Journal of Comparative Economics 33 (2005) 814–834
www.elsevier.com/locate/jce
Fiscal federalism in rentier regions:
Evidence from Russia ✩
Raj M. Desai a,∗ , Lev Freinkman b , Itzhak Goldberg b
a Edmund A. Walsh School of Foreign Service, Georgetown University,
37th & O Streets, NW, Washington, DC 20057, USA
b The World Bank, Washington, DC 20433, USA
Received 12 January 2004; revised 9 August 2005
Available online 10 October 2005
Desai, Raj M., Freinkman, Lev, and Goldberg, Itzhak—Fiscal federalism in rentier regions: Evidence
from Russia
When does sub-national fiscal autonomy prompt regional growth and recovery and, under what conditions, does it have adverse effects? We argue that unearned income streams, particularly in the form of
revenues from natural resource production or from budgetary transfers from the central government, transform regions dependent on these income sources into rentier regions. Governments in these regions can
use local control over revenues and expenditures to shelter certain firms, i.e., natural resource producers
or loss-making enterprises, from market forces. Using fiscal data from 80 Russian regions from 1996 to
1999, we test this hypothesis in both cross-sectional and panel specifications. Our results indicate that tax
retention, which is a proxy for fiscal autonomy, has had a positive effect on regional reform and investment
since the break-up of the Soviet Union. However, we also find that this effect decreases as rentable income
streams to regions increase. Journal of Comparative Economics 33 (4) (2005) 814–834. Edmund A. Walsh
School of Foreign Service, Georgetown University, 37th & O Streets, NW, Washington, DC 20057, USA;
The World Bank, Washington, DC 20433, USA.
 2005 Association for Comparative Economic Studies. Published by Elsevier Inc. All rights reserved.
JEL classification: H2; H7; P2
Keywords: Taxation; Intergovernmental relations; Socialist systems
✩
The findings, interpretations, and conclusions presented in this paper do not necessarily represent the views of the
World Bank, its Executive Directors, or the countries they represent.
* Corresponding author.
E-mail addresses: [email protected] (R.M. Desai), [email protected] (L. Freinkman),
[email protected] (I. Goldberg).
0147-5967/$ – see front matter  2005 Association for Comparative Economic Studies. Published by Elsevier Inc. All
rights reserved.
doi:10.1016/j.jce.2005.08.004
R.M. Desai et al. / Journal of Comparative Economics 33 (2005) 814–834
815
1. Introduction
The ability of rulers to tax local capital arbitrarily can limit significantly new investment and
economic growth. By contrast, stable and predictable tax policies have often presaged historically the spread of the rule of law and the rise of modern capital markets (North and Weingast,
1989). For theorists such as Madison (1941) and Montesquieu (1989), the separation of powers,
not merely between branches of government, but between levels of government provides a constitutional check on the capacity of central authorities to expropriate local wealth. As Madison
understood, governments at all levels are capable of expropriation; a fiscal division of powers
between central and sub-national government is necessary to constrain the powers of each. Thus,
a potential consequence of fiscal federalism is to afford sub-national governments incentives to
promote local economic growth.
Following the massive fiscal decentralization in Russia in the early 1990s, economic reform
and recovery has varied widely across sub-national regions. As the regional governments began
to exert greater authority over local resources, enterprises, and fiscal policy, the federal government’s ability to police the common market came under considerable strain. Although several
regions have liberalized their economies, enforced hard budget constraints, and protected private
property, others have actively resisted these reforms despite efforts in recent years to re-assert
federal authority.
The diverse Russian experiments in decentralization afford an opportunity to understand the
conditions under which federalism promotes local reform and growth. This paper identifies
some of the conditions under which federalism is, alternatively, market-preserving or marketsubverting in the Russian Federation. Our argument is an extension of the rent-seeking vs.
reform discussion found in models of self-interested governments by Dalmazzo and de Blasio (2001) and Esanov et al. (2001) applied to Russian regions. Our central premise is that
governments face a choice between appropriating rents and protecting investors. Because local firms’ tax and wage payments are the main sources of regional revenue and employment,
greater fiscal autonomy may encourage local governments to limit their own rent-extraction
from these firms. However, we posit that, where regional economies depend less on locallygenerated revenues and more on appropriable rents from resource extraction or from budgetary
transfers from the central government, the fiscal incentives to protect investors disappear rapidly
and regional governments are more likely to pursue these rent-seeking opportunities and use
firms as sources of political benefits. In sum, we hypothesize that those regions that retain a
greater share of locally-generated taxes will be more likely to reform and grow and that the
effect of fiscal autonomy on regional economic performance will be declining as appropriable
rents expand. Cross-regional and regional panel data from 80 Russian regions support these arguments.
The next section outlines our basic argument. Section 2 summarizes the evolution of intergovernmental fiscal relations in Russia. Section 3 examines the effects of rentierism on regional
reform and economic performance. Section 4 presents our findings from cross-sectional data.
Section 5 extends these results by correcting for possible endogeneity and by examining these
relationships in panel data. The conclusion draws policy implications for continuing reforms
of inter-governmental relationships in Russia and presents a brief comparison of Russian and
Chinese reforms.
816
R.M. Desai et al. / Journal of Comparative Economics 33 (2005) 814–834
2. Fiscal federalism in the Russian Federation
The devolution of power to the Russian regions following the collapse of the USSR was
primarily a response to a weakening central government that could no longer meet expenditure obligations (Lavrov, 1996). Under these conditions, regions lobbied for greater autonomy
through specific agreements with the center, while the federal government continued to limit
its own expenditure responsibilities. Partly due to this spontaneous decentralization in the early
1990s, the actual degree of fiscal autonomy varies across regions in the Russian Federation.
Through these agreements, regional governments have been afforded significant, but uneven,
control over local tax and non-tax revenue and over local spending.
In the second half of the 1990s, the degree of fiscal autonomy varied considerably. Moreover, fiscal autonomy was not associated closely with regional enterprise reform in that period.
Figure 1 shows the relationship between fiscal autonomy and the change in the average share of
enterprise subsidies in budgetary expenditures between 1996 and 1999. Decreases in subsidies
are positive, while increases are negative, to proxy for the hardening or softening of regional
budget constraints. In this diagram, we rank regions by their average tax retention ratio, from
lowest to highest, for the period 1996 to 1999. On this basis, we calculate each region’s percentile rank in the sample, which is shown along the horizontal axis so that the sample deciles
are easily visible. We then construct a ten-region rolling average of changes in subsidies so that
the first observation in Fig. 1 is the mean for regions 1 through 10 and the second is the mean of
regions 2 through 11, and so on. Finally, this rolling average is plotted against the region’s percentile rank in tax retention. Overall, no significant correlation between the change in subsidies
and the retention rate is observed.
We argue that fiscal autonomy should benefit regional economic performance. Fiscal autonomy creates powerful incentives for local governments to expand the local tax base by promoting
enterprise reform, by restraining their own rent-extraction from firms, and by policing against asset stripping by firm managers and owners. However, these incentives may be weakened by the
presence of two different types of appropriable rents, namely, income from natural resource extraction and budgetary transfers from the central government. Regions dependent on these types
of income have been transformed into rentier regions, in which the incentives to reform are
outweighed by the benefits of ensuring that firms become a permanent source of rents. The conditions under which large, usually formerly state-owned, enterprises have influenced laws and
economic policies in ways that reward these firms are discussed in Hellman et al. (2003). Firms
engage in this type of rent-seeking behavior in order to perpetuate the regimes that restrict new
entry into their markets, to preserve their opportunities for arbitrage between reformed and unreformed parts of the regional economy, and to protect themselves from regulatory interference by
the central government. In many cases, Russian regional governmental policies are characterized
by this form of state capture, in which policies and regulations are enacted to benefit influential
enterprises (World Bank, 2004).
Regional politicians are not pure captives of these firms, nor do they merely collect private
benefits in the form of campaign contributions in return for their provision of favorable policies.
On the contrary, they are typically active participants in a mutual relationship in which, through
law, regulation, or intimidation, firms may be forced to provide concrete political benefits to
regional governments in the form of employment and non-wage benefits for both employees and
the general public (Desai and Goldberg, 2001). In particular, the absence of a well-functioning
social safety net, either nationally or sub-nationally, has prompted regional governments to use
firms as surrogate sources of social services. Given the fiscal legacies of the Soviet state, Russian
817
Fig. 1. Fiscal autonomy and regional budget constraints.
R.M. Desai et al. / Journal of Comparative Economics 33 (2005) 814–834
818
R.M. Desai et al. / Journal of Comparative Economics 33 (2005) 814–834
regional governments have limited experience with independent public expenditure management.
In addition, post-Soviet Russian regions have inculcated close links between regional political
elites and managers of the largest regional companies.
3. Rentier experiences and fiscal autonomy
A rentier state is financed not from the domestic tax base but from externally generated revenues or rents. If income from these rents constitutes a significant portion of total government
revenues, the rentiers wield considerable political influence. Public officials will maintain themselves by appropriating these rents for public expenditures, as well as for private use, and have
little incentive to expand the local tax base by reforming the local investment climate for two
reasons. First, a high degree of unearned income is likely to prompt more intense rent-seeking
behavior among politicians, whose incentives are influenced by these potential income streams.
Empirical assessments of this voracity thesis, e.g. by Tornell and Lane (1999), show that sudden terms-of-trade improvements, the new discovery of natural resources, or a large stream of
foreign aid can prompt a gold rush in which competition over newly available rents creates a
diversion from more productive activities in the economy. Second, revenues from rents can lead
to the detachment of public institutions from their tax base. As the income streams from rents
become the primary source of public revenue, public institutions are less likely to rely on sound,
credibly-constrained, tax policies and more likely to become predatory in nature, relying on the
expropriation of these income streams (Levi, 1988). Rather than being a boon to the state, relief from the need to develop a sufficient local tax base leads to unaccountable, inefficient, and
uninformed bureaucracies.
The economic and political consequences of rentierism are observed most commonly in the
natural resource curse. However, similar effects are found in countries that derive significant
income in the form of remittances from foreign workers or from foreign aid flows. Sachs and
Warner (2001) show that resource-abundant countries have historically had lower growth rates
than economies with fewer resources. Leite and Weidmann (2002) and Ross (2001) demonstrate
that these countries tend to be governed by more corrupt, less democratic polities. States relying on either natural resource income or the remittance income of foreign workers also tend to
have poorer-functioning fiscal institutions, as Chaudhry (1997) discusses. If the sources of public revenue are few and concentrated, especially strong incentives for collusion arise between the
state and those who control income sources in order to maintain the status quo, as Moore (1998)
argues. Moreover, since natural resource endowments represent specific assets, rents from these
assets are more readily captured by local elites than would be the case for more mobile assets
(Collier and Hoeffler, 2002). Finally, recent assessments show that foreign aid flows create similar adverse incentives for public officials in aid-dependent countries.1
Two major rentable sources of income, namely, revenues from the production and export of
natural resources and revenues from budgetary transfers from the central government, are im-
1 Excessive dependence on foreign aid makes national governments less financially dependent on their own taxpayers.
Thus, such dependence may generate similar incentive distortions as do natural resource rents. Respective governments
may find it more efficient to spend more resources on extracting additional aid from the international community than on
developing their own tax base, as Moore (1998) argues. Similarly, cross-national evidence suggests that foreign aid can
erode the quality of governmental institutions in heavy borrowers, as Knack (2001) shows.
R.M. Desai et al. / Journal of Comparative Economics 33 (2005) 814–834
819
portant for the Russian regions.2 At the sub-national level, potential resource rents are largest in
the oil, gas, and mineral-rich regions, where both private and state-owned resource companies
provide significant private benefits to politicians and significant non-wage benefits to the local
population. Indeed, the Russian government is believed to under-tax the country’s major energy
and fuel companies.3 Consequently, a relatively small fraction of natural resource rents are captured by the public sector in the Russian Federation. Most of these rents are appropriated by the
private interests that control the natural-resource firms, as Popov (2001) and Esanov et al. (2001)
describe.
On the other hand, transfers are primarily a reflection of the central government’s efforts
to reduce cross-regional income differences through redistribution to the poorer regions. Under Russian inter-governmental revenue sharing rules, shortfalls in regional expenditures are
compensated for by transfers from the central government. Although formula-based in principle, these arrangements have been applied with greater variability across regions. In practice
since constitutional elaboration of the revenue-sharing mechanism is minimal, the Russian system lacks a stable set of predictable rules for tax sharing, transfers, and expenditure assignments.
Instead of being based on ex ante specified rules, inter-governmental finance in the Russian Federation is based primarily on the bargaining capacities of the individual regions (Freinkman et
al., 1999). During the 1990s, fiscal arrangements between regions and the center changed annually. Through a combination of extortion, bribing, and lobbying, certain regions managed to
supplement their resources with additional funds from the federal government. Particularly culpable were the autonomous republics and okrugs, the regions that contain Russia’s ethnic and
linguistic minorities and whose governors often used the implicit threat of local Chechnya-style
secessionist movements to extort greater sums from Moscow, as Treisman (1999) documents. At
least until 2001, a large portion of the total flow of transfers was determined by regional influence
peddling, according to Sinelnikov et al. (2001).
Transfer-dependent regions commonly have considerable social liabilities and a depressed
enterprise sector. The largest ultimate beneficiaries of federal transfers in these regions are the
large loss-making firms, which are typically voucher-privatized or partially state-owned firms
that require some form of budget subsidy for survival. Such enterprises remain the largest local
employers; they are run by manager-owners who, with the support of regional governments, have
shielded their firms from competition. Thus, the collusion between administrators in poorer regions and managers of bankrupt enterprises is funded mainly from the federal budget through the
inter-governmental transfer system. In other words, local coalitions in support of the status quo
are strengthened by the acquisition and distribution of federal transfers among local constituents.
Research on the economic impact of regional fiscal policies in Russia is divided on the subject
of fiscal federalism. On the one hand, several empirical analyses find that sub-national autonomy
improves the economic performance of regions. Berkowitz and DeJong (2003) conclude that
decentralization, combined with reformist preferences among regional governments, spurs new
enterprise formation and economic growth. Ahrend (2002) suggests that dependence on federal
transfers distorts the incentives of regional governors; those who have a comparative advantage
2 Similar rents were available to local apparatchiks in the USSR when large highly profitable firms, e.g. metallurgical
plants, were established in regions through a system of investment grants funded by the national budget. When such
firms became a dominant source of regional budget revenue, their role was not much different from the major oil and gas
extracting companies in the resource-abundant regions today.
3 In 2000, the Russian Federation generated approximately one half of the tax revenues per ton of extracted oil of that
of OECD oil producers, i.e., Denmark, Norway, and UK, according to Speck and Matusevich (2002).
820
R.M. Desai et al. / Journal of Comparative Economics 33 (2005) 814–834
in getting transfers devote less effort to protecting their local tax base. Transfers lead to less
growth, but this effect disappears when geographical and political factors are added. At the same
time, Kuznetsova (2001) finds little evidence of a robust link between the incidence of subsidies
in regional budgets and regional growth. A recent survey of 200 firms in 20 regions by the
Center for Economic and Financial Research (CEFIR, 2003) shows that local inter-governmental
fiscal relations within regions, in particular, the extent to which additional local revenues are not
taxed by the regional-level budgets, reduces the local regulatory costs of business. Conversely,
Zhuravskaya (2000) argues that limited fiscal autonomy encourages governments to over-tax and
over-regulate local firms.
On the other hand, several analyses find that fiscal autonomy enables sub-national governments to perpetuate soft-budget constraints for local enterprises. Litwack (2002) argues that
imperfect central control of sub-national budgets may drive regional governments towards greater
rent-seeking. Similarly, in a survey of over 800 public officials in the Russian regions, StonerWeiss (2005) finds that regional governments have acquired significant autonomy from the
center, often granted to dampen secessionist tendencies, that limits the capacity of the federal
state to govern the provinces. However, politicians and enterprise insiders in several of these
regions have used their independence from Moscow to stall, and even reverse, reforms and to
maintain local firms as sources of rents for both politicians and managers.
On average, about one-third of total regional and municipal expenditures in the middle of the
1990s were linked to the financing of various enterprise subsidies, according to Freinkman and
Haney (1997). The main recipients of these benefits were concentrated in the housing, utility,
transport, and agricultural sectors; they were typically among the largest region-specific companies. Consequently, sub-national governments have actively obstructed enterprise restructuring
by the perpetuation of soft budget constraints and, in some cases, through formal limitations on
particular enterprise activities, as McKinsey & Co. (1999) documents. Regional subsidies may
take the form of preferential tax treatment, discounts on utility bills, and favored status in public
procurement. All of these are intended to prevent companies from shutting down and laying off
employees and to place potentially productive companies at a cost disadvantage.
Similarly, Lavrov et al. (2001) claim that complicated informal agreements between regional
administrations and the largest local enterprises have been the general rule during the 1990s.
Such enterprises are subject to implicit taxation in the form of mandatory financing of local
public services, e.g., health, education, housing, and road maintenance. In return, these firms receive various privileges, such as explicit and implicit tax exemptions, debt relief, and protection
against bankruptcy and competition. Finally, Radayev (1996) describes another form of implicit
sub-national taxation, namely, volunteer contributions to various extra-budgetary funds (EBFs),
which are established through the sponsorship of sub-national governments. Local small and
medium enterprises face a higher risk of being forced to make payment to these funds. While
EBFs are technically illegal, governments have been using indirect forms of control over the
funds for several years. Finally, wealthier regions spend relatively more on subsidies to local
enterprises and richer regions have pursued less efficient policies, notably higher tax burdens,
according to Kuznetsova (2001). Nevertheless, little empirical analysis confirms the impact of
policy-induced distortions on the incentives of Russian sub-national governments. With the exception of Zhuravskaya (2000) and a report by the Center for Economic and Financial Research
(CEFIR, 2003), both of which examine revenue-sharing schemes between local and regional
governments, most research consists of case studies of particular regions. What has not been
examined systematically is the impact of fiscal decentralization on private sector development
and growth across regions.
R.M. Desai et al. / Journal of Comparative Economics 33 (2005) 814–834
821
4. Data and basic results
Our sample covers 80 Russian regions between 1996 and 1999.4 The fiscal data, including
those on retention rates and inter-governmental transfers, were provided by the Moscow Center
for Fiscal Policy, which generated these indicators based on the original database of the Russian
Ministry of Finance. All other regional indicators are obtained from the Russian Federation’s
State Committee for Statistics (Roskomstat).
To investigate cross-regional variation in regional economic performance based on regional
data, we use the following form for our regressions:
y = β0 + β1 R + β2 Q + β3 (R × Q) + β4 X + ε,
(1)
where y is an indicator of regional performance, R is the tax retention rate, Q is a measure of
rents from non-tax revenue, X is a vector of control variables, and ε is a random disturbance.
Identifying a single informative and regionally-comparable indicator of reform progress is problematic. For example, the rate of regional economic growth reflects some short-term effects of
policy reform. However, in Russia a major portion of regional growth has been driven by a small
number of the largest firms in the energy and natural resource sectors. This growth does not
depend much on sub-national policies but rather on oil and mineral prices, as well as on the
privatization of these firms. The changes in enterprise subsidies provided by regional governments are a useful proxy for the nature of budget constraints on local firms, and more broadly,
for the degree of regional policy distortions. However, during the 1990s, the main source of soft
budget constraints in Russia was not subsidies from regional governments, but rather the federal
government’s tolerance of tax and utility-payment arrears, as Pinto et al. (2000) demonstrate.
Enterprise privatization is considered a core component of the transition, but even this indicator is flawed. Privatization through vouchers, which was the instrument used for the majority
of medium and large Russian firms before 1996, may be a useful proxy for regional reform despite some evidence of poorly-functioning corporate governance arrangements for these firms
in Black et al. (2000). However, privatization using a range of other methods, in particular, the
non-competitive tenders and trade sales employed for the largest firms, may not reflect reformist
tendencies at the regional level. Finally, other outcome variables, such as levels of foreign direct
investment (FDI) or small-medium enterprise (SME) penetration, have their own limitations,
given the peculiarity of the Russian industrial structure. In Russia, FDI inflows have been concentrated in the energy sector and the linkage between FDI and regional economic policies is
unclear. Similarly, while claims are made that SMEs have been an engine of growth in transition
economies, Slinko et al. (2005) show that SME development and regional policy reform are not
closely related in Russia.
With these caveats in mind, we choose four indices of regional performance. First, we consider
the cumulative change in gross regional product (GRP) between 1994 and 1999 deflated by
the regional price deflator. Second, we take the difference in the average share of enterprise
subsidies in regional budgetary expenditures between 1996 and 1999 with decreases in subsidies
being positive and increases negative. Third, we use the number of privatized companies between
1996 and 1999 as a percentage of registered companies in 1996. Finally, we consider the log of
cumulative foreign direct investment per capita from 1996 to 1999. Cross correlations for these
4 The sample excludes the Chechen Republic and eight autonomous okrugs, which are the smallest Russian regions
having limited statistical coverage.
822
R.M. Desai et al. / Journal of Comparative Economics 33 (2005) 814–834
Table 1
Cross-correlations of dependent variables
Regional recovery
Decline in subsidies
Private companies
Foreign direct investment
Regional recovery
Decline in subsidies
Private companies
Foreign direct investment
0.0209
0.23650.027
0.05880.624
0.29010.012
0.0771
0.39030.001
0.24260.037
0.0101
0.31550.008
1.6536
Notes. (i) The p-values for correlation coefficients are superscripted. (ii) Variances are listed along diagonal.
dependent variables are reported in Table 1. The relatively low correlations suggest that each
variable accounts for some particular aspect of overall regional performance and reform in the
Russian Federation.
We examine the effects of two different sources of rents, namely, natural resource production
and transfers from the federal budget in separate estimations. The former, denoted Resources,
is proxied by the percentage of regional output from fuel and energy production, according to
the Roskomstat sectoral classification. The latter, denoted Transfers, is the share of regional budget revenues from federal transfers; it includes all main channels through which federal funds
were transferred to regions, namely, equalization transfers and smaller federal programs such
as budget subventions.5 For the whole sample, the non-weighted average share of the energy
sector in industrial output is about 20%, but, for 31 regions, this share exceeds one third. Similarly, although the average Russian region received 16% of its budgetary revenues from federal
transfers, this share was above one third for 18 regions. The control variables are the log of the
regional population, denoted Population, to control for size, the regional export-to-GRP ratio,
denoted Exports, and the log of per-capita GRP in 1994, denoted Income. We also include an
administrative-status dummy variable, denoted Autonomous, which is coded 1 if the region is an
autonomous republic or okrug and 0 otherwise. The retention rate, transfers, and export-to-GRP
are expressed as averages for the period from 1996 to 1999.6
Our variable for fiscal incentives is the retention rate of tax revenues, denoted Retention,
which measures the share of locally generated revenues kept with the regional budget. However,
this variable reflects only official taxes, which are collected and accounted for in regular government budgets. Because a high incidence of informal regional taxation remains off-budget, our
retention indicator underestimates the degree of regional control over the local tax base. Since
we are interested in the impact of inter-governmental fiscal arrangements on regional incentives
to support growth, revenues hidden from the federal government outside of the regular regional
budget are likely to have limited impact on growth incentives because of their unstable and temporary nature. Large EBFs could help regions to solve particular current budget or economic
problems but any incentive effect is likely to be negative. Regions with large EBFs would be
more inclined to invest their resources in diverting tax revenues from the formal to informal
system of public finance and be less concerned about expansion of the local tax base. Regional
averages for retention rates, resource wealth, and transfers, are presented in Appendix Table 1.
5 The regional budget variables relate to the consolidated regional budget; this figure includes both the budget of
regional governments and budgets of municipalities located within the region.
6 We also considered a range of other control variables reflecting different demographic, social, and industrial characteristics of the regions, such as age structure, the share of urban population, R & D employment, and the availability of
housing. However, all such variables were statistically insignificant.
R.M. Desai et al. / Journal of Comparative Economics 33 (2005) 814–834
823
Table 2
Natural resource wealth and regional performance: basic cross-regional results
Dependent
variable
Regional
recovery
(1)
Regional
recovery
(2)
Decline in
subsidies
(3)
Privatized
companies
(4)
Foreign direct
investment
(5)
Population
0.0308
(0.0220)
0.6693***
(0.2053)
0.0799
(0.1079)
0.0294
(0.0230)
−0.0617
(0.0427)
0.0614
(0.0462)
0.0384*
(0.0224)
1.1788**
(0.4608)
0.9215*
(0.5364)
−1.2886**
(0.6017)
0.0351
(0.0236)
−0.0647
(0.0423)
0.0635
(0.0453)
0.0007
(0.0105)
0.3212**
(0.1484)
0.3444*
(0.1876)
−0.4324*
(0.2066)
0.0048
(0.0063)
−0.0094
(0.0149)
0.0442**
(0.0218)
0.0545**
(0.0205)
0.4639*
(0.2689)
2.9696***
(0.9014)
−4.4603***
(1.3492)
−0.0186
(0.0161)
−0.0521
(0.0319)
0.0969***
(0.0352)
0.1264
(0.2020)
3.0924*
(1.6338)
19.7100*
(11.6496)
−28.7285
(17.6684)
0.2819
(0.1852)
0.1775
(0.4061)
1.3153***
(0.3953)
77
0.1988
0.1432
4.58
(0.0006)
77
0.2121
0.1431
5.58
(0.0000)
74
0.2095
0.0505
3.06
(0.0075)
68
0.3390
0.0859
6.84
(0.0000)
70
0.1957
1.2416
4.07
(0.0010)
Retention
Resources
Retention × Resources
Exports
Autonomous
Income
Observations
R-squared
Root MSE
F -test
(prob.)
Notes. (i) Robust standard errors are presented in brackets. (ii) The coefficients on the intercepts are not reported.
* Statistical significance at the 10%.
** Idem., 5%.
*** Idem., 1%.
Table 2 contains the results of ordinary least squares (OLS) estimations. Column one presents
a benchmark regression equation without interactive terms. The retention rate coefficient has the
expected positive sign, but the effect of regional natural resource wealth is not significant. In the
second specification, we add the interactive term. The sign on the retention coefficient remains
positive and significant and the resources coefficient now has a positive and significant sign that
suggests a positive relationship between natural resource wealth and regional growth. However,
the negative sign on the interactive term indicates that the effect of fiscal autonomy is decreasing in natural resource wealth, as hypothesized. Ceteris paribus, an increase of one standard
deviation in the tax retention rate, which is the difference between the Novosibirsk Oblast and
the autonomous Chuvashia Republic, boosts the regional recovery rate by 10 percentage points.
Similarly, increasing resource wealth by one standard deviation with no other changes, which is
equivalent to the difference between the Ryazan Oblast south-east of Moscow and the resourcerich region of Tyumen, increases cumulative economic recovery by over 20 percentage points.
However, the combined effect of these increases in both retention rates and resource wealth leads
to a 25 percentage points decrease in the recovery rate.7
7 The percentage increases or decreases in these performance indicators are calculated as follows. Using the basic
interactive regression (Eq. (1)), we calculate a predicted value by setting all variables to their sample means initially.
Then, we increase only the retention rate by one standard deviation and recalculate the predicted value. Resetting the
retention rate to its mean, we recalculate the predicted value this time increasing only the resource or transfer variables
824
R.M. Desai et al. / Journal of Comparative Economics 33 (2005) 814–834
In addition, column (2) of Table 2 indicates that larger regions are also associated with greater
recovery rates. Subsequent specifications in Table 2 consider alternative regional reform measures. Greater tax retention and greater resource wealth are associated with greater declines
in producer subsidies, more privatization, and more foreign direct investment.8 However, the
marginal effect of retention on hardening budget constraints and on privatization declines with
greater regional resource wealth. Combining one standard-deviation increases in retention rates
and resource wealth increases the portion of regional budgets devoted to enterprise subsidies by
8 percentage points and decreases the percent of privatized firms by over 30 percent.
Rents from the extraction of natural resources and budgetary transfers from the central government serve similar fiscal functions is that they allow sub-national authorities to raise revenue
without increasing their reliance on the tax base or improving tax collection. In Table 3, we
replace the natural-resource wealth indicator with a measure of transfers. The direct effect of
retention rates on regional recovery and hard budget constraints remains positive. The effect of
transfers is also positive when the interactive term is included, although it is not significant for
recovery. Once again, the interactive term is strongly significant and negative for these two mea-
Table 3
Budgetary transfers and regional performance: basic cross-regional results
Dependent
variable
Regional
recovery
(1)
Regional
recovery
(2)
Decline in
subsidies
(3)
Privatized
companies
(4)
Foreign direct
investment
(5)
Population
−0.0474**
(0.0197)
0.2985**
(0.1454)
−0.8976***
(0.1299)
−0.0173
(0.0179)
0.0045
(0.0352)
0.0386
(0.0385)
-0.0420**
(0.0195)
0.6903**
(0.2873)
0.0751
(0.5749)
−1.3900*
(0.7792)
−0.0063
(0.0215)
−0.0088
(0.0362)
0.0343
(0.0389)
−0.0085
(0.0085)
0.3241***
(0.1103)
0.4417**
(0.2188)
−0.7961***
(0.2993)
0.0027
(0.0055)
−0.0047
(0.0154)
0.0438*
(0.0232)
0.0096
(0.0236)
−0.2986
(0.2297)
−0.5913
(1.3448)
0.6956
(1.9022)
−0.0343
(0.0211)
−0.0386
(0.0342)
0.0888**
(0.0370)
−0.1138
(0.2453)
−0.5396
(2.4756)
−5.9580
(4.5767)
6.0490
(6.5459)
0.1797
(0.1987)
0.1388
(0.4281)
0.0002***
(0.0001)
78
0.4177
0.1217
17.62
(0.0000)
78
0.4336
0.1209
15.98
(0.0000)
75
0.2481
0.0490
5.41
(0.0001)
72
0.2048
0.0965
3.13
(0.0066)
74
0.1696
1.2449
3.76
(0.0017)
Retention
Transfers
Retention × Transfers
Exports
Autonomous
Income
Observations
R-squared
Root MSE
F -test
(prob.)
Notes. (i) Robust standard errors are presented in brackets. (ii) The coefficients on the intercepts are not reported.
* Statistical significance at the 10%.
** Idem., 5%.
*** Idem., 1%.
by one standard deviation. Then, we estimate the result when both the retention rate and the appropriate rent variable are
increased simultaneously by one standard deviation.
8 In subsequent estimations, we also include the log of average SME output per capita from 1997 to 1999 as a proxy
for SME penetration. These specifications do not alter our basic results so that they are not reported.
R.M. Desai et al. / Journal of Comparative Economics 33 (2005) 814–834
825
sures. An increase in transfers from that of the relatively independent Samara province to that
of the more transfer-dependent Tuva Republic increases economic recovery by 12 percentage
points. However, the combined effect of increasing both transfer-dependency and tax retention
by one standard deviation lowers output recovery by 19 percentage points and increases enterprise subsidies by over 11 percentage points.
5. Robustness and extensions
Using transfers as a regressor leads to a potential endogeneity problem. In the Russian system of revenue sharing, transfers are designed to alleviate longer-term income differences. In
addition, reforms in the system of inter-governmental finance failed to eliminate completely bargaining over transfers by central and regional authorities. For example, autonomous republics
with ethnic minorities exact greater transfers than oblasts, and the central government often uses
transfer funds to co-opt wayward regional governors, according to Treisman (1996). Moreover,
Sinelnikov et al. (2001) show that the transfer allocation in Russia contains a statistically significant bias that benefits regions located in the far northern parts of the country. Hence, the ratio
of total transfers received to the overall current revenue is not likely to be strictly exogenous. To
correct for this potential source of bias we re-estimate Eq. (1) using a two-stage least squares
regression in which we instrument the transfers variable.
The results in Table 4 focus on estimations that yield significant results in Table 3, i.e., regional
recovery and decline in subsidies. In the first two columns, we re-estimate columns (2) and (3) of
Table 3 using lagged values of transfers, specifically, average transfers during 1994 to 1995 for
each region as an instrument. Retention retains its positive effect and the interactive term retains
its negative sign, although only the first of these is significant for regional recovery. In the second
column improvements in hard budget constraints are aided by greater local tax retention, but
the combined effect of greater retention with greater federal transfers to regions actually softens
budget constraints due to the interactive term.
In columns (3) and (4) in Table 4 we drop two variables with insignificant coefficients, i.e., per
capita regional income and the administrative status dummy variables. Moreover, these may be
useful instruments for transfers. In Russian inter-governmental finance, federal transfers were designed to support depressed regions but were often obtained due to the greater bargaining power
of autonomous regions. The transfer formula also incorporated the extent to which the enterprise
sector was already subsidized, as Freinkman and Haney (1997) pointed out.9 In columns (3)
and (4), we use as instruments those variables from which actual transfers to regions are derived, namely, initial income, the level of subsidies, and the status as an autonomous region.
Consequently, we include per capita income in 1994, budgetary subsidies in 1996, population,
and the administrative-status dummy in the second equation to examine their possible influences
on budgetary transfers, either through a formula-driven revenue-sharing scheme or through the
leverage of poorer, high-subsidy, autonomous regions in bargaining with federal authorities over
the size of annual transfers. When we allow for the possibility of endogenous transfers, the basic
results reported in Table 3 hold. Tests of over-identification do not indicate correlation between
the instruments and the error term; hence, the instruments are valid.
9 The evolution of the transfer system from one based entirely on ad hoc negotiation to one based on a more or less
explicit formula may also make the patterns of transfer allocation in only 1994 to 1995 a poor instrument for transfer
distributions in the later period.
826
R.M. Desai et al. / Journal of Comparative Economics 33 (2005) 814–834
Table 4
Budgetary transfers and regional performance: two-stage least squares
Dependent
variable
Regional
recovery
(1)
Decline in
subsidies
(2)
Regional
recovery
(3)
Decline in
subsidies
(4)
Population
−0.0503***
(0.0183)
0.6179**
(0.2682)
−0.1981
(0.5070)
−0.9779
(0.6651)
−0.0068
(0.0207)
0.0323
(0.0375)
−0.0002
(0.0339)
−0.0104
(0.0085)
0.3070***
(0.1050)
0.3791*
(0.2116)
−0.7011**
(0.2902)
0.0025
(0.0051)
0.0438
(0.0217)
−0.0026
(0.0145)
−0.0517***
(0.0184)
0.6529**
(0.2607)
−0.1675
(0.4703)
−1.0580*
(0.6236)
−0.0067
(0.0218)
−0.0102
(0.0083)
0.3001***
(0.1096)
0.3628*
(0.2124)
−0.6823**
(0.3010)
0.0022
(0.0050)
77
0.4301
0.110
16.68
(0.0000)
0.610
(0.4349)
74
0.2473
0.046
5.31
(0.0001)
2.569
(0.1090)
77
0.4235
0.110
21.11
(0.0000)
1.104
(0.8936)
74
0.2470
0.046
6.37
(0.0000)
3.981
(0.2635)
Retention
Transfers
Retention × Transfers
Exports
Autonomous
Income
Observations
R-squared
Root MSE
F -test
(prob.)
Sargan–Hansen
(prob.)
Notes. (i) The coefficients are obtained from a two-stage least squares estimation using a covariance matrix of the equation
errors adjusted for small samples. (ii) The instrument in columns (1) and (2) is average federal transfers to regions during
1994 to 1995. The instruments in columns (3) and (4) include: lagged transfers, per-capita regional income, administrative
status, and the share of enterprise subsidies in regional budgets. (iii) Robust standard errors are presented in brackets.
(iv) The coefficients on intercepts are not reported. (v) The Sargan–Hansen test is a test of over-identifying restrictions.
* Statistical significance at the 10%.
** Idem., 5%.
*** Idem., 1%.
All of our central explanatory variables, i.e., retention rates, resource income, and budgetary
transfers, are highly time-variant. In Fig. 2, we rank all region-year observations according to
private investment-to-GRP and rescale all observations according to their percentile rank in the
sample. We construct two 40-observation rolling means, one for resource wealth, and one for
budgetary transfers. As Fig. 2 indicates, resource wealth and transfers appear to increase among
the top quartile of regions ranked by private investment to GRP, but the effect overall is more
ambiguous. As a test of the robustness for specifications with time-variant variables, we modify
Eq. (1) to estimate regional investment for cross-regional time-series data covering 80 regions
between 1996 and 1999. We use the following autoregressive equation:
Iit = β0 + β1 Rit + β2 Q + β3 (R × Q)it + β4 Xit + β5 Iit−1 + εit ,
(2)
where I is the ratio of private investment to GRP and all other variables are defined as in Eq. (1).
We include a lagged dependent variable to control for serial correlation of the errors, and estimate Eq. (2) using two different methods. First, using OLS with two-way fixed effects, we
model the error term as εit = ηi + µt + νit , where η and µ are time- and region-invariant effects, respectively, and ν is a random disturbance. Along with the regional dummies in these
R.M. Desai et al. / Journal of Comparative Economics 33 (2005) 814–834
827
Fig. 2. Resource wealth, budgetary transfers, and private investment.
specifications, we include year dummies to control for unobserved time effects. These are particularly important in 1998 and 1999 following the ruble devaluation and the resulting fiscal and
monetary adjustments. Second, we estimate (2) using pooled OLS, but correct the error terms to
reflect contemporaneous correlation across panel cross sections, as suggested by Beck and Katz
(1995, 1996).
In these specifications, we include three separate dummies for administrative status to control
for any unexplained differences in investment rates across autonomous republics, autonomous
okrugs, and the two largest cities, i.e., Moscow and St. Petersburg. In all regressions, we include
as a proxy for human capital endowment the number of university-educated students per 10,000
persons, denoted University. To account for the effect of subsidies on firm-level investment,
we also include enterprise subsidies as a percent of budgetary expenditure, denoted Subsidies.
Finally, we include a time trend to control for unexplained trends in investment between 1996
and 1999. Table 5 reports the coefficients for estimations using resource wealth as a measure of
rents. These results corroborate those in Table 2. Higher tax retention rates and greater resource
income increase investment, but the combined effect remains negative. Finally, we apply the
instrumental variables approach used in Table 4 to these data in order to test whether budgetary
transfers substitute for the rentier effect of resource wealth. As in Table 4, we control for possible
endogeneity of transfers by instrumenting this variable using lagged values of transfers. These
results exhibit remarkable similarity to some of the cross-regional results in Table 4. As expected,
828
R.M. Desai et al. / Journal of Comparative Economics 33 (2005) 814–834
Table 5
Natural resource wealth and regional investment: panel results
Population
Retention
Resources
Two-way fixed
effects
(1)
Two-way fixed
effects
(2)
Panel-correct std.
errors
(3)
Two-stage least
squares
(4)
−1.4168***
(0.3035)
0.6011**
(0.2618)
0.7149*
(0.3850)
−1.5520***
(0.2768)
0.5856**
(0.2407)
0.8032**
(0.3663)
0.0377***
(0.0063)
0.1628**
(0.0848)
1.0880***
(0.3091)
−1.9514***
(0.3164)
0.6254**
(0.2509)
1.7458***
(0.6115)
Transfers
Retention × Resources
−1.7419***
(0.5478)
−1.7499***
(0.5084)
−1.4783***
(0.4241)
Retention × Transfers
Subsidies
University
Exports
−0.0740
(0.0468)
−0.1107*
(0.0639)
−2.7181***
(0.8422)
−0.0563
(0.0479)
−0.1412***
(0.0159)
0.0422***
(0.0092)
−0.1495***
(0.0106)
0.0425***
(0.0091)
0.1117***
(0.0311)
0.0128
(0.0120)
0.0808***
(0.0294)
−0.1229***
(0.0178)
0.0804***
(0.0157)
−0.1432***
(0.0087)
0.0426***
(0.0096)
78
228
0.7365
43.79
(0.0000)
79
234
0.7313
57.55
(0.0000)
79
234
0.6218
101.70
(0.0000)
79
237
0.7157
3115.80
(0.0000)
−0.0898*
(0.0484)
−0.0945
(0.1108)
0.0262
(0.0167)
Republic
Okrug
City
Trend
Private investmentt−1
Regions
Observations
R-squared
Wald/F -test
(prob.)
Notes. (i) The dependent variable is private investment to GRP. (ii) Two-stage least squares uses lagged transfers as an
instrumental variable for transfers. (iii) The standard errors for columns (1) and (2) and the panel-corrected standard
errors for column (3) are in parentheses. (iv) The coefficients on country and time dummies and on the intercepts are not
reported.
* Statistical significance at the 10%.
** Idem., 5%.
*** Idem., 1%.
the coefficients on the retention rate and transfers variables are positive, but their combined effect
remains negative.
6. Conclusion
In this paper, we find statistical evidence to support the hypothesis that fiscal incentives of
the Russian regions are an important determinant of regional economic performance. Since the
R.M. Desai et al. / Journal of Comparative Economics 33 (2005) 814–834
829
mid-1990s, some degree of centralized control has been reasserted in the Russian Federation,
along with more uniform rules for revenue sharing and expenditure assignment. These reforms
have achieved mixed results. In general, re-centralization has improved the transparency of intergovernmental fiscal relations, but it has increased federal control over regionally generated tax
revenues. Our results indicate that an increase in the retention rate, which is the share of locally
generated taxes that are left with the regional budgets, is accompanied by stronger economic
growth, harder budget constraints, and greater private investment in most regions. We also find
evidence that greater retention is associated with more privatization and more foreign investment.
This confirms the view that tax sharing arrangements play a critical role in establishing incentives
for regional development and reform. Regions facing a higher retention rate tend to be more
interested in developing the local tax base and, consequently, they are more likely to improve the
business environment and to support new business entry and genuine enterprise restructuring. By
contrast, excessive fiscal centralization is detrimental to regional reform and investment.
Our results also indicate that the impact of expanding regional fiscal autonomy is not uniformly beneficial for the local economy. If a large portion of regional revenues are derived from
unearned income streams rather than local taxation, the effect of fiscal autonomy weakens. We
have identified two types of regions facing this risk of excessive competition for rents, namely,
regions with high shares of revenues derived from the fuel and energy sectors and regions that
are highly dependent on federal transfers. In both cases, the effect of fiscal autonomy on regional
growth declines as the share of non-tax revenues in the form of resource production or budgetary
transfers rises. Our results indicate that increasing tax retention rates has the strongest effect on
regional economic performance in regions that are relatively free of rent-seeking possibilities.
The diverging economic performances of China and Russia may be explained partly by the
incentives created by inter-governmental financial policies. In contrast to the Russian experience,
provincial authorities have supported market reforms and have promoted local entrepreneurship
in China. According to Montinola et al. (1994) and Jin et al. (in press), China’s remarkable rate
of economic growth over the past twenty five years is due to Chinese-style fiscal federalism that
gives strong fiscal incentives to the sub-national governments to develop a sustainable source of
local tax revenues by supporting new private businesses and enterprise restructuring. As a result,
local firms in China avoid excessive administrative interference because sub-national governments shield firms from predation and over-regulation in order to protect their local tax base.
In contrast, Russian inter-governmental financial policies have failed to provide local governments with any effective degree of control over revenues. Thus, their incentives to improve tax
collections or expand the local tax base have been weak. Therefore, limited local control over
local finances has impeded the growth of new firms that have fallen prey to predatory taxation,
regulation, or corruption.
According to Blanchard and Shleifer (2001), Chinese economic reforms have occurred against
the backdrop of political centralization, whereas reform in Russia proceeded alongside the chaos
of political decentralization and democratization. In China, unlike the Russian Federation, the
party-state nexus remains intact and its ability to reward and punish sub-national actors is
undiminished. More importantly, the capacity of the central government to collect accurate information from provincial governments remains high, as Huang (1994) documents. By contrast,
inter-governmental relationships in Russia are characterized by overlapping divisions of authority, instability, and opaqueness because both regional and central politicians have struggled to fill
the void brought about by the collapse of the Soviet state. Our findings are consistent with both
explanations of the success in Chinese-style fiscal federalism, i.e., local autonomy combined
with political centralization. As Chinese provinces have benefited from higher and more stable
830
R.M. Desai et al. / Journal of Comparative Economics 33 (2005) 814–834
tax retention, the central government has preserved administrative control over these provinces
by strictly limiting opportunities for rent seeking.
Our results identify certain limits of fiscal autonomy in resource-rich or transfer-dependent
regions. The main policy implication is that fiscal autonomy may be insufficient to transform
regions in federal Russia into truly market-preserving entities without improvements in revenueside management in the regions. Our results may support an argument in favor of an asymmetrical
federalism across regions based on differing economic conditions.10 Although additional fiscal
autonomy is justifiable for most Russian regions, regions that are prone to rent-seeking require
special fiscal regimes. For transfer-dependent economically depressed regions, some restrictions
on fiscal autonomy are useful to create an opportunity for broader central government control
over their expenditure policy. This recommendation is consistent with proposals for external
fiscal management of fiscally depressed and bankrupt regions.11 In the case of resource-wealthy
regions, a much higher degree of centralization of natural-resource revenues may be required
at the federal level. In addition, these revenues should not necessarily be distributed to poorer
regions as budgetary transfers; rather they should be used to finance public infrastructure projects
in low-income regions through federal investment programs.
Acknowledgments
The authors are grateful to Galina Kurlyandskaya and Natalia Golovanova from the Center for
Fiscal Policy in Moscow, and to Boris Kopeikin from the Institute of Urban Economy for their
considerable help in collecting the fiscal data. Jorge Ugaz provided invaluable research assistance
and Zakia Nekaien-Nowrouz edited the final manuscript. Financial support was provided by the
Edmund A. Walsh School of Foreign Service and the World Bank’s research preparation grants
program, RPG No. RF-P074267-RESE. The authors thank Harley Balzer, Jackie Coolidge, John
Litwack, Alexander Morozov, Helga Muller, Anders Olofsgård, Alexander Plekhanov, Stepan
Titov, Dana Weist, Deborah Wetzel, Tarik Yousef, the participants of a World Bank seminar, and
two anonymous reviewers for their comments on earlier drafts.
Appendix Table 1
Basic indicators for Russian regions: averages from 1996 to 1999
Region
Admin.
status
Population
(thousands)
Adygeya
Agin Buryat
Altay
Amur
Arkhangel’sk
Astrakhan
Bashkortostan
Belgorod
Bryansk
Buryatia
2
3
1
1
1
1
2
1
1
2
450
79
2667
1019
1485
1027
4113
1488
1460
1043
Resource
production
6.1
20.5
48.3
23.7
60.8
48.4
10.5
10.4
46.9
Tax retention
rate
67.3
40.7
69.2
70.9
73.2
60.0
75.1
66.5
58.4
68.9
Budgetary
transfers
45.8
69.8
31.2
33.5
13.6
24.1
1.4
6.6
20.7
28.6
(continued on next page)
10 Lapidus (1999) and Bahry (2001) discuss the concept of asymmetry in Russian fiscal federal relations.
11 Yasin (2002) discusses various such proposals.
R.M. Desai et al. / Journal of Comparative Economics 33 (2005) 814–834
831
Appendix Table 1 (continued)
Region
Admin.
status
Population
(thousands)
Resource
production
Tax retention
rate
Chuvashia
Chelyabinsk
Chita
Chukotka
Dagestan
Evenkia
Gorny Altay
Ingushetia
Irkutsk
Ivanovo
Kabardino-Balkaria
Kaluga
Kaliningrad
Kamchatka
Karelia
Kalmykia
Karachay-Cherkessia
Kemerovo
Khabarovsk
Khakassia
Khanty-Mansi
Kirov
Komi-Permyak
Komi
Koryak
Kostroma
Krasnodar
Krasnoyarsk
Kursk
Kurgan
Leningrad
Lipetsk
Magadan
Mariy El
Moscow
Mordovia
Moscow City
Murmansk
Nenets
Nizhegorod
North Osetia
Novosibirsk
Novgorod
Omsk
Orenburg
Orel
Penza
Perm’
Primorsky
Pskov
Rostov
2
1
1
3
2
3
2
2
1
1
2
1
1
1
2
2
2
1
1
2
3
1
2
3
3
1
1
1
1
1
1
1
1
2
1
2
4
1
3
1
2
1
1
1
1
1
1
1
1
1
1
1359
3679
1273
79
2110
19
203
357
2768
1240
792
1090
945
393
774
317
436
3015
1539
583
1358
1607
1155
154
31
794
5071
3069
1331
1103
1679
1246
243
762
6550
941
8632
1008
47
3687
667
2748
736
2173
2228
905
1544
2982
2206
815
4392
16.3
13.4
45.6
44.5
43.0
67.9
74.7
71.5
79.7
69.1
75.5
38.9
23.6
67.7
67.1
70.0
59.9
63.9
73.7
79.1
33.5
63.0
69.8
64.4
78.2
55.6
65.0
65.8
63.3
76.4
69.0
62.2
73.7
65.5
71.2
63.3
75.0
76.9
63.1
51.3
64.3
45.0
71.7
69.8
56.7
65.8
62.3
70.7
63.6
62.0
65.1
64.4
60.6
68.9
65.9
62.8
82.2
25.7
22.4
21.0
14.9
27.5
21.3
10.7
71.5
14.2
42.6
31.3
25.6
15.5
8.3
35.8
22.2
12.8
26.8
22.2
37.4
10.9
24.5
21.1
11.9
13.1
16.6
18.0
10.8
15.1
19.2
11.8
34.3
51.0
14.2
18.9
31.0
24.7
19.9
30.5
Budgetary
transfers
15.6
6.9
23.2
40.7
52.5
46.4
45.7
51.1
5.6
28.5
44.9
26.1
9.2
31.1
14.4
45.9
38.6
14.0
16.9
11.6
1.2
18.2
6.7
45.4
36.5
29.0
16.1
2.2
12.6
23.1
6.2
1.5
26.8
27.2
5.7
31.6
3.1
21.2
21.7
4.9
43.9
8.9
24.6
11.4
16.9
18.3
23.7
4.1
24.0
35.0
17.6
(continued on next page)
832
R.M. Desai et al. / Journal of Comparative Economics 33 (2005) 814–834
Appendix Table 1 (continued)
Region
Admin.
status
Population
(thousands)
Resource
production
Tax retention
rate
Budgetary
transfers
Ryazan’
Sakha
Sakhalin
Samara
Saratov
Smolensk
Stavropol’
St. Petersburg
Sverdlovsk
Tambov
Tatarstan
Taymyr
Tomsk
Tula
Tuva
Tver’
Tyumen’
Udmurtia
Ul’yanovsk
Ust’-Ordunsky Buryat
Vladimir
Vologda
Volgograd
Voronezh
Yamalo-Nenets
Yaroslavl’
Yevreyskaya
1
1
2
1
1
1
1
4
1
1
2
3
1
1
2
1
1
2
1
3
1
1
1
1
3
1
2
1301
615
996
3306
2721
1151
2684
4738
4645
1286
3775
44
1072
1776
311
1625
3210
1636
1480
144
1625
1335
2694
2480
496
1430
204
41.8
48.0
22.1
16.7
33.1
23.0
34.7
10.3
13.5
18.6
45.6
27.0
24.3
56.8
64.8
73.5
52.7
62.1
65.7
61.7
57.4
66.5
64.0
82.9
79.4
64.1
64.7
74.9
67.4
71.0
59.4
63.0
75.6
62.8
74.5
61.9
62.1
62.5
56.2
75.4
16.8
25.1
22.0
1.8
16.5
13.4
22.1
2.3
2.7
20.9
2.1
19.2
11.0
17.0
49.2
17.2
7.4
16.2
12.4
50.1
18.5
6.6
10.9
15.9
0.7
9.7
42.8
1687
29.7
65.2
21.2
Sample Averages
38.7
14.6
36.8
32.5
93.2
31.1
16.0
15.3
8.5
29.4
21.4
Source: State Committee for Statistics and Ministry of Finance, Russian Federation.
Notes. (i) Administrative status codes are as follows: 1 is for oblasts; 2 for republics; 3 for okrugs, and 4 for cities. (ii) All
averages are for 1996 to 1999 and are expressed in percentages, except where noted. (iii) Resource production is the
share of fuel and energy production in regional output. (iv) Transfers are budgetary transfers as a percentage of regional
governmental revenue.
References
Ahrend, Rudiger, 2002. Speed of reforms, initial conditions, political orientation, or what? Explaining Russian regions’
economic performance. Working paper No. 2002-10. DELTA, Paris.
Bahry, Donna, 2001. Rethinking asymmetrical federalism. In: Khakimov, Rafael (Ed.), Federalism in Russia. Tatarstan
Academy of Sciences/Kazan Institute of Federalism, Kazan.
Beck, Nathaniel, Katz, Jonathan N., 1995. What to do (and not to do) with time-series cross section data. American
Political Science Review 89, 634–647.
Beck, Nathaniel, Katz, Jonathan N., 1996. Nuisance vs. substance: Specifying and estimating time-series/cross-section
models. Political Analysis 6, 1–36.
Berkowitz, Daniel, DeJong, David, 2003. Policy reform and growth in post-Soviet Russia. European Economic Review 47, 337–352.
Black, Bernard, Kraakman, Reinier, Tarasova, Anna, 2000. Russian privatization and corporate governance: What went
wrong? Stanford Law Review 52, 1731–1808.
Blanchard, Olivier, Shleifer, Andrei, 2001. Federalism with and without political centralization: China versus Russia.
IMF Staff Papers 48, 171–208.
R.M. Desai et al. / Journal of Comparative Economics 33 (2005) 814–834
833
Center for Economic and Financial Research (CEFIR), 2003. Monitoring the administrative barriers to small business
development in Russia. CEFIR, Moscow.
Chaudhry, Kiren Aziz, 1997. The Price of Wealth: Economies and Institutions in the Middle East. Cornell Univ. Press,
Ithaca, NY.
Collier, Paul, Hoeffler, Anke, 2002. Aid, Policy, and Growth in Post-Conflict Societies. World Bank, Washington, DC.
Dalmazzo, Alberto, de Blasio, Guido, 2001. Resources and incentives to reform. Working paper 01/86. International
Monetary Fund, Washington, DC.
Desai, Raj M., Goldberg, Itzhak, 2001. The politics of Russian enterprise reform: Insiders, local governments, and the
obstacles to restructuring. World Bank Research Observer 16, 219–240.
Esanov, Akram, Raiser, Martin, Buiter, William, 2001. Nature’s blessing or nature’s curse: The political economy of
transition in resource-based economies. Working paper 65. EBRD, London.
Freinkman, Lev M., Haney, Michael, 1997. What affects the propensity to subsidize: Determinants of budget subsidies
and transfers financed by the Russian regional governments in 1992–1995. Working paper 1818. Policy Research,
World Bank, Washington, DC.
Freinkman, Lev M., Treisman, Daniel, Titov, Stepan, 1999. Subnational budgeting in Russia: Preempting a potential
crisis. Technical paper 452. World Bank, Washington, DC.
Hellman, Joel S., Jones, Geraint, Kaufmann, Daniel, 2003. Seize the state, seize the day: State capture, corruption, and
influence in transition. Journal of Comparative Economics 31, 751–773.
Huang, Yasheng, 1994. Information bureaucracy and economic reforms in China and the Soviet Union. World Politics 47,
102–134.
Jin, Hehui, Qian, Yingyi, Weingast, Barry, in press. Regional decentralization and fiscal incentives: Federalism, Chinese
style. Journal of Public Economics.
Knack, Stephen F., 2001. Aid dependence and the quality of governance: Cross-country empirical tests. Southern Economic Journal 68, 310–329.
Kuznetsova, Olga, 2001. Economicheskaya politika regional’nykh organov vlasti (Economic policy of regional authorities). In: 3rd International Conference on Public Sector Reforms, St. Petersburg.
Lapidus, Gail W., 1999. Asymmetrical federalism and state breakdown in Russia. Post-Soviet Affairs 15, 74–82.
Lavrov, Alexei, 1996. Fiscal federalism and financial stabilization. Problems of Economic Transition 39, 83–94.
Lavrov, Alexei, Litwack, John M., Sutherland, Douglas, 2001. Fiscal Federalist Relations in Russia. OECD, Paris.
Leite, Carlos, Weidmann, Jens, 2002. Does mother nature corrupt? Natural resources, corruption, and economic growth.
In: Abed, George T., Gupta, Sanjeev (Eds.), Governance, Corruption, and Economic Performance. International Monetary Fund, Washington, DC.
Levi, Margaret, 1988. Of Rule and Revenue. Univ. of California Press, Berkeley.
Litwack, John M., 2002. Central control of regional budgets: Theory with applications to Russia. Journal of Comparative
Economics 30, 51–75.
Madison, James, 1941. The particular structure of the new government and the distribution of power among its different
parts. In: Hamilton, Alexander, Madison, James, Jay, John (Eds.), The Federalist: A Commentary on the Constitution
of the United States. The Modern Library, New York.
McKinsey & Co., 1999. Unlocking Economic Growth in Russia. McKinsey Global Institute, Washington, DC.
Montesquieu, Charles de Secondat, 1989. The Spirit of the Laws (Cohler, Anne M., Miller, Basia C., Stone, Harold S.,
Trans.). Cambridge Univ. Press, Cambridge, NY.
Montinola, Gabriella, Qian, Yingyi, Weingast, Barry, 1994. Federalism, Chinese style: The political basis for economic
success in China. World Politics 48, 50–81.
Moore, Mick, 1998. Death without taxes: Democracy, state capacity, and aid dependence in the Fourth World. In: Robinson, Mark, White, Gordon (Eds.), The Democratic Developmental State: Political and Institutional Design. Oxford
Univ. Press, Oxford.
North, Douglass C., Weingast, Barry R., 1989. Constitutions and commitment: The evolution of institutions governing
public choice in seventeenth-century England. Journal of Economic History 49, 803–832.
Pinto, Brian, Drebentsov, Vladimir, Morozov, Alexander, 2000. Give macroeconomic stability and growth in Russia a
chance: Harden budget payments by eliminating non-payments. Economics of Transition 8, 297–324.
Popov, Vladimir, 2001. Reform strategies and economic performance of Russia’s regions. World Development 29, 865–
886.
Radayev, Vadim, 1996. Malyi bisnes i problemy delovoi etiki: Nadezhdy i real’nost (Small business and the problems of
business ethics: Hopes and reality). Voprosy Ekonomiki 7, 72–82.
Ross, Michael L., 2001. Does oil hinder democracy? World Politics 53, 325–361.
Sachs, Jeffrey, Warner, Andrew M., 2001. The curse of natural resources. European Economic Review 45, 827–838.
834
R.M. Desai et al. / Journal of Comparative Economics 33 (2005) 814–834
Sinelnikov, Sergei, Kadochnikov, Pavel, Trunin, Ilya, Shrebela, Yekaterina, 2001. Vliyaniye Mezibyudzetnykh Transfertov na Fiscal’noye Povedeniye Regional’nykh Vlastei v Rossiiskoi Federatsii (Impact of Inter-Governmental
Transfers on the Fiscal Performance of Regional Authorities in the Russian Federation). Institute of Economic Transition, Moscow.
Slinko, Irina, Yakovlev, Evgeny, Zhuravskaya, Ekaterina, 2005. Laws for sale: Evidence from Russia. American Law
and Economics Review 7, 284–318.
Speck, Stefan, Matusevich, Alexander P., 2002. Taxation policy reform in the Russian Federation: Challenges and obstacles for implementation of environmentally related taxes. Typescript. OECD, Paris.
Stoner-Weiss, Kathryn, 2005. Whither the central state? The regional sources of Russia’s stalled reforms. In: McFaul,
Michael, Stoner-Weiss, Kathryn (Eds.), After the Collapse of Communism: Comparative Lessons of Transitions.
Cambridge Univ. Press, New York.
Tornell, Aaron, Lane, Philip R., 1999. The voracity effect. American Economic Review 89, 22–46.
Treisman, Daniel, 1996. The politics of inter-governmental transfers in post-Soviet Russia. British Journal of Political
Science 26, 299–335.
Treisman, Daniel, 1999. After the Deluge: Regional Crises and Political Consolidation in Russia. Univ. of Michigan
Press, Ann Arbor, MI.
World Bank, 2004. From Transition to Development: A Country Economic Memorandum for the Russian Federation.
The World Bank, Washington, DC.
Yasin, Yevgeny (Ed.), 2002. Bremya Gosudarstva i Ekonomicheskaya Politika: Liberal’naya Alternativa (The Burden of
State and Economic Policy: A Liberal Alternative). The Expert Institute, Moscow.
Zhuravskaya, Ekaterina V., 2000. Incentives to provide local public goods: Fiscal federalism, Russian style. Journal of
Public Economics 76, 337–368.