World Development Vol. 40, No. 8, pp.

1511–1521, 2012
Ó 2012 Elsevier Ltd. All rights reserved
0305-750X/$ - see front matter
www.elsevier.com/locate/worlddev
http://dx.doi.org/10.1016/j.worlddev.2012.04.005

Fiscal Decentralization and Development Outcomes in India:

An Exploratory Analysis
KALIAPPA KALIRAJAN
The Australian National University, Canberra, ACT, Australia

and

KEIJIRO OTSUKA *
National Graduate Institute for Policy Studies, Tokyo, Japan
Summary. — This paper attempts to quantify the impact of fiscal decentralization in India on its social infrastructure and on rural devel-
opment. Overall, the results in this paper indicate that Government of India within a federal framework has been fostering development
equitably across its states, particularly through health and education expenditures aimed at improving human capital development. In
this context, the importance of the two tier centre-states decentralization should be noted. However, the third tier of local governance,
particularly the state-rural local bodies decentralization has been dismal and has not achieved significant results across states, which
warrant the attention of the central and state governments.
Ó 2012 Elsevier Ltd. All rights reserved.

Key words — fiscal decentralization, development outcomes, Panchayati Raj, Asia, India

1. INTRODUCTION
One of the central questions in Development Economics is
why some countries have remained poor for a long period of
time, though the general policy approaches to combat poverty
are well understood. A corollary to the above question is why
the similar package of policies differs in efficacy across coun-
tries. Similarly, an identical set of policies differs in effective-
ness across provinces or states within a country. 1 Drawing
on the Post-Washington Consensus, it may be argued that
the inability to achieve similar results from policy packages
across states is due to the constraints posed by country-specific
organizational or institutional factors. One of the major insti-
tutional factors directly involved in the delivery of public ser-
vices and in implementing development policies is the degree
of decentralization. Therefore, a key to enhancing the efficacy
of policies lies in a better understanding of the extent and pro-
cess through which various forms of decentralization contrib-
ute to development (Appendix I).
What do we understand by decentralization? Decentralization
can best be understood as a political process in the sense of the
devolution of resources, tasks and decision-making power to
democratically elected lower-level authorities, which are largely
or wholly independent of central government (World Bank,
2000a). It is rational to argue that decentralization facilitates
time-specific and location-specific knowledge to implement
policies that influence people’s welfare. Decentralization in
political, fiscal, and economic systems affects development out-
comes in a number of ways. First, decentralized provision of so-
cial and physical infrastructures should correspond with the
diverse demand conditions in different regions and match their
resource endowments better than central provision. Even with
regard to the provision of quasi-public goods, identification of
target groups of beneficiaries is easier and implementation of
policies more effective when undertaken by decentralized gov-
ernmental units (Ostrom, Schroeder, & Wynne, 1993). Thus,
1511
even when there is some controversy over the redistributive role
of sub-national governments, their desirability in implementing
poverty alleviation policies is generally agreed upon (Brown and
Oates, 1987; Ladd and Doolittle, 1982; Pauly, 1973).
Second, competition among sub-national jurisdictions may
promote innovations and enhance productivity. At the same
time, as the decentralized governmental units function within
a large country-wide unified market free from impediments
to the movement of factors and products, it can provide a con-
genial environment for the efficient functioning of the market
economy. Thus, drawing on the decentralization theorem
introduced by Oates (1972), in an ideal decentralized system,
existing resources will be allocated to yield the maximum pos-
sible output (locating on the production possibility frontiers)
and the competitive environment including inter-governmen-
tal competition will be conducive for technological progress
(shift in the frontier). In reality, the possibility of decentraliza-
tion failing to overcome regional and local dimensions of pov-
erty and inequality (Prudhomme, 1995; Rodden, 2002) may
not be ruled out, mainly because the decision making power
on local developmental initiatives is often highly dependent
on central government bureaucracy. Further, decentralization
may increase the probability of empowering local elites in cap-
turing larger share of public resources at the cost of the poor
(Dreze & Sen, 1996).
With the recent developments in the political arena where
coalition government is at the center, and different regional
parties who are mostly the partners of the coalition are at
the states, the case study of the Indian experience of decentral-
* The authors are thankful to Dr. M. Govinda Rao, Director of NIPFP,
New Delhi, Dr. Shashanka Bhide, Dr. Kanhaiya Singh of NCAER, New
Delhi, and four anonymous reviewers of this journal for their valuable
comments and suggestions on an earlier draft of this paper. Final revision
accepted: February 20, 2012.
1512 WORLD DEVELOPMENT
ization provides an important context of understanding the
ways in which decentralization can influence overall socio-eco-
nomic welfare and rural development. The Government of In-
dia passed a series of Constitutional reforms in 1993 to
democratize and empower local administrative institutions be-
yond the two tiers of the center and the states. The third tier of
local governance comprises a three-tiered structure of rural lo-
cal bodies, and three levels of urban local bodies as per the
73rd and 74th Constitutional amendments, respectively. 2
Since then, the experiences of different states with respect to
decentralization vary a lot mainly due to differences in fiscal
decentralization, which is a major instrument of overall decen-
tralization.
As fiscal decentralization was not initiated by states them-
selves, but was thrust upon them by the center through the
1993 Constitutional reforms, the attitudes toward decentral-
ization varied across states. Though the literature on fiscal fed-
eralism explains the economic gains from decentralized
decision making and related issues (Rao & Singh, 2005), there
are very few empirical studies examining the causal relation-
ship between fiscal decentralization and development out-
comes in India. 3 Much of the demonstrated gains are in the
nature of assertions or qualitative statements. This study at-
tempts to analyze and quantify the impact of fiscal decentral-
ization in India on its social infrastructure that needs to be
supplied by governments as they are not optimally provided
by the private sector and on rural development where about
70% of the population lives. In this study, the ratio of per ca-
pita expenditure of Panchayati Raj (rural) institutions to per
capita expenditure of urban local bodies is used as a measure
of states-rural local bodies decentralization in the context of
third tier level of local governance.
The following section discusses some important issues con-
cerning fiscal decentralization in India. The next section exam-
ines the measures of spatial disparity in social infrastructure
development across states. The following section analyzes
the impact of decentralization on social infrastructure and
on rural development. A final section brings out the overall
conclusions of this study.
2. FISCAL DECENTRALIZATION IN INDIA—MAJOR
ISSUES 4
The basic framework of inter-governmental relationships in
Indian federation is given by the Constitutional assignment of
functions and sources of finance. The seventh schedule to the
Constitution of India specifies the Union list—the exclusive
domain of the central government, the State list—the exclusive
domain of the state governments, and the Concurrent list
where both levels have joint jurisdiction.
An important feature of the Constitutional assignments in
India is the centripetal bias in the assignment of powers. First,
almost all broad based taxes, with the sole exception of sales
tax, are assigned to the central government. Second, effec-
tively, states do not have independent borrowing powers.
When a state is indebted to the center, it must seek the center’s
permission to borrow. As all the states are indebted to the cen-
ter, their market borrowing is determined by the Union Min-
istry of Finance in consultation with the Planning Commission
and the Reserve Bank of India. Also the states cannot incur
overdrafts with the Reserve Bank of India for more than 10
continuous working days. The Constitution not only assigns
overwhelming powers to the central government, but also
has overriding powers in the event of a conflict in a concurrent
jurisdiction.
The centripetal bias is not merely the consequence of Consti-
tutional assignment. The adoption of planned development
strategy and allocation of resources according to priorities
determined by the planning agency have also contributed im-
mensely to the centralization of economic power. Although
planning was originally to be implemented in a decentralized
manner, the involvement of sub-national governments in re-
source allocation has in practice been limited. Comprehensive
central planning is the negation of federalism; and planning
even in a mixed economy framework has significantly contrib-
uted to the concentration of economic power with the center.
In the event, the potential benefits of decentralization have not
been realized. It is worth noting that decentralization else-
where with the similar characteristic of the center having more
power over the states has positive outcomes. For example,
though the central government maintains control of certain
major expenditures, such as education and health, South Afri-
ca’s governance system has been restructured, which allow for
the policy implementation on a local level. Such decentraliza-
tion has significantly contributed to the post-apartheid unifica-
tion (World Bank, 2000b, chap. 5).
Achieving the potential benefits of fiscal decentralization in
the Indian federation has been further obstructed by the vari-
ous fiscal and regulatory impediments to the movement of fac-
tors and products across the country (Rao, 2004). Some of
these impediments were created for the economic management
of supply shortage of essential commodities. The fiscal imped-
iments were the consequence of the free-riding behavior of the
states, particularly the attempt to export the tax burden to
nonresidents. These impediments, along with high levels of
protection, distorted relative prices and created a high cost
economy.
Another reason for centralization has been the failure to
specify the powers and functions of governmental units below
the state level, until the 73rd Constitutional amendment was
given effect recently. The state governments, following the Bal-
wantrai Committee report, adopted the local self government
unit in rural areas (Panchayati Raj), right from the village le-
vel. However over the years, in many of the states these insti-
tutions became inactive with elected councils being superseded
for indefinite periods. In some states, notably in Andhra Pra-
desh, Karnataka, and West Bengal there was an attempt at
rejuvenating these local self governments at the village level
by activating them politically, making greater devolution of
powers and finances to them and in some cases, consolidating
them into more viable economic and administrative units.
However, frequent changes in the ruling party in Karnataka
and Andhra Pradesh rendered these experiments abortive,
though in West Bengal it has continued. With the 1993 Con-
stitutional amendment, however, it is now mandatory to have
a uniform system of Panchayati Raj throughout the country.
The 11th schedule of the Constitution specifies a list of 29 sub-
jects to be carried out by local governments (Appendix II).
Many states have already transferred responsibilities for the
29 subjects listed in the 11th schedule of the Constitution
through legislation, in reality not all subjects have been carried
out by rural local bodies in those states.
For example, among the South Indian states, Karnataka,
and Kerala have decentralized their local governance effec-
tively by providing responsibilities to rural local bodies for
all 29 subjects not only through legislation, but also through
actual implementation (Appendix III). On the other hand,
Andhra Pradesh and Tamil Nadu have paid less attention to
decentralization. Though Andhra Pradesh has agreed through
legislation to transfer 17 out of 29 subjects to rural local
bodies, in reality only nine subjects have been given until
 FISCAL DECENTRALIZATION AND DEVELOPMENT OUTCOMES IN INDIA: AN EXPLORATORY ANALYSIS
recently. With respect to other states in India, Madhya
Pradesh, where 23 out of 29 subjects have been transferred
from state to Panchayati institutions, paid more attention to
decentralization and human development than any other
states in East and West India. In East India though West
Bengal and Kerala had similar political parties in power for
several years, the former’s progress in state-rural local body
decentralization has not been on par with that of Kerala,
where 29 out of 29 subjects have been transferred from the
state to rural local bodies effectively. However, in another East
Indian state of Orissa, 16 subjects have been transferred from
the state administration to rural local bodies, particularly to
Zilla Parishads. Though the North Indian state of Haryana
has legislated that it would also follow the models of Karna-
taka and Kerala to transfer 29 subjects listed in the 11th Sche-
dule of the Constitution, it is not clear yet when exactly it will
be carried out, as only 10 subjects have been actually under-
taken by rural local bodies. (Planning Commission, 2006).
Though there are several reasons for this varied interest in
decentralization across states, one major reason is the number
of elected representatives, such as Members of Parliament
(MP) and Members of Legislative Assemblies (MLA). The
MP Local Area Development Funds and MLA Local Area
Development Funds allotted by the center and state govern-
ments are often used to dilute the powers of rural local bodies
such as Panchayats (AnilKumar, 2009).
The 1993 amendment also provides for regular elections to
these Panchayats and objective methods of devolution of re-
sources by appointing State Finance Commissions (SFC).
However, SFC’s recommendations generally are not taken in
tandem by most states due to various factors including lack
of confidence in the analytical procedures followed by SFC.
At present, the shortcomings of lack of proper data and
expertise in SFC are being rectified gradually in many states
(Planning Commission, 2006).
The above discussion brings out two important features.
First, the potential economic benefits of fiscal decentralization
in the Indian federation could not be achieved fully due to the
centripetal bias in the constitution and the centralization of
economic powers due to planning and other developments.
Of course, it is extremely difficult to quantify the loss of poten-
tial output or welfare due to the absence of the desired degree
of fiscal decentralization. Second, there has been a significant
inter-state variation in the degree of fiscal decentralization and
other institutional situations. This has impacted differently on
states. Therefore, the analysis of different states should lead to
a better understanding of the economic dynamics of fiscal
decentralization.
Another important feature of the Indian fiscal federalism is
the existence of severe vertical and horizontal imbalances
(Rao, 2005). States have been assigned expenditure responsi-
bilities far in excess of their revenue resources. In particular,
the Constitution assigns predominant responsibility of build-
ing social and physical infrastructure for rural development
to the states. Also, being closer to the people, the states have
to cater to the needs of the poor much more responsively than
the central government. Furthermore, the capacity to raise
revenues varies widely among the states. Variation in per capi-
ta incomes even among the 14 general category states is almost
1:4. 5 General purpose transfers are required to offset these
vertical and horizontal imbalances. Additionally, for services
with a high degree of inter-state spillovers or those in the nat-
ure of merit goods, specific purpose transfers are necessary to
ensure availability of certain minimum levels. To ensure this,
specific purpose transfers, preferably with matching require-
ments are necessary.
1513
In India, both general purpose and specific transfers are gi-
ven, but their design and implementation leave much to be de-
sired. The Finance Commission, a statutory body, and the
Planning Commission give general purpose transfers, but nei-
ther has been able to design the transfers to offset fiscal disad-
vantages of the states with low revenue capacity and high unit
cost of providing public services (Rao, Shand, & Kalirajan,
1998). In fact, the two agencies sometimes have worked at
cross purposes. The bulk of the transfers given by both agen-
cies is on the basis of general economic indicators and is not
targeted to offset fiscal disadvantages of the states with low
revenue capacity and high unit cost of public services. The
practice of filling projected budgetary gaps by the Finance
Commissions has created disincentives for prudent fiscal man-
agement. Thus, the general purpose transfers have not served
the objectives of either equity or efficiency to the desired
extent.
There are a number of problems in the design of specific
purpose transfers given for poverty alleviation. The prolifera-
tion of specific purpose transfers for a variety of purposes has
constrained the availability of resources for spending on pov-
erty alleviation and has spread the resources thinly across sev-
eral programs. The volume of resources available for transfer
to poverty alleviation is inadequate to make any impression on
the poverty problem (Kalirajan & Singh, 2010). Nor is the de-
sign of the transfer system targeted to alleviate poverty (for de-
tails, see Rao & Das-Gupta, 1995). Thus, the design and
implementation of inter-governmental transfer systems in
India leave much to be desired. This is another factor creating
the divergence between the potential and actual rural develop-
ment outcomes and there are inter-state differences in these
divergences. Further, inter-governmental transfers create dif-
ferent degrees of fiscal autonomy to different states and these
influence development outcomes differently. When states in
India are compared for fiscal autonomy and development out-
comes, it may be observed that local institutions of decentral-
ization are more effective in states where land reforms and
social movements have been successful in promoting social
egalitarianism (Bardhan, 2002).
3. IS THERE SPATIAL DISPARITY IN SOCIAL
INFRASTRUCTURE DEVELOPMENT?
As the major objective of decentralization is to provide easy
access to local public goods that improve social infrastructure,
we first examine the status of some of the indicators of social
infrastructure across states. An effective functioning of decen-
tralization across states is expected not to widen spatial dispar-
ity in social infrastructure development. We adopt the measure
of weighted coefficient of variation to examine the trends in
spatial balance in the basic social development indicators: In-
fant Mortality Rate (IMR), Life Expectancy at birth (LE), Lit-
eracy Rate (LR), Telecom Density per thousand population
(TD), and per capita Electricity Consumption (EC) in kwh,
and the results are given in Tables 1 and 2. The trends clearly
show that there is a reduction in disparity in literacy rate, and
life expectancy at birth.
In the case of infant mortality rate, the trend appears to show
a slightly increasing disparity in later years (Table 1). For
example, in 2006, Madhya Pradesh had the highest infant mor-
tality rate with 74 deaths in 1000 births, followed by Orissa at
73 and Uttar Pradesh at 71. On the other hand, the mortality
rates in Kerala, Maharashtra, and Tamil Nadu were 15, 35,
and 37 respectively, as these states have effective primary health
care delivery systems reaching down to the poorest section of
1514 WORLD DEVELOPMENT

Table 1. The pace of development and spatial balance in development: selected human development indicators
Year Life expectancy at birth Year Infant mortality rate (per Year Literacy rate
thousand)
Average CV Average CV Average CV
1983 60.2 0.7398 1971 121 0.0565 1971 32.8 0.0830
1988 61.1 0.7299 1976 123 0.0650 1981 42.2 0.0655
1990 59.2 0.7541 1981 107 0.0681 1991 51.2 0.0575
1991 59.7 0.7472 1986 95 0.0685 2001 63.9 0.0370
1993 60.5 0.7376 1991 77 0.0708 2002 64.2 0.0322
1995 61.2 0.7298 1994 53 0.1237
1999 62.1 0.7197 1998 66 0.0670
2000 62.3 0.7175 2005 54 0.0680
2001 62.5 0.7157 2006 62 0.0824
2003 63.0 0.7107
Note: The coefficient of variation is calculated over data for 15 major states of India. The CV is estimated as a weighted measure using population shares
of the states as weights.

Table 2. The pace of development and spatial balance in development: selected infrastructural development indicators
Year Telecom density (% popln.) Year Electricity consumption (kwh)
Average CV Average (per cap.) CV
1980 0.38 0.3350 1975 97 0.1157
1985 0.40 0.1661 1980 121 0.1322
1987 0.47 0.1717 1983 150 0.1335
1988 0.48 0.1720 1985 168 0.1298
1989 0.52 0.1726 1987 196 0.1352
1990 0.55 0.1670 1990 232 0.1285
1991 0.62 0.1654 1991 247 0.1244
1992 0.71 0.1603 1992 263 0.1268
1993 0.83 0.1592 1993 288 0.1197
1994 1.00 0.1606 1994 306 0.1234
1995 1.20 0.1629 1996 336 0.1235
1998 1.68 0.1275 2002 359 0.1597
1999 2.37 0.1499 2004 387 0.1523
2000 2.68 0.1498 2006 396 0.1488
2003 4.24 0.1555
2004 5.69 0.1618
2006 10.74 0.1250
2007 15.85 0.1139
See Note to Table 1.

Table 3. Trends in state’s own revenue receipts and revenue expenditures

Year State’s own receipts to total receipts (%) State’s rev. expend. to total expend. (%) State’s own rev. receipts to state’s rev. expend. (%)
1990–91 35.2 54.6 53.1
1995–96 39.2 57.0 58.6
1999–00 38.6 56.4 49.8
2000–01 37.8 56.0 48.6
2001–02 40.2 56.9 50.0
2002–03 38.28 54.31 56.01
2003–04 37.54 56.27 53.83
2004–05 38.10 56.28 60.05
2005–06 37.07 56.59 56.07
Source: Table 7 in Rao, Sen, and Jena (2008).

population (Kapoor, 2009). Further, what is alarming is that
rural infant mortality is higher than urban infant mortality rate
in all states. On examining the per capita expenditure on social
sectors, which includes health at village level, through Panc-
hayati Raj institutions given in Table 6, it is evident that per ca-
pita expenditure on social sectors at the rural level have been
very low in Madhya Pradesh (Rs. 110), Orissa (Rs. 57), and Ut-
tar Pradesh (Rs. 44) compared to the best performing states of
Kerala (Rs. 742) and Maharashtra (Rs. 821). Also, malnutri-
tion due to poverty and other factors such as poor access to
hospitals due to lack of infrastructure such as roads and trans-
portation in rural areas, and the number of scheduled caste and
 FISCAL DECENTRALIZATION AND DEVELOPMENT OUTCOMES IN INDIA: AN EXPLORATORY ANALYSIS
Table 4. Finance Commission (FC), Planning Commission, and nonstatutory transfer of resources from the center to states
(A) Vertical transfer
Transfers from center to states as percentage of gross revenue receipts of the center
Period Finance Commission transfers Other transfers Total transfer
(4 + 7)
Share in central Share in Total transfers Grants through Planning Nonplan grants Total other (%)
taxes (%) grants (%) through Finance Commission (%) (nonstatutory) (%) transfers (5 + 6) (%)
Commission (2 + 3) (%)
1 2 3 4 5 6 7 8
VII FC 22.39 1.96 24.35 12.11 1.66 13.77 38.11
VIII FC 20.25 2.52 22.77 13.56 1.54 15.1 37.86
IX FC 21.37 3.42 24.79 14.48 1.06 15.54 40.33
X FC 21.4 2.34 23.75 10.57 0.63 11.19 35.79
XI FC (first 2 years) 20.93 5.20 26.13 10.39 0.82 11.21 37.2
(B) Horizontal transfer
Criteria Weight (%)
Formula of 12th Finance Commission (2005–10)
Population 25.0
Income 50.0
Area 10.0
Tax effort 7.5
Fiscal discipline 7.5
Gadgil formula used in X plan (2002–07)
Population 60.0
Per capita income 25.0
Tax efforts 7.5
Fiscal management 7.5
Source: Authors’ compilation from various Planning Commission plan reports and Finance Commission reports.

1515
1516 WORLD DEVELOPMENT
scheduled tribes living are seen as major reasons for high infant
mortality in some states (Kapoor, 2009).
The trends in infrastructure development, which include
telephone density and per capita electricity consumption, also
show a general improvement, particularly in very recent times
(see Table 2).
Thus, there has been an improvement in recent times in four
out of the five development indicators considered, and the pro-
gress appears to be more spatially balanced than before recog-
nizing the importance of the center–states decentralization.
The movement toward more equitable levels of performance
amidst rising development indicates that there are factors
facilitating spatially equitable social infrastructure develop-
ment. Which are the factors contributing to such equality? Is
fiscal decentralization a significant contributing factor? An-
swers to these questions are attempted in the following pages.
4. FACTORS FACILITATING SPATIALLY EQUITABLE
DEVELOPMENT
The central government intends to ensure equal access to
basic services such as primary education, primary health care,
infrastructure such as roads, drinking water, electricity and
communication services across states. State governments then
are responsible for implementing a higher proportion of gen-
eral government spending than in most developing countries
with the exception of China. However, Table 3 shows that
states could finance their current expenditures or revenue
expenditures by their own revenues only by about 56%. There-
fore, the center shares some of its revenues from sources such
as income tax and excise duties with states. The norms for allo-
cation of resources from the center to states are mostly “for-
mula” based, which depend on factors such as tax efforts of
states without any distinction of economic backwardness. 6
There are two main institutions that are constitutionally cre-
ated for every 5 years by the President of India to transfer re-
sources from the center to states and they are the Finance
Commission and Planning Commission. Both institutions first
fix the amount of tax revenues that needs to be transferred
from the center to states, which is called “vertical transfer,”
and then distributes the allocated amount to states, which is
called “horizontal transfer” using different formulas. The
Planning Commission also does specific purpose transfers
for various central schemes implemented by different minis-
tries of the central government without using any formula,
which are called “nonstatutory” transfers (Rao and Chelliah,
1995). Table 4 shows the vertical transfer from the center to
states over the years, which indicates the overall transfer of
about 38% of center’s revenues to states. The formula used
for horizontal transfer by the XII Finance Commission
covering the period 2005–10 is given in Table 4 along with
the Planning Commission’s Gadgil formula used in the X Plan
(2002–07). 7 The Gadgil formula that is binding the Planning
Commission need not have any sanctity for the Finance Com-
mission (Rao, 2005).
Whether the instrument of decentralization has worked in
achieving spatial equitable distribution in social infrastructure
development in the Indian context? We examine this dimen-
sion of the process in the following section.
(a) Disparities in state government expenditures
What has been the pattern of expenditures across states
since the beginning of a policy regime which has been more
reliant on markets to deliver growth than in the past? We have
used revenue (or broadly, current expenditures) of the state
governments for analysis as they make up about 77% of total
expenditures today. Within revenue expenditures, “develop-
ment expenditures” make up to 80% of the spending. The
development expenditures refer to expenditures on various so-
cio-economic development programs in the social sectors and
economic sectors.
The “development expenditures” of the state governments
are grouped into two broad categories: social services and
economic services. The social services include health and edu-
cation. The economic services include development programs
in different sectors of the economy particularly in infrastruc-
ture. For example, revenue expenditures on programs relating
to agriculture, roads, electricity, and industries are grouped
under economic services. These are not capital expenditures
but relate to expenditures on operation and maintenance of
on-going programs.
Data were obtained from the Handbook of Statistics on
State Government Finances (RBI, 2004) and previous publica-
tions on State Finances by the Reserve Bank of India (RBI).
The state level population estimates are obtained by interpo-
lating decadal Census estimates to obtain per capita expendi-
tures for the states. The values in current prices are deflated
by the wholesale price index to obtain expenditures and GSDP
in real or constant prices.
The trends in government development expenditures, are
examined in a regression model in the framework of the “con-
vergence analysis.” The methodology of income convergence
analysis of Barro and Sala-i-Martin (1995) provides a useful
tool to examine whether the disparity across states in state
government expenditures is increasing.
The conditional convergence regression model we used is:
     
1 PCREt 1
ln ¼ a0 þ a1 ln PCREt0 þ a2
t  t0 PCREt0 t  t0
 ln PCY t0 þ a3 DCI þ u ð1Þ
where PCRE = per capita real state government expenditures.
PCY = per capita real gross state domestic product (average
of 3 years ending in year t or t0 in all cases except for 1980–
81 where the average is for 3 years beginning in 1980–81; the
averages are taken to remove abnormal data points).
t0 = beginning year of the time period. t = ending year of
the time period and rest of the variables are as defined earlier.
DCI = fiscal decentralization index, which is the share of
state’s revenue including statutory transfers to state’s expendi-
tures.
Given our interest in welfare indicators, we have extended
the analysis to cover four types of development expenditures
of the state governments: (1) expenditure on economic services
within development expenditures; (2) expenditure on social
services within development expenditures; (3) expenditure on
medical and public health services within social expenditures;
and (4) expenditure on education services within expenditures
on social services. We have examined the pattern of expendi-
ture for the periods of 1993–94, when the Constitutional re-
forms were passed to empower the third tier of local
governance, to 1999–2000.
Our interest is to examine if the expenditures across states
are likely to be “converging” or “diverging” over time. The
convergence would imply that the disparity is likely to be
declining and divergence would imply the opposite. In the
above equation, if the coefficient “a1” is positive, the per capita
expenditures of the different states would be moving at differ-
ent rates, with the “higher expenditure state” increasing expen-
ditures faster than the “lower expenditure states.” Therefore,
there would be no convergence of per capita expenditures
 FISCAL DECENTRALIZATION AND DEVELOPMENT OUTCOMES IN INDIA: AN EXPLORATORY ANALYSIS 1517

Table 5. Results of regression analysis of inter-state disparity in state government expenditures

Expenditures PCREt0 PCYt0 STPOP DCI
Eco. services 0.3129(0.2952) 0.2882(0.2416) 0.1762**(0.0807) 0.1774**(0.0878)
Social services 0.2627**(0.1292) 0.1955**(0.0899) 0.1881**(0.0920) 0.1375**(0.0640)
Pub. health 0.2772**(0.1355) 0.2174**(0.1080) 0.1825**(0.0875) 0.1448**(0.0723)
Education 0.2028**(0.1004) 0.2072**(0.1033) 0.1954**(0.0961) 0.1220**(0.0589)
**
Significant at the 5% level.

Table 6. Per capita expenditure through urban local bodies and Panchayati Raj Institutions in India, 2002–03.
State Per capita expenditure through Per capita expenditure through
urban local bodies (Indian Rupees) Panchayati Raj Institutions (Indian Rupees)
Himachal Pradesh 854 59
Madhya Pradesh 762 110
Tamil Nadu 727 153
Manipur 720 37
Kerala 693 742
Andhra Pradesh 657 898
Gujarat 653 783
Punjab 639 108
Chhattisgarh 614 355
Rajasthan 523 382
Maharashtra 489 821
Goa 443 419
West Bengal 426 30
Uttaranchal 398 46
Karnataka 329 1147
Uttar Pradesh 315 44
Orissa 395 57
Tripura 287 253
Jharkhand 192 0
Jammu & Kashmir 192 851
Assam 174 3
Bihar 162 40
Haryana 143 24
Meghalaya 124 25
Mizoram 33 34
All states average 463 316
Source: (Basic data) Report of XII Finance Commission, 2005.

across states and the disparity would increase over time. If the
coefficient is negative, then the per capita expenditures would
be converging or disparity would decrease over time. If the
coefficient is zero, then again disparity would not be rising.
The role of per capita real GSDP in the equation is to control
for overall “initial conditions” of the state economy. If the
average income (per capita GSDP) is larger in one state as
compared to the other, the expenditures may also be higher
because of the availability of larger resources (through own
taxes) to that state. Once we control for this variable, the pat-
tern that remains should reflect the influence of the other fac-
tors, including the devolution of resources from the central
government to the states, on the tendencies of the states to
spend. In order to control for this latter aspect of central gov-
ernment transfer and also the capacity of the state to collect
revenue, the Decentralization Index (DCI) has been used as
another controlling variable.
The results obtained from applying the ordinary least
squares method of estimation to Eqn. (1) with different depen-
dent variables with concerned independent variables explained
above are summarized in Table 5. The expenditures on “social
services” show a statistically significant convergence pattern
more consistently than the expenditures on “economic ser-
vices.” In fact, the expenditures on “economic services” show
no convergence. Thus, state government expenditures on so-
cial sectors appear to be pro-spatially equitable than the
expenditures on “economic services.” It is a known fact that
India lacks in infrastructure when compared with China.
Infrastructure such as highways and airports are not well
developed in states like Bihar, Madhya Pradesh, and Orissa,
while they are reasonably developed in states like Tamil Nadu,
Karnataka, and Maharashtra. Why does such diversity in
infrastructure exist across states? The answer points to several
factors including socio-political factors and it is beyond the
scope of this paper.
Within the “social services,” expenditures on both “health”
and education’ services show statistically significant conver-
gence for in the post-reform periods indicating that the mech-
anisms driving these expenditures are based more uniformly
on the needs of population across all regions of the country.
The coefficients of decentralization index are positive and sig-
nificant at the 5% level for social services, medical and public
1518 WORLD DEVELOPMENT
health, and education expenditures. These results support the
thesis that greater center–states two tier level of decentraliza-
tion improves overall economic growth, which in turn facili-
tates higher expenditure on social services. 8 These results are
in conformity with studies on decentralization in other coun-
tries. For example, Faguet (2004) found that decentralization
in Bolivia increased the responsiveness of public investment by
local governments to local demands. His empirical tests dem-
onstrated that patterns of investment in human capital and so-
cial services, such as education and water supply and
sanitation, improved significantly after decentralization. 9 Fur-
ther, the significant and positive coefficients of DCI in these
equations imply that state governments do pay special atten-
tion toward the improvement of human capital, though the
central transfer to states does not give any special priority
for education and health in states (Table 4). The initial condi-
tion of average per capita GSDP does not influence the
dynamics of expenditures on economic services, though it sig-
nificantly influences expenditures on social services.
The above results indicate that the two tier level of decentral-
ization appears to be an important instrument facilitating con-
vergence of “social services” expenditures across states.
However, what is more important is to examine whether decen-
tralization has contributed to rural development within the
third tier of governance (states-rural local bodies decentraliza-
tion) because 70% of India’s population still lives in rural areas.
Therefore, next, we examine the impact of states-rural local
bodies decentralization on rural development within the third
tier of governance, which has bearings on the effective func-
tioning of overall decentralization. Table 6 shows states’ per
capita expenditure through Urban Local Bodies and Panchay-
ati Raj Institutions in India during the period 2002–03. The
average figure for Urban Local Bodies-all states per capita
expenditure was Rs. 463, while the average sum for Panchayati
Raj-all states per capita expenditure was Rs. 316. Though the
average figures do not differ very much, the figures across
states vary widely. 10 Many states’ per capita expenditure
through Panchayati Raj Institutions was much below the all
states average. Thus, the state-wise figures do provide an indi-
cation about whether Panchayati Raj Institutions are func-
tioning effectively or not. Karnataka State has been in the
front in terms of effective functioning of Panchayati Raj Insti-
tutions, which is followed by Andhra Pradesh, Maharashtra,
Gujarat, and Kerala.
We use the Gross State Domestic Product from agriculture
and allied sector (GSDPA) at constant (1999–2000) prices as a
proxy for rural development and we use the state level data
published by the Central Statistical Organization. We use
the ratio of per capita expenditure of Panchayati Raj (rural)
institutions to per capita expenditure of urban local bodies
(RPEUL) as a proxy measure for three tier level of decentral-
ization. We use the panel data from 25 states covering the peri-
ods of 2000–01 to 2002–03. The following regression model
was formulated:
ln GSDPAit ¼ a þ b1 ln RPEULit þ b2 ln Fertit þ b3
 ln Laborit þ b4 ln Rainit þ b4 ln Areait þ uit ð2Þ
where Fert is the amount of fertilizer used in tons; Labour is
the number of agricultural labor force; Rain is the amount
of rainfall in millimeters; and Area refers to the gross cropped
area in hectares. In Eqn. (2) the subscripts “i” and “t” refer,
respectively, to the state and year. Eqn. (2) was estimated
using the fixed effects estimation using data from the period
of 2000–01 to 2002–03 and the estimates of the equation are
as follows:
ln GSDPA ¼ 3:5421 þ 0:0288 ln RPEUL þ 0:1826 ln Fert
ð0:8543Þ ð0:0223Þ ð0:0815Þ

þ 0:2177 ln Labor þ 0:4289 ln Rain þ 0:5627 ln Area

ð3Þ
ð0:0982Þ ð0:2122Þ ð0:1735Þ
All the estimated coefficients shown in Eqn. (3) are significant
at least at the 5% level except in the case of the variable
RPEUL, which is significant at the 10% level only. As our
main interest is to examine whether the third tier level of local
governance, which is the states-rural local bodies decentraliza-
tion, has contributed to rural development across states in In-
dia, we only explain the result of the coefficient of the variable
RPEUL. The above results indicate that the contribution of
the ratio of per capita expenditure of Panchayati Raj (rural)
institutions to per capita expenditure of urban local bodies is
dismal. In other words, the states-rural local bodies decentral-
ization has not been conducive to rural development in many
states in India. Nevertheless, the estimation confirms that
decentralization contributes positively to rural development
by facilitating a very small increase in rural development
across states. However, the magnitude of increase is not
impressive, given the emphasis for the third tier of decentral-
ization in the 10th and 11th Five Year Plan statements. The
implication is that decentralization needs to be more effective
in promoting and facilitating the activities of Panchayati Raj
Institutions and the impetus should come from the state gov-
ernments rather than forcing it from the central government.
5. CONCLUSIONS
The economic and Constitutional reforms of the 1990s
have given more space to the markets in the allocation of re-
sources as compared to the state relative to the pre-reform
days. What implication does this have to the spatial equity
in development? The state continues to be responsible for
the supply of public goods including basic human capital
and infrastructural development services across the country.
Analyses presented in this study suggest that the mechanisms
by which state governments provide for resources for such
services do not continuously lead to higher inter-state dispar-
ity. If this pattern is a result of equitable sharing of central
resources by the states or the effective functioning of decen-
tralization, this element of state behavior is important in
keeping the inter-state disparities from widening. Further,
the results also point out that the expenditures on basic ser-
vices such as health and education are pro-spatially equitable
than the economic services.
Overall, the results in this paper indicate that Government
of India within a federal framework has mechanisms that fos-
ter development equitably across its states, particularly
through health and education expenditures aimed at improv-
ing human capital development. In this context the impor-
tance of decentralization should be noted. However, the
slowly rising disparities in economic services across states,
which has implications for increasing disparity in some social
sectors such as infant mortality, warrant the attention of the
central and state governments. Further, as decentralization
could contribute more to rural development, the structural as-
pect of decentralization needs to be re-examined. Therefore,
drawing on Hayami (2001), it is conjectured that such tenden-
cies arise mainly due to lack of appropriate and efficient insti-
tutions at the state governments levels in India, which
indicates the need for further institutional reforms. For exam-
ple, many states have not taken effectively the recommenda-
tions of their state finance commissions, whose responsibility
 FISCAL DECENTRALIZATION AND DEVELOPMENT OUTCOMES IN INDIA: AN EXPLORATORY ANALYSIS
is to recommend the extent and type of fiscal decentralization
to the third tier level of decentralization, due to various factors
including lack of confidence in the data and analytical proce-
dures followed by SFC. This emphasizes the need for institu-
tional reforms at state level. Also, the MP Local Area
Development Funds and MLA Local Area Development
Funds allotted by the center and state governments, which
are often used to dilute the powers of rural local bodies such
NOTES
1. A comprehensive study that has analyzed the different responses of the
states to the reforms initiated by the Central Government of India is by
Howes, Lahiri, and Stern (2003).
2. However, the focus in this paper is only on rural areas.
3. There are studies in the literature examining the impact of fiscal
decentralization on economic growth. Martinez-Vazquez and Mcnab
(2003) have provided a comprehensive review on the relationship between
fiscal decentralization and economic growth. They argue that a direct
relationship between fiscal decentralization and economic growth remains
an open question. Nevertheless, Davoodi and Zou (1998) through cross
country analysis found a negative relationship between fiscal decentral-
ization and economic growth in developing countries.
4. Fiscal decentralization or autonomy of a state is given by the extent to
which it is able to finance its public expenditure from the revenue sources
assigned to it. Thus, one measure of fiscal decentralization will be the
share of a state’s own revenue in its expenditures. However, as
the statutory transfers are mandated in the Constitution and given on
the basis of recommendation by the Finance Commission, an independent
Constitutional body, the revenue from shared taxes and statutory grants
can be considered on par with that of the states’ own revenues. Therefore,
an alternative measure is given by the share of state revenue including
statutory transfers to denote fiscal decentralization.
5. The Planning Commission of India has named the states into “special
category states” and “general category states” for the purpose of
providing financial assistance from the centre. A special category state gets
preferential treatment in federal assistance and tax breaks because of
harsh terrain, backwardness and other social problems. Remaining states
are called “general category states.” These states are Andhra Pradesh,
Bihar, Gujarat, Haryana, Karnataka, Kerala, Madhya Pradesh, Maha-
rashtra, Orissa, Punjab, Rajasthan, Tamil Nadu, Uttar Pradesh, and West
Bengal (see Appendix I).
REFERENCES
Anilkumar, V. (2009). Federalism and decentralisation in India: Andhra
Pradesh and Tamil Nadu. Institute for Social and Economic Change,
Working Paper 208. Bangalore: ISEC.
Bardhan, P. (2002). Decentralization and governance in development.
Journal of Economic Perspectives, 16(4), 185–205.
Barro, R., & Sala-i-Martin, X. (1995). Economic growth. New York:
McGraw-Hill.
Brown, C. C., & Oates, W. E. (1987). Assistance to the poor in a federal
system. Journal of Public Economics, 32(3), 307–330.
Davoodi, H., & Zou, H. (1998). Fiscal decentralization and economic
growth: A cross-country study. Journal of Urban Economics, 43(2),
244–257.
Dreze, J., & Sen, A. K. (1996). India: Economic development and social
opportunity. Delhi: Oxford University Press.
Grover, V., & Arora, R. (1996). Encyclopedia of India and her states (Vol.
1). New Delhi: D&D Publishers.
1519
as Panchayats need to be restructured to make them comple-
mentary to resources given to local bodies. In this context, it
is worth noting the following statement that appears in the
Planning Commission’s 11th Five Year Plan: “Much higher
levels of human development can be achieved even with the gi-
ven structure of the economy, if only the delivery system is im-
proved” (Planning Commission, 2008, p. 2).
6. Jha (2007) has provided a concise discussion on the issues concerning
the principal constituents of the resources transfer formula between the
first and second layers of governments in developing countries. He has
argued that the relative weights on these constituents will often be
determined by country-specific circumstances.
7. In the Indian case, decentralization has come about more from the
center than the states. Many states did not find the need to decentralize
below their administrative level until the Constitution was amended in
1993, which also insisted the need for instituting State Finance Commis-
sion. Decentralization was initiated to provide easy access to public goods
locally. There were 247,033 rural bodies known as Panchayats and 3682
urban bodies in 2005. Nevertheless, Constitution grants strong powers to
the central government, including the control of the central executive over
state legislation, and the right to take over state administration in a state
of emergency (Rao, 2005).
8. Zhang and Zhou (2001) observed a statistically significant negative
effect of fiscal decentralization on Chinese provincial economic growth,
and a statistically significant positive effect of decentralization on state
growth in India.
9. These results are in contrast with those of some studies. For example,
West and Wong (1995) argued that increased fiscal decentralization led to
unequal provision of basic services such as education and health in China
in sub-provincial units in three different provinces.
10. No clear pattern can be observed in transfers at local level. There is
no doubt that the level of development of the states is a factor influencing
the success of decentralization in terms of raising revenues. The indication
is that India needs to improve its institutional framework to be conducive
to the success of decentralization at local level.
Faguet, J. P. (2004). Does decentralization increase government respon-
siveness to local needs? Evidence from Bolivia. Journal of Public
Economics, 88(3–4), 867–893.
Hayami, Y. (2001). Development economics. Oxford: Oxford University
Press.
Howes, S., Lahiri, A., & Stern, N. (Eds.) (2003). State level reforms in
India. New Delhi: MacMillan.
Jha, R. (2007). Fiscal policy in developing countries: A synoptic view.
ASARC Working Paper 2007/01. Australia South Asia Research
Centre, The Australian National University, Canberra.
Kalirajan, K., & Singh, K. (2010). Economic liberalization strategies and
poverty reduction across Indian states. Asian Pacific Economic
Literature, 24(1), 26–42.
Kapoor, S. (2009). Infant mortality in India: District-level variations and
correlations. Department of Economics, University of California,
Riverside (mimeo).
1520 WORLD DEVELOPMENT

Ladd, H., & Doolittle, F. C. (1982). Which level of government should
assist the poor?. National Tax Journal, 35(3), 323–336.
Martinez-Vazquez, J., & Mcnab, R. M. (2003). Fiscal decentralization and
economic growth. World Development, 31(9), 1597–1616.
Oates, W. E. (1972). Fiscal federalism. New York: Harcourt Brace
Jovanovich.
Ostrom, E., Schroeder, L., & Wynne, S. (1993). Institutional incentives and
sustainable development: Infrastructure policies in perspective. Oxford:
Westview Press.
Pauly, M. V. (1973). Income redistribution as a local public good. Journal
of Public Economics, 2(1), 35–58.
Planning Commission. (2006). Democratic decentralization and Panchayati
Raj Institutions. Eleventh Five Year Plan Rural Development Working
Groups. New Delhi: Planning Commission.
Planning Commission (2008). Eleventh five year plan 2007-2012, volume
1-inclusive growth. New Delhi: Oxford University Press.
Prudhomme, R. (1995). The dangers of decentralization. World Bank
Research Observer, 10(2), 201–220.
Rao, M. G. (2004). Transition to market and normative framework of
federalism in India. Paper presented at the conference on fiscal
federalism, University of Torino, August 27–28.
Rao, C. H. H. (2005). Essays on development regional disparities and centre-
state financial relations in India. New Delhi: Academic Foundation.
Rao, M.G., & Chelliah, R. J. (1995). A Survey of research in Fiscal
Federalism in India. Indian Council of Social Science research, New
Delhi.
Rao, M. G., & Das-Gupta, A. (1995). Intergovernmental transfers and
poverty alleviation. Government and Policy, 13(1), 1–23.
Rao, M. G., Sen, T. K., & Jena, P. R. (2008). Issues before the 13th
Finance Commission. Economic and Political Weekly, XLIII(36),
34–41.
Rao, M. G., Shand, R. T., & Kalirajan, K. P. (1998). State electricity
boards: A performance evaluation. The Indian Economic Journal,
46(2), 51–66.
Rao, M. G., & Singh, N. (2005). Political economy of federalism in India.
Delhi: Oxford University Press.
RBI (2004 and selected previous issues). State Government Finances.
Mumbai: Reserve Bank of India.
Rodden, J. (2002). The dilemma of fiscal federalism: Grants and fiscal
performance around the world. American Journal of Political Science,
46(3), 670–687.
West, L. A., & Wong, C. P. W. (1995). Fiscal decentralization and
growing regional disparities in rural China: Some evidence in the
provision of social services. Oxford Review of Economic Policy, 11(4),
70–84.
World Bank. (2000a). Overview of rural decentralization in India: Vol. I.
Unpublished Report. Washington, DC: The World Bank.
World Bank (2000b). World development report 1999–2000. In Decen-
tralization: Rethinking government. Washington, DC: The World Bank.
Zhang, T., & Zhou, H. (2001). The growth impact of intersectoral and
intergovernmental allocation of public expenditure: With applications
to China and India. China Economic Review, 12(1), 58–81.

APPENDIX I. SIZE AND INCOME OF INDIA’S STATES AND UNION TERRITORIES (2005–06)

Sl. No. State/UT Population million GSDP Per capita GSDP

Rs billion Billion US$ Rs US$
1 Andhra 80.4 2360 53.32 29,369 663
2 Arunachal Pradesh 1.2 29 0.66 25,086 567
3 Assam 28.5 575 13.00 20,186 456
4 Bihar 90.2 802 18.11 8891 201
5 Jharkhand 29.1 622 14.06 21,377 483
6 Goa 1.6 124 2.80 79,389 1793
7 Gujarat 54.6 2198 49.65 40,221 909
8 Haryana 23.1 1064 24.03 45,974 1038
9 Himachal 6.6 255 5.75 38,457 869
10 Jammu & Kashmir 10.9 265 5.99 24,397 551
11 Karnataka 56.0 1680 37.94 29,999 678
12 Kerala 33.4 1190 26.88 35,601 804
13 Madhya Pradesh 65.9 1163 26.28 17,649 399
14 Chhattisgarh 22.7 519 11.73 22,873 517
15 Maharashtra 104.2 4381 98.95 42,056 950
16 Manipur 2.5 57 1.29 22,684 512
17 Meghalaya 2.5 63 1.43 25,699 581
18 Mizoram 1.0 27 0.61 27,027 610
19 Nagaland 2.5 57 1.28 22,736 514
20 Orissa 38.8 785 17.74 20,251 457
21 Punjab 26.5 1097 24.79 41,420 936
22 Rajasthan 61.8 1242 28.06 20,095 454
23 Sikkim 0.6 18 0.41 31,186 704
24 Tamil Nadu 64.9 2235 50.49 34,424 778
25 Tripura 3.4 94 2.12 27,694 626
26 Uttar Pradesh 181.9 2798 63.19 15,382 347
27 Uttaranchal 9.2 262 5.91 28,572 645
28 West Bengal 84.8 2347 53.02 27,668 625
29 Pondicherry 2.6 68 1.4525,712 588
All India 1116.1 32,757 739.93 29,350 663
 FISCAL DECENTRALIZATION AND DEVELOPMENT OUTCOMES IN INDIA: AN EXPLORATORY ANALYSIS 1521

APPENDIX II. SUBJECTS THAT STATES CAN APPENDIX III. ASSIGNMENT OF FUNCTIONS
DEVOLVE TO PANCHAYATS AS PER THE 11TH FOR RURAL LOCAL BODIES THROUGH
SCHEDULE (ARTICLE 243-G) LEGISLATION AND ACTUAL ACTIVITIES
UNDERTAKEN, 2004–05

Agriculture, including agricultural extension. State Transfer of Subjects

Land improvement, implementation of land reforms, land subjects covered
consolidation, and soil conservation. through under activity
Minor irrigation, water management, and watershed legislation mapping
development.
Animal husbandry, dairy, and poultry. Andhra Pradesh 17 subjects 9 Subjects
Fisheries. Assam 29 subjects 29 subjects
Social forestry and farm forestry. Arunachal Pradesh 3 subjects
Minor forest produce. Bihar 25 subjects 25 subjects
Small scale industries, including food processing industries. Chhattisgarh 29 subjects 27 subjects,
Khadi, village, and cottage industries. except forests
Rural housing. and drinking
Drinking water. water supply
Fuel and fodder. Goa 6 subjects 18 subjects
Roads, culverts, bridges, ferries, waterways, and other Gujarat 15 subjects 14 subjects
means of communication. Haryana 29 subjects 10 subjects
Rural electrification, including distribution of electricity. Himachal Pradesh 26 subjects
Nonconventional energy sources. Karnataka 29 subjects 29 subjects
Poverty alleviation program. Kerala 29 subjects 29 subjects
Education, including primary and secondary schools. Madhya Pradesh 23 subjects 23 subjects
Technical training and vocational education. Maharashtra 18 subjects
Adult and nonformal education. Manipur 22 functions 22 subjects
Libraries. Orissa 25 subjects 7 Subjects
Cultural activities. Punjab 7 subjects
Markets and fairs. Rajasthan 29 subjects 12 subjects
Health and sanitation, including hospitals, primary health Sikkim 28 functions
centers, and dispensaries. Tamil Nadu 29 subjects 9 subjects
Family welfare. Tripura 29 subjects 21 subjects
Women and child development. Uttar Pradesh 12 subjects
Social welfare, including welfare of the handicapped, and Uttaranchal 14 subjects 9 subjects
mentally retarded. West Bengal 29 subjects 15 subjects
Welfare of the weaker sections, and in particular, of the Source: Planning Commission (2006, pp. 34–37).
scheduled castes and the scheduled tribes.
Public distribution system.
Maintenance of community assets.
Source: Grover and Arora (1996).