D K SRIVASTAVA

I

edented growth averaging more than 8 per cent of GDP per annum since 2003-04, the draft approach paper (AP) prepared by the Planning Commission aims at 8.5 per cent of growth over the five-year period, 2007-12. While a reasonable amount of plan resources become available with high growth and fiscal correction, the AP notes that there is a problem of “time-phasing”, which leads to a backloading of plan resources that become available relatively more towards the end of the plan period rather than the initial years. It argues that in order to get a better inter-year spread of resources for the plan, a pause in pursuing the targets stipulated in the Fiscal Responsibility and Budget Management (FRBM) Act and the related rules of the central government is needed. The AP argues that some excess over the FRBM Act targets could be allowed for the central government by considering these as cyclically adjusted. These issues also link to the broader debate in India on the need and role of fiscal responsibility legislation, the beneficial multiplier effects of public investment, the crowding-in and crowding-out controversies in analyses of interdependence between public and private investment, and the impact of the fiscal deficit on interest rates [see, for example, Rakshit 2005a and 2005b; Pattnaik 2001; Pinto and Zahir 2004 and Rangarajan and Srivastava 2005].

While the AP is a well drafted document, some issues are worth taking note of. First, the “time-phasing” problem that it alludes to is an outcome of its specific assumptions about constant growth and constant tax buoyancy and the path of corrections in non-plan revenue expenditure. Second, the AP pleads for cyclically adjusted definition of the fiscal deficit but continues to work with growth rates that are constant throughout the five-year period. Third, the AP does not take into account the fiscal and economic impact of the award of the Sixth Pay Commission on central and state finances that is likely to overtake its projections. Fourth, the AP argues that a limit on fiscal deficit is a constraint but does not appreciate the benefit of fiscal correction that has already had a positive impact on the growth rate by a turnaround of the public sector saving rate. Fifth, in the AP, growth in the economy is independent of the size of the plan. The plan size is a spinoff of growth, but growth is not shown to depend on the plan size.

This article is organised as follows. Section I reviews the experience of the 1990s and the early years of the present decade in order to examine the relevance of fiscal and revenue deficit targets stipulated by the centre’s FRBM Act and the fiscal restructuring plan suggested by the Twelfth Finance Commission (TFC). Section II examines the equivalence of revenue deficit and net saving of government administration in the economy. Section III considers the issues related to working with cyclically adjusted fiscal deficit targets. Section IV discusses the prospects of growth of the Indian economy during the plan period. Section V looks at the rationale of fiscal responsibility legislation, in the light of selected international experience and the scope for flexibility in India. Section VI examines the rationale of the assumptions used in the AP projections and provides a set of alternative projections by changing some critical parameters that gives a better inter-year spread of plan resources over the Eleventh Plan period. Section VII provides concluding observations.

I Fiscal Deficit and Debt: Lessons from Recent Past

The AP claims that

The scope for increasing the fiscal deficitas a means of financing Plan expenditureis constrained by the Fiscal Responsibilityand Budget Management (FRBM) Actwhich requires the central government toreduce the fiscal deficit to 3 per cent of GDP by 2009 and also to bring the revenuedeficit to zero in that year. Similar actshave been passed by most of the states.This forces the combined fiscal deficit of the centre and the states to be limited to 6 per cent of GDP from 2008-09 onwards.Far from any increase in fiscal deficit asa means of mobilising resources we areconstrained to work with a reduction of the fiscal deficit of around 1 percentage pointof GDP to be achieved in the first two years of the Plan.

A higher level of fiscal deficit can, it is argued, result in higher level of primary expenditure, that is, total expenditure excluding interest payments. This can have a beneficial effect on growth and employment if the economy is demand constrained and resources are not fully employed. However, this strategy may not always be effective as shown by our experience in the early years of this decade when we went through a severe slow down in spite of having the highest levels of fiscal deficit relative to GDP. In those years, as shown below, high levels of fiscal deficits succeeded in reducing both primary and capital expenditure rather than increasing it. The experience after 2003-04 shows that a fall in the fiscal deficit can release resources for primary expenditure by reducing debt and interest payments relative to GDP. Table 1 indicates the profile of the fiscal deficit, interest payment, revenue receipts, total, primary, and capital expenditures all relative to GDP at current market prices for two periods, 1987-88 to 1990-91 and 1999-2000 to 2002-03. Both these periods are characterised by levels of fiscal deficit of about 9 per cent of GDP or more, on average.

The higher level of the fiscal deficit in the latter period did not necessarily deliver a higher level of government primary and capital expenditure compared to the previous period. Comparing, for example, the fiscal deficit of about 9 per cent or more relative to GDP during 1987-91 and 200004, while the fiscal deficit ratio is marginally higher in the latter period, both primary expenditure and capital expenditure are much lower, and the margin of fall is higher than the fall in revenue receipts relative to GDP. This was due to a much higher debt level relative to GDP in the latter period, resulting in a large part of the fiscal deficit getting pre-empted by interest payments leaving a much smaller balance for primary expenditure. The period 2000-03 witnessed a severe economic slow down in spite of the high fiscal deficit. The real growth rates were 4, 5.3 and 3.6 per cent, respectively, in 2000-01, 2001-02 revenue plan for the centre and states, which and 2002-03. An effective countercyclical can accommodate the revenue-intensive policy could not be put in place as it required resource needs for the social sector. fiscal deficits of levels much higher than the average of 9.4 per cent to make any

impact. By the end of 2002-03, the ratio

of government liabilities to GDP ratio had

become about 81 per cent. It is easier to mount a countercyclical policy when the The AP argues that while often grants debt-GDP ratio is low and one can are given for creating assets, it is taken as use a sharp rise in fiscal deficit from its revenue expenditure in the account of the trend level to significantly increase government giving the grants. This leads government’s primary expenditure. The to showing a higher revenue deficit in the debt-GDP ratio continued to rise until 2004-donor government’s account even though 05 reaching a peak of about 82.5 per cent the expenditure is leading to creation of before it started falling as fiscal deficits assets. However, the AP does not go on were brought down. One central objective to suggest that the definition or scope of of TFC’s restructuring plan was to bring revenue expenditure should be changed on the debt-GDP ratio down. In the first year this account. Instead, it argues that the of the TFC award period, it fell to 79.5 revenue deficit target in FRBMA should per cent. With continued fiscal correction, not be insisted upon. It observes: “A more the debt-GDP ratio has continued to fall basic approach is to question the rationale since then. Experience of recent years has of including the revenue deficit as a fiscal shown that with lower fiscal deficits, growth control measure. Unlike the fiscal deficit, has become stronger. Further, the fall of which at least has a clear economic implithe debt-GDP ratio allows through a fall cation, because it determines the increase in the interest payments relative to GDP, in public debt, the revenue deficit is a pure a release of resources for augmenting the accounting construct with no linkage to

Table 1: Combined Account of Central and State Governments – Expenditure and Deficit Trends: 1987-88 to 2004-05

Revenue Capital Interest Primary Revenue Revenue Fiscal Expenditure Expenditure Payments Expenditure Receipts Deficit Deficit

1987-88 21.73 6.56 3.67 24.63 18.86 2.87 9.08 1988-89 21.31 5.85 3.90 23.27 18.39 2.93 8.52 1989-90 22.15 6.01 4.22 23.94 18.98 3.17 8.85 1990-91 21.62 5.31 4.40 22.53 17.46 4.16 9.30 Average (1987-88 to 1990-91)(I) 21.71 5.93 4.05 23.59 18.42 3.28 8.94 1999-00 23.15 3.44 5.68 20.91 16.89 6.27 9.46 2000-01 23.23 2.92 5.88 20.27 16.75 6.48 9.21 2001-02 23.61 3.09 6.20 20.50 16.70 6.91 9.67 2002-03 23.75 2.98 6.27 20.46 17.17 6.58 9.28 Average (2000-01 to 2003-04) (II) 23.44 3.11 6.01 20.54 16.88 6.56 9.41 II-I 1.73 -2.83 1.96 -3.06 -1.55 3.28 0.47

Source: Indian Public Finance Statistics, Ministry of Finance, GoI and National Income Accounts, CSO.

III Countercyclical Roleof Fiscal Deficit

The AP argues that the fiscal deficit target in the centre’s FRBM Act should be defined in countercyclical terms so that the impact of downturns in growth, when actual growth rate falls below its trend value can be minimised by incurring a fiscal deficit relative to GDP above its trend value. This argument should be considered as acceptable but not necessarily applicable for 2007-08 and 2008-09. In Rangarajan and Srivastava (2005), it is observed that

Chart 1: Domestic Saving and Revenue Deficit

-----0.00 –1.00 –2.00 –3.00 –4.00 –5.00 –6.00 –7.00 –8.00 1993-94 1994-95 1995-96 1996-97 1997-98 1998-99 1999-00 2000-01 2001-02 2002-03 2003-04 2004-05 • Net domestic saving by government administration (percentage of GDP) +Revenue account balance (percentage of GDP)

Chart 2: Investment by Private and Public Sector Relative to GDP: Long-Term Profile

20.0

18.0

16.0

14.0

12.0

10.0

8.0

6.0

4.0

2.0

0.0

IP UB

IP VT

It is also argued that the limit of combined fiscal deficit of central and state government at 6 per cent “…should be considered as the relevant target over the cycle”.

There is a need to bring suitable changes in the rules associated with the FRBM Act to accommodate this. However, this flexibility should be used only when some clearly defined preconditions are satisfied.

The AP has argued that such a flexibility is needed for 2007-08 and 2008-09. There is nothing in the AP or in the assumptions of growth that warrants the expectation of a downturn in the first two years. The union finance minister in his letter of observations on the AP raises three points as to why a cyclically adjusted deficit target adjustment should not be taken up. “First, it will undermine credibility and be interpreted as changing the rules when the going got tough…Second, cyclical adjustment makes a large difference when a deficit caused by a slowdown is matched by a surplus generated in an upswing in the business cycle…Third, it is doubtful that the current conjuncture of the Indian economy can be characterised as the downswing of the business cycle.”

For an effective use of countercyclical policy, we require, first, that there is an effective framework for forecasting cyclical movements in the economy. Once an impending downturn induced by demand deficiency is recognised, a departure from the trend or structural level of fiscal deficit should be permitted, making sure that effective correction is done in the upswing so that the trend fiscal deficit does not exceed the permitted level. Second, a ceiling and floor of the extent of variation from the trend or mean value of fiscal deficit relative to GDP should be specified in the rules linked to the FRBM Act.

At the present juncture, it is difficult to subscribe to the view that the Indian economy is likely to face a recession in 2007-08 or 2008-09. In fact, the TFC had suggested that fiscal stabilisation should be undertaken in two phases: an adjustment period followed by a stabilisation phase. In the adjustment phase, the debt-GDP ratio should be brought down substantially to reduce the burden of interest payments. This will facilitate following a countercyclical policy when it is needed. With fiscal correction, the debt-GDP ratio is now falling steadily making it easier to take corrective action when a real need arises.

the ex ante demand for resources, that is investment and supply of resources, that is, domestic saving and current account deficit puts pressure on interest rate and/ or the exchange rate. A medium-term policy should provide for reasonable balance in the demand for and supply of investible resources, which may both be growing. This will allow for an increase in the growth rate while the interest rate remains stable.

In India, the highest sectoral investment-GDP ratio is that of the household sector followed by the private corporate sector. The public sector has been able to show the lowest investment rate in recent past. Table 3 shows the public investment rate reached its lowest level of 6.2 per cent of GDP in 2002-03. In the TFC restructuring programme, public investment should improve to about 8 per cent, consisting of 6 to 6.5 per cent for government administration and about 2 per cent for the nondepartmental enterprises. This is feasible when the revenue deficit becomes zero and all the fiscal deficit up to a ceiling of 6 per cent is spent on capital expenditure, augmented further by some non-debt capital receipts.

The private corporate sector investment has increased from the low of 5.6 per cent in 2001-02 to more than 8 per cent of GDP in recent years. There has also been an increase in the household sector investment. In fact, while the household sector saving and investment have both increased, the excess of their saving over their own investment, which is equivalent to their saving in financial form has not increased. As shown in Table 4, it has remained around 10.5 per cent. This also explains why the TFC had fixed the aggregate fiscal deficit of the central and state governments at 6 per cent of GDP. With 8 percentage points out of the available surplus of 10.5 per cent of GDP being claimed by the public sector (six for government administration and two for non-departmental enterprises), for the private sector demand

IV Growth and the Approach Paper

The AP does not show any link between growth and the plan size. Growth takes place independent of the plan size. If the plan size and GDP growth affect each other, this interdependence should be modelled in a macro framework for determining these jointly. In the AP projections, the plan size comes out almost as a residual. That is why different plan sizes are consistent with the same growth rate across years in any given simulation. The driver of growth in the AP discussion is investment, of which, plan capital outlay is a very small component. A more general treatment would make growth a function of the entire plan outlay, revenue and capital, or the entire primary expenditure of government.

Growth in AP depends on aggregate investment and the incremental capitaloutput ratio. Investment requires to be financed by the available domestic savings and current account deficit. Mismatches in

Note: * Includes a share of investment in precious goods not shown here. Source: (Basic Data): National Income Accounts.

V Role of Fiscal ResponsibilityLegislation

In India, for the central finances, a FRBM Act was enacted in 2003. Several states have also enacted fiscal responsibility legislations. The central government has also framed rules under the FRBM Act. The act and the rules have provided for the elimination of the revenue deficit by 2008-09, with 0.5 percentage point of GDP as the minimum annual reduction target, and the fiscal deficit to be brought to a level of 3 per cent of GDP, with 0.3 percentage point of GDP, as the minimum annual reduction target. The FRBM Act has some built-in flexibility in achieving revenue and fiscal deficit reduction targets as there is a provision that the specified limits may be exceeded “due to exemplary reasons such as considerations of national security and national calamity”. The act has also provided that the “Reserve Bank of India may subscribe to the primary issues to the Central Government Securities” for specified reasons.

In the international context, the main institutional reforms for controlling the growth of debt and deficit relate to

(a) formal deficit and debt rules,(b) expenditure limits, and (c) requirements of transparency. In regard to the first, apart from the Maastricht Treaty norms read with the Growth and Stability Pact, the United Kingdom introduced in 1997 a golden rule along with a debt rule (“sustainable investment rule”), capping the ratio of net debt to GDP over the cycle at 40 per cent. In Switzerland, an expenditure rule was introduced at the federal level in 2003 aiming at keeping the budget balance close to zero over the cycle and setting a ceiling for expenditure, which cannot exceed cyclically-adjusted revenue. Empirical evidence for OECD countries suggests that fiscal adjustments are more successful when based on the retrenchment of current expenditure, rather than on hiking revenue and/or cutting back public investment. McDermott and Wescott (1996) and Alesina and Perotti (1997) argue that governments that are able to cut the politically more sensitive components of the budget (public employment, social security, welfare programmes) may signal that they are more committed to sustained fiscal adjustment. They argue that fiscal consolidation based on expenditure cuts, especially transfers and government wages, are more likely to succeed in reducing the debt ratio.

There has been the argument in the Indian context, that we should operate a golden rule without putting a ceiling on fiscal deficit. This should be avoided because the annualised return on government investment is much less than the cost of borrowing for the government. Borrowing, even if used entirely for government investment, cannot be serviced fully out of budgetary returns on investment. It has to be serviced largely out of other current tax and non-tax receipts. While the TFC has recommended a golden rule, by targeting balance on revenue account, it is subject to a ceiling on fiscal deficit.

Table 5: Salient Changes in Central Finances over Plan Period

Total revenue receipts 10.02 11.72 1.69 Interest payments 3.47 2.37 -1.10 Subsidies 1.25 0.65 -0.60 Central plan revenue

expenditure 2.67 4.17 1.50 Plan capital expenditure 0.72 2.03 1.31 Total plan expenditure 4.29 7.10 2.81 Capital expenditure 1.88 3.19 1.31

Source: Basic Data.

As referred to earlier, even the golden rule in UK is accompanied by a debt ceiling.

One important advantage of a deficit ceiling is that it introduces a hard budget constraint and forces the government to prune unproductive expenditures and substitute these by productive expenditures. As long as open ended borrowing is available, it constitutes a soft, least effort source of revenue for the government of the day, which will overstate the benefits of increasing expenditure and keep postponing the hard decisions for correcting unproductive and ill-targeted subsidies.

Some important modifications are needed in the assumptions made in AP for the projections of central and state finances in order to derive the size of the central plan outlay. The modified projections eliminate the time-phasing problem. A similar exercise is not taken up for the state plan outlay since even in the AP projections, the problem of a fall in the plan outlay arises only in 2008-09, which gets eliminated by using either a higher central tax buoyancy or a higher growth assumption. Tax buoyancies: The AP makes two alternative assumptions for tax buoyancy of central revenue at 1.25 to 1.15 in different simulations. The annual buoyancies of gross central taxes in the four years preceding 2006-07 were 2.11, 1.4, 1.5 and 1.6. In 2006-07 BE, it is likely to exceed 1.4. There is thus no reason to downgrade the central tax buoyancy to 1.25 or 1.15. As the coverage of services increases and the direct taxes continue to perform well, central taxes may still do well in the plan period. However, as more and more SEZs take off, there would be some erosion of central tax buoyancy. Accordingly, we have allowed the central tax buoyancy to be equal to 1.4 in 2007-08 letting it go down to 1.20 by 2011-12. Pay and allowances: The AP assumes a constant 10 per cent growth rate in pay and allowances for government employees throughout the plan period. The main challenge in the Eleventh Plan period is going to be the surge in salaries and pensions as a result of the impact of the Sixth Pay Commission. We allow for a differential growth of 10, 15, 20,15, and 8 per cent in the respective 5 years of the plan. The governments have some degree of flexibility in managing the time of payment of arrears. Also, it is expected that the impact would be less than that of the Fifth Pay Commission, because of the merger of 50 per cent of DA that has already been given. Pensions: The AP does not use pensions as a separate item in the case of the centre. These are merged with other non-plan expenditure, which appears to grow at the same rate as nominal GDP. For pensions also, we need to allow for a surge as a result of the award of the Sixth Pay Commission. A similar growth profile as for pay and allowances is given for pensions. Share of states in central taxes: The AP has taken states’ share in central taxes as 29 per cent. The actual recommended share applicable to the divisible taxes by the TFC is 30.5. However, in the calculation of divisible central taxes several deductions are made including cost of collection and surcharges and cesses. These deductions are quite large. The share of states in gross central tax revenues are shown in the union budget 2006-07 as 25.5 and 25.7 for 2005-06 RE and 2006-07 BE, respectively. We have taken it as 26 per cent for the projection period. Non-debt capital receipts: In the AP calculations, no account is taken of non-debt capital receipts, which consist of repayments and proceeds from disinvestment, if any. Although as percentage of GDP, the amount is small, it allows capital expenditure to be marginally above the fiscal deficit levels after revenue account balance is achieved. Interest rate: The interest rate used for the calculation of interest payments in the AP projections is mentioned as the marginal interest rate, that is, interest rate at which government can undertake current borrowing. However, government makes interest payments in the current year according to the interest rates in the past at which debt was contracted, which is currently being serviced. Therefore a weighted average of interest rates of the past according the rate-composition of the outstanding debt needs to be applied to derive interest payments. If the marginal rate is less than the average rate, the average rate will fall to the extent old debt is replaced at the lower rate in addition to current borrowing as borrowing is done on gross basis including repayments falling due in the current year. The debt-swap exercises in the past have expedited the process of bringing the effective rate closer to current interest rates. The outstanding

VII Concluding Observations

The Eleventh Plan Approach Paper highlights the need for expanding the revenue plan in addition to augmenting public investment for expanding expenditure on social services. It argues for redefining revenue expenditure as capital expenditure to overcome the need for maintaining revenue deficit targets of central and state governments. Although achievable growth of 8.5 to 9 per cent releases reasonable amount of resources for the plan, the AP projections highlight some time phasing problems and ask for a pause in pursuing the FRBM Act targets or for redefining of fiscal deficit targets in countercyclical terms. The AP also argues that there is no link between revenue deficit and saving and investment. If so, an insistence on achieving a balance on revenue account is not called for. In this context, this article makes the following main points.

(1) The targets for fiscal responsibility legislations were determined in the light of the experience of the 1990s where government debt and interest payments relative to GDP increased enormously following the award of the Fifth Pay Commission. With an inordinately high level of debt, and corresponding high interest payments liabilities, governments allowed public investment to steadily fall. Further, even high fiscal deficits could not increase primary expenditure of government making it difficult to launch any countercyclical intervention even when we faced inordinately low growth rates in the first three years of this decade.

(2) With recent fiscal correction, as fiscal deficits were lowered, interest payments fell, revenue deficit fell and government savings increased. With the interest rates remaining moderate, after the inordinate rise in the late 1990s and improvement in household saving and investment, growth rates improved and the debt-GDP ratio started falling after 2004-05.

(3) With fiscal correction continuing and in conjunction with improvement in private investment, the highest ever growth rates at about 8 per cent have been experienced for four years at a stretch.

(4) There is a clear one-to-one relationship between the combined revenue deficit of central and sub-national governments and the net saving of government administration and departmental enterprises. The improvement on account of revenue deficit alone has enabled a rise in the saving rate of 4 percentage points during 2001-02 to 2004-05.

(5) As fiscal correction continues, the falling interest payments relative to GDP release revenue plan resources, which is a particular need for augmenting the social services. The falling revenue deficit leads to releasing a larger share of the fiscal deficit for plan capital expenditure.

(6) The time phasing problems of the AP projections are the outcome of specific assumptions. Some of the parameters like revenue buoyancy and phasing of salary increases are at least partially under the control of the government. These problems can be overcome, if the need arises without disturbing the fiscal adjustment already underway.

(7) There are some shortcomings in the centre’s FRBM Act. It is useful to define fiscal deficit targets in countercyclical terms as is done in many countries. The application of countercyclical intervention can be strengthened by having reliable forecasting models for anticipating downturns.

(8) If we successfully achieve a 9 per cent growth with 5 per cent inflation rate, we will eventually need to uplift the fiscal deficit target because the existing targets were derived with respect to a nominal growth rate of 12 per cent. However, this should be done after the debt-GDP ratio has fallen to levels consistent with the fiscal deficit target. It can be undertaken earlier if the level of the household sector’s savings in the financial form as a percentage of GDP increases above 11 per cent on a sustained basis and private investment is not able to absorb it.

(9) It is best to adhere to the revenue deficit target. This is specified in the FRBM Act rather than the rules. It also makes better economic sense for the government not to use up available savings for current or consumption expenditure.

m

Email: srivastava@mse.ac.in

Note

1 I had occasion to recently discuss this issue with C Rangarajan who pointed out that the slight increase in deviation in the combined revenue deficit and net saving of government administration and departmental enterprises in recent years is due to the surpluses of the local bodies. I thank him for this insight. I would like to add that any errors in the relationship highlighted in this section are entirely mine.

References

Alesina, A and R Perotti (1997): ‘Fiscal Adjustments in OECD Countries: Composition and Macroeconomic Effects’, IMF Staff Papers, Vol 44, No 2, pp 210-48.

McDermott, J and R F Wescott (1996): ‘An Empirical Analysis of Fiscal Adjustments’, IMF Staff Papers, Vol 43, No 4, pp 725-53.

Patnaik, Prabhat (2001): ‘Fiscal Deficits and Real Interest Rates’, Economic and Political Weekly, April 12-20.

Pinto, Brian and Farah Zahir (2004): ‘Why Fiscal Adjustment Now?’, Economic and Political Weekly, Vol 39, March 6.

Rakshit, Mihir (2005a): ‘Budget Deficit: Sustainability, Solvency, and Optimally’ in A Bagchi (ed), Readings in Public Finance, OUP.

– (2005b): ‘Some Analytics and Empirics of Fiscal Restructuring in India’, Economic and Political Weekly, July 30.

Ram Mohan, T T, Ravindra H Dholakia and Navendu Karan (2005): ‘Is India’s Central Debt Sustainable? Revisting an Old Debate’, Economic and Political Weekly, March 5.

Rangarajan, C and D K Srivastava (2003): ‘Dynamics of Debt Accumulation in India: Impact of Primary Deficit, Growth and Interest Rate’, Economic and Political Weekly, Vol 38, No 46, November.