Taxation Policy Since 1991 Economic Reforms

views updated


TAXATION POLICY SINCE 1991 ECONOMIC REFORMS A comparison of the current structures of India's main central government taxes with those prevailing before 1991 indicates that, following international trends, there has been a sizable scaling back of rates in income, excise, and trade taxes. During this period, states also attempted to harmonize their sales tax rates and, most importantly, introduced a value-added tax (VAT) on 1 April 2005, comprising perhaps the most important subnational tax reform since the formation of the Indian Republic in 1950. The base of the central government's service tax has been expanded steadily, though its full coordination into a national VAT remains to be accomplished. At the subnational level, an agreement among states to cut back incentives and exemptions met with partial success. The VAT should improve its adherence.

Before 1991 India's overall tax structure had been broadly inefficient and quite inequitable. By international standards, the income tax rates had been high, and there was no VAT at the central level, except on a selective basis from the mid-1980s. The consumption tax base was narrow, with services excluded from the tax base, and customs duties were very high yet riddled with complex exemptions. Selected export duties reduced the international competitiveness of traditional exports. At the subnational level, state sales taxes caused heavy excess burdens due to input taxes getting built into the prices of final commodities, resulting in tax-on-tax, or cascading. The changes in India's tax structure are generally agreed to have led to improvements in its efficiency and equity.

Nevertheless, one cost of the improvements has been the government's inability to make up for revenue loss from rate decreases because of insufficient base expansion. The central government's tax revenue collected since 1994 declined by 1 percent of gross domestic product (GDP) from what had been collected previously. Declines in customs and excise revenues were not compensated by the increase in income tax revenues. Some sunset tax exemption clauses were extended and new incentives crept in, despite the scaling back of central tax

Corporate tax rates in India for selected years
(in percent)
 1990–19911992–19931995–19961997–19982001–20022004–2005 (1)
(1) Subject to a surcharge of 2.5 percent. Capital gains are taxed at 20 percent plus a surcharge of 2.5 percent (which is exempted if reinvested in primary securities).
SOURCE: Ministry of Finance, Government of India.
Domestic company504540353535
Foreign company656555484540

incentives in the newly emerging economy. Attempts by the tax administration to expand the taxpayer net through registration drives and a new set of requirements for filing tax returns were initially successful. But further improvements would depend on the efficiency with which newly legislated information returns from third-parties is utilized and associated computer techniques implemented.

Though the states' tax collections improved somewhat, they could not fully compensate for the central government's tax revenue decline, so that the combined central and states tax/GDP ratio fell during the decade. Overall, the ramifications for the consolidated fiscal deficit and, in turn, for public debt could be significant. There are expectations that the VAT will be revenue enhancing. However, there may be an initial period of revenue loss since the VAT rates for all states are the same and, for some high-revenue states, the VAT rates are not revenue-neutral when compared to their earlier sales tax rates. As a result, the central government has agreed to compensate states for revenue loss in the initial three years.

Major Changes in Central Tax Structure

Income tax

By the mid-1990s, many developing countries had emerged from the reform process with much lower and fewer individual income tax rates, typically 15, 25, and 35 percent. Even India legislated comparable rates of 10, 20, and 30 percent in 1997–1998. Both the rates, and their number and dispersion, were reduced on efficiency grounds. Across the developing world, for example in East Asia and Latin America, corporate income tax rates were slashed. The scaling back of corporate income tax rates reflected, to some extent, the twin objectives of administrative feasibility and better tax compliance, but was motivated in particular by the forces of globalization and the increased international movement of capital. In India, corporate income tax rates for both domestic and foreign companies have been reduced to 35 percent and 40 percent respectively (see Table 1).

Insufficiency in streamlining exemptions and incentives has adversely affected the full potential of revenue productivity in both the individual and corporate income tax. The coverage of tax incentives includes savings generation, regional development, capital investment, labor employment, research and technology, infrastructure development, exports, and charities, among others. The outcome has been a thinning out of the overall income tax base. However, income tax revenue in terms of GDP has steadily improved, reflecting administrative improvements, an expansion in the taxpayer net, and, possibly, favorable supply-side effects.

While industry tends to favor income tax incentives, these incentives have tended to benefit large entities, resulting in inequity within the corporate sector. The effective corporate tax rate is, therefore, skewed among companies. Highly generous depreciation rates were scaled back in 2005–2006 and that should correct for some of this problem. There is little doubt that without base broadening, income tax revenue is unlikely to be able to make up fully for the revenue losses emanating from structural reforms of production and trade taxes.

Central excises and customs

Central excises essentially operate as a VAT that has evolved over almost two decades, with a small beginning in 1986–1987, when a VAT-type credit mechanism for selected raw materials was introduced for the production of specified goods. In 1994–1995, capital goods were made creditable. The emerging quasi-VAT structure was termed Modified VAT or MODVAT. With a further effort to reduce the main rates to only two—8 and 16 percent—it was renamed the Central VAT, or CENVAT, in 2001.

Thus, the excises have been transformed to a VAT structure that is comparable to the successes and foibles of the VATs of most countries, with its administration carried out by the Customs and Excises Department of the Ministry of Finance. The main, and quite important, difference is that while most countries that have introduced the VAT have tended to do so as a one-shot preparation and implementation package, India has done so in a seemingly deliberate, learning-by-doing approach. This provides an interesting alternative to the usual rapid approach to VAT introduction that has sometimes entailed strong opposition from the representative taxpayer in many parts of the world. Another crucial difference from other countries is that the base of CENVAT is truncated to manufacturing, given the taxation assignment of manufacturing only (and not sales) to the central government by the Constitution of India. This has led to much litigation by businesses on the definition of manufacturing to truncate its definition to avoid tax. The CENVAT base also gets eroded by various exemptions. More than 200 pages of the standard excise tariff, of some 700 pages, comprise exemptions. Each exemption has many entries, conditions, and lists, in turn containing hundreds of items in each list.

The rate structure of customs duties is widely recognized to have been rapidly scaled back over the last decade, the peak rate declining from 150 percent in 1991–1992 to 15 percent in 2005–2006. Nevertheless, the tendency to tinker with incentives and exemptions remains alive. This inherently leads to much complexity in interpretation and administration, let alone economic distortions.

State-Level Taxes

Multiple taxes and low buoyancy

The taxing powers of Indian states include a plethora of minor taxes and one major source of tax revenue, the sales tax, recently replaced by the VAT. In the major states (14 out of a total of 25), tax revenue has represented approximately 7 percent of the state domestic product in recent years. Indirect taxes include state excise duties, taxes on vehicles, purchase tax, entertainment tax, and some surcharges. The sales tax represents approximately 60 percent of total tax revenue, while excises are the second most important revenue source, in particular, on potable alcohol. Thus, Indian states have been assigned mainly indirect taxes by the Constitution. Direct tax powers include stamp duty and registration fee, profession tax, and an income tax on agriculture. The last is usually viewed as insufficiently exploited, while the profession tax is basically a fee with a low nominal ceiling imposed by the central government. Only the stamp duty and registration fee could be said to have been revenue productive among the direct taxes.

A disturbing factor has been the low buoyancy of revenue (i.e., the percentage response of tax revenue, including discretionary changes, to a percentage change in GDP) from these various taxes. A tax-reform commission of the Government of Karnataka (2001) estimated that the buoyancies of various taxes fell sharply during the 1990s. Recommendations made in selected state-level tax reform studies—including Government of Karnataka (2000), Government of Madhya Pradesh (2001), Government of Maharashtra (2000), and Government of West Bengal (2001)—offer many ideas and directions in which structural reform could be undertaken, with a focus on extending the tax base.

Replacing sales tax with a VAT

Given the primary importance of the sales tax in revenue generation and its recent conversion to a state-level VAT, the main concerns and prospects are examined here. The general dilemma for a subnational VAT is that introducing a VAT at the central level is far easier than at the state level. Countries with the intention of introducing a subnational VAT have grappled with one main problem, that of structuring the VAT as a consumption tax, generally without an appropriate solution. Either they have introduced a VAT that is not a fully consumption-type (Brazil), or one that is administratively complex (Canada); or they have desisted from introducing it at all (United States), or have been debating its appropriate form for a considerable time (Argentina).

In India, all states—through an empowered committee of state finance ministers—have agreed to have the same rate structure—4 and 12.5 percent—for the VAT and the same exemptions. However, each state is allowed to have ten additional exempted items of local importance from a list of about forty items. Some goods such as petroleum products are outside the VAT base with a floor rate of 20 percent. Immediately after the VAT's introduction, however, the states felt compelled to add to the exemption list reflecting popular demand. This was especially to counterbalance the opposition-ruled states that decided not to introduce the VAT at the last moment. Out of a total of thirty-five, twenty-three states (and centrally administered Union Territories [UTs]) have introduced the VAT. Two UTs should introduce them soon, while two did not even have a sales tax. The remaining eight states that did not introduce the VAT are states with opposition governments.

The states have gone halfway in their attempt to move to a destination based VAT, though interstate trade continues to be taxed at 4 percent as before. Under the VAT, states will not give input tax credit in their own states for inputs bought in another state. However, when a good is exported to another state, input tax credit will be given against such export. It is anticipated that input tax credit for inputs imported from another state should be in place in 2006 or 2007. However, a solution would need to be found regarding how to capture the lost revenue currently collected from this source. A computerized system for exchange of information among states, under development, would also need to be operational prior to implementation.

In sum, the states have made an impressive beginning in the introduction of their VAT, and as in the case of the central government's CENVAT, future evolution in its structure should result in its sophistication. The central government, in its catalytic role, has demonstrated its willingness to participate in the states' tax reform process. The combination demonstrates exemplary fiscal federal cooperation.

Tax Administration

Three issues in tax administration are closely connected to the success of any tax policy reform: expanding the taxpayer register; computerization; and implementation of the state-level VAT. Arguably the most successful action that has been undertaken in the area of central tax administration has been an impressive expansion of the taxpayer net for the income tax. In the mid-1990s, the taxpayer roll included some 14 million taxpayers, of which 10 million to 11 million were current. However, a rudimentary calculation of the potential number of taxpayers would be as follows. Of the total population of 1 billion, the taxable population is approximately 300 million. With an average household size of 5, that would imply 60 million potential taxpayers. Discounting 10 million for taxable agricultural households would result finally in a net 50 million taxable households. Thus only about 20 percent of potential taxpayers were within the taxpayer net.

In the second half of the 1990s, a voluntary disclosure program required individual income earners who possessed certain characteristics, such as ownership of property and telephones, and trips undertaken abroad, to register even if their taxable income was nil. The characteristics were further expanded with time so that more individuals would be required to register. By 2000 the taxpayer register had increased to over 20 million. Thus, within a relatively short period, the number of potential taxpayers doubled, an objective that had been unattainable for decades. In 2005 the register contained approximately 30 million, of which about 25 million are understood to be current.

While it has been found that the number of assessees has not constrained tax collection, nevertheless there is a need to allocate adequate administrative resources to bring the medium to small taxpayer into the tax net. The strategy must include a credible threat of audits for all taxpayers. This draws attention to the second issue of the extent of resources the tax administration can devote to administering the returns of relatively small taxpayers. If the taxpayers that were rapidly brought into the tax net realized that their chances of being assessed or audited were very small, then having a larger taxpayer register may not result in any significant increase in revenue collection in the long run. Needless to say, having a large taxpayer unit (LTU), which facilitates payment of all taxes by large taxpayers through a single window, is also important. India does not yet have an LTU but the intention to set one up has been announced.

In order to more effectively tackle the problem of tax evasion in a modern tax administration, a computerized information system needs to be quickly developed. For example, the income tax department collects a wide array of evidence during the course of any investigation. In addition to information from taxpayers' returns and other information returns, a large volume of information is collected during assessment, searches and seizures, and survey operations. Third-party information has been legislated from various sources, but this could lead to revenue enhancement only if such information is successfully collated, disseminated, and verified. Currently the income tax department has initiated massive information-technology transformation that is experiencing teething problems. Complete implementation will take a year or two. Customs operations, on the other hand, are focused specifically at import-export points, catering mainly to businesses, so computerization of customs procedures has moved further on. Excise or CENVAT entails cross-checks of invoices among buyer and seller. It poses more difficult challenges, and systems development is in a nascent, discussion stage regarding alternative models of computerization.

The success of implementation of the state VAT is dependent on the computerization of VAT procedures of various states. States have progressed at differing speeds in this area. The central government is lending a helping hand to some of the smaller states in the form of turnkey projects through computer training, installation, and implementation. A further challenge remains in the development of a comprehensive information exchange system among states that should enable the states to cross-check tax data across state borders, once the VAT evolves to the destination principle. The cost of developing such a system is being shared between the center and the states.

The loss in the tax/GDP ratio could be made up and further resource mobilization successfully achieved only through an expansion of the tax base. Thus exemptions and incentives for both direct and indirect taxes at the central level need to be scaled back significantly. On the administration side, computerization is imperative. At the state level, there is generally a high expectation that, through a broad tax base, and by its extension to the retail arena, the VAT would be revenue productive in the medium term.

In the long term, the goal for the reform of consumption taxation must be a national VAT. Many strides have been made, though much remains to be done. The process should not stop with the introduction of the state level VAT operating side by side with the central CENVAT. A national two-tier VAT, based on the destination principle, would comprise both the central and state levels. The center and states would appropriately share services as a tax base, reflecting cross-border or intraborder consumption, respectively. The challenge would be to achieve cooperation in information sharing between the central government and the states, and among the states themselves, for the concurrent VAT to operate successfully. Only then could the national VAT comprise a comprehensive consumption tax that would cover both goods and services at all subnational levels.

Parthasarathi Shome

See alsoEconomic Reforms of 1991 ; Economy since the 1991 Economic Reforms ; Fiscal System and Policy since 1952


Dasgupta, Arindam, and Dilip Mookherjee. Incentives and Institutional Reform in Tax Enforcement. New Delhi: Oxford University Press, 1998.

Ebrill, Liam, Michael Keen, Jean-Paul Bodin, and Victoria Summers. The Modern VAT. Washington D.C.: International Monetary Fund, 2001.

Government of India. Report of the Group of Officials and Experts on Taxation of Interstate Sales. New Delhi: Ministry of Finance, 1996.

——. Report of Expert Group on Service Tax. New Delhi: Ministry of Finance, 1999.

——. Tax Policy and Tax Administration for the Tenth Plan. Report of Advisory Group. New Delhi: Planning Commission, 2001.

Mukhopadhyay, Sukumar. Manufacturing in Central Excises. New Delhi: Centax Publications, 1995.

Ponniah, Aurobindo. "India: Budget for 2004/2005." Asia Pacific Tax Bulletin 10, no. 4 (2004): 210–211.

Shome, Parthasarathi. India's Fiscal Matters. New Delhi: Oxford University Press, 2002.

Shome, Parthasarathi, ed. Tax Policy Handbook. Washington, D.C.: International Monetary Fund, 1995.

——. Value Added Tax in India: A Progress Report. New Delhi: National Institute of Public Finance and Policy, Centax Publications, 1997.

Shome, Parthasarathi, Pawan K. Aggarwal, and Kanwarjit Singh. "Tax Evasion and Tax Administration: A Focus on Tax Deduction at Source." In Fiscal Policy, Public Policy and Governance, edited by Parthasarathi Shome. New Delhi: National Institute of Public Finance and Policy, Centax Publications, 1997.

Tait, Alan A. Value Added Tax: International Practice and Problems. Washington, D.C.: International Monetary Fund, 1988.