provision for changes
in village government. These changes could change quite substantially the way that Indonesian villages are
run, although vagueness in the wording of the laws means that in practice there may be little change.
The spirit of the legislation points towards greater democracy at the local level: Law 22 of 1999 states that
the basis for village government regulations will be “diversity, participation, real autonomy, democratization
and people’s empowerment”.14 The law introduces a new village representative body, the Badan Perwakilan
Desa (BPD), whose members will be elected directly by villagers. This contrasts sharply with the body it
replaces, the Lembaga Musyarawah Desa (LMD), whose members were effectively appointed by the village
head. Under the new system, the village head can be held accountable at BPD meetings, and the BPD has
the right to ask that the village head be removed. This could dilute quite substantially the present
concentration of village power in the village head.
The BPD was made… after the regional autonomy system came into place. The BPD are elected by us. That gives a
greater degree of control. Now, the people have more power… the BPD system also gives us a way to resolve
problems we have. There’ll be fewer problems of corruption now because we can check and control them.
Villager in Pagar Batu village, North Sumatra
However, despite their spirit, the laws provide few specific guidelines for how local empowerment will
materialize. For instance, specific regulations surrounding election to, the operation of and powers of the
BPD have been left unclear, so whether its formation effectively increases popular control over village
government may be determined on an ad-hoc basis. The law also gives wide discretion to district
governments in determining village affairs.15 This means that there could be wide variation in village
government: some areas could have democratic, accountable local governments whereas others could
maintain or even strengthen their hierarchical and autocratic governance structures.
The impact of decentralization on village life is thus uncertain. Village structures could become more
authoritarian in some places, limiting accountability and increasing yet further the possibilities for
corruption. Alternatively, they could become significantly more democratic, opening up avenues of
transparency and accountability and raising the perceived costs of corrupt behavior.
14 Law 22 of 1999, cited in Antlöv, Hans. Village Governance in Indonesia – Past, Present and Future Challenges. Paper presented at the
PERCIK conference “Dynamics of Local Politics in Indonesia”: Yogyakarta, July 2000.
15 See Evers, Pieter, Resourceful Villagers, Powerless Communities (Rural Village Government in Indonesia). Jakarta: World Bank/Bappenas
Local Level Institutions Study, DRAFT, 2000
Village life in Indonesia
The experience of village life and of previous village projects is a crucial factor in influencing how actors in
KDP perceive their interests and rights. It affects how actors calculate the costs and benefits of corrupt
behavior and affects the will and ability of villagers to protest if they suspect foul play. Thus any
understanding of how to fight corruption in KDP, of what motivates its agents, and of what factors enable
corruption must rest on an understanding of what life in an Indonesian village is like.
It is not an easy understanding to reach. Indonesia is diverse and so generalizations are hard to make. Yet
despite this, there are some family resemblances among villages that enable a rough picture to be drawn.
A village (desa) is the lowest unit of government administration in Indonesia. Many villages came into
existence only with Law 5 of 1979, which imposed uniformity in form and structure onto Indonesian
communities. Beforehand, many of these were smaller and organically-formed.16 Because villages are
political and not social units, their boundaries may cut across traditional boundaries of land drawn
according to adat (local custom). They are divided up into hamlets (dusun), groupings that, being small and
generally homogenous, are often socially more cohesive than villages. Hamlets are sometimes physically far
apart from one another, and levels of communication among different hamlets vary. Sometimes one
hamlet may consist of people from an ethnic group entirely different to that of another hamlet in the same
village, usually because one consists of migrants from another part of Indonesia. In Java, many villages are
extremely large. This limits cohesiveness and control.
A village is led by a village head (kepala desa), who is paid by the government. In rural villages, the village
head is elected by villagers but ratified by the district head. In urban villages (kelurahan) the head (lurah) is
appointed. Village regulations ensure that almost all power is concentrated in the village head, who is
accountable not to villagers but only upwards to the district head.17 Because running for office is
expensive, village heads tend to be from richer families in the village, although not, in recent years, from the
traditional land-owning elite, for whom the gains to be had from the post are often not worth the effort.18
The cost of running for office places pressure on village heads to use their time in office to recoup as much
money as possible.
Apart from the village head, there are a few other officials—mainly the hamlet chiefs—and a couple of
village councils, which are deliberative fora for villagers to communicate their views on village issues and
priorities. For the most part, though, power rests with the village head. The councils have little decisionmaking
power and the village head and his officers tend to predominate in them 19
Three aspects of Indonesian village life have an especially important effect on the workings of village
corruption. The first of these is the influence of a heavy, centralized state bureaucracy, government rural
policy and state administrative mechanisms upon the incentives for corruption in villages. The second is
16 Law 22 of 1999 replaces Law 5 of 1979. The preamble to the law considers that Law 5 of 1979 is ‘no longer suitable with the spirit of the
1945 Constitution and the significance to acknowledge and to respect the privileges of Regions ’ (my italics). From Law 22 of 1999 text,

17 See Evers, Resourceful Villager, Powerless Communities for details.
18 An interesting discussion about how this applies in Java can be found in Frans Hüsken’s article ‘Village Elections in Central Java: State Control
or Local Democracy?’ in Antlöv, Hans and Cederroth, Sven. Leadership on Java. Surrey, UK: Curzon Press, 1994. Hüsken studied 50
candidates for village head in 9 villages in Central Java in 1989 and found that “the traditional elite families are hardly represented…Asked why
they had withdrawn, their answer was perfectly simple and straightforward: the advantages of being a lurah no longer outweighed the burden of
the job”.
19 Usually ‘his’. Village heads are rarely female.
the effect of a corrupt and inaccessible legal system upon village accountability and corruption. The third is
the way in which development projects themselves have been run.
The state, politics & village corruption
Villages are institutional battlegrounds for control over people and resources. Historically, the state has
given material and institutional backing to village elites in an attempt to co-opt potential opposition. State
backing of some groups over others has created serious problems. One of these is that it has created the
conditions for widespread corruption.
The environment in an Indonesian village is one where corruption flourishes easily and is difficult to root
out. It is peculiarly embedded. To understand why, think of a stable structure locked into place by several
pillars. If enough pillars are removed, the structure collapses. But removing only one or two simply
distributes the load to the others.
Corruption in a village is somewhat the same. The central pillar of support is the omnipresence and
omnipotence of the state. Indonesian villages are characterized by the massive presence of both the
government bureaucracy and, through the ‘territorial command’ structure that matches it, the military.
Almost all of an Indonesian villager’s dealings with the outside world or with government involve
bureaucratic procedures of one kind or another.20 Such high levels of red tape breed opportunities for
village officials to gain small—or, in the case of land transfers, large—amounts of money through
embezzling funds and extorting fees for services. The participating officials have a monopoly over these
procedures and have wide discretion over their actions.
The history of impunity acts as the second pillar of support, enabling village officials to exploit such rentseeking
opportunities. Village officials know that the sanctions for corruption have been few and rarely
enforced. This lessens the perceived cost of corrupt behavior. For their part, villagers know that
whistleblowers have few protections and low likelihood of redress. They also know that most previous
attempts at fighting corruption have been futile. This increases the costs associated with an attempt to fight
corruption, meaning that in most cases the strictly rational course of action for any individual villager is
apathy—in other words, to do nothing. This enhances even further the culture of impunity.
[The villagers] are used to old-style bureaucrats, who are like kings, who think they are like kings, who are never wrong.
And nothing will happen. In their experience, if they report problems to a bureaucrat, there will be no response, so why
Civil servant, Kecamatan Palipi, North Sumatra
20 Hans Antlöv describes the multitude of paperwork required in his article ‘Village Leaders and the New Order’ in Antlöv and Cederroth,
Leadership on Java. “Let me exemplify the variety of official letters by the experience of a young man, Dadang, who wanted to join the Navy.
Dadang knew this would take much time and energy. He armed himself with twenty photos and a letter of recommendation from his father, a
school headmaster. First, Dadang had to go to the sub-district Bureau of Education to get his high school diploma validated. He took this back to
the village, and asked for a letter of “Good Behaviour”, a letter of “Self-Cleanness”, a Birth Certificate, and a letter of residency. All these letters
had to be signed by the headman, the Police Community Guidance Officer and the Army Village Guidance Officer. He took the signed letters to
the Majalaya police and army commands where the letters were co-signed by police and army officials. They were brought back to the village
office, where Dadang obtained a letter of recommendation from the village secretary, sanctioning his application. All this was then brought to the
sub-district office in Majalaya, where the deputy leader, after being asked by Dadang’s father, wrote a similar letter of recommendation. All this
was necessary just to get drafted for the test. The whole procedure had cost Dadang and his father 25,000 rupiah. When Dadang later failed the
test, his friends said that he forgot to bribe the drafting officers.”
A third pillar of support is deference and fear. The average Indonesian village is characterized by hierarchy,
with a tradition of not questioning those in power. Its traditional social arrangements might not be
hierarchical, as this varies across Indonesia, but it almost certainly has a hierarchical political culture.
Indeed, a visitor to a village at least during the period of the New Order government might think it
curiously de-politicized, with a paternalistic political culture and an emphasis on bowing to ‘consensus’—a
direct result of New Order policies towards rural areas.21 Such hierarchical political relations limit popular
pressure on the powerful.
Holding the structure together is an administrative setting that gives villagers few formal avenues to
complain. Although elected by villagers, the village head is formally accountable only upwards to the
district head. This gives him wide discretion over his actions and lowers the costs associated with
corruption. Indeed, it may increase the costs associated with not acting in a corrupt way, since the village
head may be put under vertical pressure to collude with his superiors. Complaints are resolved by
musyawarah, a process of discussion and consensus–building. But it is the village head who in the first
instance organizes the musyawarah, and only if popular pressure is extraordinarily strong does he have an
incentive to pursue problems in which he or his peers are implicated.22
Rural justice & village corruption
Justice is for the rich man, the man who has power.
Villager, Kotabumi, Lampung
The failings of the formal justice system lock this structure of village corruption even further into place.23
Some of these weaknesses are peculiar to the Indonesian legal system and its particular set of problems.
Others are a function less of the Indonesian system but of the special problems of rural life. Both types of
weakness have a strong effect on the workings of corruption in Indonesian villages.
The heart of the problem is that, rather than being a buffer against corrupt behavior, the legal system is an
active and knowing participant in it. In a better-functioning system, the courts might be expected to act as
arbiters of last resort, able¾albeit with slowness, inefficiency or other flaws¾to act independently and
with some degree of fairness. The presence of an institution able, even imperfectly, to act independently of
vested interests is especially important in an environment where the other mechanisms of accountability
have failed.
This is not true of the Indonesian courts. Instead, the courts are where impunity for corruptors is realized.
The Indonesian legal system is perceived to be thoroughly corrupt. It is commonly assumed that all court
decisions are for sale: that everything is negotiable for parties paying the highest bribes or with the most
21 See Annex I: Historical Background
22 When we asked what steps they would take to resolve a problem, most villagers we met replied that they would go to see the village head.
When asked what they would do if the problem involved the village head, most could not think of a response.
23 I would like to thank Pieter Evers especially for contributing background information and analysis of the workings of the rural justice system.
important contacts.24 This level of corruption extends to the police, who are charged with finding evidence
in court cases. As a result, trust in the police and the courts is low, corruption cases are rarely brought to
court and, when they are, a pretext often is found to drop cases or otherwise skew the court process. The
result is that potential corruptors in villages know that the risk of being charged legally with corruption is
slim and the risk of punishment even slimmer. Villagers know this too.
Why would I go to the police [for anything]? If someone stole my chicken, I’d have to pay the same as a pig in bribes.
Villager, Wamena, West Papua
A further problem is collusion and lack of independence. The relationships among legal and local
government officials tend, especially in rural areas, to be close and characterized by personal ties. This
erodes the separation of institutional functions necessary to preserve the independence of the judiciary.
The chief of the local court is often included in the musyawarah pimpinan daerah (consultation of leaders of
the region), an informal consultation among local government, the prosecutor, police officers and the
military high command. Such consultations provide an opportunity for these officials to pressure the local
court chief to act in accordance with their interests.25 As a result, even when corruption cases are brought
against government officials, they rarely succeed. The overlap of functions both reflects and strengthens
the culture of favor and patronage that exists among local officials. It is one that infuses all levels of the
government hierarchy, and serves to embed even further the culture of impunity for corruption.
A third problem is the lack of witness protection. There is no witness protection program in the
Indonesian legal system, either formally or in practice. Accordingly, threats and intimidation of witnesses is
common. Protecting witnesses is especially difficulties in rural communities where anonymity is unknown:
in such an environment, it is difficult to ensure that whistleblowers are not, in the words of one specialist,
‘knifed in the paddy-field at midnight’.
These weaknesses of the Indonesian legal system are compounded by problems that stem from the special
nature of rural communities: distance and cost. The court system in Indonesia extends down to the district
level only, and so the most accessible courts are found in district capitals. From many rural villages, these
are difficult and costly to reach: from remote areas, this may take up to a day or more. The transportation
cost of attending a trial is accordingly high. Even participating as a witness for the information-gathering
exercises of the local prosecutor costs money. This cost is borne by the witnesses themselves, and the
government itself has no budget for such operational costs. All of these problems mean that the cost in
rural areas of any villager-led attempt to use the legal system is prohibitively high, which provides a strong
disincentive for participation in it.
A final obstacle to accountability in Indonesian villages is lack of awareness. The level of awareness of
rights, legal processes and legal resources among the Indonesian public is low. In 2001, the Asia
24 This is borne out by the empirical evidence that is available concerning people’s attitudes towards the courts. The 2001 Partnership for
Governance Diagnostic Survey of Corruption in Indonesia found that the judiciary was considered the third most corrupt institution in Indonesia.
(Partnership for Governance Reform in Indonesia. A Diagnostic Study of Corruption in Indonesia. Jakarta: Partnership for Governance Reform
in Indonesia, 2002.) An Asia Foundation 2001 survey found that 57% of respondents felt the judiciary was corrupt (Asia Foundation. Survey
Report on Citizens’ Perceptions of the Indonesian Justice Sector – Preliminary Findings and Recommendations. Jakarta: Asia Foundation,
25 Internal World Bank note (Note on Rural Judiciary), Pieter Evers, 2002.
Foundation found in a survey of Indonesian citizens that only 56% of respondents were able to identify a
single right to which they were entitled. In rural areas, that figure dropped to 33%.26 Such lack of
awareness makes the environment in villages even more conducive to corrupt behavior.
Development projects & village corruption
Development projects themselves have contributed significantly towards corruption in Indonesian villages.
Their funds have historically been a fertile source of enrichment for local government officials and village
elites. In 1997, the World Bank estimated that despite acknowledged difficulties in quantification, between
20-30% of the government’s development budget funds were lost due to leakage, Bank-sourced funds
There are several reasons why development projects have contributed to corruption in villages.
Historically, the Indonesian government used development as a form of patronage: foreign aid funds were
used as a direct means of co-opting potential opposition in villages, and development projects were used as
a way of keeping communities happy. The government’s budget arrangements also create a strong
incentive for individual implementing agencies to misuse foreign aid funds. Each agency’s projected
revenue is divided into a ‘routine’ and ‘development’ budget. ‘Routine’ budgets are funded from
government sources and ‘development’ budgets from external sources. Although ‘routine’ budgets are
designated not for development but instead for paying civil servants’ salaries and other routine operational
costs, they are almost always woefully inadequate for their purposes. This creates a strong incentive at
ground level to use development funds to cover routine agency costs and to supplement civil servants’
Foreign aid funds have also been relatively large and easy to manipulate. They have been easy to
manipulate partly because of internal donor pressure to disburse loans; partly because many development
projects are large and difficult to supervise; and partly because of a tacit agreement on the part of donors
that some diversion could be tolerated if the projects achieved their primary targets of getting infrastructure
Government agencies implementing development projects thus had relatively wide discretion over a large
sum of money and were faced with a strong incentive to misuse it. They were able to use this money,
however, not only to supplement civil servants’ salaries, but also to buy patronage.
The government’s style of development directly supported the culture of impunity for corruption.
Development projects have traditionally been conducted in a top-down fashion, with outside assistance
imposed onto a village without the active participation of its members. There was almost no local
involvement in development planning. Villagers thus became passive subjects of development, not active
agents in it. The projects were thus perceived by the communities as something to wait for, handed down
from an external government over which they had little control, rather than as something to which they had
a right and over the manner of whose budget expenditure they could protest. Indeed, according to one
26 Asia Foundation. Survey Report on Citizens’ Perceptions of the Indonesian Justice Sector – Preliminary Findings and Recommendations.
Jakarta: Asia Foundation, 2001.
27 Internal World Bank memorandum, Stephen Dice, 1997
28 All of this information is taken from Partnership for Governance Reform in Indonesia. Corruption in Indonesia’s Foreign Aid Programs:
Naivety, Complacency and Complicity within Donor Agencies. Jakarta: Partnership for Governance Reform in Indonesia, 2001, DRAFT
experienced development practitioner in Indonesia, a common phrase used by villagers to describe
development assistance to their communities was ‘Bantuan sudah di-drop’, or ‘The help has been dropped’.29
Furthermore, the projects existed in an environment where all decisions, from tax to service-delivery, were
made in such a top-down manner. In a system that is almost completely formalized and non-participatory,
the only means of control and redress for complaints are the formal channels. But in Indonesia’s case, the
formal channels did not work: they could be manipulated easily to serve the interests of the powerful. In
this system, any protest against an individual project necessarily translated as protest against the whole
system, an infinitely graver offence. This meant that impunity became institutionalized.
Indonesian villages thus suffered from a failure of institutions and an almost total failure of incentives, in an
environment with large inflows of development funds and a culture of impunity for embezzling those
funds. This embedded even further the conditions for corruption.
29 From talk with a former head of Oxfam in Indonesia, September 2001.
II. Corruption in KDP
What is KDP?
KDP is a large, decentralized community development project in Indonesia. It covers more than 20,000
villages in approximately 27 provinces across the country. Its aims are
· to alleviate poverty by raising rural incomes
· to strengthen community institutions
· to improve local governance
KDP’s key principles are community participation, transparency and sustainability. It works by giving
villagers decision-making power over funds, which can be used either as grants for village infrastructure or
as loans for income-generating ventures. The project is designed to give villagers the chance to prioritize
village development needs and a say over how village resources are used. Villages compete with one
another for project funds, giving villages an incentive to propose well-designed projects of demonstrable
benefit. The three main components of the project are unallocated grants to communities, technical
assistance, and monitoring & supervision.
The project works by giving a block grant directly to each participating sub-district (kecamatan), which
usually contains about twenty villages. Each village then prepares a proposal for entry into an inter-village
funding competition at kecamatan level. Villages can submit two proposals if the second comes from
women’s groups. KDP is ‘open-menu’, so proposals can be for a wide range of activities, from building
roads or water systems to providing seed capital for small businesses. The important thing is that all
villagers are given the opportunity to participate in the proposal-preparation process, and that decisions are
democratically made. There is no outside ‘guidance’ of decisions.
Once villagers have prepared their proposals and had them verified for technical feasibility, they take them
to an inter-village meeting where representatives from each village decide collectively which proposals will
be successful. The block grant to the kecamatan is usually not big enough to fund proposals from all
villages, so choosing the proposals involves a process of competition, bargaining and compromise.
After successful proposals have been picked, funds are released directly to collective village accounts at the
kecamatan level. From here, money is withdrawn and given either to village implementation teams to start
buying materials for infrastructure, or to groups of borrowers to start setting up their economic ventures.
If funds are used for income-generating ventures, borrowers must pay them back at market rates of
interest. The repaid funds go back into a kecamatan- level revolving fund, to be made available to other
groups of borrowers. The idea is that funds grow and are recycled to other members of the community.
At all stages of the project there are checks, balances, and mechanisms whereby villagers can hold the
project accountable. There are multiple channels of accountability in the project. Primary among these is
that there are different channels for villagers to access information. In addition to the communities, the
project is monitored regularly by project facilitators, local NGOs and AJI, an association of independent
journalists. Each of these acts as a discrete center of support and information for villagers. Each also is
relatively independent from local government and village elite control.
In addition, the project has other mechanisms of accountability. Among them are that:
· Project formats and contract documents do not allow for single signatures.
· All financial information about the project must be made transparent and publicized widely at all times.
This takes place mainly through village meetings, local media, and posting information on community
· Accountability mechanisms whereby villagers can oversee project funds are built into the project cycle.
What happens if there is a complaint?
One important element of the project is
the independent, centralized system of
assistance and complaints follow-up. As a
government project, KDP is formally
implemented through government
channels, although there is no government
control over funds. At each level of
government, from the national to the
village level, a government representative is
formally responsible for KDP. This
management system is matched at each
level by independent KDP staff – the KDP
consultants and facilitators, who at village
level are elected and at higher levels are
recruited. At national level, the project is
administered by the government’s KDP
Secretariat, matched by a KDP body called
the National Management Consultants
If there is a problem in a village, it can be
reported directly to the national level of
KDP, at which point it sets into motion an
automatic complaints resolution process
(Fig 1). Although this process is not
designed as a substitute for local problemsolving
mechanisms, it does mean that,
when pursuing their complaints, villagers
have redress to powers that are
independent of local power dynamics. This is a crucial factor in giving villagers control and ensuring that
local government and local elites do not have carte blanche over their actions.
Facilitators play a special role in KDP. There are both village facilitators (FDs) and kecamatan facilitators
(FKs). Village facilitators are elected. They receive a small honorarium to help spread information about
Fig 1.
KDP Complaints Procedure
Complaints to
Complaints to PO
Box (anonymous)
Complaints to
KDP Secretariat
Open and
supporting data
Solve through
consultant channels
Solve through
govt. channels
Suggest type of
Implement action
Get feedback from field.
If problem not solved,
suggest another course of
Compose official
draft letter
Prepare official
field trip
KDP, ensure broad participation, especially by marginalized groups in the village, and to help serve as a
help and contact point. Kecamatan facilitators are recruited. They run the project at sub-district level
together with their government counterparts. They are responsible with village facilitators for spreading
information and ensuring participation in KDP. They also serve as a contact point between the villages and
the project, and help with the complaints-resolution process. Community members often find it easier to
complain to their local facilitator rather than directly to the KDP Secretariat. Out of 124 complaints
received by KDP in its first two years, 89 were from KDP staff and only 13 from community members
The role facilitators play in limiting corruption in KDP is crucial. First of all, they act as conveners for the
project, by getting villagers to come to project meetings. The extent to which villagers participate depends
largely on the quality of facilitation. If the facilitator succeeds in ensuring wide participation and
transparency, it makes it much more likely that corruption will be protested against. Second, facilitators are
an independent source of information, both for villagers and for project staff higher up. They can give
villagers information about projects and KDP procedures, and they can help check facts if accusations are
made. Many facilitators are from NGO or reformist backgrounds, and so can act as independent sources
of other information pertinent to the villagers’ experience of political change. Third, the facilitators have a
power base independent of government and local elites. They need not fear redress if they get involved in
pursuing corruption cases: indeed, it is their mandate. Their role in limiting corruption in the project is thus
This does not, however, mean that facilitators are never involved in corruption themselves. Approximately
13% of corruption cases reported to KDP are alleged to involve facilitators. When facilitators themselves
are involved in corruption, it makes it especially difficult for villagers to report it. It thus becomes especially
important that villagers know that corruptors will be pursued in KDP, even if the corruptors are KDP
Fear of complaining: why do it?
Fear and intimidation play their part in KDP. Although villagers are encouraged to report corruption cases
to KDP, and although they are assured that the allegations will be pursued, there is no guarantee of safety
for those who report corruption. KDP may discourage but it cannot prevent retribution for
whistleblowers, and several cases of intimidation have been reported to KDP. Intimidation takes place
outside the village level too: KDP’s independent monitors at regional level – NGOs and AJI, the
association of independent journalists – have also been threatened.
The question of motivation then remains. Given the history of impunity in villages, and given that villagers
may put themselves as risk for reporting corruption, why do villagers complain? We don’t know. But a
few pertinent facts point to the importance again of costs and benefits. Villagers rarely complain as
individuals. To do so would be to put themselves at risk. They do tend to complain as groups. This
spreads risk, but does not dilute the benefits of complaining. Villagers tend also to use facilitators as a
means of complaining. And, although facilitators too are subject to intimidation, their risk is smaller
because they are backed by KDP. The motivation of groups of villagers to complain depends largely on
how confident they are that action will be taken if they do. This depends mainly on the strength of
30 Details of reported complaints can be found in the KDP Annual Report. Kecamatan Development Program. Second Annual Report
1999/2000. Jakarta: Kecamatan Development Program National Secretariat: September 2000.
socialization in the villages. In some sense KDP has been a test case; villagers say they have reported
corruption because they have been assured that this time, it will be different: cases will be pursued and
sanctions applied for wrongdoing.
Sanctions for corrupt behavior are built into the project cycle and , crucially, are made known ahead of
time. KDP facilitators and consultants run the risk of being dismissed or criminally charged if exposed
with corrupt behavior. Government officials may be fired or, more commonly, rotated to less desirable
posts, hierarchically or geographically. This can be an unpleasant experience for local civil servants,
especially if they have extended families and other ties in their communities.
The World Bank and the Government have other sanctions at their disposal in KDP. They can cancel
contracts with firms that manage and recruit local KDP staff. They can, if necessary, disqualify particular
districts or provinces whose corruption problems are particularly bad. Finally, as a last resort, the World
Bank can delay, suspend or block disbursement of the loan. The use of sanctions has proved especially
important in fighting corruption cases in KDP. One of its biggest challenges is how to apply sanctions that
are credible enough to act as a disincentive towards corruption.
Different types of corruption in KDP
Corruption cases in KDP range from the imaginative—20 chickens being valued at a somewhat expensive
US$2000—to the mundane but common— kickbacks to village officials and collusion in the procurement
of goods. Most of the cases can be put into a few broad categories. The categories are not exclusive;
indeed, most cases of corruption have elements of more than one.
One of the most common types of corruption in KDP is mark-ups on budgets, which usually involves
some element of collusion. Most of the time, the budget mark-ups are reversed: what happens is not that
budgets are inflated in advance, but that once budgets are agreed, implementation teams substitute cheap
materials for those originally specified and pocket the difference in cost.
Panasakan: pocketing funds for implementation
Villagers in Panasakan, a small village in Central Sulawesi, noticed that the amount of wood delivered to the village
seemed small compared to the amount allocated for the wood in the budget. When they conducted a field audit, they
concluded that 8 cubic meters more of wood, worth about Rp 3.6 million, should have been delivered to the village. In
this case, it was suggested that the head of the village implementation team had bought less wood than specified in the
budget and had pocketed the difference. He was accused of colluding with wood seller to verify his story. This particular
case is interesting because it was complex: on his part, the head of the implementation accused the villagers themselves
of stealing some of the wood, and at a meeting to discuss the problem some villagers did indeed admit doing so. They
claimed, however, to have taken only just a tiny portion of the amount missing. It is an interesting case of complexity: in
some corruption cases all parties may have dirty hands, including those who are relatively powerless.. The case also is
a good example of the point, made often in the corruption literature, that it can be difficult to find proof of corruption. In
this case, the receipts, which were the only possible source of proof, were allegedly forged.
Another common type of collusion is when false borrower groups are formed to get KDP credit.
According to KDP credit rules, a group that wants access to credit may borrow funds only if it has existed
as a group for a year prior to borrowing the money. This is to strengthen social capital and to increase the
probability of repayment. Often, though, people are persuaded to pretend to belong to a borrower group,
but the money borrowed collectively by the ‘group’ finds its way only to one or two people, usually already
relatively well-off.31
A third type of corruption seen in KDP consists of bribes and kick-backs to officials. These are rarely
overt, but will come in the guise of money to ‘process’ papers for KDP or other such administrative need.
Success in obtaining these bribes depends on villagers’ ignorance about KDP’s principles and an
expectation on their part that KDP will be like previous development projects in the village.
Tips and bribes in West Sumatra
In Sungai Beremas, West Sumatra, the village head allegedly received two types of bribes connected with KDP: several
small-scale bribes of Rp 50,000 each so that he could ‘process papers’ for villagers to gain access to KDP credit, and
four bribes of Rp 500,000 each that he was accused of asking for directly by knocking on a few villagers’ doors in the
middle of the night, wearing his government uniform and asking for the money. When we spoke to members of the
financial management unit in the kecamatan , they insisted at first that the smaller bribes were perceived as ‘tips’, not
bribes. It was just ‘uang lelah’ they said, and was standard practice. They claimed also that, because the money came
from the villagers’ pockets and not from KDP funds, that it did not count as misuse of money in KDP, even though they
admitted, after our questioning, that probably if the villagers had not paid the money, they would not have received KDP
credit. ‘Oh please’ responded the team leader of the kecamatan facilitators later when told about these ‘tips’. ‘Do you
think the villagers would just give money to the village head if they didn’t have to?’
There are also instances of nepotism in KDP. Often, TTDs, technical advisors chosen by villagers, are
chosen through nepotism: friends or relatives of village elites may be chosen as technical advisors even
though their qualifications, if any, are inappropriate. This lack of technical expertise means that
infrastructure is built badly or even dangerously.
Many of the corruption cases in KDP, especially the larger ones, are complex and involve many different
types of corruption. They fit most appropriately under the umbrella of straightforward theft. Jayawijaya
district in West Papua (Irian Jaya) is a good example. In Jayawijaya approximately Rp 3 billion went
unaccounted for, through the theft and collusion of the district-level technical consultant, contractors,
suppliers and the district co-ordination team, possibly with the involvement of local KDP consultants and
local government: in short, a complex, large-scale operation.
KDP thus operates in a context where corruption is made easy by bureaucracy, impunity, fear, and a lack of
accountability. It is a context, though, that is shifting: political changes may be altering local power
dynamics in ways that reduce the fear of protest, and decentralization legislation may be opening up
avenues of representation and accountability.
How much corruption is there in KDP?
Measuring corruption is difficult. A closer look shows why. At first glance, KDP seems to have startlingly
low levels of corruption for its size. By the end of August 2000, only 124 complaints had been made to the
31 One indicator that a group might be false is when each member of the group has ostensibly borrowed the same amount of money: it is unlikely
that a genuine calculation by group members of their individual borrowing needs and capabilities would yield such a uniform result. However,
this isn’t necessarily an indication of corruption. Often false groups are formed so that all their members may get money, not just a few.
national KDP Secretariat concerning alleged corruption, contravention of procedure or inappropriate
government intervention in the project.32 If taken at face value, such low figures would be unprecedented
for a project operating in so many thousands of different villages across Indonesia.
The figures alone, however, do not tell much: they exclude cases that do not reach the national level of
complaints resolution, and, most importantly, leave out cases not reported at all. In other words, there is a
significant selection bias.
Other indicators of corruption are similarly problematic. Because procurement in KDP is decentralized,
there is no way of knowing whether differences in project costs across locations reflect different levels of
corruption or simply reflect different prices. Furthermore, a project of the type and size of KDP has not
yet been attempted in Indonesia, so strict comparisons with other projects are hard to pin down. Any
measures must therefore necessarily be rough.
Nevertheless, rough and ‘dirty’ indicators indicate that there is less corruption in KDP than in other
projects in Indonesia. The best indicator we have is cost. It costs KDP approximately 20-30% less than
other projects to deliver the same kind of benefit.33 Although it is hard to gauge with certainty why KDP is
so much cheaper, it is almost certain that corruption-related factors play a role. KDP funds pass through
fewer intermediaries than funds from other projects, and KDP infrastructure is built by villagers and not
contractors, giving fewer opportunities than in other projects to skim off funds. A second indicator that
there is less corruption in KDP than in other projects is that villagers say there is less corruption in it. This
suggests that involving communities in managing their own projects, and using simple, direct means of
transferring funds, is a useful project design element in limiting corruption.
Anti-corruption, politics and protest: the white shroud of Sidoardjo
According to one kecamatan facilitator in Central Sulawesi, villagers in Sidoardjo village were pleased to see that KDP’s
emphasis on transparency and participation gave good quality, cheap results. So pleased were they, in fact, that they
grew to expect such standards from all their projects. When a contractor for a government-funded village project to
build a bridge bought poor-quality wood with the funds, the villagers decided to take action. They took the wood and
wrapped it in a white cloth so that it looked like a shroud. A group of them then got together and delivered this ‘wood
coffin’ to the regional parliament. They stood in front of the legislators, chanting. ‘We don’t want any more projects like
this,’ they said, pointing at the white shroud. ‘We only want projects that involve the people, like KDP.’
Incentives for corruption in the KDP project cycle
A major part of this report consists of an attempt to map the opportunities for corruption in KDP by
examining the project cycle step-by-step to analyze the incentives for corrupt behavior for each major actor
Throughout the project cycle, there are three important areas of risk. The first is that village elites will try to
dominate the project, mainly so that they have an opportunity later in the cycle to secure kickbacks. The
second is that technical assistance will be chosen through some form of nepotism and not on merit.
Although the amounts of money involved here are relatively small, this has serious ramifications for the
32 Kecamatan Development Program. Second Annual Report 1999/2000.
33 This figure is drawn from a study conducted on infrastructure in KDP. Kecamatan Development Program. Studi Teknis Sarana/Prasarana.
Jakarta: Kecamatan Development Program, 2001.
quality and safety of infrastructure built. The third is that village implementation teams misuse funds to buy
materials. Again, apart from the value of the money misused itself, this usually has detrimental impact on
the quality of infrastructure.34
The other major source of corruption lies outside the project cycle. A KDP investigation found that there
was substantial misuse of operational funds for KDP, which constitute 3% of total project funds.
Facilitators for KDP at kecamatan level and above are recruited and managed by consulting companies hired
to administer the project. The administrative funds allocated are supposed to cover facilitators’ salaries,
transport costs, and other operational expenses. Many of these companies, however, misuse these funds,
pocketing in some cases up to half the money they receive. They cover their tracks by issuing false invoices
to the project, claiming reimbursements, for instance, for computers that were never bought. KDP II, the
second phase of KDP, will not use such firms and will administer the project directly.
This is an overview of the results of the incentive-mapping exercise:
(1) Dissemination of information & selection of
· But risk of nepotism in choosing consultants &
elite capture for later kickbacks
(2) Socialization: villagers learn how the project
works and what their rights are
· But risk of elite capture
(3) Villages prepare proposals High
· Kickbacks with selection of TA; collusion; budgetmark-
ups; false groups formed to get micro-credit
(4) Technical teams look at the proposals to see if
they are feasible
· But incentive to try to influence results of
assessment in village’s favor
(5) Villages choose proposals that will be funded Low
· But incentive to simply split funds among villages:
leads to bad projects
(6) Preparations for implementation: funds are
· Main threat is collusion among three signatories of
bank withdrawal to take a cut of funds before
transferring to village
(7) Implementation:
· Materials are bought
· Village labor is mobilized
· Roads & bridges are built
Extremely high
· Most common source of corruption in KDP (apart
from operational funds & consultants’ payments) is
with procurement of materials. Implementation
teams buy cheaper materials than those specified
and pocket the difference.
(8) Post-implementation:
· Loans repaid; revolving funds set up
· Maintenance of infrastructure
· Loan repayments managed badly; risk of
corruption with user fees for maintenance.
34 Sometimes it means that projects cannot be built at all. In one village which wanted to build a clean water system, for instance, the village
implementation team bought pipes that were smaller in diameter and cheaper than those originally specified. The pipes, though, were suitable
only to cover electricity wires and not for water, so they could not be used at all.
Preventing corruption in KDP: using designs to change the opportunity structure
KDP contains several design features that work to limit corruption. The basis of all of them is that
KDP hands management control of the project to villagers.
Handing control to villagers has broad benefits. It strengthens social capital and promotes local ownership,
empowerment and sustainability. But more pertinently, it enables corruption to be limited in a project
where close supervision is almost impossible— a project spread out over more than 20,000 villages
nationwide. It does this by creating a structural incentive: if the people who control the project are those
who benefit most from it, they have an interest in not letting money disappear.
It cannot, however, do it without support and without a project design that limits the opportunities for
corruption. The features of KDP’s design that are most effective in limiting corruption are:
Community control & simplicity · Direct transfer of funds
· No local government control
· Villagers control budgets
· Financial formats are simplified so villagers can understand
Socialization · Villagers learn how the project is supposed to work, what
their rights are and what to do if they are unhappy.
Transparency · All financial information is made public and publicly displayed
in villages
· Complaints database will be published in newspapers
Limited monopoly · All goods to be procured require at least three quotations
Limited discretion · All financial transactions require at least three signatures
Accountability mechanisms · Regular village meetings to account for funds. Disbursals can
be suspended if misuse of funds is suspected.
Monitoring & follow-up · Regular project monitoring, complaints tracking & follow-up
· Independent monitoring by civil society groups & journalists
1. Community control & simplicity
KDP hands control of the project from local government to villagers and, by making the project simple,
enables villagers to participate. First, funds are transferred directly from the center to kecamatan-level bank
accounts, with no local government control over funds. It also gets rid of all the bureaucratic procedures usually
associated with development projects, such as intermediary forms and approvals. This represents a radical
break from past village projects. By getting rid of red tape and direct government management, it
minimizes in one step many of the opportunities to skim funds from the project. Second, villagers control
project budgets, and financial formats are simplified so villagers can understand them. Villagers have an
incentive to prevent corruption in KDP projects and, because financial formats are simple, they can see
clearly when money seems to have gone missing.
2. Socialization
I think the villagers are brave in reporting problems because of socialization in KDP. We have told them that ‘this
program is your program’. They’ve been waiting a long time for a program like this. Before, they never had one.
Kecamatan Facilitator, Baolan, Central Sulawesi
Socialization is the process by which villagers in KDP learn their rights. It takes place through a long
process of information dissemination. About six months before the start of the project, KDP facilitators
start disseminating information about KDP by holding open village meetings; meeting with small groups of
villagers, including women-only groups; talking about KDP at local gatherings, including prayer meetings
and local savings groups; and publicizing information about KDP on village notice-boards. In this way,
villagers are able to learn about their rights in KDP and about what to do if they are dissatisfied with it.
Most kecamatan and village facilitators feel that socialization is the key to preventing corruption in KDP.
When villagers don’t know their rights, it creates the space for corruption:
Powerlessness and corruption in Sungai Beremas
Pak Darmen [not his real name], a poor fisherman, borrowed money from KDP to expand his fishing business. But the
night he received the money, the village head knocked on his door in the middle of the night and asked him to hand over
Rp 500,000, a third of the total amount. The village head told Pak Darmen that the money was for some other people
who also deserved it, though didn’t say who these people were. Pak Darmen felt sorry for the other poor people, but he
also needed the money himself. Still, he didn’t think he could protest. This village head, a temporary replacement, was
an official in the office of the sub-district head (camat).
‘All I knew was that this person was from the camat’s office,’ said Pak Darmen. ‘He was wearing his uniform. When he
asked for the money… I thought I had to give it to him.’
Later, some other villagers told Pak Darmen that he had been stupid. They told him that in KDP he didn’t have to give
the village head any money, and that the village head had fooled him.
‘When I heard this,’ said Pak Darmen, ‘I felt sad.’
From interview with Pak Darmen, Sungai Beremas village, West Sumatra
3. Transparency
Socialization provides communities with the impetus for controlling corruption in their projects.
Transparency provides them with the information to do it. Most corruption cases reported to KDP’s
secretariat have stemmed initially from complaints from villagers themselves. According to KDP rules, all
financial information must be publicly disclosed. This includes details of budgets for original proposals;
information about salaries and honoraria for all project participants; market prices of materials for
infrastructure; details of funds borrowed as micro-credit and rates of repayment; details of bank
withdrawals; and receipts and accounts for all project monies spent. Putting information in the hands of
participants is the key to KDP’s anti-corruption strategy.
This is not always easy in practice. Requiring that all financial information be made transparent directly
threatens the material interests of those who traditionally have benefited from the ‘informal payments’ of
previous development projects. And, since people tend not to demand what they do not know,
communities tend not to insist on transparency. There are thus concrete obstacles to achieving good levels
of transparency in the field. In more than one kecamatan in Aceh, for instance, the camat tried to ban the
publication of financial information on community notice-boards across the sub-district. In other cases,
financial information is posted publicly, but in a way that is dull and hard for villagers to understand.
The question then becomes how to institute transparency in practice at community level. KDP has taken
several measures to try to do this. First, it has banned the issuing of written invitations to village meetings.
Written invitations send a message of exclusivity rather than inclusiveness; people that are not formally
invited feel they cannot come. KDP insists instead that meetings are arranged by public announcement.
Second, KDP has moved away from written media towards oral media. It has also moved away from
disseminating information at village-level meetings and towards hamlet-level meetings, which tend to be
better-attended, especially by poor people. Furthermore, following a local institutions study, it has moved
towards using local institutions for transparency: using prayer groups, for instance, to distribute
information. At the heart of this is the idea that villagers are best at giving information to other villagers in
ways that are interesting and make sense.
4. Limited monopoly
The provision of goods and services in KDP is privatized. Supplies for building infrastructure are procured
not centrally but through each individual village implementation team. Each team is free to buy materials
from whichever supplier they want, as long as they get three price quotations first and read the quotations
out at a village meeting. There is thus no monopoly over the provision of supplies in KDP. This keeps
costs down and also limits supply-side corruption and price inflation.
5. Limited discretion
Discretion refers to the extent to which any individual actor has exclusive control over his actions. Where a
single actor has wide discretion, corruption is made easier. For instance, if one person, or one small team,
is responsible for buying materials for a project, they can easily pocket project funds if they are not obliged
to account for monies spent to others.
Discretion is limited at most stages of the KDP project cycle by requiring all financial transactions to have
at least three signatures. This makes it difficult for any one actor to misuse money. The areas most at risk
of corruption are those where discretion is not limited enough: for instance, with the procurement of goods
for building infrastructure. KDP is trying to limit discretion in villages further by encouraging the
formation of village teams to check and verify any monies spent. Such teams would be composed entirely
of villagers and would be unpaid. This is to create the right incentives for joining the teams and so that the
teams do not become sites of material reward and potential corruption.
6. Accountability mechanisms
Built into KDP are mechanisms of accountability whereby those involved in running it have to account
downwards to villagers for all monies spent. The primary vehicle for this is the Musbangdes Pertanggungjawaban,
a village meeting to account for funds. Funds for KDP cannot be released in a single tranche to the
collective village accounts; instead, they are released in a staggered three-stage process. After 50% of the
project has been implemented, villages have to hold a Musbangdes Pertanggungjawaban. If villagers feel that
money is misused, the next stage of disbursement can be blocked until the problem is resolved. This
creates pressure on KDP participants not to misuse money and increases exponentially the chance that if
they do they will be found out.
Participants in KDP often do not take its accountability mechanisms seriously. However, a significant
proportion of cases reported to KDP have arisen because people entrusted with project resources have had
their account of how project monies were spent refused at a Musbangdes Pertanggungjawaban.
Socialization, transparency, accountability and follow up: the case of Tangga Batu Dua village
In Tangga Batu Dua, a village near Lake Toba in North Sumatra, villagers were suspicious that the village
implementation team had misused project funds allocated to build a clean water system. Villagers had learned about
KDP’s principles at the first KDP village meeting and had participated in forming the proposal for funding. They knew
which types of pumps were specified for the system in the original budget, and knew how much they should cost.
The villagers became suspicious, though, when the village implementation team came back with pumps of a cheaper
brand than that specified in the proposal. ‘We were suspicious that the money had been misused,’ said one villager,
‘[but] the implementation team didn’t want to discuss with the people what they done with the money. But we knew they
were supposed to. So we thought something was wrong.’
The villagers decided to do something about the problem. They organized themselves and formed an audit team. The
audit team checked all of the books of the village and saw that, indeed, there seemed to be some kind of discrepancy.
They saw also that the implementation team had been paying themselves a salary over and above their official
honoraria. According to the villagers’ calculations, approximately Rp 13 million in total had been misused.
At the village meeting to account for funds, the villagers presented their information and refused to approve the account
made by the village implementation team for how monies had been spent. The villagers fired the implementation team,
formed a new one, and decided that the old implementation team should repay the money. They tried to follow up the
problem also through KDP channels with the involvement of local government.
A month later, a decision was made at a meeting between the implementation team, the audit team, the kecamatan
facilitator and local government officials that the old implementation team should repay the money or suffer possible legal
Three months later, the money still had not been repaid. The villagers decided at this point to take legal action. At the
time of writing the report, the villagers had been advised by KDP facilitators to seek help from a legal aid foundation to
support their efforts to resolve the problem.
6. Monitoring & follow-up
The final element of KDP’s project design that works well to limit corruption is monitoring and follow-up.
KDP has a comprehensive monitoring system that is both internal and external. Internally, the project is
monitored by the villagers themselves, KDP field consultants, national-level KDP staff, local government
officials, the government’s Finance and Development Supervisory Agency (BPKP), and World Bank staff
on supervision trips. The project is also regularly scrutinized by independent external monitors, consisting
of local NGOs and AJI, the Association of Independent Journalists. All complaints are made public.
The Complaints Handling Unit at the KDP Secretariat follows up corruption cases and maintains a
comprehensive database of complaints. It does not have the capacity to investigate single-handedly every
corruption case that arises, though it endeavors to follow up as many as possible. Instead, it publicly
investigates cases at random. This sets a crucial precedent. When villagers and village elites see that
complaints are followed up, it breaks the cycle of years of impunity. It gives villagers a future incentive to
protest against corruption, and, by increasing the probability of punishment, tilts the cost/benefit
calculation of corruption away from corrupt behavior. When village elites know there is a good chance of
being caught, engaging in corrupt behavior becomes a much more risky enterprise.
The involvement of different, independent actors in the information and monitoring system breaks the
monopoly over information in KDP. The communities, KDP staff, NGOs and the Association of
Independent Journalists are independent channels of information, all of which can be cross-checked against
one another. However, despite their independence from another, these institutions form a coalition of
actors that can work in concert to support anti-corruption in KDP. This in turn strengthens civil society, a
crucial component of anti-corruption activities.
Bukit Kemuning: strengthening civil society through fighting corruption
In Bukit Kemuning, Lampung, a broad range of independent KDP actors was mobilized to work in coalition to resolve a
corruption case.
The local sub-district head (camat) was accused of distributing funds for the first installment of Cycle Two of KDP directly
to village heads and keeping a portion for himself. He was accused of calling six village heads to his office one-by-one to
give them their village’s ‘share’ of KDP funds, and of keeping Rp 20-30 million from each village as a cut to himself. One
of the village heads, however, sent out a general letter reporting that this had happened. He sent this letter to the local
newspapers and reported the case to the police. The other five village heads later sent letters of complaint too.
The local government coordination team, together with local KDP consultants, reported the case to the Complaints
Handling Unit in the National Management Consultants of KDP. At this stage, the governor of the province sent a letter
to the district head, the Bupati, asking him to resolve the problem or risk administrative sanctions. Later, however, it was
discovered that the bupati and the local head of the district attorney’s office were related to the camat. The KDP
facilitator was discovered to be implicated in the case and was fired, and the local Inspectorate investigated the case but
did little to resolve it.
Within a few months of the case being reported, there had been little progress on the case. So the NMC tried to garner
support for the case from a broad range of actors. First, it approached the local legal aid NGO, Lembaga Bantuan
Hukum (LBH) Lampung, to see if it could help. The six village heads involved got together and gave power of attorney to
this NGO to act on their behalf in prosecuting the case. And during the following month, LBH and one of the village
heads met with the police, the District Attorney’s office, the local parliament and press in order to press for action on the
case. In July, six months after the case had originally been reported, villagers from five different villages protested in
front of the district attorney’s office demanding action on the case. Finally, the district attorney agreed to call the camat in
for questioning. In June of 2002, after months of pushing by LBH and media interest in the case by local journalists, the
camat was found guilty of corruption by the local court, but at the time of writing this report had not yet been imprisoned.
Fighting corruption in KDP: making the design operational
The previous section of the paper discusses which features of KDP’s project design work best to prevent
corruption. This section discusses which strategies work best in the field to combat corruption.
1. Enforce sanctions
The intervention that contributes most to the overall success of KDP’s anti-corruption efforts is that
sanctions are enforced. The most effective tool that KDP has for doing so is that it can drop specific areas
from the project if widespread corruption is found and no serious attempt is made by the local government
to stop it. KDP funds commonly represent a significant stream of local revenue and, because the project is
heavily ‘socialized’ and it funds highly visible infrastructure such as small roads and bridges, KDP is highly
visible at the local level. The threat of dropping funds from a particular area thus carries some weight.
Sanctions are not enforced uniformly in KDP and there is room for the project management to make an
even more concerted effort to do so. Nevertheless, KDP has enforced sanctions in a few clear and highly
visible cases: the entire district of Jayawijaya in West Papua was dropped from the project, funds for the
entire province of South Sulawesi were suspended, and the district of Tapanuli Selatan in North Sumatra
was in the process of being dropped at the time of writing the report.
The evidence that these actions have had a serious impact, particularly on the perception of the boundaries
of corrupt behavior in KDP, is overwhelming. Project staff report that few things have enabled them to
convince local governments that KDP will not tolerate corruption more than the evidence that KDP has
followed through on its threats to cut funding elsewhere: in other words, ‘put its money where its mouth is’.
Enforcing sanctions credibly thus has a dramatic effect on the calculation by potential corruptors of the
risks of punishment for corrupt behavior.
Two aspects of this are noteworthy. First, it highlights the principle, widely accepted in the anti-corruption
literature, that ‘frying big fish’ is a particularly effective way of fighting corruption.35 Applying sanctions to
some important players breaks the cycle of impunity by showing that no player is immune from
punishment by virtue of being powerful. The second is more specific to KDP but is easily applicable to
other multi-level projects. KDP is able to apply sanctions at the level above that at which the offence has
taken place. For instance, if one district has been particularly corrupt, KDP can threaten to drop the entire
province from the project, which gives the provincial government an incentive to act to combat corruption.
In other words, the project is able to play with the incentives acting upon different parts of the project
structure in order to combat corruption, even when the ostensible interest of each part of the structure lies
not in fighting corruption but elsewhere (in this case, the provincial government’s interest in not losing
funding for all districts because of corruption in one of them). 36
2. Publicize
Publicity is crucial. It makes communities involved, raises the social costs of corruption, makes it more
likely that people will protest, and reduces the risk of protest. Indeed, most anti-corruption measures, such
as increasing sanctions, are useless without publicity. Publicity also increases the cost of failure. If
communities see that an anti-corruption drive is unsuccessful, it will make them even more convinced that
corruption is an unchangeable feature of life.
An important part of a successful anti-corruption drive is to start with something easy and visible.37 If a starting
point can be picked where the chances of success are good, it encourages people’s involvement and
35 See Corrupt Cities.
36 This is made easier by the fact that KDP is such a big project and that it carries political weight of its own.
37 This point is made by Klitgaard in Corrupt Cities.
strengthens the chances of success later on. Similarly, if a powerful and well-known figure is involved in
corruption and can be held publicly to account, it sends a powerful message that the anti-corruption
measures are not simply cosmetic.
Preventing corruption by powerful figures
In Kecamatan Palipi, North Sumatra, the new sub-district head (camat) was accused of ordering her subordinate, the
‘PjOK’, to attempt several types of corruption in KDP on her behalf. These included trying to get the camat’s husband a
place on the KDP financial management committee; cutting 15% of project funds so that they could be distributed
among local government officials; and using KDP funds to buy a cow to be slaughtered and divided among staff of the
camat’s office for Christmas.
The PjOK refused and reported the camat to KDP. This civil servant had been involved in KDP prior to the new camat’s
arrival. She said that she felt she could report the camat because in KDP, she knew there would be follow-up
“I knew that if I reported the problem it would go to NMC and there would be follow-up,” she said. “With KDP, it is the first
time that civil servants are taught that they can write a letter to report problems like this one. Before, who would be brave
enough to do this? Before, we just had to obey the boss… now it’s different. There has been a change. There has
been a big improvement.”
The camat denied the accusations. The PjOK said that it would be difficult to fire or press charges against the camat
because the only proof was her word. Nevertheless, members of the NMC, the KDP Secretariat and the World Bank
paid special visits to Palipi to try to resolve the problem and to warn the camat not to attempt any similar actions in future.
Community members in Palipi have also formed a special team to monitor the performance of the financial management
unit, consisting of the kecamatan secretary, NGO representatives, and local villagers.
When thinking of publicity, attention must be paid to its method. Distributing posters and posting
information on notice boards is often not the best way to reach people. For instance, in Jayawijaya district,
West Papua (Irian Jaya), facilitators told a KDP supervision team that very few people looked at
community notice-boards, partly because of low literacy levels and a strong traditional oral culture. Instead,
attention must be paid to what methods of publicity work best locally. In KDP, this has meant thinking
about how villagers lead their lives, how news travels organically in villages, and which points in villagers’
lives can be harnessed to distribute information. The strategy arising from this has focused on a few key
areas. First of all, it has focused on using media appropriate to the community. This includes not just the
type of medium, but also the way the information is conveyed in it. For instance, KDP encourages local
groups to produce posters in local languages and showing people from local ethnic groups. Second, it has
focused on training village facilitators well. Facilitators are usually well-known village figures that have good
networks of information and contacts. Third, it has focused on using pre-existing, local institutions to
spread information. This means using prayer meetings, savings groups and other gatherings to distribute
3. Involve local stakeholders
The key to combating corruption in the field is to involve local participants in the anti-corruption drive – in
KDP’s case, villagers. Once villagers are involved, it creates a momentum for resolving the corruption
problem much bigger than ever could be sustained by any external body. It multiplies the anti-corruption
push. Also, when local stakeholders are involved, they resolve problems in ways that are meaningful to
them rather than in ways that simply accord with project rules. For example, many locally pursued
corruption cases in KDP are not followed up through the law but through local adat conflict-resolution
mechanisms. Sometimes, after doing so, villagers will decide that a case is resolved even though the
accused has not repaid all the stolen money. They may decide, for instance, that the accused has suffered
social sanctions so serious that the case may be considered resolved. In these case, it doesn’t matter that all
the money has not been repaid; what matters is that villagers have gone through the process of combating
corruption and have reached a solution that is meaningful in their local context and that acts as a deterrent for
future corruption. This is what constitutes sustainability.
Local resolution of problems
In Rao Mapat Tunggal, a village in West Sumatra, villagers accused their village implementation team of misusing KDP
funds allocated for building a village road. Villagers were suspicious because, although bought rocks for the road were
specified in the original road proposal, they knew that the rocks and earth used for the road had simply been taken from
the land surrounding the road site. So they suspected that some of the money had been pocketed. Furthermore, only
10% of the road’s budget remained, but the road was far from being 90% complete.
The villagers decided to take action to resolve the problem. With the help of the village head, they organized a village
musyawarah. At this discussion, they elected a team to get data and follow-up the problem. They also informed KDP.
Although the implementation team admitted misusing money at the first meeting, the problem took a long time to resolve.
There were no fewer than 13 special village meetings to resolve the problem. After about four or five meetings, the head
of the implementation team ran away from the village. By this time, it had been decided that he was the main culprit of
the misuse of funds, since the other members of the team were not fully involved in its actions.
After several rounds of debate, the villagers decided to finish building the road through contributing their labor for free.
‘At first some of the villagers didn’t want to do this,’ said one of the village officials, ‘but in the end they thought it was
better to finish the road rather than to go round and round in circles trying to get the money back. They knew that if they
really waited for this, the road would not get finished’.
After building the road, they considered whether they should nevertheless pursue the problem through the law. But after
some consultations, they decided not to, because by this time, their ‘emotions had died down’ and the road was
complete. Furthermore, the main culprit feared the severity of his impending social sanctions so much that he had left
town. When asked why, we were told that he would have suffered complete social ostracism, to the extent that he would
not have been able to participate in village musyawarah, parties, weddings and other celebrations¾indeed, ‘all the
activities.. that people have’. The villagers thus considered the case closed.
4. Piggyback onto local institutions
Involving local institutions in the anti-corruption drive is an important element in making it sustainable. If
the anti-corruption strategy is able to involve local conflict-resolution mechanisms, such as adat law systems,
and get local religious and cultural leaders involved, it has a greater chance of success. Again, this is because
it is more likely that the solutions reached will be meaningful in their local context. It also gives the
communities a sense of ownership over the anti-corruption drive, which is crucial to ensuring sustainability.
At the very least, an anti-corruption strategy will have difficulties if it attempts to work against local
If we have problems here, we usually follow it up through the family. We also follow problems up through the adat
system. The way it works is that we get the families of the clans together – the old people – and then we have a
musyawarah. People hardly ever follow up their problems through the law. We don’t do it because we feel united. We
don’t want to go to the law because we are all one family here. The adat system works well for what happens on the
ground. The legal system is what works for the country.
– Villagers in Pagar Batu, North Sumatra
There is one caveat. It should not be assumed that because institutions are local they are democratic,
beneficial to poor people, and benefit women or other marginalized groups. Often they are hierarchical,
benefit the powerful, and work in ways to encourage corruption. With this in mind, however, if they can be
involved in the right ways, it is more likely that the anti-corruption drive will succeed and be sustained.
5. Understand local power dynamics and attempt to get allies
Successful anti-corruption strategies start with an understanding of local power dynamics. It is important
to identify leaders and try to involve them in the anti-corruption fight. It is also important to understand how
local power dynamics work to see if they can be exploited for the end of anti-corruption: are local leaders,
for instance, concerned about securing a political power base? Could they benefit from being publicly
associated with an anti-corruption drive?
In KDP, anti-corruption strategies have worked best when those fighting corruption have had allies. These
allies need not necessarily be from government or be high up, but it helps. Where sanctions have been
applied successfully in KDP, especially against local government officials, they have usually come from
higher levels of government. If the ‘hearts and minds’ of leaders can be won over, the anti-corruption
strategy has a much greater chance of success. KDP’s experience with fighting corruption has been that
once government officials are won over, they tend to become anti-corruption champions, even though
their material interests are threatened. This paradox can perhaps partly be explained by a recognition on
the part of government officials of Indonesia’s changing political context and a desire to secure a broad
power base. Local rivalries can sometimes work to aid the fight against corruption:
Rivalries and other social factors
In Hutaraja village, North Sumatra, the village head reported a corruption case to KDP. The village head suspected that
members of one of the village councils, the LKMD, had misused money allocated to build a road in the village. After
reporting the problem to KDP, villagers formed an audit team, which found that approximately Rp 6 million seemed to
have been misused. At the community meeting to resolve the problem, villagers decided that members of the family of
the village council head should be forced to finish building the road using their own free labor, to compensate for the
funds that seemed to have been misused. But this wasn’t enough for the village head, who reported the problem to the
police and to the district lawyer.
What motivated the village head? We don’t know precisely. But the kecamatan facilitator told us that in Hutaraja there
was a certain degree of rivalry between the clan of the village head, Sibagariang, and the clan of the village council head,
Hutauruk. The decision of the village head caused some conflict. Most members of the Sibagariang clan supported
the village head, an members of the Hutauruk clan were opposed, though not without some overlap. The facilitator drew
us a symbolic map of the village, showing who was for and against the action of the village head:
In this case, rivalry among different village clans meant that the corruption case was pursued more vigorously than it
would have been otherwise. Interestingly, though, many of the villagers did not agree with the law being involved. “The
villagers wanted to keep the problem in the village,” said the facilitator. “It’s their culture. And it’s the same among all the
Bataks, where clan ties are very strong. People don’t like going to the law – the clan would be ashamed if one of their
own was reported to the police. Clan members don’t like it when there seems to be misuse of money, but they also don’t
like going to the law.” This is an example of the relative importance of social sanctions over legal sanctions.
= Sibagariang (clan of village head)
= Hutauruk (clan of village council head)
Modeling corruption: analysis of a case in KDP
A well-known model helps us to break down the elements of corruption. According to the World Bank,
‘the opportunity for corruption is a function of the size of the rents under a public official’s control, the
discretion that official has in allocating those rents and the accountability that official faces for his or her
decisions.38 Discretion here refers to the exclusive power the official has to allocate the rents, a function of
whether others are involved and how many checks and balances there are. Accountability refers to the
likelihood that the agent will be caught and held responsible, a function, at least at local level, mainly of
transparency and information.
This is a version of the well-known Robert Klitgaard model
where the level of monopoly power and amount of discretion encourage corruption, and where the level of
accountability discourages it. ‘If someone has monopoly power over a good or service and has the
discretion to decide whether someone gets that good or service or how much a person receives, and there
is no accountability whereby others can see what that person is deciding, then we will tend to find
corruption,’ he argues.39
The model tells us that we tend to find corruption where monopoly is high, discretion is high and
accountability is small. However, it should be used as a rough guide only, not an equation. It is hard to
measure monopoly, discretion and accountability, and this formula is not intended to be quantifiable.
Instead of treating it as a mathematical or economic formula, we should treat it simply as a guide for
helping us to break down the different elements in any given corruption case. Here is an example. of how
to do it.
Tangga Batu Dua : the story analyzed
The case of Tangga Batu Dua, where villagers took to task the implementation team accused of misusing money to build a
clean water system for the village, lends itself usefully to analysis. The corruption was two-fold. First, the implementation team
bought a brand of water pump and other materials cheaper than that specified in the original proposal. Second, the team
members paid themselves a daily salary in addition to their official project honoraria.
There are two calculations involved: first, the calculation on the part of the implementation team that they could get away with
misusing the funds, and second, the analysis of the case itself, and which elements of the project made it impossible for the
team to misuse the funds. The first involves a calculation of benefits and costs, based on the risk of discovery and punishment.
The second depends on the level of monopoly, discretion and accountability in the project.
1. Benefits and costs: the village implementation team’s calculation
The incentives calculation of the village implementation team is a function of their perception of the costs and benefits attached
to corrupt behavior rather than the ‘actual’ costs and benefits, so any analysis of incentives is speculative. Nevertheless,
various factors could have come into play:
(a) Benefits. The benefits, or the size of the ‘rents’ under the control of the implementation team, were relatively substantial.
38 World Bank, Helping Countries Combat Corruption: the Role of the World Bank, p.12
39 Klitgaard, Corrupt Cities, p. 27
Rp 93 million had been allocated for the village’s KDP projects, Rp 84 million of which was for the materials and labor for
building the four water units. By buying cheap materials and keeping the difference, and by paying themselves an
additional salary, the team was accused to have collectively misused Rp 12 million among its five members.
(b) Costs. It is unlikely that the members of the implementation team would have perceived the costs of stealing the money
to be high. The five members of the implementation team were the village head; the village council (LKMD) head, and
three others. None of these are classified as KDP staff so, unlike, facilitators, they cannot be fired by KDP. They thus
faced three types of sanctions: (1) for the village- and village-council head, administrative government sanctions for
attempted corruption, such as being fired; (2) for all, legal sanctions; and (3) for all, social sanctions.
· Administrative sanctions. With a long history of government impunity for corruption, it is unlikely that the village
head and village council head would have feared the enforcement of administrative sanctions.
· Legal sanctions. In view of (a) the high cost of pressing charges; (b) the difficulty of proving corruption; (c) the
complications and delays of a weak judicial system; and (d) the widely-held and accurate perception that police
officers and judges can be bought-off easily by the highest bidder, it is unlikely that the threat of legal sanctions would
have been taken seriously by members of the implementation team.
· Social sanctions. All members of the team were relatively powerful within the village and so it is unlikely, given the
history of impunity for corruption in villages, that they would have viewed with alarm the threat of social sanctions.
Here, though, they may have made a miscalculation, not having taken into account the extent to which villagers
believed that they could protest against corruption in KDP. The culture of the area in North Sumatra, Batak culture, is
reputed to be relatively egalitarian compared with other traditions in Indonesia, so it is more likely that the culture of
impunity rather than a culture of social deference would have prompted the team not to take seriously the threat of
social reprisals.
(c) Probability of being caught. The village implementation team recorded their extra salary payment in the village books
and defended these payments on the grounds that their official honoraria were supposed to cover certain tasks such as
arranging meetings only, and not day-to-day activities associated with the project, for which they claimed they were
entitled to payment. So it seems that with regard to these specific payments, they felt justified and so did not make a
calculation based on being caught. It is unclear what they thought with regard to the supposed pocketing of funds through
buying cheaper materials than those specified in the original plan. The team members would have been aware that they
would have to account for how monies were spent at the Musbangdes Pertanggungjawaban (village meeting to account
for funds). The most likely scenario is that the implementation team was overconfident that they could hide the misuse of
money and, because of their experience of development projects in which there was a culture of impunity, underestimated
the extent to which they would be forced to explain their actions.
2. Monopoly, discretion and accountability: the case disaggregated
(a) Monopoly. What was the level of monopoly power over the goods and services involved in the case?
There are two sites of control. One is the monopoly any single supplier had over the provision of goods and services to
the implementation team. Two is the monopoly the village implementation team had over providing those goods to the
· Provision of goods to the implementation team. Here, monopoly was limited. Part of KDP’s design is to prevent
monopoly by privatizing the provision of goods and services. No single supplier can provide materials for KDP;
rather, village implementation teams choose from the supplier offering the cheapest price for the materials.
· Provision of goods to the village. Although no single member of the implementation team had a monopoly of any
kind, the implementation team as a unit had a monopoly over providing those goods to the village. No other unit
could provide the service. Given the small scale of villages, though, it might be unrealistic to expect that two
implementation teams could be formed to perform the same service and bring bids back to the village so that
villagers could choose between the bids: the likelihood is that the members of each team would all know each other
well, precluding effective competition.
(b) Discretion. To what extent did the implementation team have discretion over their actions?
The members of the implementation team had high discretion over their actions. There were only five members of the
implementation team, so it would be relatively easy for the five to collude in misusing funds. No other team of people was
charged formally with checking prices, and there were few other checks and balances over the actions of the
implementation team.
(c) Accountability. What was the level of accountability involved? The level of accountability in this case was ultimately
relatively high, which is why the case was brought to the fore. Accountability here was a function of (1) socialization, (2)
participation, (3) transparency and information (4) the checks and balances.
· Socialization . Villagers in Tangga Batu Dua were relatively well-informed about their rights in KDP and felt that if
they reported a problem, something would happen:
Ten years ago, people were brave in reporting problems, but there was never a solution. So in the end we simply
used to keep our mouths shut. But the kecamatan facilitator told us that this problem was not like [previous village
development projects].. he told us that the project had to be run according to certain procedures, and if it didn’t, it
counted as misuse. That’s why we want to report problems. Also, we know there will be a solution if we report
problems. That what the kecamatan facilitator said… the facilitator told all the villagers about it. Everybody knows
about the project – Villager
· Participation. There was a relatively high degree of community participation in the project, which translated into a
high degree of ownership and thus a demand that it should work well. Each KDP village meeting was apparently
attended by at least 40 people. All villagers we met were fully versed on the details of the case and all expressed an
interest in resolving it.
· Transparency and information. Although the village implementation team was not transparent about how they had
spent project funds, villagers knew that the team should be transparent according to project rules, so they became
We were suspicious that the money had been misused because the implementation team didn’t want to discuss with
people what they had done with the money. But we knew they were supposed to¾we had learned about KDP at
the first village meeting. So we thought something was wrong. – Ibu Sitorus, farmer.
Villagers had information about project procedures, their rights, and what to do if they suspected misuse of funds.
They were also able to get information about prices of materials, a crucial element of their accusation that the
implementation team had misused the project funds. Information thus played a major role in strengthening
· Checks and balances. In Tangga Batu Dua, villagers were able to use the Musbangdes Pertanggungjawaban as a
check on the implementation team. This is a village meeting at which the implementation team must account to
villagers for monies spent. The checks and balances built into the project design thus played an important role in
limiting corruption.
III. Recommendations for improvement
Recommendations for improvements are in four parts :
1) Introduce measures to change the underlying system of incentives for actors in KDP
2) Introduce measures to limit monopoly, clarify discretion, and increase accountability for the project as a whole.
3) Introduce measures to improve training and information for KDP staff
4) Form a strategy to take into account the future context in which the project will operate.
Political assumptions
The recommendations are based on certain assumptions about the context in which the project continues
to operate:
· Weak institutions. The institutions of state remain weak, in particular the ability of the state to
enforce contracts. This means that, to a large extent, politics continues to be based around patronage.
The recommendations are thus based around incentives and information rather than agreements and
· Weak judicial system. The judicial system is weak and subject to widespread corruption. The
recommendations thus de-emphasize the law as a means of resolving corruption problems, though
acknowledge that access to legal aid should be widened if people wish to pursue their problems
through the law.
· Systemic corruption. There is widespread corruption in all institutions of state. This includes
especially the high level of administrative corruption in the civil service. Civil servants continue to be
paid an extremely low salary, and are expected to supplement their incomes by having second jobs and
through various types of ‘informal payments’. None of the recommendations are thus premised upon
a lower level of corruption in government.
· Relative freedom of the press. The recommendations are based on the assumption that the press is
freer than during the New Order period of government and can be harnessed to publicize corruption
cases. However, the report recognizes that although press freedom has increased substantially at the
national level, there is still a great deal of ‘money politics’ in journalism, and regional media especially
are still subject to elite influence and intervention.
· Increased political freedoms. The recommendations assume that ‘civil society’ is more free than
during the New Order period, and that to a limited extent, barring conflict zones such as Aceh, space
has opened up for political protest. They thus assume that if villagers see a problem and have
information about it, they will be able to harness their institutions to protest.
The key premise of the report was that corruption is primarily a problem of incentives. It argued that the
most important part of any sustainable anti-corruption strategy was to change the underlying system of
incentives such that agents were no longer motivated towards corrupt behavior. It suggested that anticorruption
strategies that focused simply on morals or on strengthening regulations were unlikely to be
This is even more compelling in an environment where corruption is systemic and the judicial system has
almost no power to enforce sanctions. In a system where corruption is widespread, the level of honesty
needed for any individual actor to take a moral stand against corruption is high, because honesty simply
does not pay.40 In a system where the judicial system is unable to enforce contracts and regulations, it
makes little sense to make them the focus of an anti-corruption strategy. Indonesia has such an
The findings from our fieldwork support our hypothesis. At the heart of KDP’s anti-corruption strategy
should be an emphasis on incentives. At each stage of the project cycle, this means instituting reforms
· decrease the benefits of corrupt behavior
· raise the costs of corrupt behavior
· close the loopholes for corruption
· increase the possibility of being caught and held accountable
The recommendations made in the main body of this report are kept general enough for them to be
applicable across different projects. Concrete recommendations relating exclusively to KDP were made
separately. Those stemming directly from loopholes in the project cycle can be found in the body of
Annex II, which consists of a detailed breakdown of different stages of the project in terms of the costs and
benefits of corruption for the main actors involved.
Two themes emerged from the incentives analysis of the project and the fieldwork: the effect of
‘socialization’, or information-spreading, on how villagers perceive their interests, and the importance of
social norms as a motivation for action. Both need to be addressed when making recommendations for
‘Socialization’, in KDP terminology, is the process through which communities learn about KDP’s
procedures, their rights, and what to do if they have complaints about the project. It consists of a long
process of information-dissemination that starts about six months before project proposals are submitted.
During this process, an emphasis is placed on anti-corruption. Facilitators stress to communities that KDP
40 Bardhan points out that the idea that corruption is an example of frequency-dependent equilibria is common in the literature. Here, the
‘expected gain from corruption depends crucially on the number of other people we expect to be corrupt’. If all are honest, it does not pay to be
corrupt, whereas if all are corrupt, it does not pay to be honest. Bardhan, “Corruption and Development: A Review of Issues”.
is ‘different’ from previous village development projects, that it ‘belongs to them’, that corruption will not
be tolerated, and that they have some redress if they suspect it.
Villagers have to take this in good faith. With the long history of impunity for corruption, there is no
reason for them to think that KDP will be any different from previous village development projects. This
makes it crucial that it is indeed different: if it is, it sets a precedent for the future; if it isn’t, it sets into stone
a culture of apathy and cynicism. Villagers are thus taking a gamble when they believe that KDP is
‘different’ and that they can protest against corruption in it.
One of the most remarkable things we observed during our field research was how much of a gamble
villagers were willing to take. Information played a major role in their taking it. It had a direct effect on
how villagers and potential corruptors perceived their interests. When villagers possessed information
about their rights and, crucially, when the potential corruptors knew that villagers had this information, it
raised the perceived cost of corrupt behavior and reduced the cost of fighting it. Where villagers possessed
information about standard KDP honoraria and average prices of materials, they had the information to
begin to prove corruption. And where villagers had information about how to make complaints in KDP,
they were able to do something about corruption.
One of the most important ways to change the incentive structure in KDP is thus to improve flows of
information about the project.
Recommendation 1: Improve information-dissemination in KDP
· Encourage innovations. KDP should encourage kecamatan facilitators to come up with innovative ideas through which to
spread information about KDP and fight corruption, focusing on using locally-relevant means of spreading information.
· More use should be made of local institutions as a means to disseminate information. Facilitators need to make more use
of local prayer meetings, arisan (savings groups) and other social gatherings as a means of spreading information. They
should exploit local, oral means of communicating.
· More use should be made of religious leaders as a means of spreading information, such as priests or ulama (Islamic
religious scholars), so that they incorporate information about KDP into lectures and sermons. Facilitators should contact
the Majelis Ulama, church groups and other local religious organizations to begin to do this.
· Facilitators should submit proposals to KDP for how they might be able to harness local cultural forms as a means of
spreading information about KDP or corruption. For instance, they might be able to make announcements at or weave
stories about fighting corruption into local dance performances, or, where it is popular, dangdut (a particular type of popular
music) songs.
· In their monthly reports to KDP, facilitator should detail the various ways that they have attempted to spread information.
These inputs should form part of the facilitators’ evaluations.
Social sanctions & rewards
The second theme that emerged from our fieldwork was the significance of social norms. Social norms
played an important role in how actors perceived their interests and in their weighting of sanctions. This
was evident in the explanations people gave for action; in many cases, when asked why someone did a
particular thing, the answer would be ‘because he felt malu (ashamed)’. It was evident also in the way
villagers framed their dislike of corruption. Most often, it was expressed in terms of the transgression of
some kind of social code rather than in terms of an individual wrong: ‘we don’t like corruption here’, rather
than ‘corruption is wrong’.
Much has been made of the lack of shame attached to corruption in Indonesia and of the effect of Javanese
culture upon norms attached to corruption. The argument usually made is that Javanese culture¾and,
through its hegemonic effect on the rest of the country, ‘Indonesian’ culture¾is particularly conducive to
corruption because it values family ties over state ties and because its conception of power is hierarchical
and based on patrimonial rule. This argument is so common that it deserves mention when discussing
This report does not examine in detail the strength of such cultural arguments. It believes, however, that
despite their initial plausibility, their explanatory power is limited at best. Such arguments do not explain
the wide incidence of corruption across different regions and time periods. Nor do they explain the
variance in corruption between areas whose cultural norms are similar. And, as Pranab Bardhan has
pointed out, such arguments can be tautological, claiming that corruption tends to be found in places
whose cultural norms favor corruption.41
Furthermore, this report has found that, in Indonesian villages, there is some shame attached to corruption
and indeed a strong current of anti-corruption feeling. Although at the national level it may seem that there
is no shame attached to corruption, villages are small and personalized enough for social sanctions to carry
weight. When people did not wish to take action against corruption, this was because of a political culture
borne of impunity rather than anything ‘perennial’¾‘why bother doing anything,’ was the argument, ‘when
nothing will happen anyway’. When people believed they could do something, they framed punishments
and rewards in social terms: the social meanings of the goods at stake mattered, not the goods themselves.
The second recommendation is thus to find ways to work social norms into the KDP incentive structure.
Recommendation 2: Work with social rewards and sanctions to make the incentive structure less conducive to
· Facilitators should investigate how local social norms can be harnessed to increase the costs and decrease the benefits of
corrupt behavior.
· Adat review by local working groups for problem-resolution. Local KDP working groups in districts should be asked to
come up with inputs on what kinds of adat systems are in place, how strong local adat systems are, what the main
features of the systems are, and how such adat systems could work with KDP to support people’s efforts better to resolve
their corruption cases. The next KDP training should establish a forum to exchange these ideas.
41 Bardhan, “Corruption and Development”.
Incentives to fight corruption
Facilitators in KDP often face an environment hostile to whistle-blowing. Facilitators depend on KDP and
not local power dynamics for their power base, so it is unlikely that such hostile environments will make
them afraid per se to report corruption cases. It does, however, mean that the cost of fighting corruption is
increased, so facilitators might not pursue cases as vigorously as they would do otherwise. The right
incentives thus need to be in place not only to deter corrupt behavior in KDP but also to encourage
facilitators to prevent and fight corruption.
Recommendation 3: Increase the incentives for facilitators to fight corruption in KDP.
· Link performance to rewards. KDP needs to link the performance of facilitators in combating corruption to rewards without
simply encouraging the under-reporting of corruption. This could be done by linking performance not with the corruption
cases themselves but with an indicator known to be connected with corruption. For instance, they could be rewarded for
‘best practice’ of some kind: how well have they done in disseminating information about KDP to villagers? To test it, KDP
could hold a KDP pop quiz competition in each district, where winning villages could receive a prize and the facilitators a
· Institute bonuses for innovative ideas. The bonuses, preferably non-monetary, should be large enough to warrant the
extra effort involved and stimulate innovation.
· Raise the moral costs of corruption. Facilitators should sign an explicit anti-corruption code of conduct, pledging not to
engage in corrupt behavior themselves and to work against it wherever possible.
Monopoly, discretion and accountability
The second set of recommendations concerns making structural changes to the project in order to limit
monopoly, clarify discretion and strengthen accountability. It follows Robert Klitgaard, who argues that a
key element of a successful anti-corruption strategy is to ‘restructure the principal-agent-client relationship
to weaken monopoly power, circumscribe discretion, and enhance accountability’.42
Weakening monopoly
As noted, the provision of goods and services in KDP is privatized so that, at least in theory, there is
competition in the provision of supplies. Sometimes, however, village implementation teams circumvent
this by buying materials from their friends or relatives rather than from the supplier with the cheapest price.
To avoid this, the regulation that in KDP at least three quotations must be read out publicly to villagers
should be enforced.
42 Klitgaard, Corrupt Cities.
Recommendation 4: Institute measures to limit monopoly.
· Facilitators should enforce the requirement that at least three quotations (unwritten if less than Rp 15 million and written if
more than Rp 15 million) are obtained for supplies. The requirement that these are read out at village meetings should be
enforced. Furthermore, facilitators should include these quotations in their monthly reports. If they cannot provide them,
the burden of proof should be on them to explain why.
Limiting discretion
KDP needs to institute certain measures to limit the discretion of any single participants in KDP, from
KDP staff to implementation teams. The anti-corruption literature suggests that limiting discretion is a
more effective way of preventing corruption than instituting regulations. Widening the range of actors that
have any say over or knowledge of any particular transaction makes it much harder to be corrupt and, in
particular, makes it harder to be collude. Any cost/benefit calculation over whether to collude depends on
the probability that the other party to the collusion will agree and will not report the attempt. It thus
depends crucially on personal ties and trust. Such expectations can be deliberately upset if the pool of
actors with a say is widened. And unlike instituting extra rules, widening discretion involves little extra
Recommendation 5: Institute measures to clarify discretion.
· Rotate KDP project staff. Project staff in KDP should be rotated periodically to different locations. If KDP project staff
are tempted to attempt to collude with each other, village elites or local government officials, they must be fairly sure of
agreement to avoid the possibility of being reported and sacked. Staff rotation will lessen the possibilities for this. It will
also ensure greater quality control over KDP: good facilitators will be able to transfer their skills to kecamatan which have
had a worse facilitators, and villagers are more likely to notice poor skills on the part of facilitators if they have had
experience with good ones. The right administrative level of rotation should be picked: rotating staff simply across different
kecamatan and kabupaten rather than across different provinces would achieve the rotation goals effectively while
avoiding the danger that not enough KDP staff will agree to be rotated.
· Enforce the recommendation that unpaid village verification teams are formed to count money and check materials.
This is simple, effective and increases village participation. Ensuring the formation of verification teams should be made
part of facilitators’ evaluation. If they are not formed, the burden of proof should be on facilitators to explain why.
· Institute bottom-up and cross-evaluation of KDP. Bottom-up evaluation could take the form of a score-card evaluation
by villagers of the project. The evaluation would have to be done in such a way that it did not create extra layers of
administration or extra opportunities for corruption, for example through bribing villagers to give good scores. Crossevaluation
could take two forms. Kecamatan facilitators could make random visits to other kecamatan to evaluate
performance. Alternatively, villager-composed audit teams could conduct spot audits of other villages. This idea has been
mooted already by the KDP II preparation team. Advice from facilitators should be sought at the next large KDP training
sessions on how such bottom-up and cross evaluation could work in practice. Close attention should be paid to avoiding
extra administration.
The most important element in making a fight against corruption sustainable is to institutionalize
accountability mechanisms. By far the most important elements of accountability in KDP are the internal,
social controls. Accountability in KDP is a function of socialization, information, participation and
transparency. Measures to improve socialization and information-dissemination have already been
Measures should be instituted to improve participation. Close attention should be paid to the participation
of marginalized groups especially, to avoid elite capture of the project. This is especially important at the
early stages of the project.
Recommendation 6: Institute measures to improve participation
· Kecamatan facilitators should work with local leaders to get them to buy into KDP and anti-corruption as a means of
distributing information and getting local ownership of the project. When doing so, they should think about the interest
local leaders would have in participating in such a project, and the constituencies the leaders can reach.
· Facilitators should focus especially on the participation of marginalized groups, such as poor women, and say how they
intend to do so in their reports to KDP.
Transparency is at the heart of KDP’s anti-corruption strategy. KDP bookkeeping is already fairly
transparent. However, transparency still needs to be improved in KDP.
Recommendation 7: Institute measures to improve transparency
· Enforce the recommendation that unit price lists of materials are published. These give villagers the information they need
to detect and pursue corruption, attack one of the stages in KDP most at risk to corruption. Almost every case we visited
was one in which villagers complained because there was some discrepancy between what materials ostensibly cost and
what they should have cost.
· Posting a price list should be made part of the evaluation of the kecamatan facilitators and kabupaten-level technical
consultants. If it is difficult to get information about prices, the burden of proof should be on facilitators to explain why.
Training and information
Throughout the history of the project, KDP facilitators have been the most important agents in fighting
corruption. The importance of facilitators, emphasized already in this report, needs to be stressed again.
Facilitators are the point of contact between ordinary villagers and the project and are often the lynchpin
for attacking cases of corruption. They need to be supported better in their efforts to fight corruption.
Recommendation 8: help KDP facilitators fight corruption by improving training and information
· For cross fertilization learning purposes, a group of some of the best facilitators should be picked as an ‘anti-corruption
talent pool’. These facilitators could work for two months a year alongside facilitators in kecamatan with high levels of
corruption to help them and transfer some of their skills and information. These facilitators could be rewarded.
· There is a clear need for training to be improved. KDP facilitators need to know how other facilitators have solved their
problems so that they can learn by example rather than by being taught. Training needs to be made more participatory.
Future strategy
One of the changes created by the decentralization legislation will be a realignment of the relationship
among governors, district heads, local politicians and their constituents. Governors and district heads will
no longer be appointed but will be elected by and accountable to regional assemblies (DPRD). This makes
the position of DPRD members much stronger. DPRD members are elected, but indirectly by party
affiliation: parties and not individuals are elected. This creates an interesting context for KDP: since
projects with less corruption such as KDP are cheaper than previous projects and have visible results, it
may be in a party’s political interest to champion them and to champion anti-corruption activities. But
because parties and not individuals are elected, the incentive for any individual DPRD member to act in
such a way is not as strong as it could be: direct accountability to voters is not there. Parties will thus have
to find favor with constituents in a way that parliamentarians will not. Nevertheless, compared with the
previous system, power will less concentrated at the top.
But there are caveats. Sub-district heads (camat) continue to be appointed and, according to the new laws,
are still accountable not downwards but upwards to the district head.43 Furthermore, as with BPD
members, the costs of running for office at DPRD level may create a huge incentive either to misuse local
funds or to use the period in office to regain some of the costs of running. This could thus create
incentives that motivate actors towards corruption. Thus, although decentralization introduces changes,
the ways in which they affect incentives for corruption have yet to make themselves clear.
The last recommendation is thus that KDP should think systematically about the incentive structures
created by decentralization: how do they affect the environment for corruption and how?
43 See Article 66, Law 22/1999. From Law 22 text.
Recommendation 9: Institute measures to anticipate the future context of the project.
· Decentralization incentives review by local working groups. Local project working groups should come up with an initial
outline on what kinds of post-decentralization scenarios they think are likely to emerge with their districts, what the
incentive structures created are, and whether there are any specific local problems to be borne in mind such as conflict.
The groups might want to consider the following factors:
· What are the main institutions of local government? Who are the main players?
· What are the main political parties and what proportion of seats does each party have?
· What is the power base of each party? Are they trying to appeal to rural voters? Is it in their interest to
champion a project such as KDP?
· How do people get political information? Has the way in which people relate to local government
changed because of decentralization?
· How active and independent are local NGOs, the media and pressure groups?
· Is the local government perceived as being reform-minded?
· What has the local government’s response been so far to decentralization? Is it seeking to implement
new laws? How will they affect KDP?
· Village level governance map by village facilitators with the help of kecamatan facilitators. If facilitators are able to, they
should come up with some sort of map of local power, with attention paid to who the local players are and how they relate
to each other. This might constitute a list of BPD members and holders of village office, an outline of powerful families in
the village, and how the players relate. (This would help kecamatan facilitators to see whether heads of borrower groups
are all members of village elites).
· Facilitators should report on whether their districts have any intention of instituting their own administrative or legal codes
that are different from the national ones. News of this kind should be reported as a matter of course. Bonuses should be
given for innovative ideas in harnessing local systems for anti-corruption.
Concluding thoughts
The primary purpose of this report was to see how corruption could be prevented in a large decentralized
community development project. Its key assumption was that corruption was primarily a problem of
incentives, and could only be fought sustainably by changing the costs and benefits of corrupt behavior. It
stressed that any analysis of the costs and benefits of corruption needed to start with an informationgathering
exercise from the field that took into account local context and local norms.
The report attempted to do this for KDP. Through field research and an analysis of incentives in KDP, we
attempted to get a sense of the way corruption works in Indonesian villages and how it can best be
prevented. Our findings supported our assumptions. We found that the best way to prevent corruption in
a project such as KDP, especially in a country with weak institutions and systemic corruption, was to
emphasize information, participation and incentives: in other words, to turn ordinary people into their own
agents of anti-corruption.
Anti-corruption shouldn’t be considered an end in itself but instead an element of a wider reform process.
KDP’s anti-corruption efforts fit well into a wider program. KDP itself is a governance program as well as
a poverty-alleviation program. KDP is in approximately one in four villages across Indonesia. This means
that it is both at the forefront of reform and powerful enough to have an influence. Indeed, there are signs
that the government likes the program and is propelling it further. The government itself, for example,
instituted an idea that KDP work on legal aid linkages for poor communities so that they have access to
advocacy services to fight corruption. This is an important marker of reform.
Such things are milestones of Indonesia’s reform. Indeed, the process of Indonesia’s reform is developing
fast. But we do not know where it is heading. There is also a great deal of backsliding, and it is very
difficult to make predictions about where Indonesia will be five years from now.
Indonesia is changing. This report claimed that two aspects of this change, political change and
decentralization, were likely to have a major effect on village life in Indonesia. It argued that political
change might be opening up avenues for further protest and that, because of ambiguities in the law, the
effect of decentralization on village governance was uncertain.
KDP needs to manage that uncertainty. In particular, it needs to think seriously about how decentralization
will affect incentive structures at village, sub-district and district level. The main change at village level, as
noted, will be the formation of village representative boards (BPD) that have the power to ask for the
removal of the village head. This could work to strengthen accountability in village government and
increase sanctions for corrupt behavior. On the other hand, the necessity of obtaining campaign finance
may increase the incentive on the part of BPD candidates to engage in corruption.
Similarly, political change and decentralization change incentive structures at sub-district and district level.
It has been noted that, because it is at village level, KDP is sheltered from some of the major national
changes taking place, which all are at higher levels of government¾KDP is ‘below the radar screen’ at
which most of the changes are taking place.44 This is true. But nevertheless, some of the changes even at
higher levels affect the corruption environment lower down.
Such immense uncertainties can mean that it is difficult to plan projects. But it also gives us opportunities.
One thing it emphasizes is that, if reform is to be sustainable, it must be conducted at the behest of people
themselves, rather than from the outside. This is the only way the momentum of Indonesia’s change can
be channeled in the longer-term towards reform. In such an uncertain environment, enabling people to
take control of the things that affect their lives is the only thing that can be done. It places an emphasis on
the micro, on letting people take their own decisions. Corruption, in a community development project, is
about starting with this micro-level, with the question, ‘what motivates people?’ Governance, too, is about
starting at this grassroots level.
This translates into an emphasis on transparency, information, linkages and access. KDP is attempting to
provide these things. The question asked at the beginning of this report was whether a program such as
KDP could channel the momentum of democracy into concrete changes into how village decisions were
made. It is too early to say whether KDP is doing this, though the beginnings look good. One measure of
how well it is doing will be how many more complaints there are from villagers about it, how many more
problems arise. This will mean that KDP will have achieved its most important objectives, strengthening
community institutions and improving local governance.
44 Opening Remarks to World Bank Board for preparation of KDP II, Scott Guggenheim, June 2001
Annex I
Historical background
Corruption is made easy in villages by a combination of bureaucracy, hierarchy and lack of accountability.
Much of this is a function of history. Dutch colonial policies acted to promote paternalistic patterns of
government, strengthen the position of the village head, and turn administrative positions into sites of
material reward. For instance, the coffee- and sugar- cultivation system called the cultuurstelsel, where a set
part of village land was set aside to produce export crops for the Dutch government, used local officials as
middlemen between village producers and the colonial government. These middlemen were paid
percentages on crop deliveries, and could extort enormous amounts of money from the exchange.45 Public
office thus became a site of material reward.
The same was true of the New Order. The New Order’s policies towards rural areas were characterized by
two conscious strategies: ‘betting on the strong’ and de-politicizing the rural ‘floating mass’.46 They had the
effect of increasing the incentive to exploit public office as a means of getting richer and of cutting the
heart out of popular opposition and control.
‘Betting on the strong’ involved attempting to co-opt rural elites by giving them privileged access to state
resources through village office. Village heads were able to profit from state land through the system of
tanah bengkok, whereby a portion of village land was set aside for the sole use and profit of the village head.
Although the size and quality of this land varied, in some villages it could be extremely lucrative. Village
officials were also given preferential access to state power, and discretion over bureaucratic procedures and
their accompanying bribes. This strategy had a dual function. By operating through traditional local elites,
the New Order was given some legitimation in villages. But by making it advantageous for those elites to
co-operate with the State, potential sources of opposition to New Order rule were co-opted. This enabled
the New Order regime to preserve its hold on power. But it also opened up opportunities for enrichment
through state office. And, since official salaries were low and because it cost so much to run for office,
pressure was high on village heads to use their term in office to amass as much money as possible.47
Another New Order self-preservation strategy was to treat the rural population as a ‘floating mass’, depoliticized,
docile and concentrated upon economic advancement. The New Order curtailed political
activity in the countryside, and promoted the notion of ‘family-ness’ (kekeluargaan), whereby a community
was seen not as a locus of political conflict but as a family, with problems resolved through consensus and a
father-son relationship between villagers and the village head. Thus problems were supposed to be
resolved through musyawarah without any attention to the power dynamics lying behind consensus, political
protest was discouraged, and latent ethnic, social and religious conflicts were suppressed.
45 See Ricklefs, M. C. A History of Modern Indonesia since c. 1300. Stanford: Stanford University Press, 1981
46 See Antlöv, “Village Leaders and the New Order” for a discussion of ‘betting on the strong’.
47 For a discussion of the cost of running for office, see Hüsken, “Village Elections in Central Java: State Control or Local Democracy?” Hüsken
found that candidates spent “amounts ranging from Rp. 15 million to Rp 90 million [then USD 8,000 to 50,000]” and that most had to rely on
money-lenders and relatives to put up the large loans needed. After a few weeks, to pay back the loans, some were “busily trying to find other
sources of finance: borrowing money from the village funds, renting out communal lands, or selling village property… The government
regulations for local reform, aiming at “rationalized” and “professionalized” village administration, might therefore have accomplished quite its
opposite by bringing out a system which forces village bureaucrats either to go bankrupt or to become corrupt.”
Annex II
Corruption map of the KDP project cycle
In this section we present the breakdown of the cost/benefit analysis of incentives for corruption in the
KDP project cycle. The cycle is broken down into eight stages.
Stage 1: Socialization & dissemination of information
Opportunities for corruption: LOW
· Workshops
· Training of kecamatan facilitators (FKs)
· First inter-village kecamatan level meeting (UDKP 1)
· First village meeting (Musbangdes 1)
· Selection of village facilitators (FDs)
· Training of village facilitators (FDs)
· Information boards placed in villages.
Selection of village facilitators
There is an opportunity here for nepotism. The monetary incentives are not high: the amount of money
the village facilitator earns is reasonably low, and the amount of work the facilitator must perform is high.
However, non-monetary incentives are attached: the position carries with it a great deal of prestige and it
is perceived to be an avenue of influence over the project. In year one, village facilitators were effectively
appointed, which meant that village elites tried to gain influence through getting friends or relatives selected.
In year two, the procedure was changed. Village facilitators are now selected by a two-stage competitive
process: candidates are elected by villagers, and the kecamatan facilitator & government kecamatan-level
project manager (PjOK) then make an appointment. Since this change there have been very few reported
cases of nepotism.
of information
Socialization in hamlets and
activity suggestions collected
Selection of activities for
village proposals
Feasibility/verification stage
(maintenance, repayments)
Preparations for
Funding decisions (selection
of proposals to be funded)
Stage 2. Socialization in hamlets and activity suggestions collected
Opportunities for corruption: LOW
· Inventory of village groups & poorer villagers
· Sub-village & group meetings
· Special meetings with women’s groups
· Updating information on village boards
· Further dissemination of project information
No money changes hands at this stage, so the opportunities for corruption are low. However, village elites
and other actors may see this stage as an opportunity to gain influence over the project, primarily in order
to obtain later kickbacks. They may try to intervene to determine what suggestions are proposed.
This applies to other stages of the KDP process. At any stage, although there may be no immediate
monetary incentives, elites may try to gain influence as a means to secure future kickbacks.
Stage 3. Selection of activities for village proposals
Opportunities for corruption: HIGH
· Second village meeting (Musbangdes II)
· Village technical assistance (TA) for proposal preparation selected
· Formation of team to prepare village proposals
· Preparation of proposals with designs and budgets
· Updating information on village boards
Selection of village technical assistance for proposal preparation
· Opportunity for nepotism/kickbacks
The official process for selection of village technical assistance is as follows: The kabupaten-level technical
consultant (KM-Kab) has a list of people pre-certified to give technical assistance. The people on the list
should be qualified. Villagers choose a candidate from that list and pay him/her market rates. The choice
is approved by the KM-Kab. The person chosen has to be from outside the village barring special
The process too often does not work the way it should. Unqualified people get on the KM-Kab list,
villages choose technical assistance even when they don’t need it, under-qualified friends/relatives of the
village elites are chosen, or candidates may offer kickbacks in order to be chosen. The root of the problem
is that often the KM-Kab is reactive and either does not have a list of qualified candidates or simply puts
names of candidates onto the list without checking their qualifications.
The problems with selecting technical assistance for proposal design are the same as the problems with
selecting technical assistance for project implementation. However, the amounts of money involved for
implementation are much higher and so the problem is correspondingly more serious.
Incentives & disincentives for main actors:
Kabupaten technical consultant/engineer
The incentive for the kabupaten engineer to put
unqualified people on list, barring a genuine lack of
locally qualified people, is low unless there are
kickbacks involved. If there are kickbacks, the
incentive is the amount offered.
The disincentives are high: the kabupaten engineer
can be found out and fired easily, and the facts are
easy to check. The provincial-level consultant
(KM-Prop) can monitor the kabupaten engineer.
incentive is the amount offered.
This finding translated into certain measures being recommended to KDP to ensure that only qualified
candidates made their way onto the list of candidates.
Technical assistance
The incentives for a potential technical assistance
candidate to offer kickbacks in reward for being
chosen are high: he/she has the opportunity to
make money through this.
The disincentives are small. If a candidate
attempts to offer a bribe to get picked for technical
assistance and it gets turned down, it’s extremely
unlikely that the candidate will suffer either major
social approbation or any official sanction. The
probability that he/she would be charged with any
crime is almost zero. The only disincentive is that
the bribe might be accepted but that there would
be no result.
Details of candidates for technical assistance for the kabupaten should be posted on information boards
so that villagers can see who is available for what type of project. If villagers can see that there are TA
candidates with qualifications appropriate to the village’s project, they are much less likely to accept an
unqualified candidate, regardless of whether the candidate is from that village.
The incentive to be involved in collusion here is
fairly small, except where kickbacks are involved.
There have been some cases of kecamatan
facilitators receiving kickbacks for the selection of
village-level technical assistance. This is especially
easy when kabupaten-level supervision and control
is weak.
The disincentive is that the kecamatan facilitator
would be fired if discovered. However, unless
some involved admit collusion, it can be difficult
to prove.
The incentive for a villager to consent to having
an unqualified person as TA for proposal
preparation is not monetary, but springs from a
sense of village loyalty when the unqualified
person is from the same village, or a desire to
avoid conflict.
The disincentive is that if a proposal is bad it will
not be funded.
The question here is whether the disincentive in this case outweighs village loyalties. This will depend
on how well the kecamatan facilitator has done his/her job in convincing villagers that only good-quality
proposals will be funded – in other words, that there will be competition and that funds will not simply
be divided up among villages in the kecamatan.
PjOKs/local government staff
The incentive to be involved in collusion
regarding technical assistance again is low, except
as far as potential kickbacks are concerned.
The disincentives for involvement are, however,
also low. Disincentives may be lower than for an
FK, depending on how seriously corruption
allegations are taken by the local government
apparatus. In kecamatan where it is taken seriously,
the disincentives to participate will be high; they
include the risk of punishment. In places where
such sanctions do not exist, the disincentives for
involvement are low.
Village elites
The incentive for village elites to collude in
arranging for a friend/relative to act as village TA
is reasonably high, as they receive social benefits
(gratitude from a friend) and possibly even
material benefits (in the case of kickbacks). The
incentive for them to consent to an unqualified
person from the village being chosen over a
qualified person from another village is also high:
they can avoid being seen as supporters of
outsiders to the village and thereby avoid conflict
The disincentives for them to consent in
collusion are the threat of villagers’ anger if
discovered. This will be higher in places where the
kecamatan facilitator has done a good job in
socializing villagers. It will be lower in places that
are hierarchical. The disincentive is also that the
village proposal gets turned down, but in this case
the cost is borne over the whole village and not
just the individual. The disincentive for village
elites to consent to having an unqualified person
and preserve their power base.
from that village be chosen over a qualified person
from another village is villager anger if they don’t
co-operate: they may be seen as disloyal. This will
be lower in cases where the kecamatan facilitator
has been able to explain to villagers that only
competitive proposals will be funded and that
therefore the quality of technical assistance is of
paramount importance.
This finding translated into a recommendation that the burden of proof be shifted so that villages
needed to make a good case for hiring technical assistance from someone not from the approved list. It
was recommended that the option be dropped and granted only if villages presented a special case to
Formation of team to prepare village proposals
Some of the same factors exist: village elites may see this as an opportunity to gain influence over what gets
proposed, not for immediate monetary gain but for the opportunity of obtaining kickbacks as a subsequent
Preparation of proposals, with design and budgets
There are opportunities here for collusion, budget mark-ups, kickbacks, formation of false groups
(a) For infrastructure:
· Here the opportunity for corruption is for collusion among the project-preparation teams to inflate
budgets. However, it is far more common in KDP for implementation teams to substitute cheaper
materials than those specified in the original budgets and pocket the difference than for project
preparation teams to inflate original budgets.
· The incentive to inflate an original budget is reasonably low: a project that looks too expensive might
not be approved at the second kecamatan-level inter-village meeting (UDKP II), so if the team wants
the proposal to succeed, which is a prerequisite for any funds to flow, they need to make it seem as
cost-effective as possible. The incentive will be made even lower when the kecamatan facilitator has
done a good job in convincing all project participants that the KDP process is indeed competitive and
that only good-quality, cost-effective proposals will be funded
(b) For economic activities:
· Here the opportunity for corruption is for false or new groups to be formed to borrow money. The
incentives to do vary across actors, but are reasonably high across the board.
Incentives & disincentives for main actors:
The incentive for facilitators to collude in forming
false groups is low: they do not benefit directly,
and they run the risk of being accused of doing
their jobs poorly if any supervision team finds that
there are many false groups. However, they may
be under pressure from communities to allow new
groups to form proposals, since villagers may
argue that there simply aren’t any pre-existing
groups of the kind specified. This is common.
The disincentive for facilitators is the risk of
being discovered and fired.
Village elites
The incentive to form false groups (where false
groups mean not simply new groups but groups
where the bulk of the money goes to the group
head—often a member of the village elite—and
then each member gets a smaller amount) to get
money is high: they can potentially profit relatively
substantially. The incentive will be lower in
villages where the kecamatan facilitator has done a
good job in socializing KDP’s principles to all
The disincentives for them to attempt this are
low. There are no formal sanctions and it is
unlikely that any of these cases will be followed up
through the law/government. The main
disincentive is local and social: if villagers discover
that they could have profited more by following
the rules, they may become angry and enforce local
(probably social) sanctions. This is more likely to
happen in places that are politicized, have a low
level of hierarchy, and where levels of socialization
are high.
The incentive for them to participate in forming
entirely false groups, where they receive a smaller
proportion of money than the village head, is
reasonably high if they are unaware that they
would receive more money otherwise – i.e. if there
has been bad socialization. The incentive for
them to participate in forming new groups is
extremely high: if they don’t have pre-existing
groups and they nevertheless follow the rules, they
have no chance whatsoever to gain access to KDP
There are few disincentives, unless it is made
clear to villagers that the process is competitive
and that any group that seems likely to be false is
unlikely to receive project funds, and unless it is
made clear to them that sanctions for nonrepayment
of loans will be applied rigorously.
funds. So, unless villagers want infrastructure,
there is a structural tendency towards the
formation of false groups.
This finding translated into a recommendation that much more emphasis be placed in KDP on the
avoidance of false groups, especially through the kecamatan facilitators.
Stage 4. Feasibility/Verification Stage
Opportunities for corruption: LOW
· Formation of verification team to see if proposals are technically feasible
· Field visits & inspections
· Team meetings to compile results and recommendations
· Feedback to villages & groups
Formation of verification teams
Members of verification teams are appointed at kecamatan level by the kecamatan facilitator and kabupatenlevel
technical consultant, and are paid a salary from ‘operational funds’ based on local market rates and
their qualifications. The possibility for nepotism and kickbacks does exist here, but there have been
relatively few cases of it. Where there are cases of it, it is for reasons unrelated to corruption, such as a lack
of locally-trained candidates.
This is because the monetary incentives for kickbacks are relatively small, and it is easy for this kind of
corruption to be found out and exposed at UDKP II level.
Field visits & inspections; team meetings to compile results & recommendations,
feedback to villages & groups
There is a formal possibility here for kickbacks so that verification teams judge proposals more easily. The
main opportunity here, though, is for non-monetary influence.
· If the verification process works the way it should, the incentive for villages to try to influence the
verification teams should be very low: the verification teams are supposed to be giving villages advice
on how to improve their proposals, not simply saying whether their proposals are good enough to be
approved or not. However, often the process does not work the way it should, and verification teams
simply review proposals and give a verdict of ‘layak’ (feasible) or ‘tidak layak’ (not feasible). Because of
this, and because villages know this, a strong incentive is created for villages to influence the
verification team to pass verdict in their favor. The incentive is enormous: if proposals don’t get rated
‘feasible’, the village will not get any money.
This finding translated into recommendations to improve the verification process, including a
recommendation that a small review team in villages simply sign and state that they were satisfied with the
work of the verification team.
Stage 5. Funding Decisions (Selection of Proposals to be Funded)
Opportunities for corruption: LOW
· Second kecamatan inter-village meeting to select projects for funding and to select financial management units (UPK).
Second kecamatan inter-village meeting
Here there are few opportunities for straightforward corruption because the mechanism involves so many
people and is so transparent. However, there are opportunities here for the circumventing of competition
– for simply dividing money up among villages.
This finding led to the recommendation that more attention be paid to preventing the simple division of
money at this stage of the project cycle. For the third year of KDP, changes to original budgets are
forbidden at this stage, to avoid this problem.
Selection of financial management units (UPK)
Members of financial management units are selected by the second kecamatan meeting by election in an
open forum, which makes the level of discretion involved virtually non-existent. The opportunity for
nepotism is thus fairly low, though not unknown.
Stage 6. Preparations for implementation
Opportunities for corruption: HIGH
· 3rd village meetings (Musbangdes III) to discuss results of UDKP II, elect community implementation team (TPK) and
decide on village technical assistance
· Updating information on village boards
· Training for implementation teams, UPKs, village technical assistance (TA)
· Administration and release of funds
· Release of funds: from the state treasury in 3 stages (40%, 40%, 20%), from collective village accounts as needed (not
by pre-set %)
3rd village meetings: election of village implementation team, decide on village technical
There are opportunities here for nepotism, especially for the purposes of later collusion. The same factors
apply as those above for the selection of village technical assistance in proposal preparation, but the
amounts of money involved are higher. We have experienced more problems with selecting technical
assistance for implementation than in project preparation.
Administration & release of funds
The process for the release of funds is as follows: First, 40% of the total allocation for the kecamatan gets
released from the state treasury to the kecamatan-level collective village bank account (that should be in the
name of the collective account of the village council (LKMD) – ‘Rekening Kolektif LKMD’). In the villages’
budgets, there should be a rough schedule of when they need what funds. Based on this, the kecamatan
facilitator makes a collective schedule for the villages: for instance, in Week 1, Rp. 10 million gets taken out
in total, Rp. 2 million to go to Village A, etc. Every withdrawal from the bank needs three signatures: the
kecamatan facilitator, the head of the financial management unit, and the head of one of the LKMD. The
cash then is given to the LKMD head (Ketua 1 LKMD), who disburses it to his/her village’s
implementation team.
Opportunities for corruption:
(a) From the State Treasury Office (KPKN) to village collective bank accounts
We have not experienced many problems here. This is because the mechanism is simple and it would be
immediately noticeable if some was cut. In other words, the process is transparent and discretion is limited.
All actors know how much money should be transferred into the bank accounts, and the kecamatan
facilitator, financial management unit head & village council head are able to see how much money goes
into the bank accounts. There are few opportunities for collusion among the KPKN and the trio of the
kecamatan facilitator/financial management unit head/village council head because there is little contact
among them and, because accounts have to be made transparent, they would be easily discovered.
(b) From collective bank accounts to village council head
The opportunity here is for collusion among the kecamatan facilitator, financial management unit head and
village council head so that each gets a cut of the money before it goes down to the individual villages.
There are relatively few cases of this. This is because each actor has limited discretion (three people are
involved) and, unless both of the others agree, each runs the risk of being reported. Even if they do agree,
it is easy for them to be discovered – unless there is collusion with the bank also, it is easy for a village audit
team to check how much was taken out of the bank, and it should be possible for the teams to check how
much money the village implementation team received, if the implementation team keep records and also
do not collude. Thus each actor’s motivation depends on how certain they are of agreement with the
others, how big the amount of money involved is, and the sanctions he/she will suffer if found out.
Incentives & disincentives for main actors
Kecamatan facilitator
The incentive for the kecamatan facilitator to
collude is small. Their salaries are relatively high
and the sanctions for corruption are high (being
sacked). The incentive will be made even smaller
if sanctions are enforced all the time and this is
publicized among facilitators.
The disincentive is the risk of being discovered
and fired.
Head of financial management unit
The incentive for the UPK head to collude is
bigger than the kecamatan facilitator’s because the
honorarium is smaller and the sanctions for
corruption are lower.
The disincentive is the possible risk of social and
other local sanctions.
Head of village council (Ketua 1 LKMD)
The incentive for the head of the village council
to collude is small: it is in interest of the village
council head to take a cut of the money not at this
stage, where the kecamatan facilitator & UPK head
are involved, but at the subsequent stage, where
there are fewer checks and balances.
(c) From village council head to village implementation team
There are two opportunities for corruption here: (i) for the village council head to keep a cut of the money
before giving it to the village implementation team or (ii) to collude with the village implementation team in
keeping a cut of the money.
Village council head taking a cut: incentive is medium.
· The incentive of the head of the village council here is reasonably high: it is measured by the amount
of money he will stand to gain. The incentive is higher when the village council head does not follow
the KDP process and instead deals directly with suppliers and/or works with the village
implementation team to do so; in this case it is easier for him to take a cut of money without it being
· The disincentives are the possible social sanctions if discovered and the threat of being fired or forced
to resign. This in turn rests on how open the village is and how well the kecamatan facilitator has done
his or her job in socializing villagers and ensuring transparency. For instance, if village implementation
teams know how much money they should be receiving, they are unlikely to accept less without
(ii) Head of village council colludes with implementation team for a cut of money: incentive is
· The incentive for each actor is the amount of money each stands to gain
· The disincentives are the threat of sanctions if discovered. The head of the village council runs the
risk of being fired or forced to resign. Village implementation team members also run the risk of being
fired by villagers and thus losing their honoraria. Their risk of being discovered depends on the
kecamatan facilitator’s success in socializing villagers, and could be increased considerably if the small
village checking teams (‘Victors’, in unofficial KDP terminology) are set up.
Stage 7. Implementation
Opportunities for corruption: EXTREMELY HIGH
· Mobilization of village laborers
· Procurement of materials and equipment (less than Rp. 15 million demands 3 unwritten quotations, more than Rp 15
million demands 3 written quotations)
· Implementation of village activities, loans disbursed
· Supervision, monitoring, reporting and monthly meetings of facilitators
· Village meetings to account for funds (when about 50% of funds expended).
Mobilization of village laborers
Opportunity for corruption here is for the village implementation team to take a cut of the village laborers’
· The incentives are medium: they are measured by the amount of money the village implementation
team stands to gain
· However, the disincentives are high because the risk of being found out is high (many laborers are
involved but it only takes one laborer to report a problem to cause difficulty) However, this depends,
though, on the kecamatan facilitator’s success in socialization and transparency, and in how well the
books of the village are kept. Often, information in the books of the village implementation team is
kept in a confusing style: for instance, they may specify how much each labor ought to receive each
day, how many days worked and the total amount allocated, but fail to specify how many laborers are
involved. Such ‘opacity’ clears the space for corruption.
Procurement of materials and equipment; implementation of village activities; loans
The opportunity for corruption here is extremely high and cases are extremely common. This is the
most common type of corruption in KDP. The process is supposed to work as follows: the village
implementation teams receive money from the village council heads to procure materials and equipment.
Materials and equipment should already be specified in the village’s original proposal. The implementation
team is responsible for buying only and exactly the materials and equipment specified, at the best possible
price. For this end, they need to get three quotations (written if the materials are worth more than Rp 15
million, unwritten if not) from three different suppliers and announce the bids at an open village meeting
before picking the one with the best price.
Village implementation teams almost never do this. The most common case of corruption is when they
substitute cheaper materials for those specified in the budget and pocket the difference: for instance, they
may substitute smaller pipes than those specified, or a different type of wood. They may also collude with
the supplier so that the supplier issues them with false receipts.
When questioned about their activities, village implementation teams often reply that it is difficult to find
the materials that are specified in the budget, so they are unable to get three quotations or may have to
make substitutions. This in itself does not pose a problem. The problem arises when substitutions are
made without the prior consent of villagers, without technical assistance to advise whether cheaper
materials can be substituted, and without informing villagers of both the changes and the amounts of
money involved.
A second major source of corruption is when village implementation teams do not use direct suppliers at
all, but instead contract all the work out. This is explicitly against KDP rules and creates an extra layer of
cost, inefficiency and corruption.
Incentives and disincentives for main actors:
Village implementation team (TPK)
The incentive for members of the
implementation team to collude & take a cut of
project funds is high. Their discretion is low,
because there are at least three members of the
TPK, but accountability is also low – they need
only explain their actions at the village meeting to
account for funds (Musbangdes Pertanggungjawaban)
and, since most villagers are not experts either on
the market price of goods or of different types of
equipment, it is easy to explain away discrepancies
in terms of changed prices or lack of supply. They
have a large amount of money at their disposal –
up to 40% of project funds, or more in cases when
the village needs to buy an expensive piece of
machinery – and can potentially profit quite
The disincentives are the perceived sanctions that
they would suffer if discovered, combined with the
risk of discovery. These sanctions are usually local
and outside legal channels; however, there may be
fairly stringent social and/or adat-related sanctions.
The main mechanism by which they can be
discovered is through the Musbangdes
Pertanggungjawaban, held after 50% of funds have
been spent, where the implementation teams have
to account for their actions, and where villagers
sometimes refuse to accept the implementation
team’s version of events. If the FK has been
successful in socializing villagers, if original budgets
are made transparent and if price lists of materials
are made public, it hugely decreases the incentive
for the implementation teams to skim off funds
and increases the risk that they will get caught.
The incentive for suppliers to collude is high: they
may receive a cut of funds, and there is virtually no
risk that they will be discovered or that, if
discovered, they will suffer any sanctions.
Stage 8. Post-implementation stage
Opportunities for corruption: MEDIUM
· Formation of maintenance teams
· Training for village councils and maintenance teams
· Repayment of loans and re-lending of repaid funds, on instructions of kecamatan meeting
· Maintenance of infrastructure
· Evaluation
· Village meetings for hand-over of completed projects, to account for funds used.
Repayment of loans and re-lending of repaid funds:
(a) Repayment of loans
The process for the repayment of loans is fairly flexible. Villages are expected to set up their own systems
for managing repayments. Officially, each borrower group is collectively responsible for paying its loan
back to the financial management unit. The official agreement for paying back the loan is usually with the
group head. In year one, funds had to be repaid directly to the financial management unit, who then could
re-lend the money to other borrower groups from different villages in the kecamatan. However, this often
acted as a disincentive to repay: low inter-village loyalties meant that the social incentives to repay were low,
and often borrowers did not believe the repaid funds would stay in the kecamatan or that they would not be
misused. So for year two, the system was changed. It was agreed that, with the approval of the UDKP II,
funds could be paid back to the village and they could be revolved to borrower groups from that village
only. The role of the financial management unit in this case would be not to collect but to repayments.
With both systems, there are few sanctions for non-repayment of loans.
· There are some opportunities for misuse of funds. For instance, heads of borrower groups may collect
repayments from group members but not repay all of the funds to the financial management unit.
Where the kecamatan facilitator has done a good job in ensuring transparency, the incentive to do so
will be small, as members of borrower groups should be able to see the total amount of money that
should have been repaid to the financial management unit versus the amount they have collectively
paid back to the group head.
· The main point of leakage, though, is not straightforward misuse of funds but leakage due to the
cyclical effects of corruption. There are few official sanctions in KDP for non repayment of funds.
One of the incentives for villagers to repay money is supposed to be that the funds can then be
revolved to other borrower groups in the same village or kecamatan. This is supposed to create social
pressure to repay funds. However, repayments are very low. In part, this is because villagers are used
to corruption, and believe that if they repay all the money they owe, it will simply get misused.
Incentives and disincentives of main actors:
Heads of borrower groups
The incentive for heads of borrower groups to
misuse funds is the amount of money they would
stand to gain from keeping a portion of funds.
The disincentives are possible social sanctions if
discovered, coupled with any village-instituted,
local sanctions. They depend on the risk of being
found out, which will be higher in places where the
kecamatan facilitator has done a good job in
ensuring that the repayments system is transparent
– in this case, villagers will know how much is
supposed to be paid.
There is a small possibility that borrowers may collude with each other and/or the group head in not
repaying money.
The incentive is the amount of money they save
by not repaying.
The disincentives depend on what kinds of
sanctions, if any, exist for individual borrowers for
non-repayment of funds. Where sanctions apply,
they are much less likely to collude.
The disincentives depend on what kinds of sanctions, if any, exist for individual borrowers for nonrepayment
of funds. Where sanctions apply, they are much less likely to collude. The most likely leakage
scenario, though, is not a concerted collusion effort, but rather that borrowers will be not feel a compulsion to
repay if they think the money will be misused once paid back.
(b) Re-lending of repaid funds
Once KDP funds have been repaid once, the funds pass out of the KDP remit. Possibilities for corruption
once funds have been repaid are high. Funds are supposed to be managed by the financial management
unit so that they can revolve and be lent again to villagers. The more times this happens, the looser the
controls become, and the more scope there is for corruption by all actors: the financial management unit,
kecamatan-level government officials, village officials, and heads of borrower groups. Government auditors
seem interested only in the initial project investment, not in revolving funds.
Because funds at this stage have passed beyond KDP’s remit, it does not strictly concern KDP.
Nevertheless, it is worth mentioning because the possibilities for leakage are so high. If KDP has worked
as well as it should – and as well as it can – then the possibilities for corruption even at this stage should be
small: as more and more villagers participate in the project and know KDP processes, the less likely they are
to accept the misuse of money.
Maintenance of infrastructure
There are opportunities for corruption here. Maintenance of infrastructure often involves the collection of
fees. For instance, villagers may charge user fees for a road built with KDP funds, or they may collect fees
from vendors for the maintenance of a KDP-built market. The fees may then be misused and managed in
a non-transparent way.
The management of revolving funds and the long-term maintenance of infrastructure is outside the KDP
project cycle and thus the strict remit of KDP project staff. However, because KDP’s objectives are not
simply to alleviate poverty but also to strengthen community institutions and improve good governance,
this finding nevertheless translated into a recommendation that KDP staff should consider the creation of
sustainable, transparent systems of managing revolving funds and maintaining infrastructure as important
an objective as the initial setting-up of loans and building of infrastructure. It was suggested that their tasks
and evaluation criteria should reflect this.
Selected reading list
Antlöv, Hans and Cederroth, Sven. Leadership on Java. Surrey, UK: Curzon Press, 1994.
Antlöv, Hans. Village Governance in Indonesia – Past, Present and Future Challenges. Paper presented at the
PERCIK conference “Dynamics of Local Politics in Indonesia”: Yogyakarta, July 2000.
Asia Foundation. Survey Report on Citizens’ Perceptions of the Indonesian Justice Sector – Preliminary Findings and
Recommendations. Jakarta: Asia Foundation, 2001
Bardhan, Pranab. “Corruption and Development: A Review of Issues,” Journal of Economic Literature., Sept.
1998, XXXV, pp.1320-1346.
Chandler, David P, et al. In Search of Southeast Asia. United States: University of Hawaii Press, 1987.
Economist Intelligence Unit. EIU Country Profile 2000: Indonesia. London: Economist Intelligence Unit,
Evers, Pieter, Resourceful Villagers, Powerless Communities (Rural Village Government in Indonesia). World
Bank/Bappenas Local Level Institutions Study, DRAFT, 2000
Geertz, Clifford. The Religion of Java. Chicago: University of Chicago Press, 1976.
Gross, Bruce; van Wijk, Christine; and Mukherjee, Nilanjana. Linking Sustainability with Demand, Gender and
Poverty. Water and Sanitation Program: December 2000.
International Crisis Group. Indonesia: Keeping the Military under Control. International Crisis Group: 2000.
Retrieved August 7, 2001
International Institute for Democracy and Electoral Assistance (International IDEA). Democratization in
Indonesia: An Assessment. Stockholm, 2000.
Kecamatan Development Program. Second Annual Report 1999/2000. Kecamatan Development Program
National Secretariat: September 2000.
Kecamatan Development Program. Studi Teknis Sarana/Prasarana. Jakarta: Kecamatan Development Program,
Klitgaard, Robert; Maclean-Abaroa, Ronald; and Parris, H. Lindsey. Corrupt Cities. Oakland, CA: Institute
for Contemporary Studies Press, 2000.
Mauro, Paulo. “Corruption and Growth,” Quarterly Journal of Economics, Aug. 1995, 110 (3), pp. 681-712.
Oldenburg, Philip. “Middlemen in Third-World Corruption: Implications of an Indian Case,” World Politics,
Jul. 1997, 39 (4), pp. 508-535.
Partnership for Governance Reform in Indonesia. A Diagnostic Study of Corruption in Indonesia. Jakarta:
Partnership for Governance Reform in Indonesia, 2002
Project for the Establishment of an Anti Corruption Commission for Indonesia. A Review of Current
Initiatives of Anti-Corruption in Indonesia. Jakarta: De Speville and Associates, 2000.
Ricklefs, M. C. A History of Modern Indonesia since c. 1300. Stanford: Stanford University Press, 1981.
Rose-Ackerman, Susan. Corruption and Government: Causes, Consequences and Reform. Cambridge: Cambridge
University Press, 1999.
Schwarz, Adam. A Nation in Waiting: Indonesia’s Search for Stability. St. Leonards: Allen & Unwin, 1999.
Transparency International. Corruption Perception Index 2001. 2001. Transparency International. Retrieved
July 1, 2001
United Nations. Common Country Assessment: Indonesia. Jakarta: United Nations, 2001.
World Bank. The Anti-Corruption Handbook. Washington, DC: Social Policy Secretariat, Regional Office of
the Vice-President for East Asia and the Pacific (EAPVP) and East Asia Environment and Social
Development Unit (EASES), 2000
World Bank. Anticorruption in Transition: A Contribution to the Policy Debate. Washington, DC: World Bank,
World Bank. Helping Countries Combat Corruption: the Role of the World Bank. Washington, DC: Poverty
Reduction and Economic Management (PREM) Network, 1997.
World Bank. Helping Countries Combat Corruption: Progress at the World Bank since 1997. Washington, DC:
Operational Cores Services (OCS) and Poverty Reduction and Economic Management (PREM) Network,
World Bank. Indonesia Country Assistance Strategy FY 2001 – 2003, Report No. 21580-IND. Washington,
DC: Indonesia Country Management Unit, East Asia and Pacific Region; International Finance
Corporation, East Asia and Pacific Department, 2001.
World Bank. Indonesia Policy Performance Assessment. Jakarta: World Bank, 2001.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: