The paper by Elena Ianchovichina et al (December 29, 2007) presented a framework for subnational fiscal sustainability analysis and applied it to Tamil Nadu. However, the empirical application presented is unclear and so raises doubts on the veracity of their projections of the future. The recent trends in the state's fiscal situation do not corroborate their analysis and conclusions.
DISCUSSIONmarch 15, 2008 EPW Economic & Political Weekly76Fiscal Sustainability of Tamil NaduM Suresh BabuThe paper by Elena Ianchovichina et al (December 29, 2007) presented a framework for subnational fiscal sustainability analysis and applied it to Tamil Nadu. However, the empirical application presented is unclear and so raises doubts on the veracity of their projections of the future. The recent trends in the state’s fiscal situation do not corroborate their analysis and conclusions.M Suresh Babu (sbab77@hotmail.com)is with the department of humanities at IIT Madras.In their interesting paper titled ‘Sub-national Fiscal Sustainability Analysis: What Can We Learn from Tamil Nadu?’ (EPW, December 29, 2007) Ianchovichina, Liu and Nagarajan (hereafterILN) present a framework for analysing the sustain-ability of subnational fiscal policies. Tak-ing the case of Tamil Nadu the framework is empirically implemented and some like-ly future scenarios are projected. While ILN elucidate the broad contours of analys-ing subnational fiscal sustainability and the departures from a national analysis, the empirical application is plagued with vagueness raising doubts about the vera-city of their projections of the future. As much of their discussions consider data till 2003, I wish to bring to notice some re-cent data on the state’s fiscal situation which raises scepticism about the analysis and conclusions. As baseline simulation is crucial for the ensuing analysis, a clear exposition of its arrival is warranted. The discussion of ILN leads one to assume that the base line is arrived using the medium term fiscal plan (MTFP) of 2004. However, according to section 3 (1) of the Tamil Nadu Fiscal Responsibility Act, 2003 which was subse-quently amended to bring it in line with the requirements prescribed by the Twelfth Finance Commission, the govern-ment is required to place before the legis-lative assembly a MTFP along with the budget. And section 3 (2) of this Act requires that theMTFP shall set forth a multi-year rolling target for the fiscal indi-cators like revenue deficit and fiscal defi-cit, while clearly indicating the underlying assumptions made to arrive at those pro-jections. Hence, when the MTFP of an al-ternate year is used for arriving the base-line projections one ends up with a differ-ent set of results. For example, if one uses the most recentMTFP the primary surplus (denoted as x in Table 2 of ILN) portrays a different picture. Primary surplus which was projected as a deficit of Rs 643.42 crore for the year 2005-06 actually turned to be surplus of Rs 2,304.67 crore. Subse-quently the primary deficit deteriorates to Rs2,275.90 crore in the budget esti-mates for 2007-08 and it is projected to reach Rs2,395.46 crore in 2009-10. Hence the assumption of a steady primary sur-plus for the period from 2009 to 2023 looks heroic in light of the fluctuations witnessed in the earlier periods. Important QuestionsThe simulation results raise a number of important questions. The sanctity of look-ing at outcomes over a 23-year period it-self comes up first as it does not coincide with plan duration, Finance Commission’s terms and the Fiscal Responsibility Act’s objectives. In the subnation context such a time horizon seems to be too long as the centre-state relationship itself might come up for reviews. In this context, outcomes in the medium term assumes significance and a more realistic analysis could be for a shorter time span. The rationale for MTFP insteadofLTFP itself is to focus on outcomes in the medium term and not in so long time span. This is evident as the Act in the case of Tamil Nadu requires the state to contain the revenue deficit below5per cent of the total revenue receipts in 2007-08 and eliminate this deficit by 2008-09 and adhere to it there-after.As envisaged in the MTFP to elimi-nate the revenue deficit by 2008-09 and to maintain this position in 2009-10 itself come up for discussions. Hence 2026 looks far away. The simulated values appear to come out of a black box as the methodology used for arriving them does not find a place in the discussions. As there exists a possi-bility of simultaneity in macroeconomic relationships, the results are sensitive to the procedure employed. The relationship between growth, the budget constraint and “macroeconomic space” for investment is one where the direction of causation is not always unidirectional. Hence any straightforward estimation could lead to biased inferences. However, much cannot be commented on the results of ILN as the
DISCUSSIONEconomic & Political Weekly EPW march 15, 200877methodology used to infer the relation-ship is not spelt out. In growing econo-mies, the “fiscal space” for investment al-ways exists as the own tax revenue as a percentage of the state’s domestic product increases and this improves the debt serv-ice capacity. But textbook macroeconomics also tells us that growth could be acc-elerated with higher investments, even borrowed. Following ILN, subnational gov-ernments are constrained to borrow and invest as the threshold debt stress ratio is frozen at a particular level. In fact, when economies grow this threshold would also rise giving more space for additional re-source mobilisation. In any case in the era of fiscal discipline the growth in public in-vestments are slow and even the available “fiscal space” is left unutilised providing more space for private investments. The simulation results reveal only one side of the story as the authors themselves note that “this simulation does not take into account any positive effects on state revenue”. Hence the results are the out-come of a partial analysis which exagger-ates the future risks without taking into account the buoyancy in growth. The end result of such analysis is the sensitive re-sults that are amplified even with a minor deviation in the assumptions as in the case of primary deficits, which points tothe possibility of the debt burden of 50.4 per cent of gross state domestic product (GSDP) by 2026-27. While this pointsto one extreme scenario, the other one generated shows that avoiding increases in the debt burden is possible if the economy’s growth increases by 1.6 percentage points over the 2008-09 growth rate, which looks plausi-ble given the state of the economy now. Providing these extreme scenarios would hamper drawing any definite conclusions from the sensitivity analysis. Fiscal PrudenceAs noted byILN since April 2004, some of the reforms have been rolled back. Meas-ures like the restoration of free power to farmers and access to public distribution system were not in line with the fiscal dis-cipline advocated by international agen-cies. However, since then we find no signi-ficant deterioration in the fiscal situation of the state. On the contrary there have been some signs of fiscal prudence. Table 1 provides five indicators to shed light on the fiscal situation and management in re-cent times. As is evident, market borrow-ings which were used to finance the gross fiscal deficit has been reduced substantially. More than half of the fiscal deficitwasfinanced by market borrow-ings in 2005-06, which has been sub-stantiallybrought down to 33 per cent by 2007-08. It is not necessary to state the long-run implications of this, as the debt stress ratio could be well below the thresh-old levels given the current rate of growth of the economy providing more fiscal space for investments. Equally important is the reduction of the share of total out-standing liabilities of the state. Since 2005-06 there has been a decline in the total outstanding liabilities as percentage of the GSDP. Certainly the growth of the GSDP is one factor contributing to this de-cline. But it also shows that the reliance on borrowings to finance expenditure has de-clined indicating that the present growth does not come with cost in the long run. The introduction of the value added tax (VAT) and stamp duty reforms had raised doubts about the size of the state’s own tax revenue. As is evident from Table 1, the state’s own tax revenue has been keeping pace with the growth ofGSDP. It can also be noted that in an attempt to keep the deficits under control the rate of growth of non-plan non-development expenditure has been significantly brought down. As a single indicator of fiscal discipline this points to the state’s efforts to comply with the objectives of MTFP. More importantly the utilisation of borrowed funds in this ac-count has reduced as the market borrow-ings itself have come down. The decline in the rate of growth of net interest payments from 22 per cent to 2 per cent leads to two inferences; (a) control of non-plan non-development expenditure has led to the reduction in market borrowings especially of a shorter duration slowing the growth of interest outflows, and (b) a large portion of the arrears cleared in 2002-03 has reduced the debt stress levels. Whether this tendency can be continued or not depends on the maturity profile of the state’s liabilities. Table 2 provides the maturity profile of the state government’s outstanding govern-ment securities and market loans. This projects a medium-term (10 years) view of the fiscal situation of the state. As is evident, 2012 to 2014 could be years of fiscal stress for the state as a good percentageofboth theoutstandingliabilities needs to be re-tired in these years. The overall picture looks favourable to the state as a small fraction of government securitiesalong with high interest market loansmaturein the immediate years and even the shares of both the liabilities mature towards the end of the 10-year period. It should also be noted that the average interest rate on the market loansreduceswithmaturity time. With some fiscal discipline already in place, by way of reducing market borrowings to finance the deficit, a further deterioration looks unlikely in the long run even with variations intheinterestrate. Thus recent trends in the state’s fiscal situ-ation do not corroborate with the projected outcomes of ILN. I wish to draw two lessons: (a) Rolling back of the so-called critical reforms is not harmful in times of faster growth; one could view such reforms only as austerity measures in times of crisis. (b) Ambitious long-run projections based on unfounded assumptions and methods would result only in painting a bleak picture of the future when the present looks promising. Table 1: Some Indicators of Current Fiscal Situation 2005-062006-072007-08Market borrowings used to finance gross fiscal deficit (%) 53.9 35.4 33.4Total outstanding liabilities as % ofGSDP 30.2 28.4 27.8Non-plan non-development 12,854 15,439 16,570 expenditure (Rs crore) (20.1) (7.3)Net interest payments 3,761 4,617 4,704 (Rs crore) (22.7) (1.9)State’s own tax revenue (as % GSDP) 11.02 11.63 11.27Figures in brackets are variations over the previous year. Figures for 2005-06 are accounts, 2006-07 RE and 2007-08 BE. Source: compiled fromState Finances: As Study of Budgets of 2007-08, published by the Reserve Bank of India, November 2007. Table 2: Maturity Profile of State’s Major Liabilities Maturity Profile of Maturity Profile of State Outstanding State Government Market Government’s Securities Loans as of March 31, 2007 (% to Total) (% to Total)2007-08 4.965.17(12.05)2008-09 4.92 5.63(11.84)2009-10 5.34 6.44(11.43)2010-11 7.89.02(10.53)2011-12 8.51 8.88(7.67)2012-13 14.220.97(6.43)2013-14 12.63 13.03 (6.44)2014-15 14.78.54(6.67)2015-16 13.14 10.36 (7.8)2016-17 13.79 11.96 (7.59)Sources: Same as Table 1. Figures in brackets are the average interest rates.
DISCUSSIONmarch 15, 2008 EPW Economic & Political Weekly76Fiscal Sustainability of Tamil NaduM Suresh BabuThe paper by Elena Ianchovichina et al (December 29, 2007) presented a framework for subnational fiscal sustainability analysis and applied it to Tamil Nadu. However, the empirical application presented is unclear and so raises doubts on the veracity of their projections of the future. The recent trends in the state’s fiscal situation do not corroborate their analysis and conclusions.M Suresh Babu (sbab77@hotmail.com)is with the department of humanities at IIT Madras.In their interesting paper titled ‘Sub-national Fiscal Sustainability Analysis: What Can We Learn from Tamil Nadu?’ (EPW, December 29, 2007) Ianchovichina, Liu and Nagarajan (hereafterILN) present a framework for analysing the sustain-ability of subnational fiscal policies. Tak-ing the case of Tamil Nadu the framework is empirically implemented and some like-ly future scenarios are projected. While ILN elucidate the broad contours of analys-ing subnational fiscal sustainability and the departures from a national analysis, the empirical application is plagued with vagueness raising doubts about the vera-city of their projections of the future. As much of their discussions consider data till 2003, I wish to bring to notice some re-cent data on the state’s fiscal situation which raises scepticism about the analysis and conclusions. As baseline simulation is crucial for the ensuing analysis, a clear exposition of its arrival is warranted. The discussion of ILN leads one to assume that the base line is arrived using the medium term fiscal plan (MTFP) of 2004. However, according to section 3 (1) of the Tamil Nadu Fiscal Responsibility Act, 2003 which was subse-quently amended to bring it in line with the requirements prescribed by the Twelfth Finance Commission, the govern-ment is required to place before the legis-lative assembly a MTFP along with the budget. And section 3 (2) of this Act requires that theMTFP shall set forth a multi-year rolling target for the fiscal indi-cators like revenue deficit and fiscal defi-cit, while clearly indicating the underlying assumptions made to arrive at those pro-jections. Hence, when the MTFP of an al-ternate year is used for arriving the base-line projections one ends up with a differ-ent set of results. For example, if one uses the most recentMTFP the primary surplus (denoted as x in Table 2 of ILN) portrays a different picture. Primary surplus which was projected as a deficit of Rs 643.42 crore for the year 2005-06 actually turned to be surplus of Rs 2,304.67 crore. Subse-quently the primary deficit deteriorates to Rs2,275.90 crore in the budget esti-mates for 2007-08 and it is projected to reach Rs2,395.46 crore in 2009-10. Hence the assumption of a steady primary sur-plus for the period from 2009 to 2023 looks heroic in light of the fluctuations witnessed in the earlier periods. Important QuestionsThe simulation results raise a number of important questions. The sanctity of look-ing at outcomes over a 23-year period it-self comes up first as it does not coincide with plan duration, Finance Commission’s terms and the Fiscal Responsibility Act’s objectives. In the subnation context such a time horizon seems to be too long as the centre-state relationship itself might come up for reviews. In this context, outcomes in the medium term assumes significance and a more realistic analysis could be for a shorter time span. The rationale for MTFP insteadofLTFP itself is to focus on outcomes in the medium term and not in so long time span. This is evident as the Act in the case of Tamil Nadu requires the state to contain the revenue deficit below5per cent of the total revenue receipts in 2007-08 and eliminate this deficit by 2008-09 and adhere to it there-after.As envisaged in the MTFP to elimi-nate the revenue deficit by 2008-09 and to maintain this position in 2009-10 itself come up for discussions. Hence 2026 looks far away. The simulated values appear to come out of a black box as the methodology used for arriving them does not find a place in the discussions. As there exists a possi-bility of simultaneity in macroeconomic relationships, the results are sensitive to the procedure employed. The relationship between growth, the budget constraint and “macroeconomic space” for investment is one where the direction of causation is not always unidirectional. Hence any straightforward estimation could lead to biased inferences. However, much cannot be commented on the results of ILN as the 
DISCUSSIONEconomic & Political Weekly EPW march 15, 200877methodology used to infer the relation-ship is not spelt out. In growing econo-mies, the “fiscal space” for investment al-ways exists as the own tax revenue as a percentage of the state’s domestic product increases and this improves the debt serv-ice capacity. But textbook macroeconomics also tells us that growth could be acc-elerated with higher investments, even borrowed. Following ILN, subnational gov-ernments are constrained to borrow and invest as the threshold debt stress ratio is frozen at a particular level. In fact, when economies grow this threshold would also rise giving more space for additional re-source mobilisation. In any case in the era of fiscal discipline the growth in public in-vestments are slow and even the available “fiscal space” is left unutilised providing more space for private investments. The simulation results reveal only one side of the story as the authors themselves note that “this simulation does not take into account any positive effects on state revenue”. Hence the results are the out-come of a partial analysis which exagger-ates the future risks without taking into account the buoyancy in growth. The end result of such analysis is the sensitive re-sults that are amplified even with a minor deviation in the assumptions as in the case of primary deficits, which points tothe possibility of the debt burden of 50.4 per cent of gross state domestic product (GSDP) by 2026-27. While this pointsto one extreme scenario, the other one generated shows that avoiding increases in the debt burden is possible if the economy’s growth increases by 1.6 percentage points over the 2008-09 growth rate, which looks plausi-ble given the state of the economy now. Providing these extreme scenarios would hamper drawing any definite conclusions from the sensitivity analysis. Fiscal PrudenceAs noted byILN since April 2004, some of the reforms have been rolled back. Meas-ures like the restoration of free power to farmers and access to public distribution system were not in line with the fiscal dis-cipline advocated by international agen-cies. However, since then we find no signi-ficant deterioration in the fiscal situation of the state. On the contrary there have been some signs of fiscal prudence. Table 1 provides five indicators to shed light on the fiscal situation and management in re-cent times. As is evident, market borrow-ings which were used to finance the gross fiscal deficit has been reduced substantially. More than half of the fiscal deficitwasfinanced by market borrow-ings in 2005-06, which has been sub-stantiallybrought down to 33 per cent by 2007-08. It is not necessary to state the long-run implications of this, as the debt stress ratio could be well below the thresh-old levels given the current rate of growth of the economy providing more fiscal space for investments. Equally important is the reduction of the share of total out-standing liabilities of the state. Since 2005-06 there has been a decline in the total outstanding liabilities as percentage of the GSDP. Certainly the growth of the GSDP is one factor contributing to this de-cline. But it also shows that the reliance on borrowings to finance expenditure has de-clined indicating that the present growth does not come with cost in the long run. The introduction of the value added tax (VAT) and stamp duty reforms had raised doubts about the size of the state’s own tax revenue. As is evident from Table 1, the state’s own tax revenue has been keeping pace with the growth ofGSDP. It can also be noted that in an attempt to keep the deficits under control the rate of growth of non-plan non-development expenditure has been significantly brought down. As a single indicator of fiscal discipline this points to the state’s efforts to comply with the objectives of MTFP. More importantly the utilisation of borrowed funds in this ac-count has reduced as the market borrow-ings itself have come down. The decline in the rate of growth of net interest payments from 22 per cent to 2 per cent leads to two inferences; (a) control of non-plan non-development expenditure has led to the reduction in market borrowings especially of a shorter duration slowing the growth of interest outflows, and (b) a large portion of the arrears cleared in 2002-03 has reduced the debt stress levels. Whether this tendency can be continued or not depends on the maturity profile of the state’s liabilities. Table 2 provides the maturity profile of the state government’s outstanding govern-ment securities and market loans. This projects a medium-term (10 years) view of the fiscal situation of the state. As is evident, 2012 to 2014 could be years of fiscal stress for the state as a good percentageofboth theoutstandingliabilities needs to be re-tired in these years. The overall picture looks favourable to the state as a small fraction of government securitiesalong with high interest market loansmaturein the immediate years and even the shares of both the liabilities mature towards the end of the 10-year period. It should also be noted that the average interest rate on the market loansreduceswithmaturity time. With some fiscal discipline already in place, by way of reducing market borrowings to finance the deficit, a further deterioration looks unlikely in the long run even with variations intheinterestrate. Thus recent trends in the state’s fiscal situ-ation do not corroborate with the projected outcomes of ILN. I wish to draw two lessons: (a) Rolling back of the so-called critical reforms is not harmful in times of faster growth; one could view such reforms only as austerity measures in times of crisis. (b) Ambitious long-run projections based on unfounded assumptions and methods would result only in painting a bleak picture of the future when the present looks promising. Table 1: Some Indicators of Current Fiscal Situation 2005-062006-072007-08Market borrowings used to finance gross fiscal deficit (%) 53.9 35.4 33.4Total outstanding liabilities as % ofGSDP 30.2 28.4 27.8Non-plan non-development 12,854 15,439 16,570 expenditure (Rs crore) (20.1) (7.3)Net interest payments 3,761 4,617 4,704 (Rs crore) (22.7) (1.9)State’s own tax revenue (as % GSDP) 11.02 11.63 11.27Figures in brackets are variations over the previous year. Figures for 2005-06 are accounts, 2006-07 RE and 2007-08 BE. Source: compiled fromState Finances: As Study of Budgets of 2007-08, published by the Reserve Bank of India, November 2007. Table 2: Maturity Profile of State’s Major Liabilities Maturity Profile of Maturity Profile of State Outstanding State Government Market Government’s Securities Loans as of March 31, 2007 (% to Total) (% to Total)2007-08 4.965.17(12.05)2008-09 4.92 5.63(11.84)2009-10 5.34 6.44(11.43)2010-11 7.89.02(10.53)2011-12 8.51 8.88(7.67)2012-13 14.220.97(6.43)2013-14 12.63 13.03 (6.44)2014-15 14.78.54(6.67)2015-16 13.14 10.36 (7.8)2016-17 13.79 11.96 (7.59)Sources: Same as Table 1. Figures in brackets are the average interest rates.
