Vertical Sharing and Horizontal Distribution of Resources: The Equity and Efficiency Trade-off

1 Vertical Sharing of Central Taxes

With regard to vertical sharing of resources, there are two reference points from which the scheme can be examined. The first is with reference to the constitutional assignment of responsibilities. The second is with reference to the historical long-term trend of stability in the shares of the centre and the states in combined revenue receipts. As for the first point of reference, it is clear that the central government not only dominates some of the heads on the concurrent list but also spends excessively on heads that are on the state list under Schedule 7 of the Constitution. The UFCs do not want to substantially change this pattern as it has emerged over a long period of dispensation of responsibilities. Any change that they recommend is at best incremental. I nstead they try to maintain the observed stability in the shares

november 27, 2010 vol xlv no 48

THIRTEENTH FINANCE COMMISSION

of the centre and the states in combined revenue receipts. The THFC has also explicitly done this.

The literature on vertical sharing of resources has taken note of the long-term stability observed in the relative shares of the centre and the states in combined revenue receipts and in combined revenue expenditure (see, for instance, Rangarajan and Srivastava 2008). The THFC took note of this, considering it to be a desirable feature of transfer arrangements. It observed, “We are of the view that such fiscal stability be maintained during our award period”. As shown in Rangarajan and Srivastava (2008), an increase in the states’ share in shareable central taxes is needed in periods when the centre’s tax buoyancy is expected to be higher that that of the states. In particular, the three determinants for maintaining the observed fiscal stability are the buoyancies of central and state taxes and the initial share of states’ own revenues in total revenues.

The THFC observes that the buoyancy of central taxes has been higher than that of state taxes, thereby highlighting the need for increasing, albeit by a small margin, the share of states in central tax revenues. It accordingly recommends that the share of states in the net proceeds of shareable central taxes be raised from 30.5% to 32%. One important change since the Twelfth Finance Commission (TWFC) is that the goods under the Additional Duties of Excise (Goods of Special Importance) Act, 1957 have been exempted from the payment of duty under the Act from 1 March 2006. The three goods covered under the tax rental agreement – textiles, tobacco and sugar – continue to remain in the list of declared goods under the Central Sales Tax Act, 1956 and will remain subject to prescribed ceiling rates in case states decide to levy value added tax (VAT) on these commodities. Accordingly, the applicable share of states under the TWFC recommendations had been 30.5% of the sharable pool of central taxes. The THFC has increased this to 32%. No portion of this need be earmarked as attributable to additional duties of excise in lieu of sales tax.

As Table 1 shows, the product of the growth rate and central tax buoyancy over combined tax buoyancy, taking the average for 2004-05 to 2008-09, is 1.8 percentage points. As such the THFC was amply justified in raising the share of states by 1.5 percentage points. Since the observed stability refers to total revenue receipts, raising the benchmark share of the states in the

Table 1: Tax Buoyancies and GDP Growth Rate – Recent Years

Source (basic data): Indian Public Finance Statistics and National Income Accounts.

Economic & Political Weekly

EPW

c entre’s gross revenue receipts also seems justified. The only point to remember is that these historical benchmark figures are supposed to serve as a guide to the award period of the THFC, 2010-11 to 2014-15.

Although the states have been requesting that central cesses and surcharges be included in the divisible pool, the THFC does not recommend this. However, it does recommend that the centre review the current surcharges and cesses with a view to reducing their share in gross tax revenues. The share of states in combined revenue receipts does not depend only on UFC transfers but also on other channels of transfer. The Eleventh Finance Commission (EFC) for the first time recommended an indicative benchmark for all revenue account transfers, at 37.5% of the centre’s gross revenue receipts. This was progressively raised by the TWFC to 38% and to 39.5% by the THFC. Thus the THFC has increased both the states’ share of the shareable pool of central taxes and the i ndicative benchmark for all central transfers from the centre’s gross revenue receipts by 1.5 percentage points.

The vertical share also has a bearing on the horizontal distribution of resources. The higher the vertical share of the states, the lower the weight to the equalising component of tax revenue sharing, as with the distance formula for horizontal distribution.

2 Horizontal Distribution of Resources

The twin objectives that need to be served in the horizontal distribution of resources are equity and efficiency. There is a tradeoff between the two, but if the distribution of resources is “equalising”, it is supposed to serve both objectives. Equalisation transfers serve equity since they make up for the deficiency in fiscal capacity but not in tax effort. They also serve efficiency as people do not cause congestion in a limited number of locations in search for better public and merit services such as health and e ducation since service standards are equalised across states subject to the condition that citizens are entitled to similar levels of services if they are willing to pay the same level of taxes per unit of their respective tax bases or fiscal capacities. Migration is then driven by the search for better income-earning opportunities, which also promotes efficiency.

Criteria for Sharing Central Taxes

Table 2 summarises the criteria for determining the inter se shares of states in tax devolution, along with the weights assigned to them. As shown in Rangarajan and Srivastava (2008) and earlier in Srivastava and Aggarwal (1994), these criteria serve different objectives and there is an implicit axiomatic basis for the use and weighted combinations

Table 2: Criteria and Weights for Tax

of these criteria. A good de-

Devolution (percentages)

sign for transfers adminis-Criteria EFC TWFC THFC tered through tax revenue Population 10.0 25.0 25.0

sharing can be consistent Area 7.5 10.0 10.0

with the “equalisation” prin-Distance/capacity distance 62.5 50.0 47.5

ciple, which has guided

Fiscal discipline 7.5 7.5 17.5

transfers in some well-estab-

Tax effort 5.0 7.5

lished fiscal transfer systems

Index of infrastructure 7.5

such as those in Canada and

3 Design of Grants

In this section, we look at the scheme of grants proposed by the THFC. Although the Constitution envisages that the amounts of grants be specified in absolute amounts, year-wise and state-wise,

Economic & Political Weekly

EPW

this could not be done for two forward-looking incentive grants – one related to the goods and services tax (GST) and the other for measures aimed at improving outcomes. The total amount involved in these two grants is Rs 60,000 crore. This may be justified in a dynamic context like India’s. Apart from revenue gap grants, the THFC has recommended several new grants such as those related to performance, improvement of outcomes, and promotion of environment-friendly measures. Some of the earlier grants have been continued, notably those for health. For some, such as education grants, the formulation has been changed. T able 4 shows the grants-in-aid to the revenues of states recommended by the THFC for the period 2011-15.

Table 4: Grants-in-Aid to States (Rs crore)

Source: Finance Commission (2009), Report of the THFC (2010-2015), Vol 1, December.

Deficit Grant

It is the determination of revenue deficit grants that has been subjected to criticism in the literature, mainly for following a “gap-filling” approach that leads to adverse incentives. The THFC has followed an assessment procedure with normative elements that has resulted in only eight of the special category states meriting nonplan revenue gap grants (Table 5). The general category states and the other special category states – Assam, Sikkim and Uttarakhand – are a ssessed out of revenue gap grants. This is a result both of the higher tax devolution amount and the assessment norms on own tax and non-tax sources as well as permitted e xpenditures. The adverse

Table 5: Non-Plan Revenue Deficit Grant (Rs crore)

THIRTEENTH FINANCE COMMISSION

i ncentives of gap filling are therefore limited only to a few special category states and the effective d eterminant of transfer to the general category states is the scheme of sharing central tax revenues.

Education Grants

Education grants were introduced by the TWFC. The THFC has continued these but adopted the Sarva Shiksha Abhiyan (SSA) norms for determining the amounts for individual states. The SSA began in 2001-02 with a matching fund requirement of 15% from the states. Until 2006-07, the matching fund requirement was 25%. It increased progressively to 35% in 2007-08 and 2008-09 and to 40% in 2009-10. It is expected to go up to 45% in 2010-11 and 50% in 2011-12, the terminal year of the Eleventh Five-Year Plan. The THFC argues that this will provide some fiscal space for the states to meet a part of the additional resources required to implement the Right to Education Act, 2009. It has recommended a grant of 15% of the estimated SSA expenditure of each state for the award period. The north- eastern states are required to provide only 10% from their r esources as their share for the SSA. Their education grants have been accordingly determined. The health grants have, however, been discontinued.

Environment-Related Grants

The THFC has suggested a bouquet of environmental grants in view of the terms of reference asking it to take into account “the need to manage ecology, environment and climate change consistent with sustainable development”. It has recommended three areas for support – forests, renewable energy, and water sector management.

The forests of India constitute the first line of defence against pollution resulting from economic activity, whether of agricultural or industrial origin, and the TWFC had introduced a forest-related grant of Rs 1,000 crore to be distributed among the states in accordance with their share of the total forested acreage in the country. The THFC forest grant formula takes into consideration three factors – one, the share of the total forest area in the country in a particular state; two, states where the share of forested area in their total area is greater than the national average; and three, the quality of forests in each state, as measured by density. The forest grants are subject to a release and monitoring mechanism after the first two years. The grants are linked to progress on approval of working plans, and the entire amount will be r eleased only after approval of more than 80% of the working plans of a state. Until this is achieved, releases shall be in the r atio of the number of working plans approved to 80% of the number of working plans for a state.

The second of the environmental grants is for encouraging the production of renewable energy. It has an incentive component comprising two sub-components – one, an incentive for achievement in installed capacity addition (over a four-year period) r elative to unachieved potential (weight of 25%); and two, an incentive for achievement in installed capacity addition (over a four-year period) relative to the aggregate of installed capacity addition across all states (weight of 75%).

The third component of environmental grants is for water s ector management. The THFC has recommended setting up a Water Regulatory Authority in each state and specifying a minimum level of recovery of water charges. The environment- promoting grants are welcome, even though there may be complications in implementing all the conditionalities. If anything, these initiatives need to be strengthened by larger amounts and less conditionalities.

Grants for Improving Outcomes

Another set of new grants introduced by the THFC is grants for improving outcomes. There are six outcomes to which these r elate – one, reduction in infant mortality rates; two, improvement in the supply of justice; three, an incentive for issuing UIDs; four, a district innovation fund; five, improvement of statistical systems at the state and district levels; and six, compiling an e mployee and pension database. It is worth remembering that outcomes do not exclusively depend on state government efforts. They also depend on the role of two other major players, the c entral government and the private sector, at least for some of the outcomes in the above list.

Grants for State-Specific Needs

Like previous UFCs, the THFC has also recommended grants for state-specific needs covering a wide range of activities. Table 6 gives details of these grants.

4 Determining the Vertical and Equalising Content of Transfers

In the end what matters is the overall impact of the design of fi scal transfers. For this purpose, in this section we decompose the recommended transfers into three components – vertical transfers; equalising transfers; and transfers for special needs and incentives. Using the notations used in Section 3, let the per capita income

Table 6: Grants-in-Aid for State-Specific Needs (Rs crore)

Source: Finance commission (2009), THFC Report (2010-2015), Vol 1, December.

november 27, 2010 vol xlv no 48

THIRTEENTH FINANCE COMMISSION

(GSDP) of the states arranged in ascending order be denoted by y1, y2 … y and the corresponding population be d enoted by N1, N2 ...

n

N. If the vertical fiscal is measured with reference to the richest

n

state, in per capita terms, it may be defined as [e - a.y] (assuming

n

that e > a.y), where e is the per capita expenditure norm, a is the

n

average tax-effort, and y* is the per capita fiscal capacity of the benchmark state (Haryana). If e is exogenously or normatively determined, the per capita transfer to the benchmark state is given by

[e - ay*].

Since every state gets at least the amount [e - a.y] in terms of

n

per capita transfers, we may write the total vertical transfer as [e - a. y*] = [e - a. y*] ¦Ni …(3)

¦Ni All other states have a lower fiscal capacity and would get an amount, in per capita terms, higher than that obtained by the highest income state in a progressive scheme of transfers under the axiom of horizontal equity. The transfers to these states can be seen as consisting of a vertical component equal to the per capita transfer to the highest income state, an “equalising” component due to deficiency in fiscal capacity, and a residual that r eflects cost disabilities and other special need considerations. Thus, for any state i, the per capita transfer can be decomposed into three components reflecting transfers made on account of vertical imbalance; transfers on account of equalisation; and the residual component. So per capita transfers to a state can be w ritten as = (e - a y*) + (ay* - ayi) + resi for a state where resi t 0. ...(4)

t i

Here, e is the permitted per capita expenditure norm, a is average tax effort, and “res” is the residual reflecting other cost disabilities and special need considerations, as also incentive components. The term (ay* - ayi) represents equalisation transfers. The states may be divided into two groups. One would be where the per capita recommended transfer consists of three components – vertical, equalising, and special needs and incentives – as pointed out above. But there could be other states, where after vertical transfers are taken out, the balance falls short of their equalisation entitlement and nothing is left for special needs and incentives. Let the shortfall in such cases be defi. In their case, we may write per capita transfers as

t i = (e - a y*) + (a y* - ayi) - defi ...(5)

In both cases, we can multiply the per capita transfers by r espective populations to get total transfers. By adding up the two sets, we get the total transfers (TT) as

TT = ¦Niti = ¦Ni (e-a y*) + [¦Ni (ay*-ayi) -¦Ni defi ] + ¦Ni resi

Here, defi > 0 for states that get less than their equalisation entitlement and resi > 0 for states that get more than their equalisation entitlement.

The total transfers can be divided into three components, and the respective shares of each in total transfers may be written as

Economic & Political Weekly

EPW

taken as a macro proxy of the state base for own tax revenues.

It is seen that the per capita transfer for the highest benchmark state, Haryana, is Rs 1,686. The total vertical transfer is estimated to be Rs 1,82,923 crore, which is nearly 54% of total transfers. Rs 1,25,440 crore is devoted to equalising transfers, which will be able to achieve 90.5% of the needed equalisation. This is commendable. But we need to remember that equalisation r equirements will be substantially higher if they are worked out in absolute amounts for 2012-13, which is the middle year of the THFC award period. The share of transfers devoted to incentives and special needs is less than 10% of the total transfers. The equalisation transfer is calculated and added for each state as d etermined by [Ni.(a).(y* - yi)], where the symbols have the same meaning as earlier. In particular, “a” is the average tax price, which is equal to the average own tax revenue to GSDP ratio of the states, or 7.476% (as given in Annexure 7.9 of the THFC R eport). The details of the calculations and the related decomposition are given in Appendix Tables 1 and 2 (p 70). Table 7 summarises the relative shares of the three components – vertical, equalising, and special needs – in total transfers.

Several qualifications apply to these estimates. First, the equalisation benchmark is calculated with a revenue side a pproach and expenditure side considerations do not figure in it. Second, a macro base reflecting fiscal capacity like the per capita GSDP is used. Third, the highest per capita GSDP among states, excluding Goa, is used as the benchmark. Fourth, shortfalls from the equalisation benchmarks are equally weighted.

5 Concluding Observations

The THFC has designed a scheme of transfers for both its vertical and horizontal dimensions that has some of the desirable features of a suitable arrangement. For vertical transfers, the THFC

Table 7: THFC Recommended Transfers – Vertical and Horizontal Components (Rs crore)

THIRTEENTH FINANCE COMMISSION

Finance Commission, New Delhi. of service tax is also prepared putting Jammu and References – (2004): Report of the Twelfth Finance Commis-Kashmir share to be zero and adjusting the share sion, New Delhi.

Aubut, Julie and F Vaillancourt (2001): “Using GDP in

of other states pro-rata. Rangarajan, C and D K Srivastava (2004a): “Fiscal

Equalisation Calculations: Are There Meaningful Mea-

surement Issues”, Institute of Intergovernmental Re-

lations, Working Paper, Queens University, Ontario.

Barro, Stephen (2001): “Macroeconomic versus RTS Meas-tute of Public Finance and Policy, New Delhi.

ures of Fiscal Capacity: Theoretical Foundations and Im

A = R/¦Niyi/ = (Rg + Rs)/ μ (Ng + Ns) plications for Canada”, Institute of Intergovernmental and Relevance to India”, Economic & Political or a = (Rg/μ (Ng + Ns) + Rs/μ (Ng + Ns) Relations, Working Paper, Queens University, Ontario. Weekly, Vol 40, No 27, pp 3709-722.

Since, a = R/μ. Nand a = μ. NBoadway, R and F Flatters (1982): “Efficiency and – (2008): “Reforming India’s Fiscal Transfer System: A = (Rg /μg. N).(μg. N)/(μ (Ng+N) + (Rss/μ. Equalisation Payments in a Federal System of Gov-Resolving Vertical and Horizontal Imbalances”, Eco

N).(μ. N)/(μ (N+ N) ernment: A Synthesis and Extension of Recent nomic & Political Weekly, 7 June, and Working Paper

R esults”, Journal of Economics, 15(4), pp 613-33. No 31, Madras School of Econo mics, Chennai.

Courchene, T J (1984): “Equalisation Payments: Past, Pre -Srivastava, D K and Pawan K Aggrawal (1994): “Reve+ N)