Utilisation of Government Borrowings in Major Indian States – Economic and Political Weekly

The term sustainability has been used with different connotations under different circumstances in multilateral policy discussions. The fiscal sustainability of state governments is different from that of the central government, due to the differential allocation of powers in the Constitution of India. These legal restrictions leave subnational governments unable to adjust their primary fiscal balances. Yet, state governments, in order to maintain economic stability and achieve economic growth, can manipulate fiscal policy alone as the monetary policy is uniform for all states, in that it is centrally determined. As far as state government finances in India are concerned, subnational financial sustainability can be understood as the capacity to generate adequate resources to afford their expenses on a sustained basis. This involves two main dimensions: (i) fiscal sustainability, and (ii) debt sustainability. This article focuses on fiscal sustainability.

Fiscal sustainability analysis is an approach linked to intertemporal budget constraints, subject to the condition that current deficits need to be outweighed in the future by surplus (Quintanilla 2009). The borrowing operations of the government should not muddle the objectives of maintaining stability in the economy. In the case of the Indian economy, there is a paucity of resources necessary for development. Revenues, generated by way of taxes, are mostly used for the normal expenditure of the government, resulting in the governments heavy reliance on borrowed funds to finance development of the economy (Mallick 2005). If the cost or burden of borrowings is offset by the gains received, then the debt would be less burdensome. If the rise in public borrowings is accompanied by a simultaneous increase in productive capacity, investment, asset creation, and national income in the economy, then the debt will not hinder the stability of the economy. Thus, the income-generating effect of public debt makes the rising debt less troublesome (Bhatia 1994).

The actual point to be considered here then is whether the rate of growth of the economy is satisfactory or not. A lower growth rate along with higher rate of public debt is oppressive. The analysis of the net economic burden of public debt would depend on the extent to which the resources generated are used efficiently in order to enable the economy to achieve self-sustained growth. But due to inherent difficulties engaged in isolating the contribution of borrowings to economic growth and segregating the benefits, one can only analyse the purpose for which borrowed funds have been spent. Therefore, to assess the performance of public debt, one of the important factors to be considered is the utilisation pattern of government funds.

In light of the above discussion, this article shall attempt to analyse the pattern of state government finances and borrowed funds.

Database and Methodology

The data for this research has been primarily collected from the Reserve Bank of India (RBI) bulletins and annual reports on State Government Finances: A Study of State Budgets. Seventeen major Indian states have been considered for the present analysis. These include Andhra Pradesh (ANP), Assam (ASM), Bihar (BHR), Gujarat (GUJ), Haryana (HAR), Himachal Pradesh (HMP), Jammu and Kashmir (JNK), Kerala (KRL), Karnataka (KTK), Madhya Pradesh (MDP), Maharashtra (MHR), Odisha (ODS), Punjab (PNB), Rajasthan (RAJ), Tamil Nadu (TND), Uttar Pradesh (UTP), and West Bengal (WSB).

The data analysis includes a study of the following variables for the states under study: revenue deficits, fiscal deficits, loans and advances by state governments (LASG), debt receipts of state governments (DRSG), internal debt, loans and advances from central government (LACG), ways and means advances (WMA) and overdraft, small savings and provident fund (SS&PF), reserve fund, deposits and advances, and contingency fund. TheDRSGfor individual states has been calculated from 198081 to 201415. The calculations have been made by following the revised coverage used by theRBI(from 200506 onwards) for the calculation of the total budgetary liabilities of state governments, which comprises of (i) public debt, including internal debt andLACG, (ii)WMAand overdraft, (iii) public account, includingSS&PF, reserve fund, and deposits and advances, and (iv) contingency fund. All of these four components have been added up to calculateDRSG.

The data and information pertaining to various components have been statistically analysed by using percentages, ratios, averages, standard deviation, andanalysis of variance (ANOVA). An assessment of the significance of differentials in the gross fiscal deficit, as a percentage of gross state domestic product, has been made through a two-wayANOVA. Six indexes of instability were estimated for the measurement of temporal fluctuations in the utilisation pattern of total debt receipts for each of the 17 states. These indices are outlined as below:

Index1,as proposed by Sethi (2010), is based on the coefficient of variation among moving averages (as first suggested by Mahendradev [1987]).

Index2, due to Ray (1983), is estimated as the standard deviation of natural log of the ratios of successive values.

Index2= Standard Deviation (ln {Xt+1/Xt})

whereXtis the value of the variable in the current time period.

Xt+1is the value in the subsequent year.

Index3,given by Coppock (1962), uses log-variance approach, by assuming the percentage change in the values at constant rate and thereby correcting the annual changes for this as:

Index3= Antilog [(Xlog)1/2]

heren= number of observations, andm= measure of constant percentage change.

Index4,due to Glezakos (1973), basically, is the exports instability index based on absolute difference between successive values and simple linear trend value, expressed as a percentage of the mean:

.

here^1is the slope of the linear trend.

Index5,due to Glezakos (1984), is also an exports instability index based on absolute difference between successive values adjusted for exponential trend value, expressed as a percentage of mean:

.

herer= rate of growth of the variable obtained from exponential trend as:Y^t= A(1+r)t

Index6is the mean of the export instability index. It was given by Xin and Liu (2008) as

The different methodology and formulation of the indexes of instability are not bound to provide a unique picture; therefore, an intercorrelation matrix has been used to establish compatibility among these indexes as:

In order to test the compatibility among the ranks assigned to the 17 states with respect to the extent of instability in the utilisation ratio, Kendalls Coefficient of Concordance was computed by following Seigel (1988), as

wherek= set of ranks, andn= number of components.

is the sum ofTjfor all sets of rankings.

Sis the corrected sum of squares of deviations from the mean ofRj(aggregated rank overksets).

That is,

In the case of tied observations,Tjwas computed as a correction factor.

heregj= number of group of ties injthset of ranks andti= number of tied ranks inithgroup of ties.

Results and Discussion

The utilisation pattern of domestic debt is an important aspect related to the management and burden of public debt and to the development of an economy. But it is difficult to measure the real burden entailed by the growth of debt. The outstanding liabilities on account of borrowings can be analysed through the purpose for which the borrowed funds have been used towards creation of productive assets. A careful analysis of the utilisation pattern of debt receipts throws light on an important issue, whether the increasing proportion of public debt is being used for productive uses (that is, on capital outlay and loans advanced) or for unproductive or consumption expenditure. However, in the budget documents of the government, expenditure on the basis of productivity is not classified. It is rather difficult to analyse how the debt receipts received at some point of time are used for what purpose at some other point of time. Yet, it can be analysed by comparing the expenditure and receipts (both on capital and revenue accounts) of the government, which indicates the fiscal management of the government.

The higher the proportion of government expenditure on consumption, current, and administrative expenditure, the lesser resources left for productive expenditure (that is, on social, economic, and some components of general services). A higher proportion of current expenditure in total government expenditure signifies inadequacy of the government to generate adequate resources. The decline in capital outlay as a percentage of capital receipts implies that there is a relative expansion of current expenditure, at the cost of capital outlay. Also, the decline in capital expenditure, as a proportion of net capital receipts, signifies that the capital receipts are being diverted towards current expenditure, which were otherwise required to be spent on capital formation in the economy. While not all current expenditure is unproductive in nature, a major share is allocated towards defence, subsidies, and interest payments on past debt. Further, a large portion of capital receipts comprises of debt capital receipts bearing interest charges, which is further used for the meeting of current expenditure, ultimately resulting in a higher interest burden on the economy. This implies that capital receipts, which otherwise would have been used for capital expenditure, are absorbed in current expenditure.

The total revenue earned by state governments has been persistently falling short of its total expenditure for the last three decades, as they have failed to cover up its revenue expenditure from revenue receipts (Bhargava 2011). State governments try to fill their budgetary gap from capital receipts and it is evident that the major source of capital receipts is borrowings. Capital disbursements are mainly financed through the mobilisation of public debt receipts. Further, they are financed through surplus from the revenue account, deposits and advances, reserve fund, contingency fund, and recovery ofLASG.

The effective utilisation of borrowed funds can be made clear on the basis of assets and liabilities of state governments. Liabilities are required to be adequately covered by assets. However, in the case of state governments, capital outlays were observed to be less than the liabilities. Nevertheless, the whole amount of liabilities not covered by assets cannot be considered unproductive, so long as large amounts of the borrowings are for productive and developmental purposes. Therefore, what matters is not the actual size of public debt but its management in order to achieve the objectives of economic development with stability. The government mainly raises funds for three basic reasons, namely capital outlay, loans advanced, and consumption expenditure. Capital outlay is that part of the expenditure, which is mainly developmental but may also be used for the acquisition of income-yielding assets. The capital outlay portion is further categorised as general services, economic services, and social services.

Loans advanced include all kinds of loans given by state governments, public sector enterprises, government servants, etc. These loans are utilised not only for the purpose of capital formation and asset creation but also to meet the consumption or current needs of state governments (Bhatia 1994). The loans given to public sector undertakings do not give good returns, due to their inefficiency or low productivity. Therefore, most part of the debt raised has been used unproductively. So these components have not resulted in any addition to income-yielding assets. Therefore, it does not lead to any increase in the productive potential of the economy.

The assets comprise of capital outlay and loans advanced. The excess of these two components over the total liabilities of the government implies that surplus on revenue account has been used for capital expenditure. But the excess of total liabilities over capital outlay and loans advanced implies that current expenditure of the government has been financed out of the borrowings, that is, the funds have been going into the consumption expenditure of the government. Capital expenditure either results in the creation of physical financial assets or in the reduction of liabilities, whereas capital receipts, which may be debt receipts or non-debt receipts, generally implies the creation of liability.

Table 1 depicts huge variations among states in the relative shares of loans and advances plus capital outlay in total debt receipts. Utilisation (LASG+ capital outlay), as a percentage of debt receipts, was found to be the highest for Bihar, Uttar Pradesh, Jammu and Kashmir, Assam, and Madhya Pradesh respectively. For Gujarat, Odisha, Maharashtra, Karnataka, Tamil Nadu, and Himachal Pradesh, it lies between 21% and 50%. Utilisation is seen to be very less (that is, less than 20%) in Kerala, Rajasthan, Punjab, Andhra Pradesh, West Bengal, and Haryana. Further, the table reveals that Kerala, Punjab, and West Bengal have shown continuous decline in the utilisation of debt receipts since 200001. Whereas Andhra Pradesh, Haryana, and Rajasthan (also having utilisation ratio less than 20%) have shown fluctuating trends with respect to the same.

Differentials in Utilisation Pattern

Wide fluctuations have been observed to exist in the utilisation of borrowings among states and within the states, over a fairly long period of 35 years. In order to provide a comparative picture of states with regard to utilisation patterns, some basic computations have been done to examine the differentials in utilisation ratio among states as shown in Tables 2 and 3. On an average (over the years), the utilisation ratio is found to be minimum for West Bengal (31.60) and maximum for Jammu and Kashmir (134.57). However, the pattern of variability depicted by standard variation was maximum in Punjab (135.68) and minimum in Odisha (24.05). West Bengal, Kerala, and Punjab, which have a utilisation ratio of less than 20% for 201415, also have the lowest mean values among 17 states. For Punjab, the mean utilisation ratio is found to be 71.52, while the ratio is expectedly in the range of 8.95 and 134.1, associated with a confidence coefficient of 0.99. The same interval, for adjoining Haryana, ranges between 49.91 and 88.13, thus implying that position of Punjab has been relatively poor.

Table 3 reveals that the fiscal year 198283 witnessed the highest utilisation ratio in the entire 35 year-long time span, while the lowest was observed during 200304 for the states. Some improvement was seen in 200506, but it was too little and could not be sustained either, for since 201011, the mean utilisation ratio was found to be declining. The lowest levels of mean utilisation ratio have been found in the later years, starting from 200304. As basic computations of the utilisation ratio could not provide us with a unique picture, the indexes of instability have been worked out to gauge the fluctuations in utilisation ratio over time and to have a clear position of states with respect to the same.

Table 4 (p 34) exhibits the computations of various indexes of instability with regard to the utilisation ratio of states. As many as six indexes of instability were worked out. But due to the criteria of uniformity among indexes, only four indexes, namely Index2, Index3, Index4, and Index5, have been reported. The differences in the formulation of the indexes of instability were not bound to provide a unique picture. For instance, minimum instability was found in the case of Tamil Nadu as computed through Index2and Index3, in Uttar Pradesh through Index4,and Kerala as per Index5. As gauged through Index2, Index3, Index4, and Index5, the state with maximum instability was Punjab, followed by Assam and Maharashtra. The indexes could not provide us with a harmonious picture with regard to minimum instability in the utilisation pattern among states. Therefore, the nature and extent of association among different indexes have been gauged through the intercorrelation matrix.

Table 5 depicts the intercorrelation analysis performed on different values of indexes. The analysis of the table reveals a strong association between Index4and Index5(0.998), followed by Index2and Index3(0.996). Since the different indexes of instability failed to provide a concrete picture regarding the extent of instability in states with respect to utilisation pattern, we have performed ranking analysis for states on the basis of different indexes. In Table 4, rankings (given in parentheses) were assigned to different states for a given index. Then, in order to examine whether there existed any compatibility or agreement among different indexes on the basis of assigned ranks, Kendalls concordance analysis was performed on different ranks of indexes. The findings of the test provided Kendalls W coefficient (= 0.925). The testing of Kendalls coefficient turned out to be significant at 0.01% level, when tested through Chi square test. The computed value of X2at 16 degrees of freedom was 59.176, which was more than the critical value of X2at corresponding number of degrees of freedom. Significantly, Kendalls coefficient implies a high degree of concordance among the ranking of states regarding relative instability (with respect to utilisation pattern).

As evident from Table 4, on the basis of mean ranking (justified due to strong concordance therein), Punjab has witnessed high temporal fluctuations with regard to its utilisation ratio, whereas minimum instability was experienced by Uttar Pradesh. Thus, after having performed Kendalls concordance analysis, the states with minimum and maximum instability have been determined. Therefore, it is imperative to analyse the significance of the differentials existing in states, on account of the utilisation of borrowed funds. For the testing of differentials in mean utilisation ratio among 17 states, the two-wayANOVAapproach is used.

The ratios calculated in Table 2 were subjected to a two-wayANOVAapproach in order to examine the comparative performance related to the utilisation of borrowings of states. The results of the test are shown in Table 6, which reveal that significant differences exist among the 17 states in utilisation ratios over a period of 35 years. The value of theF-statistic (= 7.540) for time indicates the differentials among states in utilisation ratios at different points of time, whereas theF-statistic(= 6.926) for states signifies the differentials in the specific ratio among states for a period of 35 years. It is worth mentioning that over this period, states have shown a declining trend (Table 3), which may not be seen as a conducive sign.

The two-wayANOVAtechnique was duly coupled with Tuckeys post hoc comparisons in order to test the paired comparisons of states. Table 7 reveals a pairwise comparison among states witnessing significant mean differences in utilisation ratios. Out of 136 possible comparisons among 17 states, 24 paired comparisons turned out to be statistically significant.

As per the findings, Jammu and Kashmir witnessed the highest mean utilisation ratio in comparison to the remaining 16 states of Andhra Pradesh (by 75.23%), Assam (by 78.67%), Bihar (by 73.62%), Gujarat (by 60.22%), Haryana (by 65.54%), Himachal Pradesh (58.33%), Kerala (by 97.87%), Karnataka (by 62.01%), Madhya Pradesh (by 55.71%), Maharashtra (by 41.13%), Odisha (by 78%), Punjab (by 63.04%), Rajasthan (by 84.22%), Tamil Nadu (by 71.16%), Uttar Pradesh (by 72.12%), and West Bengal by 102.96%). The maximum gap (of 102.96% with very large P value) in utilisation ratio was found between West Bengal and Jammu and Kashmir.

As per Table 7, the mean utilisation ratio was higher for Maharashtra than that of Kerala, Rajasthan, and West Bengal. Madhya Pradesh had a significantly higher mean utilisation than Kerala (by 42.15%) and West Bengal (by 47.24%). It is worth mentioning that West Bengal has a significantly lower utilisation ratio than as many as six states, Gujarat, Himachal Pradesh, Jammu and Kashmir, Karnataka, Madhya Pradesh, and Maharashtra. This clearly indicates the relatively poor position of West Bengal on account of its deteriorating finances coupled with poor debt management. On the other hand, Jammu and Kashmir has witnessed a relatively better position, expectedly due to the status of special category state along with better management of debt servicing.

Policy Implications

This article aims at examining the pattern of utilisation of borrowings among 17 major Indian states during 198081 to 201415 by analysing broad indicators of the utilisation pattern. An attempt has also been made to examine the differentials in utilisation patterns among states during the same 35-year period by employing two-wayANOVAtechnique duly coupled with Tuckeys post hoc comparisons to establish paired comparisons among states. In order to gauge the extent of instability in the utilisation ratio for states, as many as four indexes were worked out. Due to different units of measurements in instability indexes, states were assigned ranks accordingly and an intercorrelation analysis was performed on indexes to squeeze out a unique picture regarding the agreement among indexes. The indexes were then subjected to Kendalls concordance testing to examine concordance among the rankings of states.

The analysis of indicators reveals that there has been a disproportionate rise in interest payments and current expenditure of the government, as indicated by high levels of interest payments to revenue expenditure and capital receipts, and high revenue deficits in relation to fiscal deficits. The finances of Kerala, Punjab, and West Bengal, depict that a large proportion of capital receipts is absorbed in the revenue expenditure of these states, which implies poor and unproductive utilisation of debt receipts. Basic computations of mean and standard deviation reveal that West Bengal has witnessed the lowest mean utilisation ratio and Punjab has depicted a high degree of variability as indicated by standard deviation. The position of Punjab in respect to utilisation pattern of borrowings is observed to be poorer in comparison to its adjoining state, Haryana. Similarly, the utilisation ratio is observed to be less than 20% in Kerala, Rajasthan, Punjab, Andhra Pradesh, West Bengal, and Haryana. The lowest levels of mean utilisation ratio have been found in the later years starting from 200304.

As regards the results of differentials among states in respect of utilisation are concerned, a two-wayANOVAreveals that significant differences in mean utilisation ratio exist among 17 states over a period of 35 years and as many as 24 pairs of states display significant mean difference among 136 possible comparisons. As regards the major findings, West Bengal has displayed minimum utilisation ratio and Punjab has depicted highest instability among states as gauged through different indexes of instability.

The deteriorating fiscal health of states has affected their financial strength. The above analysis shows that state government finances need the generation of additional resources in order to maintain the sustainability of their finances. State-level fiscal reforms can play a significant role, if fiscal discipline is followed by the states. Strict adherence to policy reforms and effective implementation can make a way for the same. But the policy or programmes should take into consideration the comparative situation of states. The reforms or policies need to maintain balance and harmony in the present federal set-up. Kerala, Punjab and West Bengal, facing serious fiscal deterioration, need to strictly follow fiscal rules in order to come out of the crisis. There is an urgent need to create a political will on the part of state governments to provide efficient governance and concerted efforts to ensure fiscal sustainability.

References

Bhatia, H L (1994):Indias Domestic Public Debt,Birla Economic Research Foundation, New Delhi: Bookwell Publications.

Bhargava, P K (2011):Aspects of State Finances in India,Jaipur: RBSA Publishers.

Coppock, J D (1962):International Economic Instability:The Experiences of World War II,New York: McGrawHill.

Glezakos, C (1973): Export Instability and Economic Growth: A Statistical Verification,Economic Development and Cultural Change,Vol 21, No 4, pp 67078.

(1984): Export Instability and Economic Growth: Reply,Economic Development and Cultural Change,Vol 32, No 3, April, pp 61523.

Quintanilla, J C (2009):Public Finance Sustainability in Subnational Governments, Publication No 12, United Kingdom: Debt Relief International Limited, ISBN: 978-1-903971-52-9.

Mahendradev, S (1987): Growth and Instability Foodgrains Production: An Interstate Analysis,Economic & Political Weekly,Vol 22, No 39, pp A82a92.

Mallick, H (2005): An Economic Analysis of Domestic Debt of the Central Government in India,unpublished PhD thesis,Institute for Social and Economic Change, Bangalore.

Ray, S K (1983):Growth and Instability in Indian Agriculture,Delhi: Institute of Economic Growth.

RBI (2014):State Finances: A Study of Budgets of 201314, Reserve Bank of India, Mumbai.

Seigel, S (1988):Non-Parametric Statistics for the Behavioural Sciences, New York: McGraw-Hill Book Company.

Sen, T K (2012): Recent Developments in Kerala State Finances, CDS Working Papers No 449, Trivadrum: Centre for Institute of Development Studies,https://opendocs.ids.ac.uk/opendocs/handle/123456789/3187.

Sethi, A S (2010): Some Further Aspects of Rates of Growth Computations,Journal of International Economics,Vol 2, pp 5770.

Xin, X and J Liu (2008): Geographic Concentration and Chinas Agricultural Eport Instability,World Economy,Vol 31, No 2, pp 27585.

Originally posted here:

Utilisation of Government Borrowings in Major Indian States - Economic and Political Weekly