Personal Website of R.Kannan
Students Corner - A Decade of Economic
Reforms in India - A Review

Corrective measures on the fiscal front initiated at the beginning of the 1990s produced some promising results during the first half of the decade. Expenditure growth could be curtailed leading to a decline in the fiscal deficit and the outstanding liabilities of the Government to GDP ratio. During 1990-91 to 1996-97 (excluding 1993-94), the reduction in total expenditure to GDP ratio by more than 3.5 percentage points narrowed the fiscal gap by 3 percentage points and reduced the debt-GDP ratio by over 5 percentage points. However, the fiscal consolidation even during the first half of the 1990s was brought about primarily through curtailment in capital expenditure. Decline in consumption expenditure was relatively small. From 1997-98, expenditure started rising once again, and by the year 2001-02, all the major fiscal parameters, viz., revenue deficit, fiscal deficit, and public debt rose to levels higher than those prevalent at the beginning of the reform process.

The efficacy of tax reforms for augmentation of tax revenue, expenditure correction, restructuring of public sector, public debt management policies and institutional reforms appears to be rather limited so far.

Trends in Tax Revenue

Tax reforms have generally led to a rise in tax revenue to GDP ratio across countries (Shome, 1992; Shome, 1995). In the Indian context, the expected increase in tax buoyancy a la ‘Laffer curve effect’ did not occur. Since the onset of tax reforms, the tax-GDP ratio of the Central Government has suffered a persistent decline. This has been a major drag on the reform process. The tax-GDP ratio declined from an average of 9.9 per cent during the 1980s to 9.7 per cent in the first half of the 1990s and further to 9.0 per cent in the second half of the 1990s. The pattern is, however, not the same across different types of taxes. Direct tax collection to GDP ratio rose steadily from 2.0 per cent in the 1980s to 2.3 per cent in the first half of the 1990s and further to 2.9 per cent in the latter half of the 1990s. On the other hand, the ratio of indirect tax collection to GDP declined from 7.9 per cent in the 1980s to 7.3 per cent and 6.1 in the first and second halves of the 1990s, respectively.

The decline in the tax to GDP ratio is explained by a combination of factors that led to a sharp fall in total tax buoyancy from 1.07 for the period 1981-93 to 0.96 for 1981-2001, implying buoyancy could be less than unity during the post-tax reform period 1994-2001. 3, 4 While the buoyancy of direct taxes is estimated to be higher at 1.19 for the period 1981-2001 as compared with 1.07 for the pre-tax reforms period (1981-1993), the buoyancy of indirect taxes dipped considerably to 0.88 from 1.07 in the corresponding period.

The increase in direct tax collections despite a significant cut in marginal tax rates could be attributed to the combined effect of better compliance, broadening of the tax base and increase in income. The introduction of presumptive tax, adoption of economic criteria for identification of potential taxpayers and removal of some exemptions helped in base widening. Measures such as rationalisation and simplification, both in personal tax and corporate tax, would have induced better compliance. This apart, the revision in public sector wages following the recommendation of Pay Commission is also ascribed as a contributing factor in strengthening the direct tax collection (Rao, 2000), The estimate of personal income tax buoyancy does indicate that while there was a positive impact of pay hike, it has not been statistically significant at the conventional level.5

Unlike direct taxes, rate cuts have been important factors in reducing the indirect tax collection, as there was no commensurate gain in terms of base expansion or better compliance. It was expected that the sharp cut in custom duties from a peak rate of more than 300 per cent in the period just prior to reforms to about 30 per cent in 2002-03 would lead to a net fall in custom duty collections. Fall in excise duty collections, however, came as a surprise as the rate cuts were expected to boost growth in industrial output. The less than expected buoyancy in the excise tax seems to follow from slower than expected growth in industrial output during the major part of the reform period. The rising share of services in overall GDP which largely falls outside the tax net and progressive extension of MODVAT could have affected buoyancy estimates (Mohan, 2000). Ideally, credit extended to inputs under VAT system needs to be neutralised through increase in tax rates on end-products. Instead tax rates have been scaled back leading to a fall in excise tax collection (Shome, 1997). Another important reason for reduced revenue collection from both custom and excise duties is that the reduction in rates were not accompanied by removal of concessions and exemptions (Government of India, 2001a).

A comparison of alternative buoyancy estimates with respect to GDP and the actual base indicates that though there was a marked decline in buoyancy in case of both customs and excise during the reform period, the fall was partly made up by the pick up in imports in case of customs and some base expansion or better compliance in respect of excise. In the case of custom duties, the decline in buoyancy in terms of imports was much more than in terms of GDP. On the other hand, in case of excise, there was hardly any divergence between its buoyancy in terms of manufacturing output (which is the base for excise tax) and that in terms of GDP.

Under the existing federal fiscal structure, the States’ rights to collect taxes are largely confined to indirect taxes, predominantly commodity taxes like sales tax and other indirect levies, such as State excise duties, service tax on entertainment, on betting and gambling and on passengers and goods. Direct taxes include few items like land revenue and agricultural income tax. Buoyancy estimates of sales tax (which accounts for around 60 per cent of States’ own revenue) and own tax for 15 major States during 1981 to 2000 show that the buoyancies of sales tax and own tax during the sub-period 1981 to 1993 were uniformly higher than the respective buoyancies for the full period, indicating significant fall in buoyancies during the reform period. An important reason for the fall in tax buoyancies is the competitive tax reductions by States to attract trade and industry (Government of India, 2000c). The decline in buoyancies also follows from higher growth in services, which are not adequately taxed but raises the Net State Domestic Product (NSDP). However for most of the States, buoyancies for both sales tax and own tax continued to be more than unity during the reform period (Table 4.4). Contemporaneously, for all the States combined, own tax revenue as a percentage to GDP increased from 5.1 per cent during the 1980s to 5.4 per cent during the first half of the 1990s. Though the ratio marginally declined from 5.5 per cent in 1994-95 to around 5.1 per cent in 1998-99, it improved to 5.6 per cent by 2000-01. Thus, on average, tax-GDP ratio for States during the reform period was higher than that of the 1980s.

Trends in Non-Tax Revenue

A key objective of the reform process was the augmentation of non-tax revenue by way of enhancement of user charges and returns on Government investments through restructuring of PSUs. The intention of restructuring PSUs was to improve their efficiency and thereby enhance the capacity to generate returns on Government investments. Non-tax revenue of the Central Government as a proportion to GDP recorded an improvement from 2.1 per cent in 1990-91 to 3.0 per cent in 2001-02. The trends in components of non-tax revenue reveal that increase in dividend and profits, and economic services, fully account for the improvement in Centre’s collection of non-tax revenue, as growth in other components continued to be stagnant during the reform period. Surplus transfers from the Reserve Bank, which is a major component of dividend and profits, increased from Rs.210 crore in 1990-91 to Rs.10,320 crore in 2001-02, thereby raising its share in the total from 1.8 per cent to 15.2 per cent. The size of the transfer from the Reserve Bank, inter alia, grew on account of earnings from the deployment of foreign currency assets, conversion of 4.6 per cent Treasury Bills into marketable securities and discontinuation of the practice of crediting large sums to the National Industrial Credit (LTO) Fund.

For the States, the decline in the non-tax to GDP ratio was from 1.6 per cent in 1990-91 to 1.4 per cent in 2001-02. No significant changes in their composition are discernible, barring a rise in the share of interest receipts and a distinct fall in the share of economic services.

The inability to raise user charges has been relatively more pressing for States than the Centre. The States account for nearly three-fourth of total spending on social services, and more than half of the total spending on economic services provided by the Government sector. User charges recovered from such services are much lower as compared with spending on such accounts. Our estimates of user charges for various services rendered by the State Governments indicate that cost recovery for education during the 1990s has been lowest at 1.2 per cent, followed by health (4.7 per cent) and irrigation (8.5 per cent), without any firm indication of improvement during the reform period.

Apart from low user charges on the services rendered by the Government, the sluggishness in non-tax revenue also follows from continuing inadequate returns on public investment. The outstanding investment in Central PSUs amounted to Rs.3,03,400 crore as at end-March 2000. The State Governments’ investment amounted to nearly Rs.75,000 crore in statutory corporations and another Rs.42,000 crore in the Government companies as at end-March 1999. In the case of Central PSUs, although returns on capital employed have improved from a low level of 2.5 per cent in the 1980s to 2.8 per cent in the early 1990s and further to around 5.0 per cent in the latter half of the 1990s, dividends from PSUs remain inadequate to finance future investment opportunities through internal finance. Beside operational inefficiency, the poor returns on investments in, both, Central and State PSUs also reflect the limitations of pricing policies as well as heavy implicit and explicit subsidies (Box IV.1).

In the case of State-level enterprises, returns do not cover even a fraction of their cost of funds (Government of India, 2000c). Capital invested in SEBs account for the bulk of investments made by the State Governments. Although the Electricity (Supply) Act, 1948 requires SEBs to earn a minimum yield of 3 per cent on their net fixed assets, they have rather registered a persistent negative return over the years (Box IV.2). The State Road Transport Undertakings (SRTUs), the other segment attracting major public investments, have also been a drag on the State budgets.

Resource Transfer to States from the Centre

In addition to States’ own tax and non-tax revenue, the resource base of State Governments also includes transfers from the Central Government in the form of devolution of Central taxes and grants-in-aid and contribution. Both forms of transfers to States have decelerated over time. Although successive Finance Commissions recommended devolution of higher amounts through either upward revisions of the coverage of the shareable items, or by increasing the magnitude of the States’ share (Tenth Finance Commission, 1994), the quantum of funds transferred to States as a ratio to GDP has been lower in the 1990s as compared to that of the 1980s, reflecting sluggishness or lower buoyancy of Central taxes. The Centre-State tax sharing system has been pointed out as one of the factors having disincentive and efficiency reducing effects on tax collection (Srinivasan, 2000). The main argument here is that the Centre’s efforts in collecting income tax, which is to be shared, would be less vigorous than in the case of custom duties, which accrue fully to the Central Government. The argument, however, does not receive support from the observed recent trends. Buoyancies of direct taxes estimated above showed marked improvement while that of custom duties slumped sharply during the reform period. Moreover, changes in the transfer formula by the Eleventh Finance Commission, 2000, has further reduced this possible bias, as all the taxes have become shareable with effect from 2000-01.

To sum up, the sluggishness in Central Government revenues was largely the result of lower buoyancy of indirect taxes caused by cut in tax rates without adequate expansion in the tax base. The inadequate growth in States’ revenue, on the other hand, was the result of their inability to levy appropriate user charges and tax the services sector, combined with lower Central transfers due to falling tax buoyancy.