Do Not Open

Federalism and Economic Development in India:
An Assessment∗
Nirvikar Singh+
T.N. Srinivasan†
May 25, 2006
Abstract
This paper examines India’s federal system in the context of prospects for India’s future
economic growth and development. After a brief review of India’s recent policy reforms
and economic development outcomes, and of the country’s federal institutions, the
analysis focuses on the major issues with respect to India’s federal system in terms of
their developmental consequences. We examine, in turn, the impacts of tax assignments,
expenditure authority and the intergovernmental transfer system on the following aspects
of India’s economy and economic performance: the quality of governance and
government expenditure, the efficiency of the tax system, the fiscal health of different
tiers of government, the impacts on private sector economic performance and overall
growth, and the impacts on distributional equity by income class and region. In each case,
we discuss recent and possible policy reforms. We also provide a brief final discussion of
potential reforms of aspects of India’s federal institutions. Throughout the paper, we draw
on comparisons with China’s federal system where we think this can be instructive for
analyzing the Indian case.
∗
This paper has been prepared for the SCID conference on economic reform in Asia, to be held at Stanford
on June 1-3, 2006. We are grateful to SCID for financial support. We are grateful to, without implicating,
M. Govinda Rao, Donald Wittman, and particularly Jessica Wallack for comments on an earlier draft. All
views expressed here are ours alone, and not those of any organization with which we are affiliated.
+
Department of Economics and Santa Cruz Center for International Economics, University of California,
Santa Cruz
†
Department of Economics and Economic Growth Center, Yale University and Stanford Center for
International Development, Stanford University
PRELIMINARY DRAFT, FOR COMMENTS ONLY
1. Introduction
India’s recent growth story is now much analyzed, and quite well understood.
Despite some temporary controversy over the relative impacts of economic reforms in the
1980s and 1990s – hesitant and piecemeal in the first of those decades, deeper and more
systematic in the subsequent period – the new consensus is not very different from the
old, namely, that an overall shift in economic policy toward greater reliance on the
market for resource allocation, including greater openness to the global economy, has
been an important factor in increasing India’s average growth rate from its previous low
levels. This recognition of the role of market competition does not diminish the Indian
government’s past importance in building physical infrastructure and human capital, and
in providing stability and safety nets. Nevertheless, the reform of India’s governance is
one of two major strands of current policy debates, the other being areas where further
“liberalization” of the economy is needed (e.g., small scale industry reservations,
privatization, and matters pertaining to openness to foreign capital).
Debates about governance include old concerns about corruption, affirmative

action (e.g., the latest controversy over quotas in higher education) and social safety nets
(e.g., the new Employment Guarantee Scheme), as well as newer worries about growing
regional inequality. Managing the public finances appropriately has been an obvious part
of the reform story, since fiscal deficits have been a continuing problem for well over a
decade. Within the broader context of governance, issues of federalism and
decentralization have been addressed in a somewhat piecemeal fashion. Thus, the need
for fiscal consolidation has focused considerable attention on the states’ situations in this
regard, and the central government, central bank, and central Finance Commission have
all made efforts to ameliorate aspects of the states’ fiscal crisis. At the same time, the
decentralization to local governments, put in motion by the 73rd and 74th amendments to
the Constitution, has been proceeding unevenly, and with mixed success. States have
made various kinds of efforts to attract investment, done various deals with multilateral
agencies, and wrestled with potentially major tax reforms, all the while struggling with
fulfilling their constitutional responsibilities to constituencies such as the rural poor.
Underlying all the developments in economic policymaking, and concerns about

governance, therefore, is the working of India’s federal system. It is important to
understand what this system is, what it does, and how it has been changing in response to
the forces put in motion by India’s renewed struggle to fulfill its “tryst with destiny” by
substantially improving the well-being of all its citizens in a tangible manner. In Singh
and Srinivasan (2005a) we first tackled some of these issues, and will not repeat the
details of that analysis. Here we seek a broader view, relating our discussion of the
performance and impact of India’s federal system to general theoretical issues that have
surfaced in considering the economic performance of federations, as well as comparisons
to China, which has a de facto federal system, and which has in many ways become,
appropriately enough, India’s benchmark for economic performance. In this context, it is
important to realize that India’s federal structures are a creature of the political and
economic climate of the 1940s and 1950s, and this alone warrants a careful evaluation of
reform in these structures.
1
Traditional views of federalism, going back to the 18th century, often have a
normative perspective, but this is built on a foundation of positive analysis, which
predicts how different federal structures might work. A modern analytical view combines
issues of economies of scale and scope and internalization of externalities (all favoring
centralization) with asymmetries of information (favoring decentralization) and imperfect
incentive systems (which could work in either direction) to compare different
institutional arrangements in terms of their consequences for economic performance.
Incentive and information issues influence comparisons of kinds of decentralization, in
particular the use of political competition versus administrative hierarchies for achieving
decentralization. Equity as well as efficiency concerns can be incorporated in making
normative choices or recommendations among different arrangements.
Legal and political practitioners were originally influential in shaping federalist

systems, and it is only subsequently that scholars in political science and economics have
taken up the challenge of analyzing the workings of federations. Several conceptions of
the nature of federal systems have coexisted and evolved. The idea of cooperative
federalism, which was developed by Wheare (1953), emphasized the mutual gains from
different subnational jurisdictions as well as subnational and national governments
working in concert. Similarly, Riker (1964) conceived of federations as constitutional
bargains, designed to enhance security and stability. In these conceptions, the political
leadership is assumed to be acting approximately in the interest of constituents.
Cooperation is achieved either through explicit agreements, such as treaties and
constitutions, or through efficient noncooperative bargaining that results in cooperative
outcomes. While this literature was often developed in the context of Western
participatory democracies, and had little direct application to countries such as China, the
idea of cooperative federalism did get cited in early discussions of Indian federalism.
India, of course, has a stable and resilient, though not the most effective, democratic
system.
An alternative concept of federalism stresses the benefits of competition among

subnational units, and between national and subnational governments. This competition
enhances efficiency by improving the incentives of political leaders to act in the interest
of their constituents. There can be various ways in which competition plays out.
Constituents and their assets may be able to move between subnational jurisdictions in
ways that are not possible internationally. Constituents may also benchmark their
political leaders more effectively in a federation, and replace them more effectively than
they could in a centralized nation. This idea of “competitive federalism” has been
particularly developed and integrated by Breton (1995), though it consists of components
that can be traced to many different authors, including Tiebout (1956) and Brennan and
Buchanan (1980). The latter, in particular, see decentralization to subnational
governments as a means of checking the tendency of government to become a leviathan,
swallowing all around it. However, Breton also notes that competition among
governments may be destabilizing or lead to inequitable outcomes, and does not see it as
something that is always best left unrestrained.
2
The third main academic concept of federalism is due to Weingast (1993), who
coined the term “market-preserving federalism”, or, in brief, MPF. MPF is defined by
five conditions: (1) a hierarchy of governments with delineated authorities (the basis of
federalism); (2) primary authority over local economies for subnational governments; (3)
a common national market enforced by the national government; (4) hard subnational
government budget constraints; and (5) institutionalized allocation of political authority.
MPF encompasses key aspects of competitive federalism, but goes beyond it in several
ways, particularly in conditions (3) and (4). At the same time, except in the restrictions
embodied in (3), the view of MPF is more sanguine about competition than is Breton. It
emphasizes both the decentralization and the restraint of the regulatory power of
governments vis-à-vis the market.
The concept of MPF is particularly of interest because there has been a debate
about where India’s federal system fits in the Weingast scheme (Parikh and Weingast,
1997; Rodden and Rose-Ackerman, 1997), while China is seen as firmly within the MPF
locus. In placing China this way, Weingast distinguishes between de jure and de facto
federations, India falling into the former category, and China into the latter, though in
practice any system will work as a mix of legal framework and conventions. Weingast
sees China as satisfying the conditions of MPF in practice, with decentralized political
and economic decision-making, and only a common national market somewhat lacking
(Montinola, Qian and Weingast, 1995). It is also noteworthy that the de facto nature of
Chinese federalism has allowed for more institutional and policy flexibility than in the
Indian case, though this flexibility can cut both ways, with the Cultural Revolution being
an example of a negative extreme. One may find the concept of MPF hard to pin down in
some of its details, but the themes of subnational autonomy, a common national market,
and hard budget constraints provide a useful frame of reference for considering the
evolution of India’s federal system over the last decade or two.
One of Weingast’s motivations for the concept of MPF is as an explanatory factor

for the differential economic performance of developing nations with ostensibly federal
structures. By highlighting certain key features of federalism, he draws a causal link from
the institutional structures of MPF to economic growth. A larger question is what shapes
these institutional structures – what has made China, say, closer to MPF than India?
Rodden (2006) has characterized answering this question in terms of developing a theory
of endogenous federalism, in which the institutions of federalism are explained as
emerging from underlying preferences and other structural conditions.1 This perspective,
too, provides a useful lens with which to analyze the development of India’s federal
institutions. For example, one can view some of the greater de facto decentralization that
has emerged in India in the last decade as an endogenous response to changing external
conditions, including globalization and fiscal stresses. Furthermore, there have been
responses to these changes through modifications in the legal frameworks that govern
1
Rodden himself does not undertake this exercise. Wallack and Srinivasan (2005) do offer some insights
into how one might proceed in this direction. See also Weingast (2006), in which he develops the idea of
second generation fiscal federalism, which assumes that “public officials have goals induced by political
institutions that often systematically diverge from maximizing citizen welfare.”
3
areas such as taxes, expenditures and fiscal deficits (e.g., fiscal responsibility legislations
at the national and subnational levels).
In another example, local government strengthening was part of India’s policy

agenda since independence, but only made headway, through constitutional amendments,
in the early 1990s, a period of overall institutional change and policy reform. The states
quickly ratified the amendments and passed enabling legislation, but subsequently they
have typically dragged their feet in devolving real power to local governments. This
contrasts with China, where local governments were prime movers in implementing
economic reform, and did this in collaboration with provincial governments. It may be
that administrative centralization, in the form of a well-defined party hierarchy, was more
conducive to political and economic decentralization than India’s system of divided
authority without clear demarcations. One can also view this as China’s bureaucracy (its
party cadres) having a stake in wealth creation at all levels, national and subnational,
whereas India’s bureaucracy has only seen its power to profit from rent-seeking
diminished by economic liberalization. In any case, these are precisely the kinds of issues
that are raised in our paper, and at least partially analyzed. It is also worth remarking that
in the case of both India and China, the second tier of federalism, namely the states or
provinces, represents populations and areas comparable to European countries. Therefore,
India’s lack of significant decentralization to local governments is another aspect of its
relative centralization.
The rest of the paper is structured as follows. In section 2, we provide some

background on India’s economic performance and its federal system. Section 3 discusses
the quality of governance and government expenditure, while section 4 examines the
efficiency of the tax system. Section 5 considers the fiscal situation of the center and
states, focusing on recent attempts at coordinated fiscal consolidation. Section 6 discusses
growth and equity impacts of the federal system, with special reference to the complex
channels of intergovernmental transfers. Section 7 makes some suggestions for future
reform, while section 8 provides a summary conclusion.
2. Background: India’s Economic Performance and Federal System

India has been one of the fastest growing economies in the world since it began to
reform its economic policies toward greater openness and greater market orientation
Table 1 summarizes India’s overall growth performance since 1951.. One aspect of
change in India that has been relatively less studied is the parallel developments in
governance that have accompanied and interacted with economic policy reform. At the
same time, the nature of governance in India itself shapes the kinds of policy reforms that
are politically feasible, and the pace at which they occur. A key aspect of India’s
governance is its federal system, which is often crucial in determining how economic
reforms filter down to affect the daily lives of the population.
4
Table 1: Average Annual Growth Rates for India
GNP at NNP at NNP per capita at

constant prices constant prices constant prices
First Plan (1951-56) 3.7 3.6 1.8
Second Plan (1956-61) 4.2 4.1 2.0
Third Plan (1961-66) 2.8 2.5 0.2
Three Annual Plans (1966-69) 3.9 3.8 1.5
Fourth Plan (1969-74) 3.4 3.3 1.0
Fifth Plan (1974-79) 5.0 5.0 2.7
Annual Plan (1979-80) -5.0 -6.0 -8.3
Sixth Plan (1980-85) 5.5 5.4 3.2
Seventh Plan (1985-90) 5.8 5.8 3.6
Two Annual Plans (1990-92) 3.3 3.0 0.9
Eighth Plan (1992-97) 6.8 6.7 4.6
Ninth Plan (1997-2002) 5.5 5.5 3.5
2002-03 3.9 3.9 2.2
2003-04 (P) 8.6 9.0 7.2
2004-05 (Q) 7.6 7.8 6.1
2005-06 (A) 8.0 7.6 5.9
Notes: (P) – provisional estimates, (Q) – quick estimates, growth rates till 1999-2000 use 1993-94 base
year, thereafter, 1999-2000 is the base.
Source: Economic Survey of India, 2005-06, Central Statistical Office of India.
Panagariya (2005) provides a very careful appraisal of India’s growth experience

in the 1980s and 1990s. His main objective is to reassess a “revisionist” view,2 which
argues that economic policy reforms in the 1990s were not key to India’s growth
performance in that decade. These revisionists further argue that India’s growth surge is
properly understood as beginning in the 1980s, before the 1991 and subsequent economic
reforms. Panagariya offers three conclusions about India’s growth experience in the last
25 years:
1. Growth during the 1980s was inconsistent, with the last three years of that decade
contributing 7.6 percent annual growth, without which growth in the 1980s was
only marginally better than that of the previous three decades.3
2. The high growth in the last three years of the 1980s was preceded or accompanied
by significant economic reform, including trade and industrial policy
liberalization.
3. Growth in the 1980s was fueled by expansionary policies that entailed
accumulation of a large external debt and contributed to an economic crisis.
2
This view has been articulated most prominently by deLong (2003), Rodrik (2003), and Rodrik and
Subramanian (2004a, b), though others have also made related arguments. See also Srinivasan (2004),
which provides a conceptual and empirical critique of the revisionist view. Rodrik and Subramanian
(2004a) made a distinction between “pro-business” and “pro-market” reforms, arguing that the former only
favored incumbents and not competition per se, but the success of IT and ITES suggests that this distinction
is not really a valid description of the Indian experience.
3
Wallack (2003) provides the most careful econometric time series analysis to date, of the Indian growth
experience. She finds 1980 to be the most likely structural break year for GDP, but 1987 is the choice for
GNP. Hence, as she notes, such results are not as robust as one would like. For a further discussion of the
nature of India’s growth, with additional references, see Singh (2006b).
5
Panagariya’s own conclusion from his review of policy changes and growth performance
is that “the 1991 market reforms and subsequent liberalizing policy changes … helped
sustain growth.”
Rodrik and Subramanian (2004b) use a growth accounting methodology to

evaluate and project India’s growth performance which de-emphasizes (without
dismissing) economic liberalization policies, and focus on “meta-institutions” such as
democracy and the rule of law, as well as conventional economic inputs such as human
and physical capital, and productivity growth. In emphasizing these fundamentals,4 they
highlight the infrastructure and human capital built up under the pre-liberalization policy
regime. At another level, however, their assumptions about the impacts of policy are not
that different from those of Panagariya, since they state, “policy liberalisation will
progressively erode the licence-quota-permit raj as a source of corruption and patronage
that has had such a corrosive effect on public institutions.” In addition to this indirect
effect, they also attribute productivity growth directly to reforms that removed the
“shackles on the private sector.”
If one has at least a weak consensus on the role of market-oriented reforms in

supporting India’s good growth performance over the last 25 years, there is less
agreement concerning the impacts of its political structures. India is a constitutional
democracy, now comprised of 29 states (including the National Capital Territory (NCT)
of Delhi) and six “Union Territories” (UTs). The states and the UT of Pondicherry have
elected legislatures, with Chief Ministers in the executive role. The other UTs are
governed directly by appointees of the center. Each state also has a Governor, nominally
appointed by the President, but effectively an agent of the Prime Minister. There are
directly elected parliamentary-style governments at the national and state level, as well as
nascent directly elected government bodies at various local levels. These subnational
elected bodies with explicit constitutional authorities are the essential feature of de jure
federalism. Overlapping political authorities at the central and state levels have been dealt
with through intra-party bargaining in the initial post-independence years,5 and, more
recently, through explicit bargaining and discussion. The Inter-State Council (ISC) was
created in 1990, and has become a forum where some political and economic issues of
joint concern can be collectively discussed, and possibly resolved. The ISC includes the
Prime Minister, state Chief Ministers, and several central cabinet ministers as members.
While the ISC is merely advisory, it has formalized collective discussion and approval of
important matters impinging on India’s federal arrangements, including tax sharing and
inter-state water disputes. In other cases, committees composed of state finance ministers
have provided a means for reaching collective agreement by the states.
4
Based on this analysis, these authors suggest that sustained growth of 7% or more is feasible for India.
Mukherjee (2006), in a recent paper, suggests that 8% is feasible, but not 10%, though there is no exact
connection between his qualitative discussion and these quantitative projections. Kelkar (2004) and Shome
(2006) both provide arguments that 10% growth is feasible with the right policies in place. In April 2006,
the Prime Minister, Manmohan Singh, articulated this figure as a feasible target in a speech to an industry
group.
5
The erosion of intra-party democracy, and the rise of dynastic politics, especially in the Congress party,
but also in newer regional parties that are often governed by single individuals, have made intra-party
bargaining somewhat less salient.
6
India’s Constitution, though federal, has strong unitary features. These in large
part reflected the political consensus among members of the Constituent Assembly that,
given the trauma of partition, the problem of integrating former princely states with
diverse socio-economic and administrative features into the union, and the ever-present
threat of “fissiparous” tendencies in the body politic, a strong central government was
essential. The unitary features, which in fact date back to the Government of India Act of
1935 (from which the constitution borrowed liberally), were designed to create a strong
central government, with powers to dismiss a duly elected state government, if it deemed
necessary, because in its view the ‘constitutional machinery’ had broken down in that
state. The constitution also spelled out in some detail the assignment of taxation powers
and expenditure responsibilities among states, mandated the appointment of a Finance
Commission every five years, and described the duties of the Finance Commission, the
core of which relates to sharing of central taxes under article 270 and determination of
grants for the states as provided for under 275 (TFC, 2004, p9). On the advice of the
central government, the President appoints the Commission and specifies its terms of
reference. The twelfth such commission since the adoption of the Constitution in 1950
was appointed on 1st November, 2002, and submitted its report on 30th November 2004.
Centralization of India’s governmental structures was also reflected in

bureaucratic and judicial institutions. The Indian bureaucracy is provided constitutional
recognition, through the provisions of Part XIV of the Constitution. Since each political
layer of government requires its own administrative apparatus, any bureaucracy in a
federation will have a federal character: state governments must be able to appoint and
dismiss bureaucrats to implement state-level policies. This is true for India, with a central
bureaucracy as well as an independent bureaucracy in each state. However, the key
component of the bureaucracy is the Indian Administrative Service (IAS), whose
members are chosen by a centralized process and trained together, though then assigned
to particular states. At the national and subnational levels, the judiciary constitutes a
distinct branch of government, though the legislative/executive branch influences
appointments and IAS members are vested with some local judicial authority. The
Supreme Court, at the top of the judicial hierarchy, has powers that include broad original
and appellate jurisdiction and the right to rule on the constitutionality of laws passed by
Parliament. There has been conflict between the Supreme Court and the
legislature/executive over the scope of these powers, but in specific issues of center-state
relations concerning taxation and property rights, the basic centralizing bias of the
constitution tilted the Court’s interpretation towards the center. At the state level, below
the Supreme Court, the High Courts superintend the work of all courts within the state,
including district and other subordinate courts.
Besides the constitutionally mandated Finance Commission, an extra-

constitutional body, the Planning Commission, was established by a resolution of the
central cabinet in March 1950, within three months of the adoption of the constitution.
State governments appointed their own planning commissions or boards later on. With
the 73rd and 74th amendments of the constitution relating to panchayat boards and other
7
local governmental bodies, the state governments, have appointed their own Finance
Commissions to recommend financial devolution to local bodies.
Over the course of time, the center has acquired through various channels a large
say on how the financial resources in the economy are allocated among various levels of
government and the private sector. First, the Planning Commission began making grants
to states in support of their five-year plans (which the Commission formally approved).
Second, central ministries made their own grants in support of centrally sponsored
schemes to be implemented by the states. Presumably these were meant to subsidize
states for undertaking schemes that had positive spillovers to other states and to the
economy as a whole. The TFC (2004, p11) reports that transfers to states through all
channels increased from 31.4% of gross revenues receipt of the center during the First
Finance Commission to a high of 40.3% during the sixth commission, before they fell as
a proportion to 37.3% during the first two years of the eleventh.
Control of money and finance has also been an important centralizing feature of
the Indian system. First, money creation was the exclusive privilege of the central
government so that the revenues from seignorage accrued only to it. Second, with the
nationalization of insurance companies and commercial banks in 1969, the central
government acquired a large part of the investable resources of the financial sector for its
use, directly or indirectly through cash reserve ratios (CRRs) and statutory liquidity ratios
(SLRs). Third, through selective credit controls and requirements of lending to priority
sectors, little effective room was left for discretionary lending by financial intermediaries.
Fourth, interest rates were also controlled by the central government. Of course, reforms
since 1991 have brought about substantial changes: interest rates are no longer controlled
(although the mandated rate on small savings sets a floor), CRRs and SLRs are well
below their infamously high previous levels, and public equity in insurance companies
and commercial banks has been partly divested. Still, the public sector (mostly central
government) owns 75% of the assets of commercial banks, and priority sector lending
requirements have not disappeared, though they have been somewhat relaxed.
Concentration of powers in the hands of the central government did not create
serious conflicts in the early years of the functioning of the constitution since the same
political party, the Indian National Congress, ruled the center and states. Any potential
interstate or center-state conflict was resolved within the party. With the Congress Party
losing power in some of the states, the conflicts became open. For example, the
Communist Party led Kerala government was dismissed by Prime Minister Nehru’s
government in 1961. Periodic attempts at reexamining center-state relations (e.g. the
Sarkaria Commission in 1988) have not led to any fundamental changes in the
constitution. Of course, the ISC and also the National Development Council (for
discussion and approval of five year plans) have been constructive institutional additions.
With single party governments no longer the norm at the center, the rise of regional
parties in the states, and above all the changing political landscape (an increase in the
political muscle of erstwhile discriminated groups, particularly Dalits), one may need to
rethink in a fundamental way the unitary features of our constitution.
8
The recommendations of each Finance Commission apply to the five-year period

until the next commission is appointed. However, the sustainability of existing domestic
and external debt of the central government, and the debt of state governments, cannot be
judged without a perspective on revenues and expenditures and interest rates in the
indefinite future beyond the five-year horizon of each commission. More importantly,
even if the current commission does make its recommendations based on its perspective
about the future, there is no way it can commit to preventing future commissions from
having their own and possibly different perspective about the future, even if there are no
changes in the economy in five years to warrant it. Even more so, the governments in
power who are to act on the commission’s recommendations, which have implications for
future revenues and expenditures, cannot commit future governments, even if the party in
power wins the election. In sum, myopia and short-term considerations would tend to
bias the thinking of the commission and governments.
There is also a continued problem with coordinating transfers recommended by

the Finance Commission and the Planning Commission. In the past, an attempt was made
to coordinate the approaches of the Finance and Planning Commissions by having a
member of the Planning Commission serve as a member of the Finance Commission as
well, but this had limited impact. The Twelfth Finance Commission (TFC, 2004, ch.14)
has recommended the creation of a permanent secretariat for itself and also of a research
committee to undertake relevant studies for the commission. It has also recommended
that each state set up a high level monitoring committee to ensure proper utilization of the
commission’s grants. While these are unexceptionable, more could be done to ensure
some depth and continuity in the analytical approach across commissions as long as the
present constitutional arrangements remain.
The TFC, like its predecessors, and indeed conventional public finance
economists, viewed its mandate as “to recommend a scheme of transfers that could serve
the objectives both of equity and efficiency, and result in fiscal transfers that are
predictable and stable. These transfers, in the form of tax devolution and grants, are
meant to correct both the vertical (between center and states) and horizontal (among the
states) imbalances. Correcting vertical imbalance relates to transfers from the central
government to the state governments taken together, whereas the correction of horizontal
imbalances is concerned with the allocation of transfers among the state governments.
The vertical imbalance arises since resources have been assigned more to the central
government and states have been entrusted with the larger responsibilities. The
horizontal imbalance has its roots in the differential capacities and needs of the states as
also the differences in the costs of providing services” (TFC, 2004, p.9). Clearly, the
vertical imbalance reflects in large part the constitutional provisions relating to taxes and
expenditure responsibilities. Horizontal imbalances depend not only on differential
capacities, needs and costs, but also on the efficiency with which capacities are used to
deliver services at the least possible cost. Various commissions themselves have
recognized the “gap filling approach” of some past commissions seriously eroded the
incentives to improve efficiency. This is addressed in more detail in Section 6.
9
Also, the central and state governments have vastly expanded their role in the
economy, particularly by producing of goods and services for which more cost-effective
alternatives in the private sector have always existed or have come into existence since
1950. Successive Finance Commissions have studiously avoided getting into the question
of the rationale for the existence of many public enterprises, partly as a result of their
being given narrow terms of reference.6 Their existence and operation have significant
impacts on the revenues, expenditure and borrowing of governments at all levels. Indeed,
the use of such enterprises for borrowing under state guarantees has created contingent
liabilities for the states, besides being a non-transparent device outside of the formal
budget to raise resources. The point is that a narrow traditional approach of correcting
vertical (between center and states) and horizontal (among the states) imbalances, without
examining the broader question of the social rationale for the involvement of
governments at various levels in the economy, can no longer be justified. The usual
economics of fiscal federalism provides answers to the rationale issue through public
goods theory, applied to the case of different loci of benefits (Olson, 1986), but many
public enterprises in India are engaged in activities that would not even fall under the
category of quasi-public or merit goods.7 However, one area where the last two Finance
Commissions have played a greater role is in making recommendations with respect to
overall fiscal management: their broader terms of reference were to some extent a
response to a situation where the central government executive, in an era of coalitions,
may have lacked the direct power to rein in state fiscal deficits.
In its approach to horizontal imbalances, the TFC (2004, p.10) refers to the
concept of ‘equalization’ considered in many federal countries to be “a guiding principle
for fiscal transfers as it promotes equity as well as efficiency in resource use.
Equalization transfers aim at providing citizens of every state a comparable standard of
services, provided their revenue effort is also comparable.” It goes on to add that in
Australia, the equalization principle has been defined to say that “states should receive
funding … such that if each made the same effort to raise revenue from its own sources
and operated at the same level of efficiency, each would have the capacity to provide
services at the same standard…” and that in Canada, equalization payments are meant to
“ensure that provincial governments have sufficient revenues to provide reasonably
comparable level of services at reasonably comparable levels of taxation”.
The equalization concept seems eminently sensible if there is a social consensus

on what should be included in the set of services to be provided by the government and at
what level. The conventional argument for decentralizing their provision is that state
(and local) governments are likely to be more informed of and responsive to the
heterogeneous preferences of their residents, and that competition among states could
6
In contrast to the past, recent commissions, given broader terms of reference, have played a greater role in
articulating recommended policies for fiscal federal reform. These, recommendations, together with a
process of political bargaining then influence the legislative agenda, often toward significant modification.
Fiscal responsibility legislation is an example of this process.
7
The modern term is due to Musgrave (1959), and captures the idea that some goods may be rival goods,
but have positive externalities. While Musgrave introduced the term, it can be traced back to Adam Smith.
There have been some controversies over the precise meaning, but the externality perspective is
analytically the clearest, tying in with public goods, which can also be formulated in externality terms.
10
improve the quality and cost effectiveness of services provided. However such
competition, if it takes the form of subsidies or tax concessions to attract industrial
investment could turn into a ruinous “race-to-the-bottom”. The informational advantage
of state governments with respect to the preferences of their residents may no longer be
that important in the contemporary informationally-connected world. On the other hand,
incentives still may favor decentralization, since a national government will tend to
aggregate or balance across different constituencies. Hence a state government’s
incentives may be better aligned with the preferences of constituents, if they are
heterogeneous, even if there is no difference in the informational asymmetry between the
two levels. The caveat, here, of course, is that some instances of local heterogeneity (e.g.,
a preference for discrimination against certain groups) may not be legitimate. If such
preferences are not represented so strongly at the national level, then centralization will
mitigate the problem. This was precisely the view of B.R. Ambedkar, in favoring
centralization in India’s constitution (see footnote 10).
The Indian federal system, as surveyed above, is quite close in many respects to
federations such as Australia and Canada, which share British colonial roots to varying
degrees. The practice of the finance commissions (or equivalent bodies) in those
countries, however, seems to have evolved along lines more consistent with traditional
public finance theory, particularly with respect to horizontal equalization efforts. Perhaps
the greatest deviation in India has come about through the transfers governed by the
Planning Commission, including some ministry transfers for which the Planning
Commission is charged with coordination and disbursement.8 In maintaining a reasonable
degree of fiscal responsibility (at least until the mid-1980s at the center, and mid-1990s at
the state level), however, the Indian system has been closer to these developed countries,
and much less like developing country federations such as those of Argentina, Brazil and
Mexico. China, the star growth performer among developing countries since the 1980s,
and now India’s benchmark in many ways, stands out in having a political system very
different from any of these countries. In particular, strong local and provincial fiscal
autonomy can be seen as establishing a de facto federal structure, which is better at
stimulating economic growth, even in the absence of other well-defined political and
bureaucratic institutions. This is the position of Weingast and several of his co-authors,
and will be explored further in Section 6. Here we note that the proposition that
subnational fiscal autonomy on the revenue side overrides the impact of many micro-
level institutional features, such as we have discussed in the case of India (e.g., specific
constitutional provisions, bureaucratic rules, and checks and balances), is a simple but
powerful idea. Also note that for a positive outcome, this revenue side autonomy must
work in both directions, up and down, so that it is associated with hard budget
constraints.9
8
This does not apply to all ministry transfers, some of which are determined and implemented
independently of the Planning Commission. The practice of categorical subnational grants by central
departments or ministries is common in every federal system.
9
See, in particular, Montinola, Qian and Weingast, 1995, Cao, Qian and Weingast (1999) and Jin, Qian and
Weingast (2005).
11
3. Quality of Governance and Government Expenditure

One might take the view that, since China is able to grow more rapidly without
governance that is measurably better than India’s (according to various composite
measures compiled by researchers), the quality of governance may not be an important
factor in development. On the other hand, cross-country regressions often suggest
otherwise. Similarly, some cross-state regressions for India (Dollar et al, 2002) also
support the view that some aspects of government, such as the level of harassment, which
is typically connected to bribe seeking, directly matter for business productivity. Another
factor to consider in weighing empirical work is that all the quantitative measures of
governance are limited in their ability to capture the connection between governance and
economic performance. There are certainly numerous examples of countries where
specific dimensions of poor governance have undermined or prevented growth, and, in
the extreme, predatory or failed states fail economically as well.
The MPF view is that, given basic good governance, what matters especially is
restricting inefficient government interference in the market, and the right kind of federal
institutions can be important in achieving this. Digging somewhat deeper into the
underlying property rights protection logic for restraining government leads one to focus
on the nature of government control structures. Shleifer (1995) develops an argument
based on Grossman and Hart’s (1986), analysis of property rights as residual control
rights over assets, i.e., owners can do what they wish with their assets, as long as it is not
forbidden by law. Shleifer further distinguishes between physical and legal rights, with
only the latter being protected by the courts. If there is a divergence between the two
types of rights, so that bureaucrats or politicians have extensive physical control rights
through obstruction, threats or the like, final allocations after cooperative or
noncooperative bargaining to agreement, envisaged as efficient by Ronald Coase (1960),
are not enforceable by the courts, and therefore efficient bargains may not be achievable.
If the scope of bureaucratic control extends too heavily to physical property rights, rent
seeking, corruption, and inefficiency are likely results, as discussed by Shleifer and
others. However, where bureaucratic interventions are limited to providing public goods,
correcting externalities, or doing both, there is a greater likelihood of positive effects.
Aside from the larger issue of scope of bureaucratic interventions, the incentives
that flow from the details of organizational structure are very important. The Indian
bureaucracy, as conceived by the British (and to some extent, the Mughals before them)
was designed to provide law and order and efficient revenue collection. The system was
built around district level administrators with substantial powers and discretion.
Independence saw attempts to expand the duties of these administrators to encompass a
whole range of functions ostensibly tied to economic development. State level political
control became more important than in the past, as democratic institutions replaced
hierarchical administrative control. At the same time, state-level bureaucrats saw their
salaries reduced after independence, and their core (the IAS and IPS) remained tied to the
central bureaucracy. This continuation of the British “steel frame,” albeit in a modified
fashion, has preserved a considerable degree of administrative efficiency in a static sense,
but not been particularly conducive to economic development. At the same time, much of
the problem lies with the workings of Indian democracy itself.
12
The framers of India’s Constitution opted for a relatively centralized, ‘quasi-

federal’ system because of concerns about unity and stability, as well as inequality and
local elite capture of resources.10 Implicit in this choice was the assumption that the
central leadership) politicians and bureaucrats) would be more skilled and more honest
than state and local politicians. Initially, therefore, state governments functioned basically
as “corporate divisions” of the central ruling party, with local governments having little
or no role to play in political or economic decision-making. The size and cultural
homogeneity of India’s major states, combined with the constitutional decentralization of
key government expenditure responsibilities, created a tension that was finally resolved
only in the 1990s, with the emergence of explicit political coalitions at the center as the
norm of governance.
The political story, revolving around the organizational decay and reduced
political influence of the once-dominant Indian National Congress, coupled with the rise
of the BJP as a national “right-wing” party, and the emergence of regional and caste-
based parties, has been extensively analyzed.11 Chhibber (1995) explains the deepening
of ‘rent-seeking’—including the persistence of the laws that make it possible—in terms
of the intensifying needs of political competition. Essentially, powers of patronage for
electoral support became more important in the 1970s and 1980s, overwhelming any
concerns about the inefficiency of the system from the perspective of economic growth.
Chhibber provides empirical evidence that central loans, food assistance and subsidies to
the states were all linked to electoral considerations. Similarly, Rao and Singh (2005),
Kapur and Mehta (2002), and others have argued that large payments were directed by
the center in the late 1990s to the states (Andhra Pradesh and Punjab) from which
regional parties that were key coalition partners originated. In this case, the support
mechanism is direct, to build a majority coalition in parliament after elections, whereas in
Chhibber’s analysis it derives from the pre-election need to mobilize state-level political
resources for national elections.
Kapur and Mehta also highlight the role played by the organization of the Indian
parliament. They trace the decline of parliamentary functioning in ways that reduce
legislative oversight of the executive, increase spending, and make legislation more
difficult. Individual members of parliament act chiefly as recipients and distributors of
10
B.R. Ambedkar, in 1939, stated, “I confess I have a partiality for a unitary form of government. I think
India needs it.” Jawaharlal Nehru, initially less of a centralizer, changed his mind by 1946. The Union
Powers Committee of the Constituent Assembly, over which he presided, wrote, “Momentous changes
have since occurred. ....Now that partition is a settled fact we are unanimously of the view that it would be
injurious to the interests of the country to provide for a weak central authority which could be incapable of
ensuring peace, of co-ordinating vital matters of common concern and of speaking effectively for the whole
country in the international sphere.” Ambedkar made this statement about local government during the
Constituent Assembly’s drafting of the constitution: “What is a village but a sink of localism, a den of
ignorance, narrow mindedness and communalism...?” See Rao and Singh (2005) for more detail and
references.
11
See, for example, Rudolph and Rudolph, (1987); Brass, (1990); and Kohli (1990). Note that even the
parties that are currently classified as “national” according to the criteria of the Election Commission do
not have a legislative presence in many parts of the country.
13
patronage, either directly or as intermediaries, and not as legislators.12 Major economic

reform bills, introduced by reform-minded governments, have tended to get stuck in
parliamentary standing committees, where the organizational norm of unanimous
approval is almost impossible to achieve. At the state level, the picture of legislative
institutions is very similar, and the quality of legislation is even lower in many cases,
though sometimes related central legislation provides a guiding framework. This
weakness in legislative organization tends to shift bargaining among states and between
the center and the states to other arenas, such as the NDC and the ISC, in contrast to the
models suggested by US experience, which highlight legislative bargaining. While
reform of legislative institutions should be on the agenda, their current weakness also
suggests that explicitly recognizing and developing the role of these other federal bodies
is worth considering.13
This situation, where individual legislators are mainly recipients and distributors
of patronage, is to some extent a natural feature of democracy, and has to be managed
through the detailed organization of government. In addition to legislative institutions,
bureaucratic institutions and their relationship to political leadership also need attention.
In the current system, there are direct political pressures on bureaucrats that distort
supposedly impartial administrative decision-making, as well as distortionary incentive
mechanisms such as frequent transfers of bureaucrats. Even in the 1950s, transfers were
used to reward and punish bureaucrats.14 In some cases, transfers are a part of an
elaborate rent-seeking and rent-distribution mechanism, where administrators and
politicians may be equally complicit, and which leads to self-selection for the
bureaucracy that parallels what has occurred in politics – those who seek monetary rents
(rather than what economists have termed “ego rents”) are more likely to seek these
positions.
Wade (1989) and De Zwart (1994) are among those who have analyzed the
working of bureaucratic transfers in India. One theoretical justification for transfers15 is
that of reducing corruption by reducing opportunities for enduring corrupt relationships
to develop between bureaucrats and their clients, but in practice the frequency, variability
and arbitrariness of transfers is much greater than would be indicated by any such
justification. The outcome is that the bureaucracy’s role in carrying out administrative
policies that are derived from underlying legislative goals is severely hampered. Since, in
a democracy, the bureaucracy is properly subordinate to the elected representatives of the
people, external monitoring of improper political interference is required. The media and
judiciary can (and do) both play this role, the former perhaps more effectively. To the
extent that the judiciary is over-centralized and itself works with inefficient institutional
organization, its role is somewhat limited.
To the extent that the fundamental governance problem, as described above, is
one of accountability, one can argue (Rao and Singh, 2001) that India’s centralized
12
This problem may also be related to the perceived decline in the qualifications of the legislators
themselves.
13
See Inman and Rubinfeld (1997), Weingast and Marshall, (1988), Hall and Grofman, (1990).
14
See, for example, Sivaraman (1991).
15
An additional, more positive rationale comes from the benefits of varied experience that come with
rotation of assignments, especially for younger bureaucrats.
14
traditional accountability mechanisms, relying as they do on hierarchical political and

bureaucratic control and monitoring, have been ineffective. A more robust federal
structure, extending political accountability more effectively at the subnational level, is
important to consider as a way of increasing the efficiency of governance.16 At the same
time, the MPF perspective emphasizes the importance of having the right restrictions on
the sphere of action of subnational governments vis-à-vis the market.
A comparison with China is also illuminating here. In the Chinese case, local
government officials were given freedom by central and provincial governments to
implement policies that attracted and sustained many new enterprises. Essentially, the
boundary between government and market was drawn in a peculiar way from the
perspective of conventional economic theory. Governments became more or less directly
involved in commercial enterprises, with interjurisdictional and international competition
providing a disciplining device for efficiency. Employment creation and social safety net
provision were enhanced by this involvement, though at a cost in terms of some other
typical governance objectives: for example, environmental protection and social equality
have both suffered. In some ways, the Chinese model fits with Olson’s (199X) idea of
efficient ‘stationary bandits.’ It is an idiosyncratic illustration of the MPF concept, and its
details are clearly no guide for India.
Nevertheless, the central theme of subnational governments (with political and

bureaucratic roles being substantially intertwined in the Chinese case) seeking to enhance
constituent welfare through market-based economic activity is an important one to
consider in the context of India. India has a more difficult problem in one respect, since
its institutions and history are not suitable for a Chinese style direct approach. There are
proper constraints on direct involvement by bureaucrats and politicians in private
enterprise, though these can be and are circumvented through family, friends and agents.
However, there is clearly scope for decentralizing decisions that affect economic activity.
The centralized industrial licensing policy that was finally gutted in 1991 was a prime
example of needless controls on economic activity. There is evidence that needless or
poorly designed subnational controls still exist, and that some national level discretion
16
More effective judicial functioning, as part of a system of checks and balances, would also be helpful.
The over centralization, under funding, and inefficient procedures of India’s judiciary together work against
effective decentralization of other branches of governance. Furthermore, the political system often ends up
substituting for the failing judicial system. Those in political power influence the judicial system through
patronage appointments, and also take over its functions. Disputes are resolved by each side appealing to
different politicians or political factions. Resolution of disputes becomes a function of the relative political
influence of the disputants and the relative political strength of the politicians. Politicians then become
above the law, since they control its direct enforcement. Not only are they free to engage in illegal activities
without deterrent, but, worse, those who are already lawbreakers have a strong incentive to enter politics.
Furthermore, these effects can be self-reinforcing: politicians self-selected by a system that protects them
from punishment have an incentive to weaken the judicial system, and the pervasiveness of this norm may
affect the number of those who adhere to it. In this context, the ineffectiveness or absence of intra-party
control mechanisms exacerbates the problem. The total effect on the environment for investment and
economic growth can only be negative, though the macroeconomic impacts are hard to measure. For a
microeconomic analysis that identifies the impact of harassment of business on productivity, see Dollar,
Iarossi and Mengistae (2002), which highlights variations across ten states in specific aspects of corruption,
and the resulting impacts on investment climate.
15
could be devolved to the states.17 A key example of subnational controls is state laws that
restrict agricultural trade. In this case, severe distortions in pricing through various
subsidies compound the problem. Some of these problems (where they restrict the
common market aspect of MPF) can be dealt with by central government action. Others
require subnational action, which may come from pressures such as competitive
benchmarking. This subnational action may include changing policies, as well as
changing institutional setups. Examples include modifying tax assignments (institutions)
as well as rates (policies), and changing the incentive structures of the bureaucracy
(selection, training, evaluation and promotion) as well as of politicians (e.g., elections at
the local level, with local politicians being given real authority and resources to act).
Essentially, the current incentives for subnational governments in India to promote
market functioning are weak, and strengthening them can help. This can be done while
preserving the traditional boundary between government and market, which is
interestingly blurred in the Chinese case.
This discussion is also pertinent for the direct locus of government action, namely
in the sphere of public goods, social insurance and income distribution. There are well-
known and long-standing problems of inefficiency in government expenditures in India.
Examples of inefficiency include the functioning of core administrations, many plan and
ministry projects, and PSEs such as the SEBs. The evidence indicates that for many of
the states, subsidies and salaries are taking a larger and larger share of expenditure,
though the states’ performance in this respect is not uniform (e.g., Howes and Murgai,
2005).18 While expenditure reform will result in losers, since public sector employees
currently enjoy monetary rents or leisure that will be lost, at least some of the leisure in
inefficient organizations is involuntary, and results in frustration rather than any utility
gain. The World Bank (2003) is quite clear in its conclusions: “The burden of weak
administration falls particularly on the poor, who suffer from skewed government
spending, limited access to services, and employee indifference.”19 The areas for
improved administration include budgeting procedures, accounting and auditing methods,
personnel policies and tax collection, among others (Tenth Finance Commission, 2004;
World Bank, 2005). These basic improvements in government financial management and
functioning are less difficult to implement than the broader institutional changes
suggested above, and the slow pace of reform may be the result as much of neglect by
leaderships more concerned with macro issues than micro reforms as it is of resistance
from government employees.
The efficiency of delivery of health and education in rural areas can also be
improved substantially, either through restructuring government efforts, or bringing in
private participants such as nongovernmental organizations or community groups. There
is substantial evidence that institutional innovations can improve efficiency (e.g., Drèze
17
See Singh and Srinivasan (2005b), for example, for more discussion of this issue.
18
An in-depth analysis of the social rationale for subsidies, and their cost effectiveness in fulfilling that
rationale, is overdue. See Mundle and Rao (1991) and Rao and Mundle (1992) for a classic analysis.
19
In this context, it has been noted that a system of explicit user charges often allows for more efficient as
well as more equitable delivery of services (e.g., drinking water, health and education: see World Bank,
2003, Chapter 3, as well as World Bank, 2005). This would clearly be a necessary part of a program of
reducing inefficient and poorly targeted subsidies.
16
and Gazdar, 1996; PROBE, 1999; World Bank, 2003, Chapter 3; Howes and Murgai,
2005). In particular, the evidence is that decentralization of accountability systems can
improve incentives for teachers if done properly. Decentralization in this manner is not
exclusive of private or NGO participation. In either case, the gains come from improved
incentives and reduced transaction costs.20 Such decentralization to improve efficiency
also does not remove all higher-level government oversight. If certain individual rights
are a national level merit good, then the central government can still monitor their
subnational provision to ensure there is not a case for direct or indirect intervention.21
This is very different from primary control for expenditure on local public goods (which
may themselves be inputs into providing basic rights) resting with the center.22 Thus,
decentralization of some government powers need not lead to local elite capture and
exploitation, as was the fear after independence. In fact, one might characterize this
possibility of improvement in governance as Governance Enhancing Federalism (GEF).
An important aspect of governance that also has a federal dimension is regulation.

Economic reform has included a move toward modern, arm’s length regulation of
industries where the natural industry structure may not be competitive (e.g.,
telecommunications), or where there are systemic dangers that can be triggered by moral
hazard or other asymmetric information problems (financial markets in particular).
National level regulators such as TRAI and SEBI have developed regulatory approaches
that are more likely to emphasize establishing and enforcing fair “rules of the game,”
than trying to decide each play or each move. In the case of electric power, federal issues
with respect to regulation are more salient, and have made progress more difficult.
Electric power is a concurrent responsibility of the center and the states. Each state has
had a State Electricity Board (SEB) that is vertically integrated with respect to
generation, transmission and distribution, and is part of the state government. Various
political compulsions and inefficiencies have led to large losses by the SEBs. States were
slow to set up their own Electricity Regulatory Commissions (ERCs). These delays
meant that reform proceeded in a somewhat chaotic manner. The state ERCs were not
able to function quickly, because of inexperience and pre-emption by earlier ad hoc
decisions. The latest national legislation (the Electricity Act of 2003) may finally help in
this sector, where coordination of regulation across governments is clearly required,
though not all states have followed through with their own reforms. The unbundling of
transmission, generation and distribution at the level of the SEBs, which is taking place in
some states, will also help to wring out efficiencies through competition and consequent
20
See Rao and Singh (2005), Chapter 13 for further details. Of course there are many areas where the state
governments must continue to play a dominant role, and where more cannot be squeezed out of the existing
expenditures by improving incentives for those responsible for the service delivery. In such cases, shifts in
expenditure and/or new resources for increased expenditure are required, but the latter option should be a
last resort, given the states’ fiscal situation.
21
For example, there may be externalities associated with Tiebout-type migration that occurs in response to
a failure to provide such rights. Migration may also be a costly option for individuals, and infeasible as a
consequence. In the latter case, there may be externalities on the demand side – one person’s lack of rights
may affect another’s utility.
22
In the United States, for example, decentralization in the form of “states’ rights” became an excuse for
perpetuating discrimination. However, federal action (e.g., sending troops to Little Rock, and passing
national civil rights legislation) did not deprive states and localities of their decision-making power in many
other spheres.
17
restructuring, though in some initial cases, the privatization has been poorly designed and
implemented.
4. Efficiency of the Tax System

Some of the greatest challenges for tax reform in India have come at the
subnational level. The initial constitutional assignment of tax powers in India was based
on a principle of separation, with tax categories being exclusively assigned either to the
center or to the states. Most broad-based taxes were assigned to the center, including
taxes on income and wealth from non-agricultural sources, corporation tax, taxes on
production (excluding those on alcoholic liquors) and customs duty. These were often
taxes where the tax revenue potential was greater, as a result of relatively lower
collection costs. The center was also assigned all residual tax powers. Initially, the central
government followed principles that emphasized extreme progressivity and narrow
targeting,23 resulting in a very inefficient tax structure (with very restrictive import quotas
and prohibitively high tariffs being a prime example), and tax administration that was
highly susceptible to corruption. One of the triumphs of economic reform in India has
been a substantial rationalization of the central government tax structure, in terms of
lowering marginal rates, simplification of the rate structure, and some degree of base
broadening. This reform agenda was first laid out in the 1991 report of the tax reform
committee headed by Raja Chelliah, and has been further developed in subsequent
reports, notably including a pair of analyses about a decade later, under the direction of
Vijay Kelkar. There are still areas where the tax structure might be improved, and there is
not full agreement on how to proceed. In the realm of tax administration, despite detailed
analyses (e.g., Das-Gupta and Mookherjee, 1998), less progress has been made, reflecting
the general political economy constraints on improving the quality of governance by
reducing rent-seeking and corruption, as discussed in the previous section.
At the subnational level, a long list of taxes was assigned to the states, but only
the tax on the sale of goods has turned out to be significant for state revenues. This
narrow effective tax base is largely a result of political economy factors (e.g., rural
landed interests were initially quite powerful in government at the state level) that have
eroded or precluded the use of taxes on agricultural land or incomes (and also of user
charges for public irrigation and even electricity) by state governments. In addition, the
separation of income tax powers between the center and states based on source
(agriculture vs. non-agriculture) created avenues for evasion, since the states chose not to
tax agricultural income. The greatest inefficiencies arose in indirect taxes. Even though
in a legal sense taxes on production (central manufacturing excises) and sale (state sales
taxes) are separate, they tax the same base, causing overlapping and cascading, and
leaving the states less room to effectively choose indirect tax rates. Also, the states were
allowed to levy taxes on the sale and purchase of goods (entry 54 in the State list) but not
services. This provided avenues for tax evasion, and delayed the design and
23
The sharing formulas developed by the early Finance Commissions, which resulted in almost all central
income tax revenue being devolved to the states, also likely distorted the pattern of central taxation, until it
was replaced by general revenue sharing in 2000, after recommendations made by the Tenth Finance
Commission.
18
implementation of a comprehensive value added tax (VAT). These issues have been the
main subject of recent policy and institutional reform initiatives, and are discussed later
in this section.
One other aspect of the initial assignment of tax powers, and its subsequent
evolution, deserves detailed attention, because it directly involves one of the conditions
for MPF, namely, an internal common market. The framers of the constitution were
aware of the need for a common market, but also included a rather broad escape clause.
Article 301 of the Constitution states, “Subject to the other provisions of this part, trade,
commerce and intercourse throughout the territory of India shall be free”. However,
Article 302 empowers Parliament to impose restrictions on this freedom in the “public
interest” – a term that is both very broad and not clearly defined in this context. This
significant escape clause was, perhaps, in keeping with the post-war situation of general
scarcity, and the ideology of centralized planning, but fiscal impediments to internal trade
continued or worsened, even as the economic conditions made them less necessary for
economic security or stability.
The most significant fiscal impediment to free inter-state trade has been the
manner of levying inter-state sales taxes. In general, sales taxes have been levied by
exporting states on the inter-state sale of goods, making the tax origin-based. On the
other hand, the constitution’s framers intended that the sales tax system in India should
be destination based. While there is no clear theoretical argument for choosing one
taxation principle over the other,24 clarity and consistency are virtues, and these were lost
in the evolution of sales taxation in India, as we explain next. The problem of inter-state
impediments to trade was one aspect of broader muddles in the tax system.
According to Article 286 of the Constitution, “No law of a state shall impose, or
authorise the imposition of the tax on the sale or purchase of goods where such sale or
purchase takes place (a) outside the state, or (b) in the course of import of goods into, or
export of goods out of, the territory of India.” This principle was gutted very early on.
Based on the recommendations of the Taxation Enquiry Commission (India, 1953), the
Sixth Amendment to the Constitution added clauses (2) and (3) to enable the central
government to levy taxes on inter-state transaction. These clauses read:
24
Oliveira (2001) provides the following summary of the issues, “Given the existence of different tax
schedules, neither of the principles discussed here is clearly superior than the other one as a second-best
solution. While the origin principle in general brings consumption efficiency, the destination principle
brings production efficiency to the economy of the community. The effects of the two principles on the
public administration do not show a sure winner either. The origin principle does not call for border
controls and tend to make auditing and, to some extent, compliance easier. The destination principle avoids
fiscal wars, net overall revenue losses and does not cause states to wish for a clearing mechanism. Taking a
closer look, we think that the destination principle is superior. The relocation of producers that is likely to
happen under the origin principle does not have a significant parallel under the destination principle, for
individuals are much less likely to move to another country or state solely on a VAT rate basis. Border
controls can be replaced by close cooperation between tax authorities from different states.” See also Rao
and Rao (2006) for the practical and distributional considerations favoring a destination-based tax in the
current Indian context.
19
“(2) Parliament may, by law formulate principles for determining when a sale or purchase
of goods takes place in any of the ways mentioned in clause (1).
(3) Any law of a state shall, insofar as it imposes, or authorises the imposition of, -
(a) a tax on the sale or purchase of goods declared by Parliament by law to be of
special importance in inter-state trade or commerce; or
(b) a tax on the sale or purchase of goods, being a tax of the nature referred to in
sub-clause (b), sub-clause (c) or sub-clause (d) of clause (29-A) of Article
366; be subject to such restrictions and conditions in regard to the system of
levy, rates and other incidents of the tax as Parliament may by law specify”.
Under these changed provisions, the central government authorized the states to levy a
tax on inter-state sales, subject to a specified ceiling rate (4 percent). Besides impeding
the free movement of goods (through check-posts), this tax on the export of goods from
one state to another converted the sales tax into an origin-based tax. This tax also has
caused significant inter-state exportation of the tax burden from the richer producing
states to the residents of poorer consuming states (Rao and Singh, 2005).
A further problem arose through an inconsistency in constitutional provisions.

Although Article 286 does not allow restrictions on inter-state transactions, entry 52 in
the State list empowers the States to levy tax on the entry of goods into a local area for
consumption, use or sale. In many states, the tax has been assigned to urban local bodies
and the tax is variously called, “octroi”, or “entry tax”. In some states, the local entry
taxes were eventually replaced by state entry taxes, with similar effects. Thus, taxes are
levied not only on the exports from one state to another but also on all imports into local
areas, including imports from other states. Note that these entry taxes are destination-
based, and show that the problem of inter-state taxation has had another side as well as
those created by the amendment to Article 286. These entry taxes have complicated the
tax system, created distortions, caused impediments to inter-regional movement of goods,
increased compliance costs and have been a source of corruption.
Evolving a coordinated consumption-tax system remains a major challenge. Rao
(2000) provided detailed recommendations with respect to issues such as rates, interstate
sales taxes, and tax administration for a dual VAT coordinated between the Center and
the states, and noted the problem created by the failure of the Constitution to explicitly
include services within the scope of states’ sales tax authority.25 Rao had suggested
moving taxation of services from the Union list, where it implicitly lay (through the
Center’s residual powers over taxes not explicitly specified in the Constitution), to the
Concurrent list via a constitutional amendment. However, the central government chose
instead to explicitly add service taxes to the Union List, via the 88th amendment to the
Constitution, enacted in January 2004. According to the new institutional regime for
service taxes, they are to be shared with the states, in a manner yet to be determined, but
by Parliament, and therefore outside the “common pool” that is divided among the states
by the Finance Commission. Indeed, it is possible that the sharing of service taxes will
be completely outside the Commission’s scope in the future, representing a reversal of
25
Se also Rao and Rao (2006) for a more recent analysis of the conceptual and practical issues that have
arisen in Indian tax reform.
20
previous measures to simplify the tax-sharing system and make it more efficient. The
Twelfth Finance Commission has properly raised this issue as a concern. One can
conjecture that the move represents an attempt to establish more central control over the
tax base, in the face of broader sharing of centrally collected revenues. However, one
could argue, as we do below and in Section 6, that the more efficient route to follow
would be to strengthen subnational tax bases and tax collection, reducing the need for tax
sharing that is susceptible to distorting marginal fiscal incentives.
In contrast to the noncooperative approach to service taxation, the center has been
largely successful in persuading the states to replace taxation of interstate sales with a
destination-based VAT, and this is well on the way to implementation (as of March
2006). This effort has taken a very long time, reflecting the difficulties of coordination of
policy changes, as well as the practical difficulties of implementation, in terms of rates
and administrative mechanisms. The process began with the Tax Reform Committee
report of 1991, which was followed by the Report on Reform of Domestic Trade Taxes in
India (NIPFP, 1994). The latter report studied three different possible models for a
coordinated consumption tax system: (i) centralizing sales taxes and unifying them with
excise duties; (ii) giving the states the power to levy all domestic indirect taxes while
reducing tax devolution; and (iii) evolving an independent dual VAT at the central and
state levels, with no credit to be paid for the payment of central taxes by the states and
vice versa. The third approach was pursued as being most practical, and as a part of a
dual VAT design, a destination-based retail stage VAT will replace existing state level
sales taxes. To achieve agreement on this shift, the center appointed a State Finance
Ministers’ Committee, which recommended the switch to the VAT in stages. The
Committee was then transformed into an “Empowered Committee” of State Finance
Ministers, which recommended floor rates by the states and Union Territories to avoid
any “race to the bottom” in tax rates. By December 2005, 21 states had agreed to adopt
the new VAT, and began that process, which is ongoing.26 The key point we make here
is that a major reform was achieved essentially by bargaining and logrolling among the
executive branches at the national and subnational levels. The legislatures, as in other
cases, tend to provide ratification rather than initiation or legislative improvement.
A destination-based VAT will remove some of the internal barriers that have
plagued the development of a true national market within India, and could also reduce tax
exporting by the richer states.27 Given the political problems associated with interstate
income divergence that has emerged in the reform period, any policy reform that
ameliorates the causes of this divergence would be welcome. Even though the growth
impacts are hard to quantify, the magnitude of the implicit redistribution could have been
as high as Rs. 66 per capita in tax exportation for Maharashtra, a high income state, in
26
Four UTs had also adopted the VAT. Seven major states that had failed to adopt the VAT by the end of
2005 were Chattisgarh, Gujarat, Jharkhand, Madhya Pradesh, Pondicherry, Rajasthan, Tamil Nadu and
Uttar Pradesh. A detailed account and analysis of the features of the new system, and the process of
adoption, is given by Rao and Rao (2006). See also World Bank (2005, Chapter 3).
27
There is the possibility that removing taxes on inter-state sales could lead to evasion through false claims
of inter-state exportation: the evidence from the European Union suggests this as a possibility that would
need to be guarded against (World Bank, 2005, Chapter 3). However, it may be feasible to implement a
scheme for taxing all sales, and rebating tax paid on inter-state sales with filing of claims: this would be
analogous to a duty drawback scheme.
21
1999-00 – this figure being over 10% of the explicit transfers the state received in that
period (Rao and Singh, 2005, Chapter 9), though closer to 2% of state own revenues, and
less than 0.4% of SGDP. In fact, studies commissioned by the Twelfth Finance
Commission allay concerns over revenue losses, and project that a properly designed
state-level VAT would prove to be revenue augmenting over the medium to long term,
with any transitory losses possibly compensated for by the center. The latest budgetary
figures on the initial impacts of moving to VAT are consistent with this conclusion.
Taxation of services illustrates a broader issue addressed by the Eleventh Finance
Commission, which made a general recommendation to give the states more power to
tax, to reduce the vertical fiscal imbalance. This approach takes some pressure off the
fiscal transfer system, allowing states that can obtain internal political support to more
flexibly tax their own constituents for delivering benefits to them. Another possible
example of such a tax reassignment would be to allow states to piggyback on central
income taxes.28 With tax sharing no longer applied to specific tax “handles”, but to tax
revenues in total, this change would give states more flexibility at the margin, where they
properly should have it. While states are already assigned the right to tax agricultural
income, their use of this tax is minimal: the separation of agricultural income merely
promotes tax evasion. Piggybacking, combined with a removal of the distinction between
nonagricultural and agricultural income,29 would represent a change in tax assignments
that could increase efficiency as well as reduce the states’ fiscal problems.
While services taxation and VAT represent the two most important aspects of
subnational tax reform, the potential reform agenda is much deeper. The World Bank
study of state finances suggests attention to the professions tax, state excise duties, stamp
duties and transport taxes, as well as to state-level tax administration. Ultimately, the real
issue is how institutional reform can be achieved – on the other hand, the technical
aspects of policy and administrative reform are relatively better understood. Our
perspective is that the tax reform process at the subnational level has proceeded through a
combination of cooperative and competitive federalism. The central government has
played an agenda setting and coordination role in this process, and the states have
managed to reach some level of agreement on coordinating tax rates and policies through
bargaining by representatives of the executive branch. Strengthening and
institutionalizing this process of bargaining could lead to a smoother reform process. The
competitive aspects of federalism enter indirectly, through competitive benchmarking,
sometimes spurred by individual states that initially go it alone (as in the case of Haryana
with the VAT). The center can also play a role in brokering agreement by offering
incentives in the form of compensation for lost revenue from a tax reform. Although this
might create some short-term moral hazard, if compensation is capped, it will not lead to
long run distortions. One idea that has not been explicitly tried is that of Rao (2000), who
suggested that packages of tax reforms (e.g., VAT plus service taxes) be implemented, in
ways that would compensate a lost source of revenue for states with a new one. This idea
may still be useful in implementing changes to tax assignments that reduce vertical
imbalances, tax evasion, and distortionary taxes (e.g., consolidating the power to tax all
28
This change would, of course, require a constitutional amendment.
29
This suggestion does not preclude provisions such as tax smoothing for farm income to mitigate the
effects of greater risks associated with agriculture.
22
income with the center, but allowing states to piggyback on central income taxes.)
Finally, all of the issues that have been raised in considering center-state tax reform apply
to local governments. Their tax bases are inadequate, and property and land tax systems
need to be developed and implemented more effectively for decentralization of
expenditure authority to the local government level to make some headway. In doing this,
the political process that governs reform needs to be given attention, including
institutions that will allow local governments to share information, benchmark, and
coordinate where possible and desirable.
A constant feature of China, in contrast to India, has been a strong, centralized,

one-party system. This institutional arrangement has dominated politics of course, but
allowed considerable experimentation and variation in terms of economic authority at
different levels of government. Before the 1980s Chinese reforms, which began the
country’s sustained growth spurt, taxes and non-tax revenues (i.e., profits from state
enterprises) were collected by provincial governments acting as agents of the center.
Expenditure planning was a centralized top-down process, as was the determination of
revenue sharing. With reform, there was an attempt to create a fiscal contracting system
to govern explicit sharing of revenues by the provinces with the center. However, the
increased economic resources of several provinces gave them incentives to manipulate
the system to shelter revenues from sharing with the center, resulting in fiscal stress at the
center. In addition to the contracted transfers, there was also discretionary borrowing by
the center from provinces, and transfer payments by the center to other provinces. At this
stage, centralization of political power asserted itself, and in 1994, the institutional
arrangements finally moved toward a more traditional tax system, with a separate
national tax collection bureau, and taxes that are shared (including a VAT), as well as
others that are assigned to one level of government or another. “Extra-budgetary”
revenues, which include tax surcharges, user fees and some state enterprise profits, were
also institutionalized, and are quite significant. Given the somewhat opaque nature of
Chinese government budgeting and national income accounting, it is difficult to conduct
an analysis of the Chinese tax system that could parallel some of the observations that we
have made about India. In keeping with our discussion of the MPF view in Section 2, one
hypothesis could be that the allocative efficiency of the tax system in a standard public
economics sense is of second order importance relative to fiscal autonomy on the revenue
side (with expenditure authority being somewhat taken for granted). Certainly, China has
grown rapidly without using any optimal tax or other economic models to guide their
design of tax policy. Nevertheless, it would be irresponsible to suggest that
microeconomic efficiency of tax rates can be neglected, and India’s moves toward reform
can be lauded, even without any useful comparison or guidance from the Chinese
experience.
.
5. Fiscal Situation of National and Sub-national Governments

Table 2 summarizes trends in central and state fiscal deficits since 1990. Fiscal
deficits began to rise in 1997-98 at both levels of government, though the rise was much
greater at the state level. Fiscal balances at both levels were severely affected by the large
pay increases granted to central government employees in 1997-98, followed by similar
23
increases at the state level the following year. The center’s balance continued to
deteriorate slowly till 2001-02, when the trend was reversed. The states’ aggregate
position stabilized after the one-time shock, and improved after 2002-03. Two other fiscal
indicators also deteriorated after 1997-98. First, the revenue deficit (i.e., balance between
current receipts and expenditures) grew as a percentage of GDP, coming down to 1997-
98 levels only in 2004-05.30 Second, the primary deficit (after taking out net interest
payments from expenditures) has grown, after the initial reduction in the early 1990s,
indicating that the problem is not simply growing interest payments, though these have
also gone up as a percentage of GDP. Fiscal deficits financed by borrowing add to the
government debt. Table 3 summarizes recent trends in the general government debt. After
some decline in the early 1990s, the stock of government debt rose steadily after 1997-98,
as a percentage of GDP, before stabilizing from 2002-03. A significant portion of this
increase was at the state level. For example, the debt-GDP ratio of the states increased
from 21 % in 1996-97 to 31% in 2002-03.
Table 2: Central and State Fiscal Deficits (Percent of GDP)
Year Center States Consolidated

Total Revenue Primary
1990-91 6.6 3.3 9.4 4.2 5.0
1991-92 4.7 2.9 7.0 3.4 2.3
1992-93 4.8 2.8 7.0 3.2 2.1
1993-94 6.4 2.4 8.3 4.3 3.3
1994-95 4.7 2.7 7.1 3.7 1.9
1995-96 4.2 2.6 6.5 3.2 1.6
1996-97 4.1 2.7 6.4 3.6 1.3
1997-98 4.8 2.9 7.3 4.1 2.1
1998-99 5.1 4.2 9.0 6.4 3.7
1999-00 5.4 4.6 9.6 6.3 3.9
2000-01 5.7 4.3 9.8 6.6 4.0
2001-02 6.1 4.2 9.9 6.9 3.7
2002-03 5.9 4.7 9.6 6.7 3.1
2003-04 4.5 4.4 8.5 5.8 2.1
2004-05 (RE)* 4.0 4.0 8.4 4.1 2.3
2005-06 (BE)** 4.1 3.1 7.7 3.4 1.8
2006-07 (BE) 3.8 n.a. n.a. n.a. n.a.
Sources: RBI Annual Reports (RBI, 2001, 2002, 2003a, 2005), RBI Bulletin (2006).
Notes: RE: revenue estimate; BE: budget estimate; * Center’s figure is actual; ** Center’s figure is RE.
The consolidated deficit indicators net out the inter-governmental transactions between the Center and
States, and do not equal to the sum of the deficits of the Center and the States. 1990s figures for the Center
exclude small savings allocated to the States, to give consistency across the accounting change related to
30
Excessive emphasis should not be placed on the revenue deficit: current expenditures include spending
on health and education, which, if effective, is investment in human capital, with significant social returns.
Analogously, some expenditures accounted as capital spending include some items that are really current –
essentially maintenance expenditures – and others that have negligible social returns. See Singh and
Srinivasan (2005b) for further details of a broad range of issues related to India’s fiscal policies.
24
the treatment of national small savings. The 2005-06 figures are given as reported, but appear to have an
inconsistency, since the consolidated figure exceeds the center and state sum. n.a. – not available
In addition to the stabilization or slight improvement in the center ands states’

main fiscal indicators, government guarantees have also been controlled, falling from
12.2 percent of GDP in 2001 to 10.6 percent in 2004, though the latter figure is
provisional (RBI, 2005). The external debt is also under control (22% of GNP in
September 2005 – classified as low by international standards by the World Bank), and,
as is well known, foreign reserves are at very comfortable levels ($139.2 billion on
January 27, 2006, according to the Economic Survey of India). On the other hand, the
future cost of the pension system remains a serious issue for the medium and long run.
While one demographic trend will help, by increasing the proportion of the population
that is of working age, another, longer life expectancy, will increase the number of years
for which pensions are paid, relative to the number of working years. The World Bank
estimates that the cash-flow deficit of the Employees’ Pension Scheme (EPS), which is a
defined benefit scheme, will grow to almost 1% of GDP over the next few decades, even
without increases in coverage. If more employees are covered by the EPS as growth
increases the relative size of the formal sector, then the potential problem will grow
accordingly.
Table 3: General Government Debt (percent of GDP)
Year Debt
1990-91 64.4
1995-96 61.2
2000-01 70.4
2001-02 75.8
2002-03 80.0
2003-04 81.1
2004-05 (RE) 82.0
2005-06 (BE) 81.3
Source: RBI (2005)
India’s states had nonexistent or negligible revenue deficits before the late 1980s.
From 1987-88 onwards, the states in aggregate have always been in revenue deficit, and
that deficit level increased from an average of 0.62% of GDP across 1993-96 to 2.53% in
2000-03.31 This deterioration was greater than the worsening in overall fiscal deficits for
the same period (2.55% to 4.07%), reflecting the crowding out of capital expenditures by
current expenditures such as subsides and salary payments. We have noted the impact of
the Fifth Pay Commission’s award, which, spilling over to the states, led to a very large
jump in the states’ wage bills. It also led to liquidity problems for the state governments,
which even had difficulty in paying their salary bills (World Bank, 2005, Box 1.1). The
states’ total primary deficit also worsened significantly, from an average of 0.69% of
31
These and other figures in this section are taken from the report of the Twelfth Finance Commission
(Twelfth Finance Commission, 2004). The analysis here draws on Singh (2006).
25
GDP over 1993-96 to 1.41% over 2000-03. The latest estimates for the revenue deficit
(1.4% in 2004-05 and budgeted at 0.7% in 2005-06) reflect an improvement, but it is too
early to confidently identify a trend.32 Disaggregating the states’ deficits reveals (Table 4)
that the source of deterioration has been increases in expenditures such as interest
payments, rather than declines in own revenues or transfers from the central government
(particularly tax sharing and grants determined by the Finance Commission). This
conclusion assumes that the “natural” income elasticity (“buoyancy”) of tax revenues is
equal to one. One could argue that the buoyancy of states’ tax revenues ought to be
greater than one, which would imply that tax revenues have failed to grow at a pace
consistent with that norm.33 Further aspects of changes in expenditure (not shown in
Table 4) have been increases in subsidies – with the power sector a major culprit – and a
squeeze on Plan expenditure, which ought to be earmarked for capital projects. Some of
the negative impact of the power sector’s problems also shows up in the decline in net
non-tax revenues (Table 4 and Rao, 2003).
Table 4: Trends in Revenue and Expenditure Components, All States (% of GDP)
Period Own Tax Own Finance Non-Finance Revenue Interest Pensions

Revenues Non-Tax Commission Commission Expenditures Payments
Revenues Transfers Transfers
1993-96 5.27 1.55 2.94 1.62 12.00 1.86 0.63
2000-03 5.44 1.26 2.88 1.23 13.34 2.65 1.25
Change 0.17 -0.29 -0.05 -0.39 1.34 0.79 0.62
Source: Twelfth Finance Commission Report (Finance Commission, 2004)
Further understanding of the states’ fiscal situation comes from examining the
performance of individual states. The fiscal deterioration for the special category states
was generally worse than that of the major states, but we focus on the latter here, as they
contain the bulk of India’s population. Data for these 15 states is shown in Table 5.34
Ranks are shown in parentheses, with a higher rank indicating a ‘worse’ number in terms
of deficit, change in deficit, or debt stock. While there is considerable variation across the
states, in terms of their fiscal positions and the level of deterioration, there is no clear
pattern. High and low income states, reforming states as well as those that have moved
slowly on reform, larger and smaller states, all have shown significant fiscal
32
These figures are from the Economic Survey of India, 2005-06, available at
http://indiabudget.nic.in/es2005-06/chapt2006/chap29.pdf. The improvement in the fiscal deficit, 4.0% in
2004-05 and 3.4% in 2005-06 has been smaller, and the debt-GDP ratio has reached about 33%, though it
may stabilize.
33
See also Rao (2003) for a detailed discussion of tax revenues of the states.
34
Following the analysis in the Twelfth Finance Commission Report, the new states of Chhatisgarh,
Jharkand and Uttaranchal are combined with their respective ‘parents’ for the purposes of the comparison
across the years.
26
deterioration. The magnitudes of the changes were not obviously affected by initial
positions. The correlation between revenue deficits in the earlier and later periods was
0.35, and the correlation between the revenue deficits in the earlier period and their
changes was in fact negative (-0.37). The corresponding correlations for fiscal deficits
were 0.67 and –0.11. The 2004-05 figures, in the last column of the Table, indicate that
this lack of any clear pattern of fiscal performance continued.35
Table 5: Comparative Fiscal Performance, Major States (% of GSDP)
Revenue Change in Fiscal Change in Debt, Fiscal

Deficit, Revenue Deficit, Fiscal 2000-03 Deficit,
2000-03 Deficit, 1993- 2000-03 Deficit, 1993- (Rank) 2004-05
(Rank) 96 to 2000-03 (Rank) 96 to 2000-03 (Rank)
(Rank) (Rank)
Andhra Pradesh -2.03 (13) -1.51 (11) -4.57 (9) -1.41 (13) 29.93 (11) -3.4 (12)
Bihar -1.87 (14) -0.04 (15) -4.52 (10) -1.67 (10) 44.35 (5) -5.6* (4)
Goa -2.44 (10) -3.89 (3) -4.68 (6) -2.38 (4) 33.54 (9) -3.9 (11)
Gujarat -4.66 (3) -4.75 (1) -5.74 (5) -3.93 (2) 37.92 (7) -6.3 (2)
Haryana -1.32 (15) -0.56 (14) -3.69 (15) -1.19 (14) 28.02 (12) -1.3 (15)
Karnataka -2.21 (11) -2.15 (9) -4.37 (11) -1.65 (11) 27.27 (13) -2.7 (14)
Kerala -4.17 (5) -2.99 (5) -5.13 (6) -1.81 (6) 37.58 (8) -5.3 (8)
Madhya Pradesh -2.05 (12) -1.44 (12) -3.94 (13) -1.78 (7) 30.42 (10) -6.3* (2)
Maharashtra -3.09 (7) -3.00 (4) -4.12 (12) -1.96 (5) 27.11 (14) -5.2 (9)
Orissa -4.91 (2) -2.91 (6) -7.84 (1) -3.21 (3) 63.68 (1) -5.6 (4)
Punjab -4.53 (4) -2.66 (8) -6.14 (3) -1.77 (8) 46.66 (3) -5.2 (9)
Rajasthan -3.87 (6) -2.78 (7) -6.05 (4) -1.54 (12) 44.88 (4) -6.5 (1)
Tamil Nadu -2.50 (9) -1.78 (10) -3.75 (14) -1.77 (8) 26.16 (15) -2.9 (13)
Uttar Pradesh -2.98 (8) -1.21 (13) -5.07 (7) -1.03 (15) 46.94 (2) -5.6* (4)
West Bengal -5.47 (1) -3.95 (2) -7.31 (2) -4.13 (1) 42.73 (6) -5.4 (7)
Source: Twelfth Finance Commission Report (Finance Commission, 2004) and RBI (2006)
* These states’ figures exclude those for the split-off states of Jharkhand (-8.1%), Chhattisgarh (-5.6%) and
Uttaranchal (-13.8%)
Some studies (e.g., Khemani, 2002; Purfield, 2003) have attempted to provide
causal explanations of state deficits through cross-section or panel (pooled cross-section
and time-series) regressions for the states. Explanatory variables include structural
variables such as the share of agriculture in Gross State Domestic Product (GSDP),
behavioral variables such as expenditure levels, and political variables such as affiliation
between the ruling parties at the state and central levels. The results are suggestive
(particularly with respect to the impact of political affiliation between the center and a
state on that state’s fiscal deficit) but not conclusive, with one unexplained issue being
the variation in states’ fiscal performance from year to year, which we have already noted
in Table 5. Hence, these regressions may not capture the essential mechanisms of state
fiscal policy making, nor uncover the underlying structural explanation of fiscal
performance. However, Table 6 does indicate some of the underlying sources of states’
differing performance. Table 6 again follows the convention of ranking from ‘worst’ to
‘best’, with ‘worst’ being low tax revenue or revenue increases, but high expenditure or
expenditure increases. This characterization neglects the potential benefits of government
35
Budget estimates for 2005-06 indicate significant fiscal improvement for Maharashtra and Gujarat, but
not for other states.
27
expenditure,36 focusing only on the narrow fiscal consequences. Bearing out the earlier
aggregate figures, we see from Table 6 that, while a couple of states have allowed own-
tax revenues to slip substantially, the major source of fiscal deterioration has been
increases in expenditures running well beyond tax revenues. The correlation between the
own-tax and expenditure ratios fell from -0.13 in the earlier period to -0.41 in the later
period, with the negative coefficients indicating, perhaps surprisingly, that higher
spending states tended to do worse in own revenue-raising. Again, there is no obvious or
simple link between the economic characteristics of the states and their relative revenue
and expenditure performance. However, we can consider various institutional
contributors to the states’ current situation.
Table 6: Comparative Revenues and Expenditures, Major States (% of GSDP)
Own Tax Change in Own Revenue Change in

Revenue, Tax Revenue, Expenditure, Revenue
2000-03 (Rank) 1993-96 to 2000- 2000-03 (Rank) Expenditure,
03 (Rank) 1993-96 to
2000-03 (Rank)
Andhra Pradesh 7.30 (9) 1.40 (12) 15.56 (10) 2.08 (9)
Bihar 4.46 (2) 0.75 (7) 18.11 (3) 1.60 (11)
Goa 6.46 (6) -1.45 (1) 17.25 (5) 0.13 (15)
Gujarat 7.71 (10) 0.20 (5) 18.37 (2) 5.85 (1)
Haryana 8.30 (13) 1.09 (9) 13.45 (15) 0.39 (14)
Karnataka 8.33 (14) -0.19 (4) 15.33 (11) 1.36 (12)
Kerala 8.11 (12) -0.34 (3) 16.11 (8) 1.18 (13)
Madhya Pradesh 6.45 (5) 1.53 (13) 16.74 (7) 3.45 (3)
Maharashtra 7.76 (11) 1.12 (10) 14.10 (14) 3.42 (4)
Orissa 5.81 (3) 1.87 (14) 22.22 (1) 5.74 (2)
Punjab 7.13 (8) 1.87 (14) 15.33 (11) 2.59 (7)
Rajasthan 6.48 (7) 0.25 (6) 18.06 (4) 2.63 (6)
Tamil Nadu 9.00 (15) 0.98 (8) 15.60 (9) 1.66 (10)
Uttar Pradesh 5.88 (4) 1.12 (10) 16.78 (6) 2.50 (8)
West Bengal 4.26 (1) -1.20 (2) 15.02 (13) 3.23 (5)
Source: Twelfth Finance Commission Report (Finance Commission, 2004)
For example, a large contributor to increases in current expenditure was the Fifth
Pay Commission award, and its political economy, can plausibly be described as follows.
Economic liberalization allowed private sector salaries to rise substantially, creating an
envy effect for central government bureaucrats. Their large pay increases had a similar
effect on state governments. While these phenomena have more to do with motivations of
status and envy, economic liberalization was a factor. It removed some elements of an
implicit social contract, in which large monetary rewards were discouraged (through
taxes as well as relatively flat pay structures), without changing other elements of the
36
In fact, as pointed out in the Twelfth Finance Commission Report, revenue expenditure has tended to
crowd out capital expenditure.
28
system.37 Thus, institutional structures that were somewhat adequate in the past are no
longer functional.
The increase in states’ fiscal deficits and debt has represented a major change in
the overall fiscal management of government, and presents a challenge to economic
reform. The challenge includes the direct impact on governance because of the
deteriorating quality of expenditures, as well as indirect impacts through a discourse that
blames reform for the states’ fiscal difficulties. We have suggested that the latter
perspective has some merit, though the proper implication, in our view, is that the
problem is one of incomplete rather than pernicious reforms. The Indian federal system is
in the middle of developing a new institutional framework for managing subnational
deficits and debt. To evaluate different approaches that have been proposed and
attempted, some brief review of concepts is helpful.
Obviously, borrowing, in addition to its normal role of funding capital

expenditures, is an expedient method of financing the excess of current expenditures over
current revenues. At the same time, borrowing shifts the burden of paying for current
expenditures from the present to future generations, who will be servicing the debt, as a
result also redistributing future current income from taxpayers to debt holders. This
impact of borrowing is quite different from borrowing to fund capital expenditure that
increases growth rates and future incomes. In the pre-reform period, access to the market
for borrowing by subnational governments was severely restricted. The RBI, as central
bank, managed the debt of all levels of government, and, in particular, it did not allow
market borrowing by state governments that were already indebted. In practice, all the
states were in debt to the central government, which made discretionary loans to the
states for capital and current expenditures. In this regime, debt of different state
governments that was incurred at the same time carried the same interest rate, regardless
of the borrowers’ fiscal situation. Moreover, the central government acted as a financial
intermediary, certainly with respect to external borrowing, but also for domestic
borrowing: it borrowed and then on-lent to the states, with or without an additional
charge. Recent borrowing by states, however, has been largely to fund revenue deficits.
Just as the pay and incentive system for government employees has come under
strain from the opening up of India’s market economy to globalization, the system of
hierarchical, discretionary control of subnational borrowing, which worked sufficiently
well under the old license-permit regime, came under strain in the late 1990s, with states
borrowing to fund revenue deficits partly caused by the large pay hikes (essentially a
form of subsidy for unproductive government employees), as well as by increasing
explicit subsidies to interest groups such as farmers, and implicit subsidies to PSE
employees (especially in the power sector) in loss-making enterprises. States also were
given more freedom to negotiate with multilateral agencies for loans for capital projects,
with the center traditionally serving as a guarantor. Central political control of states’
deficits became weaker in a situation where more regional parties were pivotal players in
37
Howes and Murgai (2005) analyze aspects of the pay and pensions of government employees. They
suggest that overall, public sector wages are too high, but this does not seem to be true at the most senior
levels – essentially, the private sector rewards performance and responsibility more closely.
29
broad central government coalitions. These regional parties were no longer subject to
incentives (such as influence in the party hierarchy) that might have operated through
bargaining among regional factions within a single political party. The last few years
have therefore seen attempts to create a new set of political institutions to once again
harden the states’ budget constraints, after the “softness” of the 1990s.
From the center’s perspective, there was some advantage to restricting the states’
borrowing and deficits. The center itself was under fiscal pressure, and there was a subset
of central government decision makers who saw policies that encouraged fiscal discipline
and long-term growth as an attractive political strategy. Initial attempts to control
subnational deficits by restricting ways and means advances from the RBI were limited
by their temporary and limited nature – essentially, they were not credible. One can also
argue that the political power of the center vis-à-vis the recalcitrant states was limited.
The Eleventh Finance Commission tried to build incentives for fiscal reform into the
transfer system, but the manner in which these incentives were structured left them too
weak to make a difference to state expenditure and borrowing decisions.
The approach that seems to have worked partially is that of commitment to

explicit targets through fiscal responsibility legislation. The central government passed a
Fiscal Responsibility and Budget Management (FRBM) act in 2003, laying down specific
targets for deficit reduction. Many state governments have followed the center’s lead
since then. The RBI provided model legislation, and the Twelfth Finance Commission
has recommended tying debt relief and restructuring for the states to their passage and
implementation of FRBM laws. There can be natural skepticism about the enforceability
of such laws by sovereign governments, or by subnational governments that can count on
being bailed out. A precondition for enforcement, monitoring, is also not trivial to
implement. The inability of such legislation to specify the quality of the fiscal adjustment
(what expenditure gets cut, or who gets taxed and how) has also raised concerns about
distortion of incentives (e.g. Rajaraman, 2005). Some have suggested an independent
scorekeeper for monitoring (e.g., Hausmann and Purfield, 2004), while others have
viewed the Finance Commission as a candidate for the job (e.g., Singh and Srinivasan,
2005). In fact, the Eleventh Finance Commission’s report gave impetus to the initial
FRBM legislation, and the Twelfth Finance Commission (TFC) has sought to extend the
scope and impact of fiscal responsibility laws to all the states. Howes (2005), in an
assessment of the initial impact of the state FRBM laws, is quite positive. He sees the
laws as having a positive effect on states’ fiscal positions, though he does not see
conclusive evidence that passing such legislation is a necessary route to fiscal
consolidation. In fact, one could conjecture that such laws, especially without sanctions
for failing to meet targets, are more a symptom or symbol of a political consensus with
respect to fiscal consolidation than a constraining factor. Howes does conclude that this
approach is useful, and rightly stresses the advantages of transparency, targeting, and
monitoring that can together support better overall governance. He also notes that
expenditure quality targets (at least constraining expenditure on salaries) can be (and
have been) incorporated in FRBM laws.
30
Despite the usefulness of FRBM laws, they do not tackle the fundamental
underlying incentive problems that can lead to poor fiscal decision-making by
subnational governments: for example, their effectiveness would rely on states not being
bailed out by the center. With respect to borrowing, the TFC (2004, p.12) has made a
significant recommendation, namely, that “The Central Government should not act as an
intermediary for future lending and allow the states to approach the market directly. If
some financially weak states are unable to raise funds from the market, the center could
borrow for the purpose of on-lending to such states, but the interest rates should remain
aligned to the marginal cost of borrowing for the Center.” This recommendation moves
institutions in the right direction for more efficient fiscal management, but there are still
weaknesses in what is envisaged. Direct access to the market usually means that states
deemed too risky to lend by the market have to pay a higher interest rate, and this in turn
would provide an incentive for such states to be fiscally more responsible and be
perceived as less risky. Any on-lending to fiscally weak states at about the market rate
for central loans would simply dilute the incentive to be fiscally strong that direct access
to the market induces. Instead, the Commission could have recommended that all states
have to access the market directly, with some relatively weak states given a bit more time
to reach that stage, receiving additional grants in the meantime.
The TFC has also developed an elaborate scheme for restructuring the states’
existing debt. Writing off debt and/or rescheduling it, whether it is external debt of
developing countries or the debt of states in India, creates a serious moral hazard – debt
relief blunts the incentives to change behavior that led to the accumulation of debt in the
first place. Although moral hazard cannot be eliminated altogether, conditionalities on
debt relief, provided they are credible, can alleviate it. The recommendations of TFC in
this area do have conditionalities, making passage of FRBM legislation with specific
debt-reduction targets a pre-condition for rescheduling and write-off. While the TFC has
sensibly avoided confining the write-off only to the worst-off states, it could have been
selective in other ways that do not create moral hazard.
One area that has remained relatively untouched, however, in legislation and
recommendations implemented so far, is the issue of captive financing at different levels
of government. We have argued previously (Singh and Srinivasan, 2005a) that the central
government can park its deficits in the public sector banks, which must hold large
quantities of government bonds. The states also have this luxury to some extent.
Furthermore, they have been relying increasingly on access to small savings and pension
funds to finance their deficits (e.g., World Bank, 2005), and unless this channel of captive
financing is blocked, market borrowing will not be effective in hardening budget
constraints. On the other hand, the TFC has recommended overall limits on borrowing by
each state, and if these can be enforced effectively (by limiting off-budget borrowing in
particular), they will help achieve hard budget constraints.
Two conceptual points still need to be analyzed in considering the evolution of

institutions to manage subnational fiscal positions. First, the role of the transfer system
and the consequent nature of the so-called “common pool” problem have been
misunderstood in the Indian context. Second, it is useful to situate the institutional
31
evolution described in this section to our earlier outline of concepts of cooperative,

competitive and market-preserving federalism.
The common pool characterization of subnational fiscal incentives is easy to

understand. If all tax revenues across subnational jurisdictions are pooled (either directly
or indirectly) and shared formulaically, independently of (explicit or implicit)
contributions, any individual jurisdiction has an attenuated incentive to take actions that
stimulate tax revenue. Hausmann and Purfield (2004), for example, emphasize the
common pool problem as a cause of India’s subnational fiscal deficits. Following Singh
(2006), we would like to suggest that the more fundamental concept is really the basic
marginal principle. Thus sharing in a common pool is not a problem if it does not distort
marginal incentives.38 In the context of deficits, debt relief and restructuring result in
marginal expenditures being paid ex post, or with a lag, from central revenues – the
common pool. Similarly, center-state transfers that are designed to fill current
expenditure-revenue gaps also represent dipping into the common pool. However, neither
of these phenomena is a necessary consequence of federalism, and neither is built into
India’s institutional structures – as opposed to its prevalent practice. Hence, some of the
reform that is necessary can come from recognizing where precedent is inefficient, and
changing policies rather than changing institutional structures (though the latter could be
required to achieve the former). Making sure that marginal expenditures are generally
paid through current or future subnational own-tax revenues should be the focus of
reform. This shifts the focus away from FRBM laws toward market discipline for
borrowing, and transfer systems that do not distort per-period marginal incentives. The
issue of the transfer system is explored further in the next section.
The institutional evolution underlying the problem of subnational fiscal deficits

can be related to the different concepts of federalism. India’s federal system was typically
perceived in the 1950s was typically perceived, or at least presented, as being
“cooperative” in nature, reflecting post-independence nationalist attitudes. In practice, the
workings of the system were more hierarchical. Thus, in the political arena, a shuffling of
regional leaders of the Indian National Congress under the “Kamaraj plan” of the early
1960s was projected as a cooperative decision, when its underlying motive was based
more on maintaining centralized hierarchical control. The current nature of Indian
federalism is, perhaps, more genuinely cooperative. The move toward state level FRBM
laws has been partly achieved by agreement, though without formal contracts. The TFC’s
recommendations have an element of hierarchical control, but the debt relief scheme can
be viewed as an explicit contract, which the states can choose to sign, rather than “an
offer that cannot be refused.” The aspects of tax reform discussed in the previous section
38
Hausmann and Purfield (2004) picturesquely illustrates the common pool problem with what happens
when an individual goes to a restaurant in a group and orders lobster, whereas if he were alone, he would
have ordered a cheaper item, chicken. However, this analogy oversimplifies, and masks the problem and,
therefore, the solution. In Hausmann and Purfield’s story, the implicit assumption is that the bill will be
equally divided. Hence the marginal cost of an individual order of lobster is split among the entire group.
Suppose instead that the marginal cost versus chicken of all the lobster orders is separated out and divided
among only those who order lobster. Then the common pool problem goes away. The key idea is that
marginal incentives must be right, so that, in the case of India’s states, they must bear the full marginal cost
of their spending.
32
also have elements of cooperative federalism in them. These cooperative elements can be
viewed as checks on destructive competition in the federal system, through subnational
tax and expenditure policies driven by what M. Govinda Rao and others have termed
“competitive populism,”39 which surfaced At the same time, one can obviously view the
move toward market-determined borrowing and overall restraints on borrowing at the
subnational level as a move toward the MPF condition of hard subnational government
budget constraints. In fact, the use of the market mechanism to achieve hard budget
constraints can be termed “market-disciplined federalism” (MDF). In tracing the process
of changes in subnational fiscal management and outcomes, one can perhaps identify
competitive and cooperative elements of federalism in this process of change, while MPF
can be viewed as the direction in which the system is moving.
Any fiscal problems that China might face are less severe than India’s. China’s
fiscal deficit did not touch 3% till 2002, and that was the result of a deliberately
expansionary fiscal policy, designed to compensate for sluggish domestic consumption. It
has fallen subsequently to about 1.5% of GDP. The consolidated figure for all levels of
government is about 4% of GDP. Debt, too, is much lower than India’s at under 40% of
GDP. Furthermore, China’s high savings rates and large trade surpluses give it a much
greater margin of safety than exists for India. A caveat to this optimism is the existence
of large off-budget deficits of state enterprises, and the poor shape that many state-owned
banks are in. In fact, the government has used foreign reserves, to restructure some banks,
to shore up the financial system. The key difference from India lies in the harder budget
constraints that have been in place for provincial and local governments. According to
Cao, Qian and Weingast (1999), those hard budget constraints have driven subnational
governments toward various forms of privatization strategies, further enhancing the
market-preserving nature of Chinese federalism. This development builds on the existing
fiscal autonomy of subnational governments. Privatization at the subnational level is
important, as it reduces the future use of the state-owned financial sector as a resting
place for bad debts.
Overall, despite the issue of off-budget deficits, it seems that China’s subnational
governments are considerably stronger than India’s both in terms of quality of
expenditure (better delivery of local public services), and overall fiscal health (smaller
subnational budget deficits).40 Again, this superficially poses somewhat of a paradox for
conventional theories of democratic responsiveness. While the operation of democratic
incentives at the local level has been weak in India, the better performance of Chinese
subnational governments is not explained at all be conventional democracy. Again, one
might explain this better performance in terms of greater alignment of incentives between
bureaucrats and politicians on the one hand, and constituents on the other – this being
achieved in China through shared stake holding in local economic development. The
Indian case has been much more one of adversarial relations between governed and
government. One might conjecture that fiscal autonomy and a stronger Wicksellian
connection between public revenues and spending is more important for economic
39
This characterization was made by Rao at a Stanford conference in 2002.
40
Bahl and Martinez-Vasquez (2005) are somewhat more cautious in their assessment of the Chinese case:
pension liabilities at the subnational level are one factor in this caution.
33
performance than the trappings of democracy. Of course, this does not devalue all the
inherent, non-instrumental benefits of democratic governance.
6. Impacts of Federal System on Growth and Equity

For decades, a major debate has proceeded with respect to the proper role of
government vis-à-vis the market in determining resource allocation, as well as how this
determination interacts with non-material aspects of society. The last two decades have
seen a shift toward acknowledging that market institutions are superior for many aspects
of resource allocation, including those which impact growth, as well as those which
affect static efficiency. While the debate is not settled in the minds of some, as evidenced
by various policy discussions and actions in India, the more relevant issues really lie
elsewhere. First, there is more room for disagreement with respect to how equity
concerns should be handled, since this introduces normative considerations that tend to
get tangled up with positive analyses of the impacts of government policies. Even here,
though, we have considerable theoretical guidance and consensus on which policies may
work best to achieve societal equity objectives, whatever those may be.
In comparison to this more settled literature on government-market boundaries,

there is less work on, and perhaps less understanding of, the effects of the organization of
governmental structures on economic activity and performance. Modern theories of
federalism are an important subcategory of theories of the economic impacts of
governance, with the concept of MPF being an example of an attempt to unify our
understanding in a normative ideal for federalism. In this context, there is a clear link
from some aspects of federal structures to their economic consequences, and these are
captured in the MPF rubric. In particular, the benefits of an internal common market, just
as is the case for international trade, are easily understood in terms of the theory of
competitive market exchange. The rationale for decentralization of expenditure authority
for local public goods has also been developed, in terms of political competition to satisfy
constituents’ wants effectively. The assignment of revenue authorities, coupled with a
system of intergovernmental transfers, creates some more interesting theoretical issues.
As discussed in Section 2, the traditional public finance focuses mostly on static

effects of transfers, and is mainly normative in nature. It deals with the question, given
some norms for minimum national standards with respect to the provision of subnational
public goods, how can the intergovernmental transfer system be designed to fulfill these
norms? In the case of India, even this basic objective has not been well met by the
transfer system. Recently, Barry Weingast and his co-authors (e.g., Careaga and
Weingast, 2001) have attempted to tackle an even more important issue for developing
countries, namely the growth effects of federal institutions governing revenue authority
and sharing. At the risk of some oversimplification, we can distinguish the two sets of
questions as follows. The standard public finance question takes the subnational
jurisdiction’s income as given, and looks at the incentive effects of tax assignments and
transfers. The growth perspective examines the effects of the tax and transfer system on
incentives to increase income (e.g., through public or private investment). Of course there
is a large macroeconomic literature on taxes and growth – the difference here is that a
34
federal structure adds intergovernmental transfers as a factor. In this case, there may be a
conflict between goals of short run horizontal equalization and long run development.
Beginning with the static issue of horizontal equity, the Indian case is one where
the impacts of Finance Commission transfers are definitely equalizing across states. This
was goal was built into the transfer formula from the first commission, and analysts such
as M. Govinda Rao have estimated the equalizing effects for various cross-sections and
time periods, as an elasticity of transfers with respect to per capita income. Rao has also
shown that including Planning Commission transfers weakens the equalizing effect. This
is so despite the inclusion of some equalizing criteria in the Planning Commission’s
formulas, which were introduced in 1969. In any case, the existence of ministry-based
transfers, and even more so of implicit transfers through subsidized and directed loans,
debt relief and restructuring, tax exportation, targeted public investment, and
administered pricing (particularly the freight equalization scheme) makes it very difficult
to estimate the overall degree of horizontal equalization that takes place within India’s
federal structures.
Focusing on Finance Commission transfers alone, one can note that there has
been a slight decrease in horizontal equalization in the Twelfth Finance Commission’s
recommendations, versus its predecessor (Rao and Jena, 2005; Howes, 2005). This was,
of course the result of explicit changes in the formula, to put less weight on per capita
income, thereby reducing the horizontal equalization achieved through the formula. Rao
and Jena calculate the exact differences in tax devolution as a result of the TFC’s formula
change. Howes shows that incorporating grants (which were targeted at the poorer states)
reduces this inequalizing effect, but does not remove it. It should be noted that, while
India’s states receive about half of their revenues through explicit transfers from the
center (about 30% of the center’s own revenues), these transfers represent about 5-6% of
average state GSDP. In total, therefore, the states receive transfers that are small relative
to their overall economies. Nevertheless, this process of apparent backing off from
horizontal equalization takes place against a background of increased regional income
inequality.
Table 7: Criteria and Relative Weights for Tax Sharing
Weight (%) Weight (%)

Criterion 11th FC 12th FC
1. Population (1971 Census) 10 25
2. Income (Distance Method)* 62.5 50
3. Area 7.5 10
4. Index of Infrastructure 7.5 0
5. Tax Effort** 5.0 7.5
6. Fiscal Discipline*** 7.5 7.5
Notes: *The distance method is given by: (Yh-Yi)Pi/Σ(Yh-Yi)Pi where, where Yi and Yh represent per capita
SDP of the ith and the average of the three highest income states respectively and Pi is the population of the
ith state. For the three highest income states, a notional distance is assigned.
35
** Tax Effort η is estimated as η = (Ti / Yi) / √(1/Yi) where, Ti is the per capita tax revenue collected by the
ith state and Yi is the per capita State domestic product of the ith state.
*** Estimated as the improvement in the ratio of own revenue of a state to its revenue expenditures divided
by a similar ratio for all States averaged for the period 1966-99 over 1991-1993 (11th FC years used for
illustration).
Source: Rao and Singh (2005), Twelfth Finance Commission Report (Twelfth Finance Commission, 2004)
These observations are not meant to imply favoring the previous status quo with
respect to the tax devolution formula. In fact, the Finance Commission’s methodology is
non-transparent in its rationale and its outcomes. Theory would suggest using measures
of structure, such as population density, overall size, topography, and economic structure
to establish minimum norms for tax and expenditure levels, which could then be used to
determine transfers that would sustain minimum expenditure norms for a state that
behaves according to the norm.41 States can then raise and spend money at the margin,
without any distortionary effect of transfers. Instead, the Finance Commission uses
various criteria in the formula itself, calculating tax shares based on this, without being
able to assess if the transfers are adequate or not (see Table 7). To some extent, shortfalls
are met through grants, but the use of ad hoc grants based on ex post gaps (after the
preliminary devolution is calculated) has the potential to completely undermine
incentives. The Finance Commission itself does not see this as a problem (Rangarajan,
2005), arguing that the gap-filling is based on normative measures. Nor does it seem to
show up in some econometric studies, though the results are not consistent across studies.
To some extent, the problem may also be more severe with Plan grants, which are, in
some ways, even more the result of bargaining, lobbying and “gap-filling.”
Some of the impact of different components of the formula can be assessed by

recalculating shares without one component or another (keeping the relative weights on
other components constant). In particular, since the fiscal discipline and tax effort
measures are very highly correlated with each other and with population (simple
unweighted correlation coefficients greater than 0.98), excluding them has very little
impact on the major states (with the exception of West Bengal, which has recently been a
consistent poor performer on these criteria). On the other hand, excluding the “area”
component has two kinds of effects. Because this variable is proportional to area for
larger states, but is truncated at a fairly high value for small states (presumably to capture
both fixed costs of administration and higher costs associated with lower population
density), excluding it actually helps the poorest state of Bihar. The very small states are
the biggest beneficiaries of including this component, but that includes high income
states such as Goa and Punjab.42 Overall, it is not at all obvious what the impact of a
change in behavior (e.g., tax effort or fiscal discipline) has on a state’s share, nor whether
41
As an illustration, a ‘need-revenue’ gap, which measures the difference between what a state ought to
spend to provide specified levels of public services and the revenue it can raise at a given standard level of
tax effort, can be calculated as Gi =⎯QCi -⎯tBi, where Gi is the gap (per capita),⎯Q is the desired
(normative) level of composite public service provided by the state per capita. Ci is the unit cost of the
public service (reckoned at justifiable costs), ⎯t is the standard tax effort, and Bi is the per capita tax base.
For need calculations unit cost components within the control of the State governments would also have to
be reckoned at justifiable levels. See Rao and Singh (2005), Chapter 8.
42
The share of Goa is almost four times what it would be without the “area” component of the weighting
scheme.
36
the incentive effects are sufficient to induce changes in behavior. One can certainly do
the former calculation. For example, if Chhattisgarh’s tax effort measure had fallen to
that of Madhya Pradesh (about 0.9% of GSDP lower, so a substantial decline of close to
15% of tax revenue), the penalty in terms of the reduction in transfers (neglecting second
order effects from recalculating relative shares) would have been about 0.9% of the
overall formulaic transfers to the state. It is difficult to say whether this would be a
deterrent, but the size of the penalty is an order of magnitude smaller than the tax
reduction, and it seems unlikely that any state’s behavior would be driven by the
incentives built into the formula. In the absence of good empirical models of state level
fiscal behavior, even after over 50 years of Finance Commissions, we can only speculate.
Understanding the growth impacts of intergovernmental transfers requires some

modeling of how subnational governments can affect their tax bases. Careaga and
Weingast (2001) use a model in which government decision-makers can either capture
rents, or increase their jurisdiction’s income, and hence its tax base. From this
perspective, the marginal subnational retention rate of all taxes levied on the subnational
tax base comes into play. Weingast (personal communication) observes that in the United
States in the 19th century, the marginal retention rate of a state was nearly 100%. Qian
and Weingast (2005) calculated this figure for China during the high growth phase of
reform, 1981-92, and estimated the average marginal retention rate for a province at 89%,
with 68% of the provinces having marginal retention rates of 100%. On the other hand,
they report a similar calculation by Zhuravskaya for Russian cities, which came up with a
retention rate is around 10%. Finally, Careaga and Weingast (2001) calculate this
percentage as 23.3% for Mexico in 1995.43
For the Indian case, this kind of calculation has not been seriously attempted.44
Note that the idea here is to look at the overall tax revenue of a state, without prior
assumptions about assignment. A simple calculation might be as follows. If a state
receives one-third of all taxes assigned to the center, and all of the taxes assigned to the
state, and the latter and former made up equal shares of the state’s revenue, then its
marginal share of the extra tax revenue generated by growth would be 50%. This assumes
that tax rates could not be adjusted, and that all tax revenues have the same income
elasticity. The complication in this calculation would be the impact of the Finance
Commission’s equalization formula. If only a single state grew, out of say 20 equal sized
states, then about half of the increased central tax would be shared with other states
(reflecting the 50% weight given to the “distance criterion,” though this overstates the
impact of that factor). Now the marginal share of this state would be somewhat under
40%. If one accepts this kind of calculation, it would suggest that the horizontal
equalization approach used in India has strong negative growth effects. Note that this
calculation has nothing to do with tax effort or fiscal discipline effects of tax sharing
43
They point out that there were periods when all state revenue was put in a common pool and then divided
by an equal sharing rule, which meant that the marginal percentage for the average state was close to 1/33,
the denominator being the number of states.
44
We are grateful to Barry Weingast for some suggestions, made in a different context, on how to go about
this calculation, as well as for pursuing this overall line of reasoning. He is blameless for the calculations
attempted here.
37
formulas. Weingast (personal communication) suggests that sharing rules be devised to

reduce the marginal take of the center, and correspondingly increase the marginal
retention of the state or other subnational government, even in cases where horizontal
equalization is being attempted. One should note, however, that the argument as
developed here is incomplete, since the center’s role in promoting economic activity is
not modeled. In fact, since the center also can increase tax revenue by increasing the
national tax base, there is a trade-off that is not captured by focusing on just the
subnational government. Of course, one can argue that much of the center’s activity is
related to public goods that are inelastic with respect to growth, and it is subnational
stimulus and support of the market that matters. But this is a question that deserves closer
examination. It also relates to the initial assignment of revenue and expenditure
responsibilities, which are presumably reflective of the comparative advantages of
different levels of government. One should emphasize again the caveat that, in the
absence of a model of the links between taxation, public expenditure and growth, one
should not rely too much on the specific numbers presented in our illustrative calculation.
In addition to redesigning the formulaic part of the intergovernmental transfer

system to directly improve marginal incentives, one can also argue that reducing the
magnitude of transfers can decrease the scope for political influence effects that distort
subnational behavior, and also improve subnational political incentives – constituents of
subnational jurisdictions can more clearly identify the Wicksellian connection between
their costs and benefits in voting on taxation. Since there is no reason for centralizing
expenditure decisions more than the status quo, reducing transfers requires further
decentralization of tax authority. This can be done through allowing subnational
jurisdictions to piggyback on some of the same tax bases that are used for center-state tax
sharing. For example, allowing states and local governments to impose income tax
surcharges would not only improve their marginal retention, but it could reduce the need
for tax sharing. This would also allow the focus of Finance Commission transfers to be
more clearly on horizontal equalization for the poorest states. In general, the greatest
weakness of local government reform as conceived and implemented so far has been the
failure of tax assignment to match expenditure assignment. The poor functioning of the
State Finance Commissions could be partly addressed by increasing the power to tax of
the local authorities. In this latter case, some expenditure authorities also need to be
decentralized more fully, since local government autonomy in this dimension remains
constrained by state and central decisions with respect to investment projects and social
insurance programs.
7. Future Reform and Prospects

In Section 2, we noted that the constitutional provisions on center-state economic
relations, including the mandatory appointment of Finance Commissions every five years
were largely set, based in part on debates during 1946-50 in the Constituent Assembly on
the draft constitution and also importantly, the colonial era Government of India Act of
1935. Obviously the circumstances then prevailing were seen to demand a strong central
38
government and led to many unitary features in the constitution.45 These circumstances
no longer prevail. Moreover, an extra-constitutional body, the Planning Commission,
was set up in 1950 at the center, with state planning commissions and boards following
later. Again, the economic conditions then prevailing, and more importantly the belief of
the then Prime Minister, Jawaharlal Nehru, and his Congress Party in central planning
(modeled after the Gosplan of the Soviet Union) and a dominant role for the state in
economic management, were the reasons for establishing the Planning Commission. Over
the subsequent four decades it became a major player in center-state economic relations
and has been making transfers to states in support of their five-year plans.
The Soviet Union collapsed in 1991 and central planning as a mode of articulating
and implementing a development strategy had gone out of fashion even earlier. The roles
of planning and planning commissions need urgent rethinking in the contemporary
context, now that markets have been allowed to play a far greater role in the economy
since the reforms of 1991. Besides emphasizing state control over the economy, the
Indian development strategy since the 1950s until the mid-1980s was extremely inward-
oriented, with across-the-board import substitution, implemented though a plethora of
controls driving the investment pattern of the public and private sector. Foreign
investment was actively discouraged and foreign borrowing was basically from
concessional loans of multilateral development banks and bilateral foreign aid. The
economy has moved away from this dysfunctional strategy with much greater openness
to external competition and active pursuit of foreign investment (direct and portfolio).
This shift was also accompanied by reforms in the financial sector along with making the
rupee convertible for current transactions. Although the Reserve Bank of India appointed
a committee in 1997 to put together a road map for making the rupee fully convertible
(i.e. for current and capital transactions) and it recommended a three year phase in for
doing so, spread over 1997-2000, no formal action was taken. The committee has
recently been revived to prepare yet another road map.
With the economy getting integrated more and more with the world economy both
in trade in goods and services and in finance, domestic fiscal and monetary policies (and
also public investment in social and economic infrastructure, to the extent that the public
sector continues to be the supplier of infrastructure services), have to be consistent with
foreign sector policies particularly with respect to the exchange rate and capital flows.
Evidence from other federations (e.g. Argentina) suggests that the political economy
conflicts of federalism in the fiscal arena, themselves rooted in faulty institutional design,
can trigger an external payments/exchange rate crisis. As Indian policy makers are
considering a road map for making the rupee fully convertible, they have to ensure that
fiscal aspects of India’s current federal system do not pose such a threat and undertake
appropriate actions to reform the system, if necessary, for this purpose.
We recognize that rewriting the constitutional provisions regarding center-state

fiscal relations will require considerable thought and debate and in any case, will take a
long time to accomplish. In the meantime there are several reforms that can be
considered for implementation.
45
See also footnote 10 and the related discussion of local discrimination issues in section 2.
39
First, it would be very healthy if states and the center have means for discussing
each other’s fiscal policies in a common forum. Currently the discussions on state plans
at the Planning Commission take place between the commission and each state
separately. The National Development Council, in which state chief ministers are
represented, on the other hand, discusses central plans. Of course, there are ad hoc for
discussion, such as the Empowered Committee of State Finance Ministers (ECSFM) and
the meetings of state finance secretaries organized by the Reserve Bank of India, that do
provide an opportunity for each state representatives (finance ministers or finance
secretaries) to be informed of and comment on other states’ policies. As is well known,
the ECSFM under the leadership of Asim Dasgupta, Finance Minster of West Bengal,
was instrumental in ushering in VAT. But to the best of our knowledge, those forums do
not discuss the central government’s fiscal policies. We propose supplementing, if not
completely replacing, these ad hoc forums by a formal Fiscal Review Council, analogous
to the Trade Policy Review Mechanism of the World Trade Organization (WTO) that
enables the members of the WTO to discuss, review and comment on each member’s
trade policies periodically.
The European Union introduced a new mechanism under which each member
country submits to the European Commission each year a national reform action plan
setting out how it intends to create jobs and growth, and in particular how it will meet
two or more specific economic targets: an employment rate 70% of working age
population and an expenditure of 3% of GDP on R&D. Although the proposal of the
former Dutch PM, Mr. Kok to use the EC review of action plans to "name and shame"
countries which talked a good game but failed to deliver and "fame" those whose
performance was exemplary, did not go far, there is something useful in this mechanism.
We propose that the Interstate Council (ISC) constitute itself into a Fiscal Review
Council (FRC) and meet at an appropriate frequency (certainly no more frequently than
once a year) to review the medium and long term fiscal policies of the state and center as
well as make recommendations. Each state and the central government would submit to
the FRC its plan (in terms of precise and specific tax and expenditure proposals) for
achieving set revenue and fiscal deficit targets (based on FRBM legislations) as included
in its annual budget and its performance relative to the target in the previous fiscal year.
The FRC would seriously review the targets and performance. A report ranking states by
the FRC and made public would be very useful. To avoid political grandstanding and to
encourage serious discussion, the meetings could be closed to the media. Whether the
recommendations should be binding on all parties or only advisory is an issue that the
ISC could decide. Because in the FRC’s deliberations, each Chief Minister gets an
opportunity to comment on and learn from other states’ and center’s policies, any
rankings and recommendations emerging from the review would not only have greater
political weight but also provide a commitment mechanism for each chief minister to
undertake reforms in his/her state which he/she may not be able to do unilaterally.
Second, the center-state transfers through the Finance Commission, Planning

Commission and the ministries have to be looked at in a unified framework. Ignoring
many details and simplifying a lot, there are essentially three types of transfers: from
40
current revenues as determined by the Finance Commission, capital transfer for financing
investment (largely the domain of the Planning Commission), and transfers for
internalizing positive externalities that one state’s fiscal actions may have on other states
and the country’s economy as a whole (currently, the domain of centrally sponsored
schemes). We suggest that: (i) the center take full responsibility for financing investment
and operational costs of projects that have spill-over across states, regardless of the
authority that implements them (center or state). We understand that the current system
of centrally sponsored schemes, under which the center provides partial funding for the
project’s investment cost and for its operational cost for a limited period has had the
unfortunate effect of that projects get started and completed but once completed are not
fully utilized because states have not provided the needed costs of operating them once it
became their exclusive responsibility to provide them. The center assuming full financial
responsibility will avoid this waste. (ii) Reconstitute the Planning Commission as a Fund
for Public Investment (FPI) for both the center and states. Its share holders would be the
state and central governments. The Fund, much like a multilateral development bank,
would appraise the projects proposed for their economic and social returns as well as
feasibility and soundness of proposed financing (from the center or state’s own resources,
borrowing from domestic and foreign sources and capital transfers from the center, if
relevant.
We agree with the TFC that the center should not be in the business of being a
financial intermediary between capital markets at home and abroad as well as external aid
agencies. The FPI will instead borrow from domestic and foreign capital markets with
state and central governments jointly guaranteeing the loans. The freedom for states to
approach capital markets directly and negotiate with foreign donor agencies could lead to
interstate imbalances in the flow of financing. On the other hand, the failure to attract
flows could also provide an incentive for the failing states to undertake policies to make
them more attractive to lenders. Still, to the extent that the projects proposed by states
are found to be worthwhile from an economic and social perspective by the Fund, it could
recommend that capital transfers from the center to make up for the failure of the states
proposing them to attract funding from other sources. However, such fund-recommended
capital transfers should not carry any subsidy (relative to the cost of borrowing to the
center) on interest rates. In other words, although we considered it, we do not
recommend that the fund also open a soft lending window similar to IDA. There is no
economic rationale for such a window. If a state is deemed ‘poor’ for reasons of
horizontal imbalance, it should be addressed through the Finance Commission transfers
and not through the fund. The logic of our proposal suggests that each state replace its
Planning Commission or Board with a fund for financing investment projects of local
authorities along the same lines as we have suggested for the central FPI. We note that
the TFC’s recommendation for limiting the role of the center as a financial intermediary
as a lender to states is being followed by the Reserve Bank. It is exploring the
development of institutions to support this shift to market borrowing, including offering
mechanisms, secondary markets for government debt, credit ratings, and methods of
regulation and monitoring. Therefore, the case of reforming financing states’ capital
expenditure through new borrowing mechanisms involves building on reforms already
taking place in the financial sector. It is clear that there are imbalances across states in
41
generating viable project proposals, identifying sources of finance and implementing

them once approved. These ‘capacity’ imbalances have to be addressed independently of
proposal for project financing and transfers.
Besides tax reform, reforms of expenditures for enhancing their efficiency in

delivering their goods are essential. For example, a number of subsidies continue to be
rationalized on distributional equity grounds. But there is no reason why such subsidies
need to incur efficiency costs. Simple measures such as ensuring recipients of subsidies
preferably pay the full cost of delivery at the margin, while they continue to receive
subsides on infra marginal purchases can be implemented (Noll and Srinivasan, 2006).
More generally, in a country as diverse as India, there is an opportunity of exploiting this
diversity for experimentation with alternative mechanisms. The TFC, after discussing the
principle of equalization, does not do much with it. Nor does it address the issues of
determining the services for which the governments are to be held responsible, let alone
linking them to constitutionally defined rights of the citizen. These issues need urgent
discussion.
Last but not least is the issue of public sector production of those goods and
services for which there is no social rationale based on consideration of scale economies,
public goods or certain kinds of externalities. Instead of delving into the social rationale,
Indian debate has degenerated into whether profit-making (at market prices) public
enterprises should be privatized or not. Given the impact of the existence and operation
of public enterprises directly and indirectly on the economy and on public finances in
particular, a serious analysis of their social rationale is called for. Protecting the
employment of the currently employed in these enterprises is not a convincing social
rationale.
8. Conclusions
Most observers, though not all, of the Indian economy agree that economic
liberalization and systemic reforms since 1991 have contributed to sustaining a growth
averaging more than 6% a year since, and that the growth at about the same rate in the
1980s, led as it was in large part by fiscal profligacy and rapid accumulation of domestic
and foreign debt, but without significant and systemic reforms, was not sustainable. The
balance of payments exchange rate crisis of 1991 that led to systemic reforms was the
inevitable consequence of irresponsible macroeconomic policy of the eighties. The
current debate on India’s growth prospects center around issues of governance and of
deepening, widening and accelerating reforms. The working of India’s federal system is
central to this debate. We discussed in Section 1 conventional theories of federalism,
mostly of Western democracies, and their relevance, if any, to India’s vibrant, resilient
but imperfect democracy.
In Section 2, we explored the background to India’s economic performance, its

federal system and alternative perceptions of the contribution policy reforms to the
growth experience since the 1980’s. The strong unitary features of India’s constitution
adopted in 1950 and the creation of the Planning Commission, also in 1950, set the
42
framework for policy making, particularly as it related to economic development for the
subsequent four decades, until the reforms of 1991. The role of the constitutionally
mandated Finance Commissions, (twelve have reported thus far) on center-state fiscal
relations are elaborated in this section. We found that by and large, they have played a
constructive role, but in the contemporary context of greater acknowledged role of
markets (in particular capital markets) and of openness to foreign trade and investment in
the economy, greater autonomy state governments and devolution to local self-governing
bodies, this role needs to be reexamined.
We discuss at some length the quality governance and government expenditure in

Section 3. The change in the body politic, away from the same party governing both the
center and states, to one at which coalitions (not necessarily stable over time) govern at
the center and states is found to have had a major impact toward good governance, not
always favorable. The decline in the quality of the bureaucracy and legislative bodies,
and perceptions of growing corruption at all levels, often with connivance between
bureaucrats and politicians are evident. However, we found that while India’s centralized
political and bureaucratic control and monitoring has neither been very effective nor
accountable to the larger public, the recent move towards devolution to local bodies has
the potential to bring about greater efficiency and accountability. We compared India
with China and argued that alignment of political and bureaucratic interests at the local
level, although the Communist Party appointed both in China, led to better performance.
Regulation has become much more important than direct state intervention in recent years
in India. The successful performance of some industries under regulation (e.g.
telecommunications) and still miserable performance (e.g. electricity) of others which are
neither fully regulated nor completely under direct state control are interesting. However,
this experience is fairly recent and short and one cannot reach firm and universal
conclusions yet on regulation versus intervention.
We turned to the efficiency of the tax system in Section 4. After reviewing the
history, we discussed recent developments with respect to the introduction of a country-
wide VAT and also the appropriation by the center of the exclusive right to tax services.
We concluded that although China has grown rapidly without using any optimal tax or
other economic models to guide its design of tax policy, it would be irresponsible to infer
that microeconomic efficiency of the tax system can be neglected and for this reason
applaud India’s moves toward an efficiency improving tax reform.
The unsatisfactory current fiscal situation of central states and by and large,
unsuccessful attempt at fiscal consolidation since the reform of 1991 are the subjects of
Section 5. Variation in performance across states is noted. We discuss at length and, in
some depth, the problem of ensuring subnational fiscal discipline in a federal system and
the attempts to instill some discipline through Fiscal Responsibility and Budget
Management legislation at the center and some states, and through the recommendation
of the Twelfth Finance Commission. We noted that China’s overall fiscal situation and
that of its subnational governments are considerably stronger than India’s both in terms
of quality of expenditure and overall fiscal health though this might pose a paradox for
conventional theories of fiscal responsiveness. Again, we offer greater alignment of
43
incentives of bureaucrats and politicians on the one hand and constituents on the other in
China as a possible explanation.
In Section 6 we consider the impact of the fiscal system on growth and equity.
We found that the complexity of the transfer system in India, with the Finance
commission, Planning Commission and central ministries involved in it, makes it difficult
to isolate their impact on growth and equity. No firm conclusion can be drawn
empirically on whether the transfers on the whole (including implicit transfers) have been
equalizing across states, nor can one be confident that they have been conducive to
growth.
In Section 7 we offer some reform proposals. First is to create a Fiscal Policy

Review Council (FPRC), which would be existing Inter State council constituting itself
as FPRC for review of state and central fiscal plans. Second is for the Center to assume
sole responsibility for what are now called Centrally Sponsored schemes. Third is to
reconstitute the Planning commission into a Fund for Public Investment.
All three proposals are meant to create a politically credible means of pushing
ahead with reforms which each state individually does not find it politically feasible to
implement and to ensure greater efficiency and equity in the process of allocation of
public funds for investment. Lastly, we make some tentative suggestion as how to limit
and improve the efficiency of subsidies and to link entitlements and rights of citizens
more clearly to the Directive Principles of State Policy in the Constitution.
References
Bahl, Roy and Jorge Martinez-Vazquez (2005), Fiscal Federalism and Economic Reform
in China, in T.N. Srinivasan and Jessica Wallack, eds., Federalism and Economic
Reform: International Perspectives, Cambridge, UK: Cambridge University Press
Biswas, R. and S. Marjit (2000), “Political Lobbying and discretionary Finance in India:
An Aspect of Regional Political Influence in a Representative Democracy,” processed,
Center for Studies in social Sciences, Calcutta.
Brass, Paul, The Politics of India Since Independence (New York: Cambridge University
Press, 1990)
Brennan, Geoffrey, and James M. Buchanan (1980), The Power to Tax: Analytical
Foundations of a Fiscal Constitution, Cambridge: Cambridge University Press.
Breton, Albert (1995), Competitive Governments: An Economic Theory of Politics and

Public Finance, Cambridge University Press, New York.
Cao, Yuanzheng, Yingyi Qian, and Barry R. Weingast. 1999. “From Federalism, Chinese
Style, to Privatization, Chinese Style” Economics of Transition. vol 7 no. 1: 103-31.
44
Careaga, Maite, and Barry Weingast, 2001, Fiscal Federalism, Good Governance, and
Economic Growth in Mexico, working paper, Stanford University
Chhibber, Pradeep “Political Parties, Electoral Competition, Government Expenditures

and Economic Reform in India,” Journal of Development Studies, Vol. 32, No. 1
(October 1995), pp.74-96
Coase, Ronald “The Problem of Social Cost,” Journal of Law and Economics, Vol.3, No.
1 (1960), pp. 1-44
Das-Gupta, Arindam, and Dilip Mookherjee (1998), Incentives and Institutional Reforms
in Tax Enforcement: An Analysis of Developing Country Experience, Delhi: Oxford
University Press.
Dasgupta, Sugato, Amrita Dhillon and Bhaskar Dutta (2001), “Electoral Goals and
Center-State Transfers in India”, processed, Indian Statistical Institute, New Delhi.
DeLong (2003), J. Bradford DeLong, “India Since Independence: An Analytic Growth

Narrative,” in In Search of Prosperity: Analytic Narratives on Economic Growth, ed.
Dani Rodrik (Princeton, New Jersey: Princeton University Press, 2003)
de Zwart, Frank The Bureaucratic Merry-Go-Round: Manipulating the Transfer of

Indian Civil Servants (Amsterdam: Amsterdam University Press, 1994)
David Dollar, Giuseppe Iarossi and Taye Mengistae “Investment Climate and Economic
Performance: Some Firm Level Evidence from India”, paper presented at 3rd annual
Stanford Conference on Indian Economic Reform (June 2002)
Drèze, Jean, and Harris Gazdar, 1996, “Uttar Pradesh: The Burden of Inertia,” in Indian
Development: Selected Regional Perspectives, ed. by Jean Drèze and Amartya Sen
(Oxford and Delhi: Oxford University Press).
Grossman, Sanford J. and Oliver D. Hart, “The Costs and Benefits of Ownership: A
Theory of Vertical and Lateral Integration,” Journal of Political Economy, Vol. 94, No. 4
(August 1986), pp.691-719
Hall, Richard L., and Bernard Grofman, (1990) “The Committee Assignment Process and
the Conditional Nature of Committee Bias,” American Political Science Review, Vol. 84,
No. 4 (December), pp.1149-66
Hausmann, Ricardo, and Catriona M. Purfield, 2005, “The Challenge of Fiscal

Adjustment in a Democracy: The Case of India,” A Sustainable Fiscal Policy for India:
An International Perspective, Peter Heller and M. Govinda Rao, ed., Oxford University
Press
45
Howes, Stephen (2005), Reforms to India’s Federal Transfer and Borrowing Regime
Proposed by the Twelfth Finance Commission: what will they mean for the states?
presentation at Conference on 'Fiscal Responsibility and Intergovernmental Finance' -
June 22-24, Hyderabad,
http://siteresources.worldbank.org/INTINDIA/Resources/Stephen.pdf.
Howes, Stephen, and Rinku Murgai, 2005, “Subsidies And Salaries: Issues in the
Restructuring of Government Expenditure in India,” A Sustainable Fiscal Policy for
India: An International Perspective, Peter Heller and M. Govinda Rao, ed., Oxford
University Press
Inman, Robert P. and Daniel L. Rubinfeld, “The Political Economy of Federalism," in

Dennis Mueller, ed., Perspectives on Public Choice: A Handbook (Cambridge:
Cambridge University Press, 1997), pp.73-105
Jin, Hehui, Yingyi Qian, and Barry R. Weingast (2005), Regional decentralization and
fiscal incentives: Federalism, Chinese style, Journal of Public Economics 89, 1719–1742
Kapur and Mehta (2002) “The Indian Parliament as an Institution of Accountability,”

Report prepared for the IPU, Harvard University (May)
Kelkar, Vijay (2004), India: On the Growth Turnpike, K.R. Narayanan Oration,
Australian National University
Khemani, Stuti, 2002; Federal Politics and Budget Deficits: Evidence from the States of
India, World Bank Development Research Group Paper, WPS 2915
Atul Kohli, Democracy and Discontent: India’s Growing Crisis of Governability (New
York: Cambridge University Press, 1990).
Montinola, Gabriella, Yingyi Qian, and Barry R. Weingast. 1995. “Federalism, Chinese
Style.” World Politics. 48(1): 50-81.
Mukherji, Joydeep (2006), Economic Growth and India’s Future, Occasional Paper 26,
Center for Advanced Study of India, University of Pennsylvania
Mundle, Sudipto and M. Govinda Rao (1991), “Volume and Composition of Government
Subsidies in India,” Economic and Political Weekly, Vol. XXVI, No. 18, May 4, pp.
1157-72.
Musgrave, R.A. (1959), The Theory of Public Finance, McGraw-Hill Book Company,
New York.
Noll, Roger, and T.N. Srinivasan (2006), More Efficient Subsidy Scheme Benefits
Consumers, Government and Economy, SIEPR Policy Brief, Stanford Institute for
Economic Policy Research, Stanford University, May.
46
Oliveira, Jesus, (2001) Economic Effects of Origin and Destination Principle for Value-
Added Taxes, working paper, School of Business and Public Management George
Washington University, April.
Olson, M. (1986), "Toward a More General Theory of Governmental Structure",

American Economic Review, May.
Olson, Mancur (1993), Dictatorship, Democracy and Development, American Political

Science Review, 87, 567-575.
Panagariya, Arvind (2005), The Triumph of India’s Market Reforms: The Record of the
1980s and 1990s, Policy Analysis, November 7, No. 554.
Parikh, Sunita and Barry Weingast, 1997, A Comparative Theory of Federalism: India,
Virginia law Review, 83(7), 1593-1615.
PROBE Team (1999), in association with Center for Development Economics, Public
Report on Basic Education in India, New Delhi: Oxford University Press.
Purfield, Catriona (2003), “The Decentralization Dilemma in India,” IMF Working
Paper, Washington: International Monetary Fund.
Rajaraman, Indira (2005), “Intergovernmental Finance and Policy Coordination:

Implications of the Report of the Twelfth Finance Commission”, presentation at
Conference on 'Fiscal Responsibility and Intergovernmental Finance' - June 22-24,
Hyderabad, http://siteresources.worldbank.org/INTINDIA/Resources/IndiraRajaraman-
ppt.pdf.
Rangarajan, C. (2005), Approach and Recommendations, Economic and Political

Weekly, July 30, 3396-98.
Rao, M. Govinda (2000), “Tax Reform in India: Achievements and Challenges”, Asia-
Pacific Development Journal, 7, 2, 59-74.
Rao, M. Govinda (2003), “State Level Fiscal Reforms in India”, paper presented at
Cornell University, conference on Indian Economic Reform, April 19-20.
Rao, M. Govinda and Pratap Jena (2005), Balancing Stability, Equity and Efficiency,
Economic and Political Weekly, July 30, 3405-3412.
Rao, M. Govinda and Sudipto Mundle (1992), “Analysis of Changes in State

Government Subsidies 1977-1987,” in Bagchi, Amaresh, J. L. Bajaj and W. A. Byrd
(eds), State Finances in India, Vikas Publishing House, New Delhi. pp. 103-143.
Rao, M. Govinda, and Nirvikar Singh (2001), “Federalism in India: Political Economy
and Reform”, paper presented at the conference, “India: Ten Years of Economic
Reform”, at the William Davidson Institute, University of Michigan, September.
47
Rao, M. Govinda and Kavita Rao (2006), Trends and Issues in Tax Policy and Reform in
India, paper presented at Brookings-NCAER conference on India, March 2006.
Rao, M. Govinda, and Nirvikar Singh (2002), “Fiscal Transfers in a Developing Country:
The Case of India”, paper presented at University of Birmingham conference on
comparative federalism, January.
Rao, M. Govinda, and Nirvikar Singh (2005), Political Economy of Federalism in India,
Oxford University Press, New Delhi.
Reserve Bank of India (2001-5), Annual Reports, Mumbai.
Riker, William H. 1964. Federalism. Boston: Little, Brown.
Rodden, Jonathan (2006), “The Political Economy of Federalism,” in Barry Weingast and
Donald Wittman, eds., Oxford Handbook of Political Economy (forthcoming, Oxford
University Press).
Rodden, Jonathan and Susan Rose-Ackerman, 1997, Does Federalism Preserve Markets?,
Virginia Law Review, 83, 1521.
Rodrik, Dani (2003) “Introduction: What Do We Learn from Country Narrratives?” in

Dani Rodrik, ed., In Search of Prosperity Analytic Narratives on Economic Growth,
(Princeton, New Jersey: Princeton University Press)
Rodrik, Dani and Arvind Subramanian (2004a),‘From Hindu Growth to Productivity

Surge: The Mystery of the Indian Growth Transition’, National Bureau of Economic
Research Paper, No 10376.
Rodrik, Dani and Arvind Subramanian (2004b), Why India Can Grow at 7 Per Cent a
Year or More: Projections and Reflections, Economic and Political Weekly, April 17,
Rudolph, Lloyd I., and Susanne H. Rudolph, In Pursuit of Lakshmi: The Political
Economy of the Indian State (Chicago: University of Chicago Press, 1987);
Shleifer, Andrei “Establishing Property Rights,” in Michael Bruno and Boris Pleskovic,
eds., Proceedings of the World Bank Annual Conference on Development Economics
1994 (Washington, DC: World Bank, 1995), pp.93 –117
Shome, Parthasarathi (2006), At the Threshold of 10 Per Cent Economic Growth?,

Economic and Political Weekly, March 18.
Singh, Nirvikar (2006a), State Finances in India: A Case for Systemic Reform,
forthcoming in, ed., S. Narayan, Documenting Reform: Case Studies from India
McMillan India.
48
Singh, Nirvikar (2006b), Services-Led Industrialization in India: Prospects and

Challenges, Draft Report for UNDP Project on Industrial Development for the 21st
Century
Singh, Nirvikar, and T.N. Srinivasan (2005a), “Indian Federalism, Globalization and
Economic Reform,” in T.N. Srinivasan and Jessica Wallack, eds., Federalism and
Economic Reform: International Perspectives, Cambridge, UK: Cambridge University
Press
Singh, Nirvikar, and T.N. Srinivasan (2005b), Fiscal Policy in India: Lessons and
Priorities, in A Sustainable Fiscal Policy for India: An International Perspective, Peter
Heller and M. Govinda Rao, ed., Oxford University Press, 393-439
Sivaraman B. (1991), Bitter Sweet: Governance of India in Transition, New Delhi:

Ashish Publishing House.
Srinivasan, T.N., (2004), Comments on Dani Rodrik and Arvind Subramanian, “From
‘Hindu Growth’ to Productivity Surge: The Mystery of the Indian Growth Transition”
http://www.imf.org/external/pubs/ft/staffp/2004/00-00/sriniv.pdf
Tiebout, Charles. 1956.”A Pure Theory of Local Expenditures,” Journal of Political

Economy, October, 64, pp.416-24.
Twelfth Finance Commission (2004), Report of the Twelfth Finance Commission (2005-
10), November, http://fincomindia.nic.in/Report of 12th Finance Commission/index.html
Wade, Robert “Politics and Graft: Recruitment, Appointment, and Promotions to Public
Office in India,” in Peter M. Ward, ed., Corruption, Development and Inequality
(London: Routledge, 1989), pp.73-109
Wallack, Jessica (2003), Structural Breaks in Indian Macroeconomic Data, Economic and
Political Weekly, October 11th, 4312-4315
Wallack, Jessica and T.N. Srinivasan (2005), Analyzing Federalism: Stylized Models and
the Political Economy Reality, in T.N. Srinivasan and Jessica Wallack, eds., Federalism
and Economic Reform: International Perspectives, Cambridge, UK: Cambridge
University Press
Weingast, Barry (2006), Second Generation Fiscal Federalism: Implications for

Decentralized Democratic Governance and Economic Development, working paper,
Hoover Institution, Stanford University
Weingast, Barry, (1993), Constitutions as Governance Structures: The Political

Foundations of Secure Markets, Journal of Institutional and Theoretical Economics, 149,
233-61.
49
Weingast, Barry R. and William J. Marshall, (1988) “The Industrial Organization of

Congress,” Journal of Political Economy, Vol. 96, No. 1 (February), pp.132-63
Wheare, Kenneth C. (1953) Federal Government, 3rd edition, London: Oxford University
Press
World Bank (2003), India: Sustaining Reform, Reducing Poverty, New Delhi: World
Bank and Oxford University Press.
World Bank (2005), State Fiscal Reforms in India: Progress and Prospects (A World
Bank Report), New Delhi: Macmillan India.
50

Do Not Open

Uploaded by

Copyright:

Available Formats

Do Not Open

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Do Not Open

Uploaded by

Copyright:

Available Formats

Federalism and Economic Development in India:

May 25, 2006

Debates about governance include old concerns about corruption, affirmative

Underlying all the developments in economic policymaking, and concerns about

Legal and political practitioners were originally influential in shaping federalist

An alternative concept of federalism stresses the benefits of competition among

One of Weingast’s motivations for the concept of MPF is as an explanatory factor

In another example, local government strengthening was part of India’s policy

The rest of the paper is structured as follows. In section 2, we provide some

2. Background: India’s Economic Performance and Federal System

Table 1: Average Annual Growth Rates for India

GNP at NNP at NNP per capita at

Panagariya (2005) provides a very careful appraisal of India’s growth experience

Rodrik and Subramanian (2004b) use a growth accounting methodology to

If one has at least a weak consensus on the role of market-oriented reforms in

Centralization of India’s governmental structures was also reflected in

Besides the constitutionally mandated Finance Commission, an extra-

The recommendations of each Finance Commission apply to the five-year period

There is also a continued problem with coordinating transfers recommended by

The equalization concept seems eminently sensible if there is a social consensus

3. Quality of Governance and Government Expenditure

The framers of India’s Constitution opted for a relatively centralized, ‘quasi-

patronage, either directly or as intermediaries, and not as legislators.12 Major economic

traditional accountability mechanisms, relying as they do on hierarchical political and

Nevertheless, the central theme of subnational governments (with political and

An important aspect of governance that also has a federal dimension is regulation.

4. Efficiency of the Tax System

A further problem arose through an inconsistency in constitutional provisions.

A constant feature of China, in contrast to India, has been a strong, centralized,

5. Fiscal Situation of National and Sub-national Governments

Table 2: Central and State Fiscal Deficits (Percent of GDP)

Year Center States Consolidated

In addition to the stabilization or slight improvement in the center ands states’

Table 3: General Government Debt (percent of GDP)

Source: RBI (2005)

Table 4: Trends in Revenue and Expenditure Components, All States (% of GDP)

Period Own Tax Own Finance Non-Finance Revenue Interest Pensions

1993-96 5.27 1.55 2.94 1.62 12.00 1.86 0.63

2000-03 5.44 1.26 2.88 1.23 13.34 2.65 1.25

Change 0.17 -0.29 -0.05 -0.39 1.34 0.79 0.62

Source: Twelfth Finance Commission Report (Finance Commission, 2004)

Table 5: Comparative Fiscal Performance, Major States (% of GSDP)

Revenue Change in Fiscal Change in Debt, Fiscal

Table 6: Comparative Revenues and Expenditures, Major States (% of GSDP)

Own Tax Change in Own Revenue Change in

Source: Twelfth Finance Commission Report (Finance Commission, 2004)

Obviously, borrowing, in addition to its normal role of funding capital

The approach that seems to have worked partially is that of commitment to

Two conceptual points still need to be analyzed in considering the evolution of

evolution described in this section to our earlier outline of concepts of cooperative,

The common pool characterization of subnational fiscal incentives is easy to

The institutional evolution underlying the problem of subnational fiscal deficits

6. Impacts of Federal System on Growth and Equity

In comparison to this more settled literature on government-market boundaries,

As discussed in Section 2, the traditional public finance focuses mostly on static

Table 7: Criteria and Relative Weights for Tax Sharing

Weight (%) Weight (%)

Some of the impact of different components of the formula can be assessed by