JURNAL KEUANGAN PUBLIK

JURNAL KEUANGAN PUBLIK
Vol. 5, No. 1, Oktober 2008
Hal 19 - 53

FISCAL DESENTRALIZATION
AND INDONESIA REGIONAL INCOME DISPARITY
(1994-2004)


Iman Widhiyanto 1




Abstract

The interest of fiscal decentralization is increasing among developing and
transition countries over past decade. In Indonesia the ‘big bang’ of fiscal
decentralization was started in 2001. There are two opposite theories concern with
fiscal decentralization. The conventional theories state that applying fiscal decentra-
lization will increase regional disparity and lessen economic growth. While the new
theories state that fiscal decentralization not only reduces regional disparity but also
increases economic growth. This paper estimates the relationship between fiscal
decentralization, economic growth, and regional income disparity in Indonesia during
period 1994-2004. The comparison data before and after fiscal decentralization are
employed. The beta and sigma convergence of economic growth among regions are
part of the analysis. Empirical evidence found that there is economic convergence after
fiscal decentralization while before decentralization there is economic divergence.
Fiscal decentralization in Indonesia encourages economic growth and alleviates
regional income disparity during period of observation.

Keywords: fiscal decentralization, regional disparity, regional development, economic
growth, Indonesia.


1 Staff of Directorate Cash Management Directorate General of Treasury Ministry of Finance Republic
Indonesia


Jurnal Keuangan Publik Vol. 5, No. 1, Oktober 2008
20
INTRODUCTION

The interest in fiscal decentra-
lization is increasing among developing
and transitional countries. The reason is
that fiscal decentralization is believed as
an effective tool to increase the efficiency
of public expenditures. Another reason is
that fiscal decentralization is sometimes
seen as a panacea for reaction to the
failures of centralization system over
past decades particularly in developing
and transitional countries. Over past
decade, most of developing and transitional
countries have either embarked upon or
stated their intention to embark upon
some types of fiscal decentralization as
an engine of economic growth. In East
Asian countries, a tendency towards
decentralization is underway in almost
every country. While in Indonesia the
‘big bang’ decentralization program
approach was applied in 2001.
The pressure of decentralization
has generally come from both internal
and external drivers. From internal
drivers, these pressures were supported
by country’s historical and experiences,
deepening democratization, increasing
levels of public services, countries
disintegration phenomena, responding to
failure centralization system and so on.
While from external drivers, the pressure
could come from country and/or institu-
tional donors who have influenced the
country recipient’s policies due to
decentralization.
Many literatures have pointed
out that fiscal decentralization maybe
dangerous in developing and transitional
countries. It causes regional governments
expanded their expenditures while
externalizing cost to others (Jonathan
Rodden 2002). Fiscal decentralization
also could increase regional disparities
based on traditional view. It was
underlined the fact that central govern-
ment’s power to redistribute income
among regional governments is higher in
the centralization system than in the
decentralization system. In contrary, new
theories argue that the benefits from
decentralization are increasing efficiency
and reducing regional disparity because
by implementing decentralization go-
vernment system, the regional govern-
ment will be pursued to increase their
own efforts in providing better public
services in its region. In line with
increasing better public services, the cost
of service delivery also will increase. On
the other hand regional governments
face limited budgets. Therefore they
have incentive to collect intensively their
potential incomes from their own
regions. Finally, the disparity among
regions after decentralization will be
smaller according to this alternative
approach.
The disparity of provincial income
per capita in Indonesia is relatively
severe and will outline the debate even
though there is a sign of convergence of
provincial income per capita 2 . Starting
in 2001, Indonesia allocated almost a
third of national revenues to regional
governments. Such condition couse
regions still depend highly on central
government transfers to finance their
expenditures. Disparity in own-source
revenues and revenue sharing trapped
Indonesia’s regions enjoy significantly
different levels of fiscal resources. There-
fore the issue of provincial/regional
income per capita disparity has not
disappeared, and the reasons for this
have not yet been fully discovered.

2 World Bank Report (2006).

Fiscal Desentralization And Indonesia Regional Income Disparity
(1994-2004)
21
Income per capita disparity, measured
by coefficient variations of Gross Regio-
nal Domestic Product (GRDP) per ca-
pita, has remained relatively the same
throughout the centralization era as well
as throughout the decentralization era.
The GRDP per capita growth
varies from 1994 to 2004. The GRDP
per capita growth rate in the rich regions
in some periods was lower than that in
the poorer regions. It is postulated in
convergence theorem that when the
countries are similar with respect to
structural parameters for preferences and
technology, then poor countries tend to
grow faster than rich countries 3 . The
speed of convergence is believed not
constant since the GRDP growth varies
between various periods. In addition to
that Indonesia also faced economic crisis
during 1997 to 1998, which increases
the difficulty to measure the speed of
convergence in this observation.
The main purpose of this paper
is to analyze the role of fiscal decentra-
lization on regional development and
regional income disparity especially in
Indonesia case after the changing
governmental system from centralization
to decentralization.
Several questions that will be
addressed in this paper are: (1) Does
fiscal decentralization aggravate regional
income disparity; (2) Does fiscal decen-
tralization matter for regional economic
development; (3) What are the positive
and negative impacts of fiscal decentra-
lization for Indonesia’s economic deve-
lopment.
By answering several questions
above, the role of fiscal decentralization
in Indonesia toward economic develop-
ment and regional disparity hopefully

3 Barro, Robbert J. (1991)
can be revealed and it will be useful
input to determine further policy.
This paper only relies on the
data from 1994 to 2004, because the
previous and ex post data are not
properly available. Even so many
variables influence the regional econo-
mic development and regional income
where the fiscal decentralization is one
of those variables, the main focus of
discussion is only for the role of fiscal
decentralization both in regional
economic development and regional
income disparity.
The discussion in this paper will
explore only up to the provincial levels,
because the data for district levels is not
properly available. Therefore the
economic development and income
disparity within regional are not
addressed in this paper.
Basically, this paper is desk study
which analyzes data from primary sources
(Statistic Indonesia and Ministry of
Finance) elaborates with other countries
experiences. The analysis will be divided
in two periods, before and after decen-
tralization by using Indonesia Provincial/
GRDP panel data from 1994 to 2004.
There will be compared the regional
economic development and regional
income disparity before and after
decentralization, where the growth of
GRDP per capita is as proxy of regional
economic development whether GRDP
per capita is as proxy of income.
In order to analyze the regional
economic development, the standard
methodologies of beta and sigma
convergence will be employed. Further-
more, for measuring regional income
disparity the Theil index approach will
be used.


Jurnal Keuangan Publik Vol. 5, No. 1, Oktober 2008
22
The examination of relationship
among fiscal decentralization; regional
economic development; and regional
income disparity is based on the
empirical data from 1994 to 2004. The
model will be estimated by performing
ordinary least square (OLS) estimation.

THEORETICAL FRAMEWORK

The centralized governmental
system which emphasizes on central
planning and developing was believed
as the right key to run governmental
system in order to develop its country
during 1980s. Lately, during over past
few decades, many governmental
systems around the world were changed.
Regardless they are left-wing, right-wing,
central, autocratic or emerging demo-
cracies has attempted to shift at decen-
tralization. At the same time the collapse
of the Soviet Union, a series of economic
crises, increasing poverty, corruption
and crime were culmination of factors
necessitated a shift of governmental system.
There are several reasons under-
lined the desire of decentralization. One
of them is political concerns. Such as in
Latin America, decentralization has been
an essential part of the democratization
process as discredited autocratic central
regimes. Under new constitutions the
autocratic central regimes are replaced
by elected government. In Africa, the
presence of multi-party political system
has been creating demand for more local
voices in decision making. In some
countries, like Ethiopia, decentralization
has been a response to pressures from
regional or ethnic group for more
controls or participation in the political
process. In the intense, decentralization
represents a disparate attempt to keep
the country together in the face of
pressures by granting more autonomy to
all localities or by forging “asymme-
trical federations” 4 . A long civil war such
as in Mozambique and Uganda also was
one of the reasons the desire to
decentralization, where opening politi-
cal opportunities at the locals level has
allowed for greater participation by all
former warring factions in the go-
vernance in those countries. In many
countries, decentralization happened
because of the absence of any meaning-
ful alternative governance structure to
provide local government service. In
some cases particularly in East Asia,
decentralization occurs caused by the
need to improve service delivery for
large population and the recognition of
the limitation of central administrative.
In so far we relaxed that the
willingness to decentralization comes
from internal side of the country. But
actually the pressure of fiscal decentra-
lization has generally come from both
internal and external drivers. Tanzi
(2002) pointed out several driving-forces
for fiscal decentralization. From internal
side, the pressure to decentralization
could be in the form of deepening
democratization within specific groups
or regions, or increase levels of wealth
and education. On the other hand the
pressure to decentralization could come
from external impulse such as donors.
The pressure from donors or aid agencies
have precipitated the environment
toward decentralization
In East Asian countries, reform
toward decentralization is underway in
almost every country (White and Smoke,
2005). While the ‘big bang’ decentra-

4 Worldbank.org at decentralization and sub
national regional economic

Fiscal Desentralization And Indonesia Regional Income Disparity
(1994-2004)
23
lization program approach was intro-
duced in Indonesia in the early 2000s,
Cambodia and Vietnam preferred a
gradual process of decentralization
Beside its advantages, decentra-
lization also has many drawbacks.
Efficiency may not occur during decen-
tralization time. Central governments
can loss control over scare financial
resources. Others argue that decentrali-
zation may end up with increasing cost
and reduce efficiency in service delivery,
increasing inequity and macroeconomic
instability. Weak administrative or
technical capacity at local government
levels may result in inefficiency and
ineffectively services delivery in some
areas of the country. These drawbacks
emerge in a situation where citizen
preferences are unlikely to be repre-
sented and there is almost no institutio-
nal capacity at the sub-national levels. In
the absence of citizen participation local
elites and traditional authority structures
may emerge.
White and Smoke (2005) propose
that national leadership is essential to
anticipate the drawbacks of decentrali-
zation. The national leader should able
to response across regions consistently
and systematically due to the potential
disadvantages of decentralization.
In several countries, the balance
combination between centralization and
decentralization is essential to reach the
effectiveness and the efficiency govern-
ment functions. Therefore it is possible
not only to run pure centralization or
decentralization system. Some government
functions could not be financed and
managed in decentralization fashion.
Central governments often control and
regulate important policy and supervise
roles, even they decentralized the
responsibilities.
There are some types of decen-
tralization including political, adminis-
trative, fiscal and market decentralization.
In order to achieve the usefulness of
decentralization, better understanding of
various concepts is needed. Sometimes
there is overlapping in defining of these
terms, but the definitions are not as
important as the comprehensive approach.
Political, administrative, fiscal and
market decentralization may appear in
different forms and combinations across
countries, within countries and even
within sectors. In this paper, the only
one that will be relaxed is fiscal decen-
tralization.

The View of Fiscal Decentralization
Traditionally, the theory and
practice of fiscal decentralization is to
understand how fiscal decentralization
affects the traditional economic objec-
tives of economic efficiency, income
redistribution and macroeconomic stability.
Fiscal decentralization is often defined
as the delegation of fiscal authority from
the central government to the local/
regional governments. Fiscal decentrali-
zation may provide greater economic
efficiency in the allocation of resources
in the public sector. Because of different
preferences in public services across
regions, the uniform provision of public
services provided by central government
will be less efficient. As result, regional
governments are better able to match
different preferences across jurisdictions.
Therefore, there is a competition among
regional governments in delivering
public services and pursuing economic
growth.


Jurnal Keuangan Publik Vol. 5, No. 1, Oktober 2008
24
Prud’homme (1995) said that
because of different individuals’ prefe-
rences and individuals’ mobility across
regions, the decentralization system will
be more efficient. Conversely, If indivi-
duals’ preferences are the same, the
uniform provision of public goods and
services is more efficient. He also
argued, though individuals’ who living
in a region have a quite similar prefe-
rences or if they lack mobility across
regions, in fact, regional governments
could be more efficient to provide public
goods and services than does by central
government.
A. James Heins (1971) defined
the process of fiscal decentralization as
process under which decisions about
scope of activity are made at the federal
level, but decisions about the design of
activity are made at lower levels of
government. This process has implica-
tion that fiscal decentralization involves
programs under which funds are
acquired by the federal government and
turned over to lower levels of govern-
ment for disposition.
Oates (1993) argued that fiscal
decentralization will increase economic
efficiency because local government’s
position is closer to local people and
will be more responsive to the local
needs and preferences. Local govern-
ment also has better understanding
about local preferences and local cost.
This economic efficiency induces resi-
dent to move and to live in the region
and community that satisfy their
preferences.
Oates (1999) said that fiscal
decentralization is in vogue, both in
industrialized and developing countries.
Nations are turning on devolution is in
order to increase their performance of
public sector services. Decentralization
increases economic welfare above that
which resulted from more uniform levels
of such services that are likely under
centralization government system. The
efficient level of output is likely varies
across jurisdictions as result from
different preferences and cost differentials.
Nonetheless, there remains to be said
that the provision of public services
should be located at the lowest level of
government. To carry out the local
government functions, it is required
specific fiscal instruments. The fiscal
instruments in a federal system could be
categorized as taxation, intergovern-
mental grants, and revenue sharing. The
basic questions emerge then which taxes
are best suited to use at the different
levels of government.
Jonathan Rodden (2002) argued
that one of the most formidable challenges
facing multitiered systems of government
is fiscal indiscipline among sub national
governments. Vertical fiscal imbalance
has a negative effect on sub-national
fiscal performance. Fiscal decentraliza-
tion is dangerous because sub-national
governments expand their expenditures
while externalize cost to others. Increasing
intergovernmental grants rarely lead to
tax reductions, and increase in transfer
stimulate much higher expenditures than
does similar increasing in local govern-
ment revenue. Central government will
place restrictions on sub-national bor-
rowing autonomy when vertical fiscal
imbalance is high. Vertical fiscal imba-
lance will only affect sub-national fiscal
performance at high level of borrowing
autonomy. When formal borrowing
autonomy is low, deficits are kept under
control by the heavy hand of central

Fiscal Desentralization And Indonesia Regional Income Disparity
(1994-2004)
25
government. Political federalism under-
mines sub-national fiscal discipline.
Simply, we have two opposing
arguments related to fiscal decentra-
lization and economic efficiency.
Decentralization system is more efficient
than centralization system because
decentralization creates competition
among regions. On the other hand,
decentralization reduces the central
government power in pursuing income
redistribution among regions. Decentra-
lization could be less efficient if the local
governments’ expenditures cannot be
covered by their own revenue and also if
competition among them lead to be
unfair. The instruments which are used
in the fiscal decentralization among
countries generally are the same, namely
taxation, grants/ sharing and borrowing.

Fiscal Decentralization and Regional
Disparities
Regional disparities can be
measured at different levels of aggre-
gation such as global, continental, inter-
national and national levels. Here we
consider at national level but focus on
within regional disparity.
Prud’homme (1995) said that
view-point of fiscal decentralization is in
economic efficiency, but another view
argues that fiscal decentralization may
increase regional disparities. Therefore
central government is needed to equa-
lize balanced resources distribution from
richer region to poorer region through
policy regulations. Conversely, from the
recent theoretical work point of view,
fiscal decentralization not only has
impact to increase and improve delivery
services but also reduce regional disparities.
Akai and Sakata (2005) proposed
that transfer resources from richer to
poorer regions in centralization system
in order to reduce regional disparities
does not guarantee in reducing disparity.
However, the transfer of resources from
the central government to poor regions
will lessen national fiscal capacity. Any
way it is a government commitment to
escape poor regions. And therefore,
fiscal decentralization encourages effi-
ciency public services and reduces
regional disparities.
Close to local residents gives
incentive to local government to be
more responsive to the local needs and
preferences. They also can implement
regional economic policies better than
central government does. Therefore
regional disparity can be reduced in
decentralization governmental system.
The decentralization reflects smaller
government activities in the smaller
regions. Hence, the chance to achieve
economic goals is higher. As a result it
will reduce the duty of the central
government and its economic power.
So far, there are two arguments
toward impact fiscal decentralization on
regional disparities. It makes this study
more interesting to reveal whether fiscal
decentralization increase or decrease
regional disparity. On the rest of this
paper we would try to find the answer
by addressing a systematic examination
the relationship between fiscal decentra-
lization and regional disparities and also
to expose such relation for Indonesia case.
Martinez-Vazquez and McNab
(2001) argued that there is possibility
misspecification of the empirical
estimation model in the fiscal
decentralization measurement. They also
proposed that there is no single or best
measure of decentralization. Clearly
fiscal decentralization is multidimen-


Jurnal Keuangan Publik Vol. 5, No. 1, Oktober 2008
26
sional because there are many aspects of
a country’s fiscal affairs that can be more
or less decentralized. Since regional
governments have more power to decide
their expenditures, it can be the case that
another country is more decentralized,
even though higher central government
revenue was transferred to its regional
government. Another case, lower degree of
decentralization was caused by uncoun-
table regional officials though greater
central government revenue was share to
regional government.
The degree of decentralization
also can be measured on government
revenues and expenditures, as Davoodi
and Zou (1998) did for across countries
data. They measured the degree of
decentralization in their study of the
relationship between fiscal decentra-
lization and economic growth.
Economists employ several
methods to measure the dimensions of
regional disparities. Some of the
methods are Coefficient of Variation
(CV), Gini coefficient, The Lorenz Curve,
Theil index and Atkinson index. All of
those models are common in the analysis
of interpersonal income disparity but they
have not been used in the studies related
to regional disparities.

Fiscal Decentralization and Economic
Growth
Oates (1993) argued that there is
much current interest in the potential
contribution of fiscal decentralization to
economic development. Delegating
higher responsibilities to local govern-
ments is seemed as to break the grip of
miss management and central planning
as effort to set self sustaining regional
economic growth. The increasing qua-
lity and quantity of public sector service
may be seen as the result of economic
development for particular region.
An essential question comes up
whether fiscal decentralization causes or
results of economic development. Is it
more complex outcome that accompa-
nies economic growth? The assertion is
whenever income rise and economic
grows in the mature level then the gains
from fiscal decentralization occur.
Decentralization more likely comes with
the achievement of higher stage of
economic development and that
threshold level of economic develop-
ment at which fiscal decentralization
become attractive appears to be quite
high 5 . From this perspective, an econo-
mic development comes first then fiscal
decentralization follows. On the other
hand, fiscal decentralization appears to
have a potentially useful role to play in
economic development. But the
translation of this potential into a real
contribution to economic growth depends
on a number of crucial conditions regarding
the responsiveness of local institutions to
local welfare which, in turn, depends
importantly on the poorer structure of
fiscal institutions 6 .
Davoodi and Zou (1998) po-
inted out that fiscal decentralization is
seen as part of a reform package to
improve efficiency in the public sector,
to increase competition among sub-
national government in delivering public
service and to stimulate economic
growth. In their research, Davoodi and
Zou concluded that there is no relation-
ship between fiscal decentralization and
growth in developed countries. On the
other hand in developing countries there

5 Bahl and Linn (1992)
6 Oates (1993) p. 241

Fiscal Desentralization And Indonesia Regional Income Disparity
(1994-2004)
27
is negative relationship between fiscal
decentralization and growth. They
explained that the composition of local
government spending is negative there-
fore it causes such negative relationship.
There is no explanation for current and
capital spending in the local government
expenditure. The conventional view states
that growth effect of welfare and current
spending. If the local governments spend
their budget in the wrong expenditure
excessively, it can lead to lower growth.
Lower growth also can be resulted from
the wrong revenue assignment among
various levels government. For example,
local governments may be raising
revenues using a tax instrument which
should have been used by central
government. Sometimes the local go-
vernments do not responsive to their
citizen preferences and needs. This can
occurs when the local government
officials are not elected by local citizens
and when local citizens may be too poor
to “vote with their feet”.
Mc Nab and Martinez-Vazquez
(2001) argued that fiscal decentralization
may indeed have a direct impact on
economic growth but the theoretical
underpinnings for this relationship
remain largely undeveloped. Not many
literatures devoted their attention in the
indirect effect of fiscal decentralization
on economic growth through the
impacts of fiscal decentralization on
economic efficiency, the regional distri-
bution of resources and macroeconomic
stability.
The methodology that is used
for analyzing economic growth due to
convergence or divergence is sigma and
beta analyses. Barro and Sala-i-Martin
(1995) and Sala-i-Martin (1996) drawn a
useful distinction between those two
types of convergence namely beta and
sigma convergence. The sigma conver-
gence shows how the dispersion of per
capita income across a group evolves
over time. Therefore, if the dispersion as
measured by the variance of GRDP per
capita decreases, there is sigma
convergence. Another words, the sigma
convergence hypothesis examines the
changes in variation of income per
capita between countries or regions. If
this variation decreases over time the
sigma convergence hypothesis can be
accepted.
Beta convergence was con-
sidered when the partial correlation
between growth in income overtime and
its initial level is negative. There are two
types of beta convergence, absolute and
conditional beta convergence. Absolute
beta convergence tests the neo-classical
hypothesis that poorer counties grow
faster than richer ones. There will be a
negative relationship between the initial
level of income and average rate growth
of income for the period under conside-
ration. Conditional beta convergence, as
opposed to absolute beta convergence,
analyses the incomes per capita of
countries that have identical structural
characteristic and converge in the long
run to their own steady states. Conditio-
nal beta convergence can also be analy-
zed by introducing variables that account
for differences among the regions or
countries.
The relationship between sigma
and beta convergence is that beta
convergence is a necessary but not
sufficient condition for sigma conver-
gence to occur, therefore sigma conver-
gence analysis is often used as a first


Jurnal Keuangan Publik Vol. 5, No. 1, Oktober 2008
28
approximation to the existence of beta
convergence 7 .
The estimating for absolute beta
convergence could be performed by
formula below;
( ) it t i t t i it y n y y e b a + + = − − − 1 , 1 , ln / ln ln

(2.1)

Where:
it y refers to GRDP per capita in the last
year of period t for regions i
1 , −t i y is the value of GRDP per capita in
the initial year of period t
t n is the number of years in that period
it
e is the error term

Intergovernmental Transfer

Intergovernmental transfer can
take two general forms. They can be
conditional grants and unconditional
grants. Conditional grants place any
various kinds of restrictions on their use
by recipients. While unconditional
grants are the lump sum transfers to be
used in any way the recipient wants. The
conditional grants are employed where
the provision of local services generate
benefits for local residents to other
jurisdictions. In contrast, the uncondi-
tional grants are typically the appropriate
vehicle for fiscal equalization purpose.
The aim of these grants is to channel
funds from wealthy jurisdictions to
poorer ones.
There are three main objectives
of intergovernmental transfer. First,
intergovernmental transfer is used to
address vertical fiscal imbalance. Pro-
blem of vertical imbalances may always
persist, since broad and substantial

7 Barro and Sala-i-Martin (1991)
taxes, and revenues from major natural
resources such as oil and gas remain at
the central level. Indeed, local govern-
ments are often left with low-yielding
income. Thus, intergovernmental transfers
are needed to achieve vertical balance.
Second, intergovernmental transfer is
beneficial to address horizontal fiscal
imbalance. In developing and transition
countries, it is common to have large
disparities between the rich and the poor
regions. This may happen due to
variations in the capacity of local
governments to generate their own
resources. Such variations depend on the
performance of potential income, the
natural resources and the local eco-
nomic activities. Hence, this will lead to
horizontal imbalance. System of grant
equalization might be justified to address
such problem. Third, intergovernmental
transfer is needed in order to address
inter-jurisdiction spill over or exter-
nalities. Some public services in a parti-
cular region sometimes create externalities
to other regions. Pollution control and
higher education are some example for
such externalities. Therefore, central
government should provide incentives or
financial resources for local government
to address this particular problem.
Actually, there are still many
advantages from intergovernmental
transfers, such as to control the overall
levels of local government expenditure,
to encourage revenue mobilization, to
stimulate local economic development,
to encourage responsibility and trans-
parency in local decision making, and to
provide emergency situation.




Fiscal Desentralization And Indonesia Regional Income Disparity
(1994-2004)
29
Evidence of Disparity and Economic
Development Cross Countries
Canaleta, Arzoz, Garate and
Orayen (2002) analyzed the relationship
political and fiscal decentralization to-
ward regional productivity inequalities
using fifteen OECD countries as sample.
They used measures of regional dispari-
ties in GDP per worker. They did not use
GDP per capita because the existent of
commuters who produce great distortion
in some regions. Also GDP per capita is
influenced by the age structure and rate
activities of the population. In their
article they used three measures. Those
are Gini, Theil and Atkinson’s index. All
of the indices satisfy Dalton transfer
principle and homogenous degree of
zero 8 . Their finding is that there is
indication that federalism and decentra-
lization matter only if more expenses
were decentralized.
Kim, Hong and Ha (2003)
estimated the effects of decentralization
of public goods and employment and
the influence of national economic
development on regional income dispa-
rity in Korea. They used Korean time
series data from 1971-1997. The main
variables were used in their research are
GDP per capita, urbanization rate and
spatial decentralization instrument. They
believed that regional income disparity
is affected by those variables. They also
included information network, transpor-
tation and water supply facilities,
employment opportunities, and educational
services. Gini coefficient of regional
incomes, educational services, employ-
ment opportunities, information networks,
transportation and water supply facilities
were employed. The model form showed

8 The Dalton transfer principle says that transfer
from a richer to a poorer region reduces disparities.
that inequality is function of GDP per
capita, urbanization rate and all of gini
coefficient above. Their study demonstrated
that GDP per capita and urbanization
have negative correlation to regional
disparities, while other variables positi-
vely correlated. Finally, in the early 1980s
the impacts of decentralization policies
on regional disparities were shown to
fluctuate, but these impacts began stable
during 1990s. Their research did not
explicitly explain the impacts of fiscal
decentralization to regional income
disparities.
Davoodi and Zou (1997) looked
for the relationship between fiscal
decentralization and economic growth
by observing cross country study both
for developed and developing country.
They provided a simple endogenous
growth model showing how the degree
of fiscal decentralization affects the
economic growth rate. They observed 46
developed and developing countries
using a cross country panel data set for
the period from 1970 to 1989. From
their sample, developed countries are on
average more decentralized than deve-
loping countries and to have a higher
per capita GDP growth rate. They found
the negative relationship between fiscal
decentralization to economic growth for
developing countries but not for developed
countries.
Jorge Martinez-Vazquez and
Robert M. McNab (2001) said that
whether decentralization affects econo-
mic growth has become an important
policy issues for developing and transi-
tional countries both for large countries,
like China and India, and for small
countries like Baltic Countries. They saw
that the overall impact of fiscal decentra-
lization on economic growth is uncertain.


Jurnal Keuangan Publik Vol. 5, No. 1, Oktober 2008
30
In term of direct impact, they expect
higher growth associated with higher
fiscal decentralization. They also saw
that there are potentially the indirect
effects of fiscal decentralization towards
economic growth. Those indirect impacts
include through consumer efficiency,
producer efficiency, the geographical
distribution of resources, macroecono-
mic stability, corruption and capture by
elites. These indirect effects will link
higher economic growth in one side, but
there are others that work in the opposite
direction.
Resosudarmo, Budy P, Vidyata-
ma, Yogi (2006) examined the regional
income disparities in Indonesia. They
focused on the growth rate of poorer
regions in terms of per capita income or
output. Their concept defines a
convergence as phenomenon in which
poorer regions have higher growth rate
than richer regions. The main analysis in
this concept aims to prove whether or
not the coefficient of initial condition on
growth is negative. So, sometime in the
future, all regions will have the same
growth rate. They concluded that; first,
disparity in provincial income per capita
in Indonesia is relatively severe; second,
there is a conditional growth convergence,
meaning that the GDP per capita of
poorer provinces grows faster than that
of richer provinces given the data 1993
to 2002; third, physical investment (both
private and public investment), trade
openness and the role of oil and gas are
significant determinants of growth.

FISCAL DECENTRALIZATION IN
INDONESIA
Before 2001, at the New Order
regime, Indonesia conducted highly fiscal
centralization with regional revenue
completely supported by, and depending
on, the central government. In 1968 only
around 7 percent of national public sector
revenue was redistributed to the regions.
This condition continued for the next 20
years. In 1990s the central government
still dominated approximately 90 percent
of national revenue, transferring to the
regional government with only about 10
percent of total expenditures (M. Tadjoe-
din at. al. 2003). Thereby, Indonesia was
categorized with worse financial system
in the world as indicated by deeply
imbalanced intergovernmental fiscal
relation.
The government’s aim to centra-
lizing revenue collection and transferring
a part of collected funds to the regions,
in principle, was to equalize development
throughout the regions without taking
into account regional prosperity, resources,
geography, interest and production
capacity. The main distribution goal was
to create regional uniformity as reflected
in per head population, equal budget
allocation, equal number of social
facilities such as health (hospitals) and
education (schools), and for delivering
other public services.
The economic crisis in 1997-
1998 and intense pressures for indepen-
dence from rich natural resources regions,
eventually, made government in the
difficult situation. There is no choice
except to accommodate region interests
by changing the governmental system
from centralization to decentralization.
Legally government, approved by house
of representative, issued the Law 22/1999
and Law 25/1999 as jurisdiction ‘umbrella’
for such changing governmental system.
The Law 25/1999 concerns with
fiscal balance between central and
regional governments, which would be

Fiscal Desentralization And Indonesia Regional Income Disparity
(1994-2004)
31
accompanied with new responsibilities
delegated to regional governments under
the Law for administrative decentra-
lization (Law 22/1999). The Law 25/
1999 is become basis for new inter-
governmental transfer arrangements as
equality reflection revenue sharing. The
equalization fund (Dana Perimbangan)
consists of three components according
to this law. The first is Revenue Sharing
at certain portion of taxes on land and
buildings, the transfer of land and
buildings, and the net-tax revenue from
the exploitation of forest, mining, fisheries,
oil and gas. The second component is
the General Purpose Grant (DAU), a
block grant which is aimed to equalize
the fiscal capacities of regional
governments to finance their purchase
expenditures. It is stated in the law that
the amount of DAU per fiscal year is at
least 25 percent of the central govern-
ment domestic revenue, and this amount
should be distributed among local
governments through formula that was
designed with considering the regional
needs and potential capacity. The third
component is the Specific Purpose Grant
(Dana Alokasi Khusus-DAK). Basically
DAK is also block grant that distributed
to finance special needs that either
cannot be included by the formula used
in DAU allocation or categorized as
national priorities and commitments.
Grants from central government
to regional governments can be catego-
rized into three forms. First form is grant
from the central government to be spent
by the regional governments that
integrated into the regional budget and
registered into regional government’s
account. The second form is grant from
central government to be spent by
regional governments, but not integrated
into the regional budget and not
registered in regional government’s
account. The third form is funds that
allocated by the lines ministries to
finance development activities in the
region.
As a result, by running the new
system the total revenue received by
regional government consists of three
items. Those are regional own revenues,
fiscal equalization funds and other legal
official revenues. Regional own revenues
is composed of regional taxes, levies
(retributions), and revenue from profit
regional own enterprises and other legal
regional own revenues. Equalization
funds consist of tax sharing, non tax
natural resources sharing, general purpose
fund and specific purpose funds. While
other legal official revenues were
composed by revenues from intergovern-
mental transfers such as grant, emergen-
cy fund such as funds used to cope with
disasters, and saving from previous year.





















Jurnal Keuangan Publik Vol. 5, No. 1, Oktober 2008
32
Table 1, Average Revenue Composition Before And After Fiscal Decentralization.
1994-2000 2001-2004
A Revenues 1 1
A.1 Own Revenues 0.21 0.13
- Regional taxes
- Regional levy
- Profit of Regional Corporation
- Other legal own revenues

A.2 Fiscal Equalization Funds 0.13 0.78
- Tax sharing
- Non tax natural resources sharing
- Specific purpose grant*
- General purpose grant*

A.3 Other legal official revenues 0.66 0.09
- Intergovernmental transfer/grant
- Emergency Funds
- Saving from previous year

Sources: Ministry of Finance, own calculation
* Not in 1994-2000

By looking at the table above, it
seems that Fiscal Equalizations increase
drastically from 13% to 78%. Actually in
1994-2000 the specific purpose grant
and the general purpose grant were not
counted since these types of grant not
available yet. Indeed 13% only reflecting
the sharing funds. The magnitude of
sharing fund before and after fiscal
decentralization can be seen at table 2
in next sub chapter. On the other hand,
other legal official revenues share to
total revenue is quite big before
decentralization. The grant and or
intergovernmental transfer contributed
69% from total other legal official
revenues. While the emergency fund
was only transferred whenever there is
such a disaster or catastrophe. The
saving from previous year also was very
small instead of negative. This saving
minus is due to the expenditures above
the revenues.
The important one from the
table 1 above is that before fiscal
decentralization, special and general
purpose funds were not available. On
the other side, the grant from central
government was more than half of total
regional revenues. Usually the purpose
of that grant had been targeted by
central government. Thereby, it was
often happen that the goal of the grant
did not match with the local needed.
After fiscal decentralization has
been held, the grant magnitude was
relatively low, because it was included
in general and specific purposes grant.
The General purpose fund was
calculated with specific formula that
takes into account demography, local
government officer number and price
index. The use of general purpose fund
was fully decentralized to local
government. Part of this fund used to pay
local officer salary. Specific purpose
fund was granted as investment stimu-
late. This fund was proposed by local
government to central government.
Detail about general and specific
purpose fund will be described in the
next sub chapter.
The share of own revenues
before fiscal decentralization is higher

Fiscal Desentralization And Indonesia Regional Income Disparity
(1994-2004)
33
than after, since averagely the total
regional revenues after fiscal decentra-
lization is bigger caused by enhance-
ment of the magnitude of fiscal sharing
and of fiscal equalization. While
regional own revenues relatively stable.
It was found in the data that the regions
which have abundant natural resources
have lower share own revenues.
Contrary, the share of own revenues for
poor regions is higher. Anyway the level
of regional own revenues is increasing
year by year. For instance, the share of
own revenues for East Kalimantan which
has abundant of natural resources
especially from oil is only 5%, while
Jakarta which have no oil has 40% own
revenue share.
In order to reduce the fiscal
transfer dependency from central
government, regional governments try to
increase their own resources. Local
government is allowed to dig up their
potential own resources that available in
its region. There are also possibilities to
increase the tax rates and levies based
on its region characteristics. Thereby the
tax rates among regions are different. To
be considered that the regulation about
regional taxes and levies should be
approved by local legislative. That is the
ability of citizens to pay the tax also
taken into account. On the other side,
the central government’s rule about
taxation also should get deliberation.
Instead of increasing tax rates and levies,
the higher portion of revenue sharing to
be allocated for original regions is better
for future fiscal decentralization.

Intergovernmental Transfer and
Equalization
Up to now after six years imple-
menting decentralization governmental
system, the decentralization process in
Indonesia is still difficult to be evaluated
whether it is successful or not. From the
central government point view, the
decentralization has been running well,
even there is a doubt about the future
because so many things have to be done
in order to keep the decentralization in
the right order, including to revise the
law 22/1999 and 25/1999. However, the
local governments less satisfied than the
central one. They think that central
government does not fully support
decentralization process. There is indica-
tion if the decentralization process is
considered to fail the recentralization
system will be used again 9 .
The economic impacts of
decentralization are considered as the
most crucial thing, since Indonesia still
on the recovery after economic and
financial crisis on 1998. Most Indone-
sian had high expectation on the suc-
cessfulness of decentralization to arise
economic condition. As describe in
previous chapter, the contribution of
decentralization on economic refor-
mation is one of the reasons of changing
governmental system. Reducing disparity
between Java Provinces and outside Java
Provinces or western part and eastern
part of Indonesia also has been desired
during decentralization process. Having
more authority on economic policy
decision making to the local government
is also expected as successfulness of the
decentralization process.
Recently the legal process of
decentralization in Indonesia is based on
the law 32/2204 and the law 33/2004.
The previous one concerns with regional
governance which revised the law
22/1999. The later concerns with the

9 Brojonegoro (2005)


Jurnal Keuangan Publik Vol. 5, No. 1, Oktober 2008
34
central and local government fiscals
balance which revised the law 25/1999.
The law 33/2004 mandates that revenue
sources for sub-national levels are from
own-source revenues (taxes and user
charges), intergovernmental transfers,
borrowing mechanism and other legiti-
mate revenues.
The highlight of sub-national
revenue sources stated in this law is
intergovernmental transfer. The aim of
intergovernmental transfer is to reduce
vertical fiscal gap between central and
regional government and to reduce
horizontal fiscal gap among regional
governments. In the Indonesia’s budget
term we call it equalization funds (Dana
Perimbangan). This equalization funds
deploys in the form of revenue sharing,
specific purpose grant and general
purpose grant.
From the figure 1 below, it is
clear that intergovernmental transfer
increased almost two and half times after
running fiscal decentralization, where
the total amount of government transfer
reflects the summation of A2 and A3 in
the table 1.

Figure 1, The Evolution Of Government Transfer From Central To Regional
Government 1994-2004 (1994 As Constant Price)
Total Government Transfers 1994-2004
0.00
5,000.00
10,000.00
15,000.00
20,000.00
25,000.00
30,000.00
35,000.00
1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004
years
billion rupiah

Source: Ministry of Finance, own calculation

Revenue Sharing
As stated in the law 33/2004,
the purpose of revenue sharing is to
address the problem of fiscal imbalance
between central and regional govern-
ment. As in some countries, the major
taxes are counted as central government
revenue. Those major taxes such as
personal income tax, value added tax
and property tax. In so far property tax
still administered in the central level.
Before implementing the law 33/2004 all
of natural resources revenues were
administered at central level such as oil,
gas, fishery, forestry and other mining.
By implementing this law, a particular
percentage of natural resources revenues
were plotted as regional income.
The mechanism for revenue
sharing was derived from the basis of

Fiscal Desentralization And Indonesia Regional Income Disparity
(1994-2004)
35
original local governments. Revenue
sharing from taxes consists of property
tax, property tax transfer fee and
personal income tax. While the revenue
sharing from natural resources, it con-
sists of shared revenue from forestry,
mining, fishery, oil and gas. In addition,
the new law on central and local fiscal
balance stipulates additional shared
revenue from geo-thermal energy and re-
forestation fund. The later is formed as
an earmarked grant for rehabilitating
forests in originating local government.
With this arrangement, the
central government had attempted to
address the vertical imbalance between
central and local government. However,
the derivation basis employed in the
revenue sharing mechanism implies that
the bulk of revenues are distributed to
originating regional governments. Hence,
this will create fiscal disparities among
regions. In the local government budget
structure those revenue sharing are
counted at A2 (see table 1). Table 2
shows revenue sharing comparison
before and after implementing the law of
central and local fiscals balance.

Table 2, Revenue Sharing Proportion Before And After Decentralization.
No. Revenues
Before Decentralization After Decentralization
Central Regional Central Regional
1. Land and building tax 10 90 0 100
2. Registration fee of land and building 20 80 0 100
3. Fee for forest benefit 55 45 20 80
4. License fee for forest exploitation 55 45 20 80
5. Land rent 20 80 20 80
6. Mining royalties 20 80 20 80
7. Fishery 100 0 0 100
8. Oil 100 0 85 15
9. Natural gas 100 0 70 30
10. Re-forestation 100 0 60 40
11. Income tax 100 0 80 20
Sources: the law 33/2004

Specific Purpose Grant (Dana Alokasi
Khusus/DAK)
DAK is allocated to specific
regions and certain program. DAK is
intended to promote the attainment of
minimum standard and to compensate
the benefit/cost spill over related to
priority capital investment. Hence, it is
confined mainly to finance physical
capital investment and limited-period
financing of operational and mainte-
nance needs.
The criteria for DAK are based
on the general, special, and technical
criteria. The general criteria should
consider the financial capacity of the
regions, while special criteria emphasi-
zes on the characteristic of the regions.
Technical criteria are more specific and
established by the line ministries. Condi-
tional term attached in DAK lead to the
limited scope of sub-national govern-
ments’ entitlement, particularly in the
usage of funds. However, the bottom up
principle applies during the process of
DAK allocation. In this process local
governments may propose programs and


Jurnal Keuangan Publik Vol. 5, No. 1, Oktober 2008
36
activities which are in line with the
national priorities.

General Purpose Grant (Dana Alokasi
Umum/DAU)
The DAU intends to address
both problems of vertical and horizontal
imbalance. Thus it is expected to
equalize fiscal capacities across regions
to finance public services. In article 27
law 33/2004 states that the total amount
for DAU as at least 26 percent of net
domestic revenues. Previously the
magnitude of DAU is 25 percent.
However, there are no clear reasons for
one percent increase. Brodjonegoro
(2005) points out that there is no proof
whether local governments need more
transfer or on the contrary might have
surplus. Thus, one might assume that
political compromises lay behind this
stipulation.
In principle, the main content of
amendment of DAU in the law 33/2004
concerns the formula for fiscal gap and
the basic amount of allocation. The
formula for fiscal gap is using the
difference of expenditure needs and
fiscal capacity. The main change in the
formula for fiscal gap between the old
and the current system are the indicator
for fiscal need. Those indicators such as
population, area, geographical condi-
tion, and income level which consider
poverty. New indicators had been added
in measuring the needs are the Human
Development Index and GRDP per
capita index.

Local Government Expenditure
Fiscal decentralization involves
shifting some responsibilities of expen-
ditures and revenues to lower levels of
governments. One important factor in
determining the type of fiscal decentra-
lization is the extent to which local
government are given autonomy to
determine the allocation of their expen-
ditures. Another one is the ability to raise
their revenues.
The composition of regional
budget expenditures in Indonesia in-
cludes; Officer Expenditure (salary etc);
Good and Services; Transport and
Accommodation; Other Expenditures;
Investment Expenditures and Unpre-
dictable expenditures. All of those
expenditures do not accommodate
central government officer expenditures
who work in particular region.
Increasing revenues due to fiscal
decentralization automatically will in-
crease local government expenditures in
almost the same portion. Even the
structural of expenditures is relatively the
same before and after fiscal decentra-
lization.
Generally, the expenditures are
divided into two activities, namely
routine expenditures and investment
expenditures. Routine expenditures are
used to maintain that government
activities still exist. Those activities are
such as payment for salaries, good and
services that support routine activities,
maintenance, transport, accommodation,
and other expenditures. While invest-
ment expenditures are used for invest-
ment activities such as procurement for
new building, highway, bridge, road,
dike, irrigation, and so on.
The share of investment expen-
ditures to total expenditures before and
after decentralization is relatively stable.
It is around 30 to 40% from total
expenditures. The rest of expenditures
are used to routine activities payment.
The investment expenditures increase in

Fiscal Desentralization And Indonesia Regional Income Disparity
(1994-2004)
37
the level year by year. It will be better if
the share of investment expenditures is
getting higher and higher since benefit
from investment expenditures not only
could encourage economic sectors
highly but also encourage other sectors.
The main difference in the local
government expenditures before and
after fiscal decentralization is not in the
share of expenditure itself, but the most
important things is the freedom of local
government to allocate those funds
according to their preferences and
necessities. Before fiscal decentra-
lization, it was often that central
government fully controlled what
regional wants. Central government
decided what regional should do with
those transferred funds. Therefore, there
is gap between regional needs and
central wants. Contrary, after fiscal
decentralization was run, local govern-
ments have fully authority to manage
their funds. By doing so, budget will
allocate in appropriate expenditures
conform to local needs.

Indonesia’s Provincial GRDP
Indonesia is one of the rich
natural resources countries, but those
natural resources not spread equally
among regions. So that the regions that
have abundant natural resources have
advantages to have higher GRDP
compare to the regions that do not have
natural resources. Among those regions
are Aceh where has huge natural gas
and oil, Riau has oil, South Sumatra has
oil, East Kalimantan oil and mining and
Papua has mining. Based on this
condition, it is better if we distinguish
the GRDP growth among regions into
two groups, the GRDP growth with and
without oil and gas. It is because the
natural resources depleted over time
while to keep the sustainability of GRDP
growth non oil and gas sector should get
have a higher priority.
There are 27 provinces in
Indonesia before 1999. Since a province,
Timor Leste, separated from Indonesia in
1999, therefore this province was
excluded from calculation. There are
also four new provinces since 2000 that
should be taken into consideration
namely Bangka Belitung, Banten,
Gorontalo and North Maluku. Bangka
Belitung was part of South Sumatra,
Banten was part of West Java, Gorontalo
was part of North Sulawesi and North
Maluku was part of Maluku before. So
that totally now there are 30 provinces
will be relaxed in this paper.
Column 3 in table A1 in the
appendix shows the annual GRDP
Growth from 1994 to 2000 per province
including oil and gas. Papua has highest
average growth before decentralization
(from 1994 to 2000) with 10.05%, while
Maluku have the lowest growth level
with -5.07%. The lowest economic
growth in Maluku provinces was caused
by horizontal conflict there for several
years. This horizontal conflict created
instability and directly lessens economic
activity. Papua province had had the
highest economic growth because of the
mining boom in Papua Island during
those periods. Nationally, before fiscal
decentralization, the highest growth is in
1994 where Indonesia economic growth
is 8.7%. The growth was going down in
the following year, and have a negative
sign in 1998, in which the lowest growth
is -6.45%. The reason is because as all
of us knew that at that time Indonesia
was hit by economic crisis. The
exchange rate of rupiah falls almost eight


Jurnal Keuangan Publik Vol. 5, No. 1, Oktober 2008
38
times than before crisis toward US
dollar. Up till now, Indonesia is still in
economic recovery. Economic growth
gradually increase, even the economic
growth was still below than before
economic crisis hit Indonesia.
Column 4 table A1 shows the
annual GRDP growth including gas and
oil after fiscal decentralization from
2001 to 2004. Averagely all of regions
have positive sign economic growth. The
higher growth rate is 7.11% in Southeast
Sulawesi, while the lowest one is North
Sulawesi with 2.51%. The Indonesia
economic growth increases gradually
after economic crisis. Between economic
crisis and before fiscal decentralization
the average growth rate is still below
0.5%, but after decentralization started
in 2001 the growth rate was jump to
3.71% and then the following years
economic growth remain increase to
4.19%, 4.52% and 5.75% respectively.
As stated above, not every
province in Indonesia has abundant
natural resources of gas, oil and several
kinds of mining. Inevitably those
resources depleted over years and will
vanish in the certain time. Therefore it is
useful to distinguish the level of
economic growth by excluding gas and
oil from the data.
Column 5 table A1 in appendix
shows the annual GRDP growth per
province excluding gas and oil. From
1994 to 2000 before decentralization,
the highest growth rate is still belongs to
Papua Province, and the lowest one also
the same for Maluku Province. The
growth rate trend is almost the same
compare with when still including gas
and oil as part of GRDP. The different
thing here is if oil and gas was included
there are only two provinces that still
have average negative growth, Aceh and
Maluku. Aceh Province has positive
growth now. During the period 1994 to
2000 nationally the average growth is
4.03%. Actually, the Indonesia’s highest
growth rate is 9.39% in 1994, but it
could not remain longer, since in the
following years those growth getting
lesser and lesser. Even the experience of
negative growth occurred in 1998 at -
4.01%. The economic growth start to
increase after 1998, but the level still
lower compare to that before economic
crisis.
By running fiscal decentraliza-
tion package there is acceleration in
economic growth which can be seen in
the column 6 table A1 in appendix. Not
one of provinces has negative growth
rate, instead of nationally the growth rate
getting greater and greater. But the
growth achievement is still below then
before economic crisis especially in
1994. The highest economic growth was
achieved by East Kalimantan with
3.64%.
It is better if we do not only
take into consideration the economic
growth by take a look at the growth
level, but also take consideration the
amount of populations in every region.
What we will address is known as GRDP
per capita. By analyzing the data of per
capita growth, it is more reliable and
reflecting the actual condition of
particular region. Column 3 table A2 in
appendix shows the growth of annual
per capita growth including gas and oil
before fiscal decentralization. Aceh and
Riau province have experience average
negative growth rate during this period
in the level of -3.29% and -0.57%
respectively.

Fiscal Desentralization And Indonesia Regional Income Disparity
(1994-2004)
39
Column 4 table A2 in appendix
as comparison to column 3 shows the
per capita growth after Indonesia running
decentralization policy. The growth rate
experience growing year by year, even it
is still below compare with before
Indonesia was hit by economic crisis.
No of single province has negative
growth anymore. The highest average
growth rate was achieved by North
Maluku with 12.67% while the lowest
growth rate was achieved by Riau with
only 0.31% averagely.
When we exclude the gas and
oil from the data, it will yield different
result. Column 5 table A2 in appendix
informs us about the annual GRDP per
capita growth excluding gas and oil.
Papua province still has the highest level
of growth since that province has
abundant of natural resources, wider
region and smaller population. The
average per capita growth excluding gas
and oil in Papua is averagely 7.64%.
While the lowest growth rate happen in
Southeast Sulawesi with 1.01% averagely.
Nationally the growth pattern remains
same, with negative growth during
economic crisis and growing up in the
following year. For the Maluku Province
case, the positive of average annual
GRDP per capita growth was caused by
the huge population migration in 1999
to other provinces, where in that time
there was horizontal conflict all around
island.
Column 6 table A2 informs us
about the annual GRDP per capita
growth excluding oil and gas after fiscal
decentralization. The national per capita
growth rate increase significantly year by
year. No single province has negative
growth after fiscal decentralization
averagely. The highest one was achieved
by North Maluku with 12.65%. The
lowest one was achieved by Papua with
0.81% since in that province horizontal
problem and separatism occurs. Those
problems directly or indirectly also
influence the economic activities not to
mention Papua has bad and poor infra-
structures compare to other provinces.

EMPIRICAL ANALYSIS
Regional Income Disparity and Its
Decomposition
Disparity is often measured by
means of an index that able to reflect the
degree of variation of the income between
regions. In this paper Theil Index is used
to measure regional income disparity.














= ∑
= Y Y
N N
N
N
T
i
i
R
i
i
/
/
ln
1
(4.1)
Where,
T : Theil Index
Ni : The total population in the region i
N : The total population in the country
Yi : GRDP in the region i
Y : The total GRDP in the country
R : Total number of regions















Jurnal Keuangan Publik Vol. 5, No. 1, Oktober 2008
40
Figure 2, Regional Income Disparity Decomposition.
Theil Index 1994-2004
0.0000
0.0200
0.0400
0.0600
0.0800
0.1000
0.1200
0.1400
0.1600
1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004
Years
Index
Inclding oil and gas Excluding oil and gas

From figure 2 above, we can see
that the level of income disparity slightly
decreases during the period under
observation when including gas and oil.
From 1994 to 1997 it increased slightly
but went down in 1998. After fiscal
decentralization from 2001 to 2004 overall
income disparity went down, and if we
have a look at that figure, the trend of
regional income disparity will decrease
for the next following years. Contrary, by
excluding gas and oil, the GRDP disparity
increased from 1994 to 1997 then went
down in 1998. From 1998 to 2001 the
GRDP disparity relatively stable, slightly
increase in 2002, went down in 2003
and again increase in 2004. It could be
said that gas and oil have significant role
in the GRDP disparity among regions in
Indonesia. It is clear since regions which
have oil and gas resources have potential to
collect higher revenues. Having higher
revenues make the particular regions
have chance to spend more their expen-
ditures, in which directly or indirectly foster
the GRDP. Consequently the availability
of gas and oil will create higher gap in
GRDP term among regions.
Dynamics of Regional Income
Variations (Sigma Convergence)
Figure 3 shows the dynamics of
variation regional income per capita
during 1994-2000. The coefficient of
variation is determined by dividing
standard deviation of GRDP per capita
over the means of GRDP per capita all
regions. The variance in regional income
per capita has been relatively stable both
including and excluding oil and gas, but
its level is higher if we include gas and
oil in the calculation. We see that hypo-
thesis of sigma convergence does not
valid on this period of observation since
the value of coefficient variations does
not decrease over time.








Fiscal Desentralization And Indonesia Regional Income Disparity
(1994-2004)
41
Figure 3, The Dynamic Of Variation Regional Income Per Capita 1994-2000.
The dynamic of variation of GRDP per capita
0
0.2
0.4
0.6
0.8
1
1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004
years
coefficient of
variation
including gas and oil excluding gas and oil


It was proven that by employing
either Theil Index or coefficient of
variations, it comes up with the same
result. In the beginning and in the end
of period observation both Theil index
and coefficient variation show almost
the same level, but fluctuate during
observation. This result will support the
conclusion that there is no sigma con-
vergence before and after decentra-
lization.

Figure 4, The Dynamic Of Variation Regional Revenue Per Capita 1994-2004.
The dynamic of variation of regional revenues per
capita
0
0.2
0.4
0.6
0.8
1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004
Years
coefficient of
variation


Figure 4 shows the dynamics of variation
regional revenue per capita during 1994-
2004. In the period 1994 to 2000, the
coefficient of variation was around 0.5
and yearly fluctuated during those periods.
Started in 2001 to 2002, the coefficient
of variation of revenue per capita
increased to the level of 0.6 to 0.8, that
is, fiscal decentralization has increased
the regional revenue per capita disparity
among regions. It was plausible since
several regions which have abundant


Jurnal Keuangan Publik Vol. 5, No. 1, Oktober 2008
42
natural resources such as oil, gas, and
mining have benefit with higher reve-
nues sharing from central government.
Any way the trend of disparity has been
going down in the following years. It
caused by central government policy to
reduce horizontal regional revenue per
capita imbalance by transferring higher
subsidies to the poorer regions.
Regression Models
Recall equation (1) in chapter II,
in order to test for regional convergence,
cross sectional ordinary least square
(OLS) approach is used with growth rate
of per capita income as dependent
variable and the initial income per
capita level as explanatory variable.

( ) ( ) i i t i i y n y y e b a + + = −
1994 1994 2000 ln / ln ln (4.2)
( ) ( ) i i t i i y n y y e b a + + = −
2001 2001 2004 ln / ln ln
(4.3)

Where:
1994 i y : GRDP per capita in province i in 1994 (base year before decentralization)
2000 i y : GRDP per capita in province i in 2000 (final year before decentralization)
2001 i y : GRDP per capita in province i in 2001 (base year after decentralization)
2004 i y : GRDP per capita in province i in 2004 (final year after decentralization)
t n : The number of years ( equal 7 before decentralization and equal 4 after then)
α , β : Parameters to be estimated
i
e : The error term

If the coefficient of β is negative then
there is economic convergence and on
the way around if coefficient of β is
positive then there is no economic
convergence.

The regression result is stated in the table
below,






Fiscal Desentralization And Indonesia Regional Income Disparity
(1994-2004)
43
Table3, OLS Estimation Regression Result For Convergence
( ) 1 , ln ln − −
t i it y y
Including Gas and Oil Excluding Gas and Oil
t n
19994-2000 2001-2004 1994-2000 2001-2004
α 0.006 0.102 -0.00002 0.047
(0.15) (2.94) (-0.01) (1.22) β 0.001 -0.014* 0.0014* -0.006
(1.62) (-2.60) (2.35) (-0.9)
N 26 30 26 30
R-square 0.086 0.195 0.165 0.028
Figures in parentheses represent the values of t- statistic
Asterisks indicate that variable whose coefficients are significant at the 0.01(**) and
0.05 (*) level.

Considering the OLS estimation
result above, actually there is economic
growth convergence after fiscal decen-
tralization fully implemented (2001-
2004) whether oil and gas taken into
account or not since the coefficient the
initial GRDP per capita is negative.
Contrariwise, before fiscal decentra-
lization (1994-2000) the economic
growth was divergence. It is shown by
the positive sign of coefficient initial
income per capita. The result stated in
table above indicates that after fiscal
decentralization, the richer regions’
GRDP per capita growth is lower than
that those poorer regions. On the way
around the condition before fiscal
decentralization was implemented.
Even absent of sigma conver-
gence the beta convergence remain exist
after fiscal decentralization. This condi-
tion proven that beta convergence is a
necessary but not sufficient condition for
sigma convergence to occur.
In order to reveal the relation-
ship between fiscal decentralization and
economic growth as proxy of economic
development, the model below is used.
i i i i X FD GRDPPCG e g b a + + + = (4.4)
Where:
GRDPPCG : GRDP per capita growth
FD : Fiscal Decentralization
X : A vector control variable
ε : The error term
i : Province α , β , and γ : Coefficients to be
estimated

Several explanatory variables
are introduced in the model to explain
the impression of fiscal decentralization
on GRDP per capita growth as a proxy
of regional economic development. In
the regression, it is differentiated between
including gas and oil and excluding gas
and oil, since gas and oil have signify-
cant role in determining both growth
rate and income per capita. The data
used in the regression are distinguished
as before fiscal decentralization (1994-
2000), after decentralization (2001-
2004), and mixed during period of
observation (1994-2004). All of data are
measured by means of period used data.
Suppose the used data is 1994-2000
then the value of the variable is
measured by summation of that variable
from 1994 to 2000 divided by the
number of years in this observation (7).
The same method is also used to


Jurnal Keuangan Publik Vol. 5, No. 1, Oktober 2008
44
calculate the value of every variable in
the 2001-2004. The variables are as follow;

• Gross Regional Domestic Product
per capita growth (GRDPPCG) is
the ratio of total GRDP over total
population in each province. GRDP
was measured by Production
approach expresses GRDP as the
total value of final goods and
services produced by all production
units in a region within a certain
period (usually one year period).
Production units are grouped as in
the International Standard Industrial
Classification of All Economic Acti-
vities (ISIC), which are: Agriculture;
Mining and Quarrying; Manu-
facturing Industries; Electricity, Gas
and Water Supply; Construction;
Trade, Hotel and Restaurant; Trans-
port and Communication; Financial,
Ownership and Business Services.
All product aggregates are valued at
fixed base year prices in 1993.
• Fiscal Decentralization (FD) is the
ratio of total government transfers
and revenues sharing to total
expenditures in particular region for
each municipal within province and
also province itself. In this case, the
level of FD is different between
before and after fiscal decentra-
lization (sees the table 2 especially
for revenues sharing).
• Human Capital Index (HDI) is cited
from Statistics Indonesia concerns
with the period of observations.
• Population Density includes all
residents in the entire geographical
territory of the Republic of Indonesia
per square kilometer. The measured
of provincial population density is
based on the wide area divided by
total population each province.

The regression result regarding equation
above can be seen in the table below;

Table 4, OLS Estimation Regression Result For Economic Growth
GRDPPCG Including Gas and Oil Excluding Gas and Oil
1994-2000 2001-2004 1994-2004 1994-2000 2001-2004 1994-2004
Constant 23.095 7.337 6.526 17.894 -5.289 3.22
(3.12) (0.66) (1.38) (3.11) (-0.54) (0.80)
FD 2.480* 7.88 3.677** 2.97** 11.965* 4.058**
(2.33) (1.30) (3.14) (3.60) (2.24) (4.10)
HDI -0.352** -0.156 -0.099 -0.27** -0.021 -0.049
(-3.12) (-1.16) (-1.41) (-3.10) (-0.18) (-0.84)
PD 0.00038* 0.0043 0.00026* 0.0003* 0.0046 0.00021*
(2.30) (1.63) (1.96) (2.36) (1.96) (1.85)

N 26 30 60 26 30 60
R-square 0.415 0.122 0.175 0.513 0.169 0.239
Figures in parentheses represent values of t- statistic
Asterisks indicate that variable whose coefficients are significant at the 0.01(**) and
0.05 (*) level.

The most important finding from
table above is that the coefficient of fiscal
decentralization is positive. Therefore fiscal
decentralization has positive impact on
economic growth. The higher the share
of fiscal decentralization on local

Fiscal Desentralization And Indonesia Regional Income Disparity
(1994-2004)
45
government expenditures, the higher the
economic growth will be. It can be
explained by looking at the coefficient of
FD in the period 2001-2004 compare to
that in 1994-2000 whether take into
consideration oil and gas or not. While if
we use pooled model (1994-2004 data)
the coefficient of FD is slightly higher
compare to that in the period 1994-
2000.
The estimated coefficient of HDI
is negative. This variable may serve as a
poor control variable. Since the HDI gap
is high among richer and poorer regions.
For instance the highest HDI is 76.1 in
Jakarta while the lowest one is 54.2 in
West Nusa Tenggara. The other control
variable is PD. The coefficient of PD is
small positive. It also faces the high gap
PD among regions. In Jakarta the PD is
almost 13.000 population per square
kilometers while in Papua only 6 popu-
lation per square kilometers. This un-
wellbalance PD may cause such
unexpected regression result.
The relationship between fiscal
decentralization and regional income
disparity can be found by employing the
model below;

i i i X FD Theilindex e g b a + + + =

Where:
Theil index : Theil Index
FD : Fiscal Decentralization
X : A vector control variable
ε : The error term
i : Province α , β , and γ : Coefficients to be
estimated

The model used is almost similar with
the previous model. The difference is in
the dependent variable, where in this
case, the Theil Index behaves as dependent
variable. While in the explanatory variables,
two variables are added. Those variables
are;
• Theil Index is influenced by the total
population and GRDP per year in
province. This Theil Index
measurement is explained in sub
chapter IV.1.
• Local Government Expenditures
(TE) is measured by the ratio of the
aggregate realization of expenditures
for each municipal within province
and also province expenditures itself
to GRDP per capita during particular
year.
• Regional Own Revenues (OR) is
measured by the share of regional
own revenues for each municipal
within province and also province
itself to total expenditures in a
particular year.
Using OLS estimation, the regression
result is described as in the table 5
below.
By analyzing the table, the
primary finding is that the coefficient of
FD is negative for all data. This finding
provides evidence that FD has negative
impact on disparity which is consistent
with the new theory of fiscal decentra-
lization. More precisely that the coeffi-
cient of FD is negatively higher in the
period of 2001-2004 both including and
excluding oil and gas compare to that
the period 1994-2000 and pooled data
1994-2004. It is undermined the notion
that the higher the share of FD to total
local government expenditures, the
smaller the income per capita disparity
among regions.
Other variables’ estimated co-
efficient can be explained as follow. The
coefficient of HDI is unstable and


Jurnal Keuangan Publik Vol. 5, No. 1, Oktober 2008
46
statistically not significant in the 0.05
level for period 1994-2000 and 2001-
2004. While the coefficient of PD is
statistically significant in the 0.05 level
and has negative relationship to Theil
Index.

Table 5, OLS Estimation Regression Result For Income Disparity
Theil
Index
Including Gas and Oil Excluding Gas and Oil
19994-2000 2001-2004 1994-2004 1994-2000 2001-2004 1994-2004
Constant 0.225 0.279 0.125 -0.0002 0.121 0.09
(0.68) (3.88) (2.56) (-0.01) (1.46) (2.20)
FD -0.0032 -0.280** -0.0057 -0.00026 -0.136 -0.003
(-0.62) (-3.82) (-0.47) (-0.05) (-1.61) (-0.27)
HDI -0.0003 0.00005 -0.0019* 0.0008 0.0002 -0.0013*
(-0.68) (00000) (-2.57) (0.02) (0.44) (-2.21)
PD -0.00002* -0.00003** -0.00005** -0.00003* -0.00003** -0.00005**
(-2.28) (-4.13) (-2.72) (-2.80) (-3.57) (-3.13)
OR -0.042 -0.306** 0.052 -0.040 -0.174* 0.033
(-1.89) (-4.04) (1.33) (-1.80) (-1.99) (1.01)
TE 0.261** 0.0064** 0.006** 0.021** 0.0059** 0.0051**
(9.76) (10.91) (3.70) (7.85) (8.66) (3.75)

N 26 30 60 26 30 60
R-square 0.87 0.90 0.39 0.85 0.85 0.43
Figures in parentheses are the absolute values of t- statistic
Asterisks indicate that variable whose coefficients are significant at the 0.01(**) and
0.05 (*) level.


The other variables is OR and TE. The
coefficient of OR is negative in the
period of 1994-2000 and 2001-2004,
while it is positive if we employ pooled
data 1994-2004. It means that the higher
the ratio of own regional revenues to its
regional expenditures, the smaller the
income per capita disparity among
regions. It also can be explained that the
coefficient of OR is higher in period
2001-2004 than that in 1994-2000, since
fully implementing higher portion of FD
in 2001 lures local government to
explore more their own revenue sources.
The coefficient of TE is positive for all
period used data and significant in the
0.01 level. It underlines the opinion that
the higher the ratio of total local
government expenditure to total GRDP
the higher the income per capita dispa-
rity among regions.

CONCLUSION

There are two opposite theories
regarding the relationship between fiscal
decentralization, economic develop-
ment, and regional income disparity.
The conventional theory says that fiscal
decentralization will decrease economic
growth especially in developing and
transition countries. Fiscal decentrali-
zation also could increase regional
income disparity among regions in the
country. In contrast, new theories argue
that fiscal decentralization would not
only increase efficiency and economic
development but also reduce income
disparity among regions within a country.

Fiscal Desentralization And Indonesia Regional Income Disparity
(1994-2004)
47
Indonesia has conducted a
decentralized governmental system
starting in year 2001. Therefore the
intergovernmental transfers allocated to
sub national government have increased
substantially at almost two and half
times higher than before the decentra-
lization. Actually the primary difference
before and after fiscal decentralization in
Indonesia is in the degree of freedom of
regional government to spend their reve-
nues, whether the revenues come from
own sources or from intergovernmental
transfer. Before 2001 the transfer reve-
nues from central government was smaller
and the objectives of the expenditure
were already specified. While after 2001
the transfer revenues from central go-
vernment is higher because of the
increasing rate of revenues sharing. The
most interest thing is that the regional
governments have independence to
allocate their fund to deal with their
need.
During 1994-2004, economic
growth rate reached a peak in 1994. The
growth rates then slowed down in the
following years and finally plummeted to
the lowest minus growth rate in 1998,
that is when economic crisis hit Indo-
nesia. Up until now, the economic growth
rate is still under recovery phase. How-
ever, the economic growth rate has not
reached at least the same level to that in
1994. Many regulations and polices had
been conducted in order to stimulate
economic growth including fiscal decen-
tralization.
Considering the economic con-
vergence, the empirical evidence found
that in the period of 1994-2000 there is
economic divergence, while in the
period of 2001-2004 there is economic
convergence. It is proven by the sign of
coefficient of initial per capita income in
the regression estimation. This finding
conforms to new theories of fiscal
decentralization. The sigma convergence
did not occur since the coefficient of
variation of income per capita fluctuated
during period of observation.
On the other hand, by imple-
menting fiscal decentralization policy,
local governments face the higher
coefficient of variation in government
revenues per capita. It is clear since the
regions whose abundant natural re-
sources have benefit from it. While the
regions whose lacks of natural resources
have no advantages.
Fiscal decentralization has
positive impact on economic growth
rate, since its coefficient is positive to
economic growth in the regression
model. This relationship is relevance
with new theories of decentralization in
which the role of fiscal decentralization
can increase economic growth.
Fiscal decentralization has nega-
tive impact on the disparity of regional
income per capita. It is proven also in
the regression model and match with
new theories of fiscal decentralization,
where it is stated that fiscal decentra-
lization also could reduce regional
income disparity.
There are some doubts in the
regression results incorporate with data
limitation, reliability and validity of the
data, and omitted variables. It is possible
that other variables which correlate with
dependent variables may influence the
result such as public and private invest-
ments, the level of education and so on.
Further research is still needed to reveal
the link between fiscal decentralization,
economic growth, and regional income
disparity.


Jurnal Keuangan Publik Vol. 5, No. 1, Oktober 2008
48

References
Akai, Nobuo and Masayo Sakata (2005),
Fiscal Decentralization, Commit-
ment and Regional Inequality:
Evidence from State-Level Cross-
Section Data for the U.S., Faculty
of Economic University of Tokyo.
Bahl, Roy and Linn, Johannes F (1992),
Urban Public Finance in Deve-
loping Countries, New York:
Oxford University Press.
Barro, Robert J (1990), Government
Spending in a Simple Model of
Endogenous Growth, The Journal of
Political Economy, Vol.98 No.5,
Part 2 pp.S103-S125.
Barro, Robert J (1991), Economic Growth
in a Cross Section of Countries,
The Quarterly Journal of Econo-
mics, Vol. 106, No.2 pp.407-443.
Barro, Robert J and Sala-i-Martin, X
(1995), Economic Growth, MIT
Press, Cambridge.
Brodjonegoro, Bambang (2005), Three
years of fiscal decentralization in
Indonesia: its impacts on regional
economic development and fiscal
sustainability, Department of
Economics, University of Indonesia
Canaleta, Carlos G, Pedro P. Arzoz,
Manuel R. Garate and Roberto E.
Orayen (2002), Decentralization
and Regional Economic Dispari-
ties, Departamento de Economía,
Campus de Arrosadia, Univer-
sidad Pública de Navarra, 31006
Pamplona, SPAIN.
Dauglas Seymour, Williams Oral (1997),
The Impact of Government
Expenditure on Econmic Growth
in The OECS: A Disaggregated
Approach, Economics and
Programming Devision, Caribbean
Development Bank/World Bank
Research Department, Eastern
Caribbean Central Bank
Davoodi, Hamid and Zou Heng-fu
(1998), Fiscal Decentralization and
Economic Growth: A Cross-Country
Study, Journal of Urban Economics
43-2 pp 244–257.
Heins, A. James, (1971) State and Local
Response to Fiscal Decentrali-
zation, The American Economic
Review, Vol.61, No.2, pp.449-455.
Hofman, Bert, Kadjatmiko, Kai Kaiser,
Bambang suharnoko Sjahrir,
(2006), Evaluating Fiscal Equali-
zation in Indonesia, World Bank
Policy Research Working Paper
3911.
Kim, Euijune, Hong Sung Woong and Ha
Soo Jung Ha (2003), Impacts of
National Development and
Decentralization Policies on
Regional Income Disparity in
Korea, The Annals of Regional
Science 37 pp 79-91.
Martinez-Vazquez, Jorge and Robert M.
McNab (2001), Fiscal Decentra-
lization and Economic Growth,
Working Paper 01–1, International
Studies Program, Andrew Young
School of Policy Studies, Georgia
State University, Atalanta.
M. Zulfan Tadjoeddin, Widjajanti I.
Suharyo, Satish Mishra, (2003),
Aspiration to Inequality: Regional
Disparity and Centre-Regional
Conflicts in Indonesia, United
Nations Support Facility for
Indonesia Recovery (UNSFIR),
Jakarta.
Oates, Wallace E. (1993), Fiscal Decen-
tralization and Economic Deve-
lopment, National Tax Journal Vol.
46 No. 2.

Fiscal Desentralization And Indonesia Regional Income Disparity
(1994-2004)
49
Oates, Wallace E. (1999), An Essay on
Fiscal Federalism, Journal of
Economic Literature Vol 37 No 3
pp1120–1149.
Paas Tiiu, Aschilitte Friso (2006),
Regional Income Disparity; Con-
vergence or Divergence within EU-
25? Paper for the 14 th INFORUM
conference, Traunkirchen, Austria
Prud’homme, Remy (1995), The Dangers
of Decentralization, The World
Bank Research Observer Vol 10
No 2, Washington.
Resosudarmo, Budy P, Vidyattama, Yogi
(2006), Regional Income Disparity
in Indonesia: a panel data analysis,
Asean Economic Bulletin.
Rodden, Jonathan, (2002) The Dilemma
of Fiscal Federalism: Grants and
Fiscal Performance Around the
World, American Journal of
Political Science, Massachusetts
Institute of Technology, Vol.46
No.3, pp.670-687.
Sala-i-Martin, X (1996), Regional Cohesion
: Evidence and Theories or Regional
Growth and Convergence,
European Economic Review 40,
1325-1352.
Sidik, Machfud (2002), Kebijakan,
Implementasi dan Pandangan ke
depan Perimbangan Keuangan
Pusat dan Daerah, Ministry of
Finance R.I
Tanzi, Vito and Ehtisam Ahmad, (2002),
Managing Decentralization.
Rountledge
White, Roland and Smoke, Paul (2005),
East Asia Decentralizes Making
Local Government Work, World
Bank, Washington.
Young Andrew T, Higgins Matthew J, Levy
Daniel (1996), Sigma Convergence
versus Beta Convergence: Evidence
from U.S. County-Level Data,
Working paper.






Jurnal Keuangan Publik Vol. 5, No. 1, Oktober 2008
50
Appendix

Table A1 Average annual real GRDP growth
Average annual GRDP Growth Average annual GRDP Growth
No. Provinces Including gas and oil Excluding gas and oil
1994-2000 2001-2004 1994-2000 2001-2004
1 Aceh -2.38 2.55 3.54 0.93
2 North Sumatra 4.26 4.90 4.56 3.43
3 West Sumatra 4.01 4.74 4.01 3.19
4 Riau 3.35 2.75 7.02 3.28
5 Jambi 4.61 5.55 3.71 3.47
6 South Sumatra 1.85 3.88 2.10 1.96
7 Bangka Belitung n.a 5.14 n.a 1.29
8 Bengkulu 3.37 4.85 3.37 2.90
9 Lampung 4.25 5.44 4.10 3.45
10 Jakarta 2.66 5.78 2.66 2.77
11 West Java 0.95 4.91 0.77 1.66
12 Banten n.a 5.27 n.a 1.32
13 Central Java 2.90 4.36 2.86 2.53
14 Yogyakarta 3.01 4.87 3.01 2.72
15 East Java 3.22 4.59 3.18 2.75
16 West Kalimantan 5.18 4.09 5.18 3.61
17 Central Kalimantan 4.38 4.70 4.38 3.36
18 South Kalimantan 5.11 4.33 5.04 3.62
19 East Kalimantan 5.25 3.16 6.14 3.64
20 North Sulawesi 1.35 2.51 1.35 1.30
21 Gorontalo n.a 6.59 n.a 1.65
22 Central Sulawesi 7.79 6.00 7.79 5.40
23 South Sulawesi 6.31 5.20 6.26 4.44
24 Southeast Sulawesi 3.87 7.11 3.87 3.72
25 Bali 4.15 3.68 4.15 3.00
26 West Nusa Tenggara 8.38 5.09 8.38 5.46
27 East Nusa Tenggara 5.07 5.26 5.07 3.85
28 Maluku -5.07 3.04 -5.06 -1.77
29 North Maluku n.a 3.18 n.a 0.79
30 Papua 10.05 2.80 10.59 5.86

Fiscal Desentralization And Indonesia Regional Income Disparity
(1994-2004)
51
Table A2 Average annual real GRDP per capita growth

Average annual GRDP per
capita Growth
Average annual GRDP per capita
Growth
No. Provinces Includin gas and oil Excludin gas and oil
1994-2000 2001-2004 1994-2000 2001-2004
1 Aceh -3.29 2.04 2.56 2.61
2 North Sumatra 3.11 3.55 3.41 3.63
3 West Sumatra 3.81 3.30 3.81 3.30
4 Riau -0.57 0.31 2.96 4.99
5 Jambi 3.52 3.57 2.62 3.46
6 South Sumatra 2.85 2.16 3.00 2.74
7 Bangka Belitung n.a 2.33 n.a 2.22
8 Bengkulu 1.95 3.59 1.95 3.62
9 Lampung 3.50 4.31 3.36 3.57
10 Jakarta 3.34 4.60 3.34 4.60
11 West Java 1.31 3.20 1.09 3.31
12 Banten n.a 3.15 n.a 2.94
13 Central Java 1.93 3.87 1.89 3.56
14 Yogyakarta 2.04 3.49 2.04 3.26
15 East Java 2.59 3.72 2.55 3.70
16 West Kalimantan 3.05 3.91 3.05 3.62
17 Central Kalimantan 1.63 4.05 1.63 3.92
18 South Kalimantan 4.04 2.47 3.97 2.51
19 East Kalimantan 3.28 0.36 4.16 3.79
20 North Sulawesi 4.44 1.30 4.44 1.30
21 Gorontalo n.a 4.58 n.a 4.58
22 Central Sulawesi 5.31 4.91 5.31 4.93
23 South Sulawesi 4.97 4.15 4.92 3.86
24 Southeast Sulawesi 1.01 5.58 1.01 5.20
25 Bali 2.68 2.27 2.68 2.27
26 West Nusa Tenggara 6.47 4.21 6.47 4.21
27 East Nusa Tenggara 3.57 3.14 3.57 2.84
28 Maluku 0.78 1.65 0.79 1.53
29 North Maluku n.a 12.67 n.a 12.65
30 Papua 7.11 0.37 7.64 0.81



Jurnal Keuangan Publik Vol. 5, No. 1, Oktober 2008
52
Table A3 Average annual real GRDP per capita (billion rupiah)
Average Annual GRDP per capita Average Annual GRDP per capita
No. Provinces Including gas and oil Excluding gas and oil
1994-2000 2001-2004 1994-2000 2001-2004
1 Aceh 2,753,678.54 2,398,784.48 1,552,323.56 1,643,083.78
2 North Sumatra 2,014,843.07 2,222,544.78 1,989,077.83 2,211,256.84
3 West Sumatra 1,735,909.08 1,988,860.30 1,735,909.08 1,988,860.30
4 Riau 4,636,243.48 4,800,445.66 1,898,255.48 2,274,867.39
5 Jambi 1,286,584.15 1,499,056.64 1,220,874.04 1,352,591.49
6 South Sumatra 1,663,842.14 1,699,027.94 1,382,012.84 1,360,020.02
7 Bangka Belitung n.a 2,205,009.85 n.a 2,202,584.42
8 Bengkulu 1,159,882.24 1,300,006.87 1,159,882.24 1,300,006.87
9 Lampung 1,008,504.12 1,169,911.06 1,005,651.08 1,143,314.56
10 Jakarta 6,941,306.13 7,785,037.31 6,941,306.13 7,785,037.31
11 West Java 1,638,213.47 1,674,107.57 1,551,950.33 1,564,308.58
12 Banten n.a 2,209,530.96 n.a 2,204,934.34
13 Central Java 1,314,443.43 1,425,845.66 1,241,832.63 1,320,542.62
14 Yogyakarta 1,639,167.12 1,724,028.17 1,639,167.12 1,719,883.82
15 East Java 1,642,490.91 1,764,928.20 1,639,752.21 1,761,989.14
16 West Kalimantan 1,758,786.23 1,937,884.06 1,758,786.23 1,932,353.41
17 Central Kalimantan 2,269,635.75 2,388,589.10 2,269,635.75 2,385,495.50
18 South Kalimantan 2,002,813.61 2,277,200.34 1,982,097.71 2,248,749.93
19 East Kalimantan 8,482,084.97 9,390,179.23 4,527,128.75 5,380,321.44
20 North Sulawesi 1,035,167.27 1,207,673.77 1,035,167.27 1,207,673.77
21 Gorontalo n.a 1,223,746.46 n.a 1,223,746.46
22 Central Sulawesi 1,030,008.34 1,227,162.07 1,030,008.34 1,227,162.07
23 South Sulawesi 1,078,975.61 1,384,871.91 1,077,125.69 1,376,113.14
24 Southeast Sulawesi 921,536.63 1,040,519.41 921,536.63 1,036,440.21
25 Bali 2,352,710.20 2,514,583.45 2,352,710.20 2,514,583.45
26 West Nusa Tenggara 873,431.37 1,230,824.85 873,431.37 1,230,824.85
27 East Nusa Tenggara 730,395.18 837,877.82 730,395.18 835,275.08
28 Maluku 1,233,041.91 1,121,172.03 1,226,471.81 1,113,587.59
29 North Maluku n.a 1,196,867.76 n.a 1,196,674.10
30 Papua 3,370,183.61 3,872,116.72 3,250,540.71 3,803,964.57


Fiscal Desentralization And Indonesia Regional Income Disparity
(1994-2004)
53

Table A4 Average annual revenues per capita (billion rupiah)
Average Annual revenues per capita
No. Provinces Before and after fiscal decentralization
1994-2000 2001-2004
1 Aceh 202,644.49 982,277.52
2 North Sumatra 132,757.21 516,790.65
3 West Sumatra 177,624.72 690,360.08
4 Riau 190,957.52 1,535,695.21
5 Jambi 182,076.03 832,979.41
6 South Sumatra 135,914.12 568,442.83
7 Bangka Belitung n.a 785,234.78
8 Bengkulu 170,633.54 647,077.11
9 Lampung 102,595.99 409,606.93
10 Jakarta 335,670.22 1,081,663.55
11 West Java 93,736.19 350,211.18
12 Banten n.a 378,878.24
13 Central Java 99,691.57 399,409.23
14 Yogyakarta 153,516.38 578,172.15
15 East Java 97,583.00 412,332.84
16 West Kalimantan 195,388.58 636,678.23
17 Central Kalimantan 316,132.33 1,082,215.62
18 South Kalimantan 257,623.80 771,947.34
19 East Kalimantan 329,811.99 3,204,819.83
20 North Sulawesi 209,377.35 701,955.99
21 Gorontalo n.a 748,008.77
22 Central Sulawesi 209,809.48 718,535.55
23 South Sulawesi 144,585.64 604,475.41
24 Southeast Sulawesi 197,013.19 761,485.34
25 Bali 249,209.84 897,674.80
26 West Nusa Tenggara 132,800.09 519,539.56
27 East Nusa Tenggara 169,522.38 703,915.20
28 Maluku 216,482.40 988,018.06
29 North Maluku n.a 1,041,481.39
30 Papua 475,552.10 2,235,848.40


-oOo-


Jurnal Keuangan Publik Vol. 5, No. 1, Oktober 2008
54