Information about Asia and the Pacific Asia y el Pacífico
Chapter

7 Reining in State Deficits

Author(s):
Catriona Purfield, and Jerald Schiff
Published Date:
August 2006
Share
  • ShareShare
Information about Asia and the Pacific Asia y el Pacífico
Show Summary Details
Author(s)
Catriona Purfield and Mark Flanagan 

A major factor underlying India’s large general government deficit and debt burden has been the deterioration of state government finances. State governments play an important role in implementing fiscal policies in India, undertaking more than half of general government spending. However, until recently they only received a little under 40 percent of government revenues. The imbalance in their finances has been aggravated by various special factors in recent years including large civil service pay increases granted by the Fifth Pay Commission in 1997, increased tax competition among the states, a rising interest burden, and low tax buoyancy during the growth slowdown of 2001–04 (Table 7.1). As a result, by end-March 2005 state government debt accounted for over one-third of the general government’s total burden of almost 80 percent of GDP. In the fiscal year 2004 states also accounted for about half of the general government deficit of 9.1 percent of GDP (up from only a quarter a decade earlier).

Table 7.1.Trends in State Finances(In percent of GDP)
1993–962002–032004/05
Average2002/032003/04Estimate
Revenue12.011.411.311.111.7
Expenditure14.615.415.415.515.3
Revenue balance−0.6−1.9−2.2−2.2−1.3
Overall balance−2.6−4.0−4.1−4.4−3.6
Debt19.332.631.933.032.8
Source: Reserve Bank of India.
Source: Reserve Bank of India.

While more recent indicators suggest that the states’ fiscal position has begun to improve, problems remain. The aggregate state debt-to-GDP ratio of 33 percent accounts for 320 percent of state own revenues while the interest bill continues to consume more than a quarter of state own revenues. The situation is even more serious for some individual states, with debt approaching 100 percent of GDP in some special category states, and debt-to-revenue ratios as high as 400 percent (in the middle-income state of West Bengal). A lack of free resources is also constraining states’ ability to maintain existing infrastructure and invest in new projects, and limiting options to expand social spending.

This chapter examines how India’s federal system may have created incentives for fiscal indiscipline at the state level and reviews the steps being taken to improve the system. First, it highlights those features of India’s system of federal relations that have given rise to macroeconomic and fiscal problems in other countries. Next, it tests empirically whether the institutional characteristics of India’s federal system have in fact contributed to states’ fiscal problems. The chapter then reviews recent policy changes aimed at addressing state deficits, assesses whether these recommendations are likely to improve aggregate state finances, and examines their implications at the individual state level. Finally, drawing on international practice, the chapter considers some additional measures that could be taken to strengthen India’s fiscal federalism framework.

India’s Federal System in an International Context

International evidence suggests that decentralization can contribute to large and persistent general government deficits. Studies by Dabla-Norris and Wade (2002), Rodden (2002), and Tanzi (2000) find incentives for responsible fiscal behavior and hard-budget constraints are undermined when the federal framework is characterized by a high degree of dependence on transfers, lack of constraints on subnational indebtedness, lack of clarity in the respective roles of each tier of government, and weak budget institutions.

In an international context, the level of states’ transfer dependence in India is relatively high. Central government grants account for a larger share of the states’ revenue and a slightly higher share of expenditure than is typical in most decentralized systems (Table 7.2). Moreover, states have little control over the use of these grants. The split in responsibility for grant allocations between the Finance and the Planning Commissions1 is also unusual, and it leads to coordination problems, creates incentives for states to overstate revenue needs, and allows larger and politically stronger states to bargain for larger transfers.2 Moreover, shared taxes account for about one-third of state tax revenues, and since these are pooled and divided across states by the Finance Commission, revenue collection incentives may suffer.

Table 7.2.Subnational Autonomy in an International Context(Average for 1990–97)
Measures of Subnational Autonomy
Tax autonomy1Tax sharing2Grants share of revenueGrant dependence3Vertical gap4Vertical gap after grants
India’s states46.932.442.434.7−38.5−22.1
Europe5
Austria51.588.122.823.5
Belgium36.145.955.155.92.81.3
Denmark47.84.843.143.1−0.1−0.1
Finland46.511.433.033.41.61.1
France45.10.035.234.7−4.3−2.8
Germany61.086.511.410.3−9.2−6.9
Sweden72.70.018.417.7−5.4−4.4
Average51.533.831.331.2−2.4−1.9
Central and Eastern Europe and former Soviet Union6
Belarus71.993.822.622.7−0.9−1.0
Bulgaria56.290.040.840.90.20.2
Czech Republic47.491.729.429.0−2.0−1.4
Estonia64.789.225.524.4−6.6−5.0
Hungary19.367.458.659.32.11.3
Lithuania67.0100.027.627.6−0.2−0.2
Mongolia49.541.341.40.30.0
Romania51.875.038.138.83.11.8
Slovenia59.790–10021.922.63.42.7
Average54.287.834.034.1−0.1−0.2
Latin America3
Argentina79.164.012.411.5−9.2−8.1
Bolivia41.993.024.419.4−4.0−3.3
Mexico65.2100.018.918.7−4.3−4.0
Peru8.470.471.02.20.0
Average48.785.731.530.1−3.8−3.8
Other
Australia35.937.438.12.41.6
Canada65.613.612.6−9.5−7.0
China753.053.436.9−0.7−0.3
Israel33.940.638.1−10.9−6.5
South Africa15.20.064.762.9−10.6−2.5
United States55.015.25.45.43.9
Average43.137.532.3−4.0−1.8
Sources: IMF, Government Finance Statistics (GFS) and International Financial Statistics; and IMF staff calculations.

Ratio of tax revenue (including shared taxes) to total subnational revenues, including grants.

Ratio of shared taxes from central government to total subnational tax revenue.

Ratio of central grants to total consolidated expenditure of subnational governments.

Deficit as a share of subnational nongrant revenue; a positive number implies a surplus.

Tax share ratios from Ebel and Yilmaz (2002).

Tax share ratios from Dabla-Norris and Wade (2002).

Tax autonomy measured from GFS data for 1995–99. Other measures use data reported in Ahmad and others (2002). Subnational governments receive 25 percent of domestic value-added tax, the business tax, enterprises income taxes on state enterprises, the personal income tax, and a number of smaller taxes. Rates on these taxes are generally decided by the center.

Sources: IMF, Government Finance Statistics (GFS) and International Financial Statistics; and IMF staff calculations.

Ratio of tax revenue (including shared taxes) to total subnational revenues, including grants.

Ratio of shared taxes from central government to total subnational tax revenue.

Ratio of central grants to total consolidated expenditure of subnational governments.

Deficit as a share of subnational nongrant revenue; a positive number implies a surplus.

Tax share ratios from Ebel and Yilmaz (2002).

Tax share ratios from Dabla-Norris and Wade (2002).

Tax autonomy measured from GFS data for 1995–99. Other measures use data reported in Ahmad and others (2002). Subnational governments receive 25 percent of domestic value-added tax, the business tax, enterprises income taxes on state enterprises, the personal income tax, and a number of smaller taxes. Rates on these taxes are generally decided by the center.

Even after the operation of the transfer system, the level of imbalance in state finances that remains is unrivaled. The state-level deficit in India still amounts to 22 percent of total states’ nongrant revenue, including shared taxes and grants from the central government. Elsewhere, state-level finances are typically closer to balance after the operation of tax sharing and grant arrangements.

Indian states close this imbalance mainly through borrowing, and India’s borrowing regime has been comparatively liberal by international standards.3 Many developing countries prohibit or ban local borrowing (e.g., China and Indonesia). Most that permit borrowing impose numerical ceilings on subnational indebtedness, usually in the form of ceilings on the debt stock or debt service. Where ceilings are absent, hard-budget constraints are often enforced by legally prohibiting central government guarantees of subnational debt and state guarantees of public enterprise debt (e.g., South Africa and the Czech Republic). India’s borrowing regime has been closer to that of advanced economies. Indian states are able to borrow domestically, although they need to seek central government approval if they are indebted to the center (which all states are). External borrowing by states is explicitly prohibited by the constitution. Until recently, the central government did not subject states to an explicit aggregate borrowing ceiling, choosing instead only to subject a state’s market borrowings to caps. With states obliged to take 100 percent of the debt issued by small savings schemes, they were able to easily finance their growing deficits.

The combination of (de facto or de jure) soft-budget constraints and high transfer dependence has generated serious fiscal problems in other countries. Brazil and Argentina are well-known examples where the central government’s inability to credibly commit to a policy of no subnational bailouts led to moral hazard problems. The ability to tap state-owned banks and enterprises has also softened budget constraints in Germany and the Czech Republic. A weak regulatory borrowing framework contributed to subna-tional debt problems in Colombia, the Czech Republic, and South Africa.

Explaining State Fiscal Performance

We turn now to testing empirically the importance of transfer dependence and soft-budget constraints in explaining states’ fiscal performance, using annual data from India’s 15 largest states between 1985 and 2000. Fiscal performance is proxied using two measures: (1) The ratio of the budget deficit in state i to total expenditure in state i (which measures the share of state expenditure not covered by revenue and helps control for the large differences in the size of state governments); and (2) the ratio of the deficit in state i to gross state domestic product (GSDP) in state i. Fiscal performance is modeled as a function of the following:

Transfer dependence. States that rely more heavily on transfers from the central government have fewer incentives to be fiscally responsible. Once the link between taxes and benefits is broken, states may try to offload the cost of extra spending on the central government and bailout expectations are likely to be higher. The degree of transfer dependence in a state is measured as the sum of grants and shared revenue received by state i in period t as a share of state i’s revenue in period t.

Soft-budget constraints. Soft sources of central government finance for states can generate moral hazard: having condoned the borrowing, it is difficult in future for the central government to deny assistance to states in financial stress. States’ expectations for a bailout are likely to increase the higher the level of their central government debts. Such borrowing dependency is measured by the stock of central government loans to GSDP in state i in period t.

The extent of decentralization. It could also be more difficult for a state to control deficits if it is responsible for implementing a greater share of general government expenditure. However, if a state collects a greater share of general government revenue, the link between expenditure and revenue is stronger, creating better incentives for fiscal discipline. A state’s expenditure (revenue) share in period t is measured relative to general government expenditure (revenue).

Contagion. Shortfalls in shared taxes can quickly lead to a widening deficit at both tiers of government. The inability of the central government to control its own deficits may also signal deficit tolerance or future offloading of unfunded expenditure mandates on states. We measure these spillover effects using the ratio of the central government deficit to central government expenditure in each period.

State-specific structural and economic characteristics. The hypothesis here is that a state would have greater difficulty controlling deficits the greater its reliance on agriculture, the larger its population, the greater the share of its population living in poverty, or the lower its real per capita income.

Table 7.3 presents the results with the upper section using the ratio of state deficit to expenditure as the dependent variable and the lower section using the deficit to GSDP.4 A Hausman test favors the random effects (Model 3) over the fixed effects specification.

Table 7.3.Estimates of State Deficits and Transfer Dependence1
Model 1 Two-StageInstrumental Variable Estimates
Least-Squares PooledModel 2 Fixed EffectsModel 3 Random EffectsModel 4 Random Effects
CoefficientStandard errorsCoefficientStandard errorsCoefficientStandard errorsCoefficientStandard errors
Dependent variable: state deficit/expenditure
Transfer dependence measure0.14610.0557***0.04630.12880.15550.0836**0.10340.0834
Borrowing dependency0.77090.1104***0.38510.27280.56370.1948***0.49310.1970**
Ratio of states expenditure to general government expenditure1.64540.4923***1.17500.5429**1.74330.4714***1.36730.4791***
Ratio of states revenue to general government revenue−2.84920.5205***−2.82670.5285***−3.13720.4857***−2.45490.5194***
Central government deficit to expenditure ratio0.98700.2290***1.07780.2264***1.16780.2127***0.91120.2230***
Share of agriculture in GSDP−0.17040.0752**−0.06590.1561−0.10990.1153−0.02920.1178
Share of population living below the poverty line−0.14970.0595**−0.34150.1191***−0.24240.0876***−0.23630.0864***
State population0.01130.01400.25150.0990**0.04310.03330.03780.0333
Per capita income0.00030.00020.00020.00020.00020.00020.00000.0002
1997-2000 dummy variable4.48971.3975***
Constant35.255016.7045**51.922119.2975***40.493716.4826**36.199816.1772**
R20.47830.12560.45100.4642
Groups14.000014.000014.000014.0000
Dependent variable: state deficit/GSDP
Transfer dependence measure0.02330.0126*−0.03140.02820.01170.0189−0.00060.0188
Borrowing dependency0.18370.0250***0.20590.0598***0.20090.0441***0.18530.0445***
Ratio of states expenditure to general government expenditure0.48290.1113***0.35830.1191***0.45190.1048***0.36550.1062
Ratio of states expenditure to general government expenditure (t − 1)
Ratio of states revenue to general government revenue−0.68560.1177***−0.58480.1159***−0.66300.1079***−0.50490.1150***
Central government deficit to expenditure ratio0.24620.0518***0.22690.0497***0.25240.0472***0.19290.0493***
Share of agriculture in GSDP−0.04090.0170**−0.08400.0342**−0.05850.0260**−0.03990.0265
Share of population living below the poverty line−0.01080.01340.02050.0217−0.01660.0197−0.01510.0194
State population−0.00670.0032**0.02050.0217−0.00100.0077−0.00250.0077
Per capita income0.00000.00000.00000.00000.00000.0000−0.00010.0000***
1997-2000 dummy variable1.03970.3098***
Constant4.00183.77608.08204.2318*4.67183.67443.65153.5973
R20.46530.14100.44520.4833
Groups14.000014.000014.000014.0000
Source: IMF staff calculations.

p(z) < 0.01; ** p(z) < 0.05; * p(z) < 0.1.

Source: IMF staff calculations.

p(z) < 0.01; ** p(z) < 0.05; * p(z) < 0.1.

The results confirm that the combination of high transfer dependence and soft-budget constraints significantly weaken state fiscal discipline: Transfer and borrowing dependency are individually significant. States that implement a larger share of general government spending have significantly higher deficits, while those that account for greater share of general government revenues have much lower deficits. Larger deficits at the central government level are also found to have resulted in significantly larger state-level deficits. With the exception of poverty, most state-specific characteristics are found to be insignificant in explaining cross-state divergence in fiscal performance. Although per capita incomes are found to be statistically significant, the impact on state deficits is very small. Model 4, which includes a time-specific dummy variable to capture the effects of the Fifth Pay Commission, finds that the pay awards—which also fed into higher pension spending—caused the ratio of states’ deficit to expenditure to ratchet upward by close to 4½ percentage points.

Transfer dependence and growing expectations of central government bailouts may have created upward pressure on state deficits over time. To test this hypothesis, Table 7.4 presents the results of using the one-step robust Arellano and Bond generalized method of moments (GMM) estimator where the key explanatory variable is the change in borrowing dependency measured by the change in loans outstanding from the central government.5 The results suggest that since the mid-1980s, the ratio of the deficit to state expenditure rose by almost 1 percent for each 1 percent of GSDP increase in states’ indebtedness to the central government (Model 5). It is also feasible that transfer dependency and bailout expectations could interact to cause the states’ deficit to rise over time. States have an incentive to increase expenditure knowing that the central government will be under pressure to provide assistance. This could be especially relevant in India where the revenue-sharing ratios are fixed for five-year intervals and states can use evidence of their past deficits to argue for larger transfers at the time of review. Model 6 includes a multiplicative term to capture such interactions and it confirms that bailout expectations, as measured by the stock of loans outstanding to the central government, coupled with transfer dependency, contributed to states’ rising deficit.

Table 7.4.Dynamic Panel Estimates of the Evolution of State Deficits1
Model 5Model 6
CoefficientRobust standard errorsCoefficientRobust standard errors
Dependent variable: Δ in state deficit/expenditure
State deficit/expenditure (t − 1)0.28400.0779***0.34160.1163***
State deficit/expenditure (t − 2)−0.07980.0938−0.08900.1017
ΔTransfer dependence measure0.04380.1751
Δ Borrowing dependency0.94160.3331***
Δ Borrowing dependency*transfer dependence measure0.01660.0067**
Δ Ratio of states expenditure to general government expenditure1.20781.08181.18321.0767
Δ Ratio of states revenue to general government revenue−2.47600.8999***−2.62570.8406***
Δ Central government deficit to expenditure ratio0.93620.3506***0.98380.3274***
Δ Share of agriculture in GSDP−0.07060.22810.07070.1980
Δ Share of population living below the poverty line−0.37010.2088*−0.28590.2250
Δ State population0.17800.1021*0.15990.0898
Δ Per capita income0.00010.00020.00010.0003
Constant0.20630.56790.24000.5819
Wald chi2(11)702.5344.9
Groups1414
Number of observations168168
Dependent variable: Δstate deficit/GSDP
State deficit/GSDP (t − 1)0.16190.0691**0.25890.0890***
State deficit/GSDP (t − 2)−0.19540.0717***−0.20480.0905**
Δ Transfer dependence measure−0.01810.0278
Δ Borrowing dependency0.36570.0818***
Δ Borrowing dependency*transfer dependence measure0.00550.0012***
Δ Ratio of states expenditure to general government expenditure0.20570.23030.19380.2416
Δ Ratio of states revenue to general government revenue−0.42840.1938**−0.50220.1866***
Δ Central government deficit to expenditure ratio0.17200.0766**0.19340.0774**
Δ Share of agriculture in GSDP−0.06460.0529−0.03800.0543
Δ Share of population living below the poverty line−0.02930.04160.00660.0489
Δ State population−0.00470.0324−0.00280.0300
Δ Per capita income−0.00010.0001−0.00010.0001
Constant0.19520.13940.18690.1604
Wald chi2(11)228.3544.1
Groups1414
Number of observations168168
Source: IMF staff calculations.

p(z) < 0.01; ** p(z) < 0.05; * p(z) < 0.1.

Source: IMF staff calculations.

p(z) < 0.01; ** p(z) < 0.05; * p(z) < 0.1.

Fixing the Problem

Mindful of the growing problem in state finances and the shortcomings in the existing federal system, the 2004 Twelfth Finance Commission (TFC) reassessed the system of fiscal federal relations. It made three key recommendations, which are now being implemented:

  • States now receive a higher revenue share and higher grants. These additional transfers are expected to help to close the gap between states’ expenditure commitments and their revenues. While the TFC made some effort to adjust for inadequate fiscal effort in its calculations of grants, this may yet present a risk in terms of incentives for fiscal responsibility.
  • States can benefit from conditional debt restructuring and relief provided they pass and then implement fiscal responsibility legislation (FRL) targeting revenue balance by 2008/09 and a 3 percent of GDP overall deficit by 2009/10. This would provide an incentive toward fiscal discipline in the states.
  • States also now confront a stricter borrowing regime, with the center setting global ceilings on borrowings and henceforth only lending to fiscally weak states. This will help address the issue of soft-budget constraints and moral hazard.

The TFC’s framework is expected to have a considerable direct impact on general government finances. The direct impact—due to a higher revenue share, higher grants, and interest relief—can be assessed relative to a baseline reflecting a continuation of revenue and spending ratios at levels realized during 2004/05 (Table 7.5). The comparison shows that:

  • States’ revenue and overall deficits could drop by as much as 0.6 percent of GDP due to higher direct and untied transfers. In any event, provisional estimates of the 2005/06 outturn (compiled from states’ revised budget estimates) suggest the state-level fiscal deficit declined to 3.3 percent of GDP in 2005/06, a reduction of only 0.3 percent relative to the 2004/05 outturn.
  • States will need to undertake considerable fiscal adjustment in future years just to maintain their deficit at 2005/06 levels. The challenge is driven by the fact that grants from the center are (1) fixed in nominal terms, and thus declining in terms of GDP; and (2) shifting in their composition away from untied gap-filling grants toward tied grants (which must be spent).
  • The central government will have to undertake more fiscal adjustment upfront to achieve the targets specified in its own fiscal responsibility legislation. In 2005/06, the center’s revenue deficit would have deteriorated by about 0.7 percent of GDP other things equal, though the overall central government deficit would have remained broadly unchanged if the center stopped lending to states as planned. In the event, however, the central government was able to overperform on the budget slightly, despite lower interest receipts from states, thanks to expenditure savings.
  • The general government fiscal deficit could deteriorate by an amount equal to the additional tied spending by the states. However, the TFC recommended that the center withdraw from spending in areas which are states’ responsibility to offset the impact of the higher state spending.
Table 7.5.Projected Impact of Twelfth Finance Commission Recommendations on Fiscal Baseline(In percent of GDP)
2005/062006/072007/082008/092009/10
States
Total deficit impact−0.59−0.44−0.40−0.34−0.31
Tax revenues0.220.220.220.220.22
Nontax revenues (tied and plan grants)0.380.400.330.250.19
Interest−0.12−0.11−0.10−0.09−0.08
Spending of tied and plan grants0.130.280.250.210.18
Center
Total deficit impact (1 + 2)0.160.140.04−0.07−0.15
1. Total revenue deficit impact0.720.730.640.560.49
Tax revenues−0.22−0.22−0.22−0.22−0.22
Nontax revenues (interest)−0.12−0.11−0.10−0.09−0.08
Transfers (tied grants)0.280.300.220.150.09
Transfers (plan grants)10.100.100.100.100.10
2. Impact on net lending−0.56−0.58−0.61−0.62−0.64
New loans2−0.82−0.82−0.82−0.82−0.82
Repayments (maximum debt relief)−0.26−0.23−0.21−0.19−0.17
General government: total deficit impact0.130.280.250.210.18
Sources: Government of India, Report of the Twelfth Finance Commission; and IMF staff estimates.

Discretionary.

As in 2005–06 budget.

Sources: Government of India, Report of the Twelfth Finance Commission; and IMF staff estimates.

Discretionary.

As in 2005–06 budget.

Scenarios for Aggregate State Fiscal Adjustment

Three scenarios are developed here, covering baseline, high growth, and low growth cases to assess whether the adjustment anticipated by the TFC is on track (Table 7.6).6 The baseline assumes growth will continue at its recent average, while the high and low cases are about one standard deviation above and below the TFC’s baseline growth assumption. In the baseline scenario, revenue buoyancy is assumed to be in line with recent trends, but rises and falls moderately under the different growth assumptions. In each scenario, states try to raise both primary revenue spending (reflecting the need for operations and maintenance and social spending), and capital spending (with a view to achieve the TFC’s target of a total investment increase of ½ percent of GDP). However, if the scenario shows a burgeoning deficit, states are assumed to restrain their spending. Finally, since the aggregate state deficit path differs across the scenarios, the total debt and interest relief received by states is allowed to vary across scenarios: over three-fourths of states (on a value-weighted basis) receive the debt relief in the high growth scenario, but less than half do in the low growth scenario.

Table 7.6.State Fiscal Adjustment Scenarios(In percent of GDP)
2004/05
Twelfth Finance Commission2009/10
Actual(TFC)TFCIMF-baseIMF-lowIMF-high
Total revenues11.711.613.213.312.813.9
Tax8.48.49.79.99.410.6
Of which: own tax5.95.96.86.76.37.1
Nontax3.33.23.53.43.43.3
Of which: own nontax1.51.21.41.31.31.3
Total expenditures15.316.216.316.316.416.5
Revenue expenditure13.113.613.213.413.813.4
Of which: interest2.82.92.02.83.22.7
Capital expenditure2.52.63.12.92.63.0
Revenue balance−1.3−2.00.0−0.3−1.20.3
Overall balance−3.6−4.5−3.0−3.1−3.6−2.5
Debt (end-year)32.830.330.831.834.129.6
Memorandum items:
Growth rate7.06.56.08.0
Real interest rate1.94.25.25.7
Sources: Government of India, Report of the Twelfth Finance Commission; and IMF staff estimates.
Sources: Government of India, Report of the Twelfth Finance Commission; and IMF staff estimates.

With these assumptions several important results emerge (Figure 7.1 and Table 7.6):

  • States, in aggregate, are broadly on track to achieve the TFC fiscal targets. Under the baseline, the revenue deficit is cut by about 1¼ percent, to ⅓ percent of GDP, and the overall deficit falls by slightly under 1 percent to just over 3 percent of GDP by 2009/10. Capital spending also rises to 3 percent of GDP.
  • However, broad achievement of TFC targets would not be sufficient to achieve a sustainable reduction in state debt ratios. Under the baseline, state debt falls initially by about 1 percent of GDP, but rises to almost 32 percent of GDP by the end of the period. Even in the TFC’s scenario, debt reductions were extremely gradual and a small drop in growth is all it takes to reverse debt dynamics.
  • Results are sensitive to growth and interest rate assumptions. In the low growth scenario, targets would be missed by a wide margin and debt would rise throughout the period. States are forced to substantially curtail investment by the end of the period to cope with a rising interest burden. In the high growth scenario, all targets are achieved with room to spare.

Figure 7.1.India: Fiscal Outcomes Under Adjustment Scenarios

Source: IMF staff calculations.

Fiscal Adjustment in Individual States

Even if states in aggregate achieve a sustainable debt level, some states may yet face significant fiscal stress. Since states face widely varying initial fiscal conditions and diverse growth prospects, some are likely to require more fiscal adjustment and/or debt relief than others to achieve a sustainable debt position. To evaluate the impact on individual states, we consider whether a state would achieve a debt stabilizing primary balance if it were to implement the full TFC fiscal adjustment (i.e., revenue balance and an overall deficit of 3 percent of GDP) up front and were also to be given all of the proposed debt rescheduling and relief in the first year.7

In calculating debt scenarios for individual states, relatively favorable assumptions are used. For each state, the historic state growth rate is used, adjusted for the difference between projected growth for the economy as a whole under the baseline macroeconomic scenario (Table 7.5) and past growth for the economy as a whole. States are assumed to face the same interest rate on new borrowing, approximately the rate on small savings (the largest source for state borrowing at the moment). This rate is assumed to evolve in line with, but remain above, market rates. In reality, however, weaker states would almost surely face a harsher borrowing environment. At the same time, if the adjustment were to be phased instead of undertaken upfront, the initial debt ratio would be higher for many states. On both counts, debt dynamics would be worse.

Even with these favorable assumptions, analysis of the general category states suggests that many of them would indeed continue to face significant fiscal stress (Table 7.7):8

  • The primary adjustment required by the TFC varies greatly across them. Some states are already in sustainable positions (e.g., Haryana), but others face gaps exceeding 4 percent of GDP (e.g., Rajasthan). For states facing large adjustments this would not likely be feasible over a five-year period without deep cuts to key programs.
  • Not all states would achieve a debt stabilizing primary balance. Four states would face a need for a further small primary adjustment of about ½ percent of GDP to become sustainable.9 If states instead can only implement a maximum adjustment equal to 10 percent of their initial spending level, then a further two states (Rajasthan and Gujarat) would also fall short of sustainability.
  • Finally, for those states that do succeed in reducing debt, their burden—as measured by debt-to-revenue and interest-to-revenue ratios—can nonetheless remain very high (e.g., 360 percent and 37 percent in West Bengal). This signals that significant crowding out of other budget items is likely.
Table 7.7.Sustainability of State Debts Post-TFC: Baseline Scenario1
DebtRequired Primary AdjustmentPrimary Balance3
2004/05Post-reliefDebt ReliefIncentive2Debt stabilizingProjectedGap4Debt/ RevenueInterest/ Revenue
Andhra Pradesh33.430.92.5−0.1−0.1−0.30.20.41.90.2
Bihar56.049.16.90.50.10.31.31.01.80.2
Gujarat37.335.32.02.01.0−0.3−0.30.12.60.2
Haryana27.226.40.8−1.9−2.5−0.5−0.20.32.30.2
Karnataka32.630.32.3−0.3−0.1−0.1−0.6−0.51.50.1
Kerala40.837.73.11.80.6−0.60.00.72.30.2
Madhya Pradesh43.540.72.84.21.50.2−0.3−0.42.00.1
Maharashtra31.029.71.32.21.7−0.1−0.6−0.52.20.2
Orissa69.064.34.72.40.50.52.11.72.70.2
Punjab53.951.82.12.51.20.51.10.72.40.2
Rajasthan58.957.01.94.32.20.21.41.22.80.2
Tamil Nadu27.926.21.7−0.3−0.20.0−0.6−0.51.90.2
Uttar Pradesh51.348.23.12.10.70.31.10.82.70.2
West Bengal44.542.71.82.51.3−0.91.42.33.60.4
Sources: World Bank; and IMF staff estimates.

Assumes that all debt relief and fiscal adjustment are to be realized up-front.

Amount of adjustment required to realize 1 percent of GDP in debt relief.

Assumes an overall deficit of 3 percent of GDP.

When a positive gap exists, the overall debt level would be declining.

Sources: World Bank; and IMF staff estimates.

Assumes that all debt relief and fiscal adjustment are to be realized up-front.

Amount of adjustment required to realize 1 percent of GDP in debt relief.

Assumes an overall deficit of 3 percent of GDP.

When a positive gap exists, the overall debt level would be declining.

The analysis also highlights that the debt relief incentive scheme, while a positive step, may prove insufficient to motivate a full adjustment effort in some states. While the central government is providing considerable resources to the states in aggregate, for many states the debt relief is minor relative to their total outstanding liabilities, and relative to the amount of primary adjustment they need to undertake (Table 7.7; see also Kurian, 2005). From another perspective, for debt relief equivalent to only 8½ percent of total transfers they are already set to receive, states would have to undertake very politically difficult policy reforms. For states where there is no present consensus, and which face a very large adjustment requirement, the incentive on offer may prove inadequate.10

Some Further Improvements to the Federal System

Given the long-standing nature of the problem in the states, and the risks to achieving TFC targets, there is good reason to consider ways in which the borrowing and incentive regime could be strengthened. On the borrowing side much can be done in the short term; however, on the incentive side, due to the constitutional prerogative of Finance Commissions over the design of fiscal federal relations, some of the more attractive options may need to await the next Finance Commission, in 2009.

India could, in the near term, take a number of measures to strengthen the borrowing regime:

  • The central government could set its borrowing ceilings more in line with FRL adjustment targets and sustainability considerations. It could also harden the rules for approval of new borrowing: as in some countries with FRLs (e.g., Brazil, Colombia, and Ecuador), and it could condition approval of borrowing on demonstrating compliance with the FRL. Ultimately credibility will be cemented only when the system is tested (e.g., when growth and revenues unexpectedly fall short) and the government does not raise borrowing limits.
  • In terms of market discipline, the various soft sources of finance need to be progressively restricted and eliminated, and information provision improved. In this context, consideration could be given to removing states’ obligation to borrow fully the resources from small savings schemes (allowing the schemes to pursue other investment options), and to bringing the administered interest rates on these schemes in line with market rates. Some countries penalize lenders for assisting subnational governments in borrowing beyond their legal limits—for instance, Brazil requires immediate repayment with no interest—and such sanctions could alter lenders’ incentives in India.
  • Finally, it would help to improve both market discipline and the credibility of administrative control if India were to legally rule out future bailouts as in the Czech Republic and South Africa.

The government could also, at the same time, pursue greater cooperation with states, with peer pressure as a sanction. Under a cooperative approach, negotiations with subnational governments determine a global borrowing ceiling, which is then apportioned to the various government entities (see Ahmad, Albino-War, and Singh, 2005). Enforcement can be through an independent intergovernmental entity (e.g., Austria) or through a subnational government association (e.g., Denmark), and should involve some form of financial sanction. In Spain, for instance, subnational governments that violate agreed borrowing ceilings are required to contribute to any fines assessed by the European Union (under the Stability and Growth Pact). In India, the central government regularly engages the states on their individual borrowing plans, and has plans to set up a borrowing council (involving the Ministry of Finance, the RBI, the Planning Ministry, and states). This council could become a forum for the determination of a global all-state borrowing ceiling, and for agreement on sanctions for exceeding state-specific targets.

In terms of the incentive regime, stronger requirements on information provision and stronger sanctions would help make it more effective:

  • A major problem to overcome in making the incentive regime more effective is information provision. Key issues would include establishing standardized definitions, accounting systems, and classifications; upgrading and standardizing information technology systems; and ensuring that states maintain databases on all financing sources. The central government could also increase the incentive for states to improve information provision by suspending grants when it is inadequate. Brazil and Ecuador have applied information-related conditionalities in their FRLs, with Ecuador requiring information provision within 15 days of the date set out in the FRL, upon penalty of suspension of grants (see Webb, 2004).
  • The center can encourage states to amend their FRLs to incorporate sanctions for nonperformance. The FRLs (or related legislation) would hold accountable those individuals who were responsible for any breaches of the law. Examples of sanctions include financial reimbursement and loss of employment. The Brazilian framework again provides an example, and while it has not been formally tested, observers see it as having changed civil servant behavior (see Webb, 2004).

Looking forward to the next Finance Commission, the center may also consider grants that are more tightly linked to fiscal performance, to provide a stronger institutional sanction. This could prove especially useful for states that ultimately do not abide by TFC recommendations. Brazil, Ecuador, and Peru have all embedded provisions to suspend grants in their FRLs due to poor fiscal performance. In Peru, the suspension is temporary (for 90 days), and this can avoid worsening the subnational authorities’ fiscal situation. Brazil is seen as having enjoyed good success with its framework, and indeed subnational balances have improved markedly since its introduction (see Webb, 2004). Constitutional issues arise in India, however, insofar as the revenue share cannot be subject to conditions. An option might then be to reduce the revenue share, and to use the freed funds to create an equivalent (and ex ante fixed) grant for each state. This grant could be suspended at least temporarily if a subnational government fails to meet its fiscal adjustment targets.

Conclusions

This chapter finds evidence that institutional factors have played an important role in explaining both the differences in fiscal performance across Indian states as well as the deterioration in their combined deficit over time. The results of the econometric analysis suggest that states with greater access to central government transfers have tended to have larger deficits, and this negative relationship has been amplified the higher a state’s reliance on central government loans. This finding reflects the fact that states may have little incentive to rein in deficits when they expect—on the basis of their past experience—central government bailouts.

The recommendations of the TFC represent a step forward in addressing these adverse incentives and the long-standing problems with state finances in India. However, other steps are needed to supplement the framework. The extra resources available to states will give adjustment a significant push, but a stronger adjustment strategy, both in the aggregate and for individual states, would improve overall and individual state prospects. Strengthening borrowing controls will be critical to outcomes, and in this context there is a need to tighten administrative controls, to remove impediments to more market discipline, and to reform the small savings schemes. At the same time, the conditionality on fiscal performance could be strengthened, with a focus in particular on sanctions at the state level for not meeting targets, information provision, and sanctions by the center (in the form of withheld grants) for fiscal performance shortfalls.

References

    AhmadEhtishamLiKepingThomasRichardson and RajuSingh2002“Recentralization in China?”IMF Working Paper No. 02/168 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

    AhmadEhtishamMariaAlbino-War and RajuSingh2005“Subnational Public Financial Management: Institutions and Macroeconomic Considerations,”IMF Working Paper No. 05/108 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

    Dabla-NorrisEra and PaulWade2002“The Challenge of Fiscal Decentralization in Transition Countries,”IMF Working Paper No. 02/103 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

    EbelRobert D. and SerdarYilmaz2002“On the Measurement and Impact of Fiscal Decentralization,”Policy Research Working Paper No. 2809 (Washington: World Bank).

    • Search Google Scholar
    • Export Citation

    Government of India2004Report of the Twelfth Finance Commission 2005–10 (New Delhi).

    KurianN. J.2005“Debt Relief for States,”Economic and Political WeeklyVol. 40No. 31 pp. 342934.

    PurfieldCatriona2004“The Decentralization Dilemma in India,”IMF Working Paper No. 04/32 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

    RaoM. Govinda and NirvikarSingh2001“Federalism in India: Political Economy and Reform,”Department of Economics Working Paper No. 484 (Santa Cruz, California: University of California).

    • Search Google Scholar
    • Export Citation

    RoddenJonathan2002“The Dilemma of Fiscal Federalism: Grants and Fiscal Performance around the World,”American Journal of Political ScienceVol. 46No. 3 pp. 67087.

    • Search Google Scholar
    • Export Citation

    TanziVito2000“On Fiscal Federalism: Issues to Worry About”paper presented at a conference on“Fiscal Decentralization” at the International Monetary FundWashingtonNovember20.

    • Search Google Scholar
    • Export Citation

    WebbSteven2004“Fiscal Responsibility Laws for Subnational Discipline: The Latin American Experience,”Policy Research Working Paper No. 3309 (Washington: World Bank).

    • Search Google Scholar
    • Export Citation
1The Finance Commission is a constitutional body established every five years to determine how federal revenues should be shared between the central and state governments, and among the states. The Planning Commission administers funds to states in support of their five-year development plans.
3The index developed by Rodden (2002) assesses the extent to which higher levels of government place constraints on subnational borrowing and whether subnational governments can tap financing via their ownership of public enterprises and banks.
4See Purfield (2004) for further details.
5The GMM estimator uses first differences to remove fixed effects in the error terms, and instrumental variable estimation, where the instruments are the lagged explanatory variables (in differences) and the dependent variable lagged twice to control for endogeneity. An Arellano-Bond test rejects the null of no first-order autocorrelation in the differenced residuals, and it is not possible to reject the null of no second-order autocorrelation.
6States are in a better initial position than the TFC envisioned: the expected 4½ percent of GDP overall deficit and 2 percent of GDP revenue deficit for 2004/05 are well above the provisional outturn estimates. However, the TFC used optimistic growth and real interest rate assumptions—about 1 percent above and ¾ percent below the respective historical averages—suggesting that forecast debt improvements could have been optimistic.
7The debt stabilizing primary balance can be calculated as the nominal interest rate minus the nominal growth rate multiplied by the initial debt ratio and deflated by the nominal growth rate.
8Data were not available to assess special category states. However, their average debt ratio in 2004/05 was 13 percent of GDP higher than the average for general category states, suggesting that the problem might be more acute for many of them. It is also noteworthy that the high growth scenario does not improve the outcome, since interest and growth rates are projected to move up in tandem.
9The residual adjustment for two of the states declines to under ¼ percent of GDP if they simply maintain their current deficit, rather than loosening to achieve TFC targets.
10Another category of state-those that have already achieved the FRL deficit targets—may choose not to participate altogether. This would allow them to raise their overall deficit (and thus capital spending), which would otherwise be capped at present levels by TFC requirements.

    Other Resources Citing This Publication