Tax reform figures prominently in India’s plans for fiscal consolidation and to generate fiscal space for infrastructure investment. Successive Indian governments have devoted considerable effort to developing a tax reform strategy: the 2003 Kelkar Task Force laid out a strategy for direct and indirect taxation and the Fiscal Responsibility and Budget Management Act (FRBMA) road map (Ministry of Finance, 2004) and Twelfth Finance Commission report (Government of India, 2004) both highlighted how tax reform could contribute to fiscal adjustment. Increases in revenue are expected to contribute almost 3 percent of GDP in fiscal adjustment at the center, and about 1 percent of GDP in the states, permitting 1½ percent of GDP in extra annual investment at the general government level.

Tax reform figures prominently in India’s plans for fiscal consolidation and to generate fiscal space for infrastructure investment. Successive Indian governments have devoted considerable effort to developing a tax reform strategy: the 2003 Kelkar Task Force laid out a strategy for direct and indirect taxation and the Fiscal Responsibility and Budget Management Act (FRBMA) road map (Ministry of Finance, 2004) and Twelfth Finance Commission report (Government of India, 2004) both highlighted how tax reform could contribute to fiscal adjustment. Increases in revenue are expected to contribute almost 3 percent of GDP in fiscal adjustment at the center, and about 1 percent of GDP in the states, permitting 1½ percent of GDP in extra annual investment at the general government level.

International experience has shown that revenue-based consolidation strategies can be successful, but are a more difficult route to take. The earlier literature on fiscal consolidation had emphasized expenditure-based adjustment (Alesina and Perotti, 1996). More recent work has cast a more positive light on revenue-based experiences (Gupta and others, 2004; and Tsibouris and others, 2006), especially when a country begins its adjustment at a low initial revenue-to-GDP ratio and has developed well-defined tax reform plans that are phased in gradually (as in India at present). However, successful revenue-based consolidations are less common than expenditure-based consolidations even at low revenue levels, mainly reflecting constraints on the capacity of tax administrations. Countries have also had more difficulties protecting revenue-based gains against new revenue erosions and expenditure incursions.1

To date, India’s revenue strategy has shown some success. Measures have been passed to broaden the corporate income tax (CIT) base, efforts have been made to improve tax administration, and, after many years of negotiations and planning, a state-level value-added tax (VAT) has been introduced. Early returns from the VAT have been encouraging, with half-year 2005/06 indirect tax receipts up by close to 15 percent in those states implementing it. Growth in CIT and personal income tax (PIT) collections have been enough of late to offset the drag to revenues from continuing trade tax reforms; since 1993, the average effective tariff rate has dropped from 34 percent to 17 percent, and trade-related revenues have fallen by 1 percent of GDP (Figure 8.1).

Figure 8.1.
Figure 8.1.

General Government Revenue and Deficit

(In percent of GDP)

Source: Country authorities; and IMF staff estimates.

The time is ripe in India for further tax reform. To meet FRBMA deficit reduction targets, the center must raise its gross tax collection ratio by another 2½ percent of GDP, to a little over 13 percent of GDP, by 2008/09. With progress on some planned measures behind schedule, and sizable spending commitments on the horizon, new measures may need to be identified. Past international studies have emphasized that tax reforms undertaken at a measured pace within a stable macroeconomic environment—the present situation in India—stand a greater chance of success.

This chapter looks at India’s tax reform options in more detail. It first examines India’s revenue performance in international perspective to identify areas of potential collection gains. Then it considers India’s existing tax reform strategy, noting where gaps remain. The next section discusses an important gap: the level of the PIT threshold. The chapter concludes by discussing strategies for reinvigorating and sustaining base-broadening tax reforms.

India’s Revenue Performance in International Perspective

International comparisons provide insight into India’s revenue collection effort and directions for future reforms. In selecting relevant comparators, account must be taken of several factors that are key for establishing taxable capacity: (1) per capita income adjusted for purchasing power parity (PPP) (it is infeasible to tax a subsistence level of income); (2) the structure of output (certain sectors, such as agriculture, are difficult to tax); and (3) openness to trade and capital flows (a more open economy faces more mobile tax bases). From an administrative angle, it is also important to consider population—the challenge of efficiently administering a tax system grows at least in proportion with the number of taxpayers.

It is useful to consider two groups of comparators for India. In the first group are countries in India’s PPP-adjusted per capita income range that are also broadly similar to India in terms of economic structure (China, Egypt, Indonesia, Pakistan, the Philippines, Sri Lanka, and Vietnam). In the second group are countries whose structural characteristics place them at a more advanced state of development (Brazil, Colombia, Mexico, Russia, South Africa, Turkey, and Ukraine). Table 8.1 highlights the differences between these two samples.

Table 8.1.

India and Other Economies: Socioeconomic Characteristics

article image
Sources: World Bank; and IMF, International Financial Statistics.

Of 208 total jurisdictions.

Median for population.

India’s revenue collections compare very favorably to the first group of comparators (Table 8.2). Total general government revenues exceed the level realized in direct comparators by almost 2 percent of GDP. The higher performance is mainly due to indirect taxation (sales taxes and excises). India is very much in line with income tax, customs and nontax revenue collection efforts in these other countries. This good performance is not driven by comparisons against the weakest members of the group—India outperforms both China and the Philippines, the two highest income comparators in the group.

Table 8.2.

India and Other Economies: General Government Revenues

(In percent of GDP)

article image
Sources: Country authorities; and IMF staff estimates.

However, against more mature emerging market countries, there are some revealing differences (Table 8.2). These more advanced economies collect some 13 percent of GDP more in general government revenues than does India. The largest difference is in payroll (social contribution) levies: all but one of the comparators has a formal social security system funded in this way while India does not. However, differences are broad based including an almost 3 percentage point gap in the PIT collection, and a gap of about 2½ percent of GDP for goods and services taxes (GST). India also collects a little less in nontax revenues, perhaps reflecting more efficient use of state assets elsewhere.

Part of the difference against mature emerging market countries will disappear over time as the structure of India’s economy shifts. If the share of agriculture in Indian output were to fall by 13 percentage points in favor of manufacturing and services (which would move its economic structure into line with more mature emerging market countries), this could yield close to 1 percent of GDP in additional revenue. However, it would take a long time for this shift in economic structure to occur—20 years, given the present differential between agricultural and nonagricultural growth rates. And there would still be a large gap in total revenues relative to the mature emerging market countries if nothing else in the tax system changed.

Medium-Term Directions and Strategies for Tax Reform

International comparisons suggest that India’s greatest revenue-raising opportunities lie in the goods and services and income taxation areas, and prospects for strengthening each of these areas are good. In fact, plans are already well laid out in India for reform of the GST, and many aspects of the PIT and CIT reform agenda have also been identified. The thrust of the proposals, consistent with international experience of successful tax reform episodes, is to broaden tax bases.

Goods and Services Taxation

There is already a general strategy to address indirect tax shortfalls. The Kelkar Commission report and FRBMA road map both highlighted the need to move toward a broad-based and integrated GST. Key steps include (1) extending the state-level VAT to all states, and incorporating services into the base; (2) eliminating the tax on interstate trade (central sales tax (CST)); (3) expanding the service tax base at the center (for instance, to incorporate further financial and legal services); (4) integrating the central VAT and services tax into a new central-level GST applied at the retail stage; (5) broadening the GST base by eliminating exemptions (including for small-scale industries and specific regions); and (6) introducing a comprehensive GST (having a common base at the central and state level, but allowing rates to be fixed separately, subject to some limitations).2 The government is already moving forward on several of these steps.3

There is ample opportunity, in moving to a broad-based GST, to close the revenue gap with more mature emerging market countries. In general, international experience has shown that the introduction of a VAT often leads to higher revenues due to base broadening and compliance improvements (Tsibouris and others, 2006; and Thirsk, 1997). Looking more specifically at India, estimates suggest that eliminating exemptions for small-scale industries and for specific regions could generate an additional ¼ percent of GDP in annual revenue (Bagchi and others, 2005). Expanding the service tax base could alone generate 1 percent of GDP (FRBMA report; see Ministry of Finance, 2004). In terms of compliance, the sharp distinction in India between a good and a service along with the availability of scale- and location-based exemptions have given companies an incentive to reorganize production and distribution processes to minimize taxes. This has led to many drawn-out disputes with the tax authorities, and to complicated and costly-to-administer rules for imputing taxable values. There is thus much scope for compliance-based revenue gains, although the precise amount and timing would be difficult to predict.4

An integrated GST is best viewed, however, as a medium-term reform. Some steps would involve difficult center-state negotiations; for instance, elimination of the CST would produce winners and losers among the states (depending on whether they are net exporters to other states), and is raising issues of compensation. Others, like service taxation at the state level, involve constitutional issues. Even once these impediments are overcome, international experience has shown that implementing a full GST can take substantial lead time, in order to put in place appropriate administrative arrangements and to train taxpayers. The challenge is magnified for India, given the need to introduce a system of joint or unified audit and establish channels for adequate information exchange between different tax administrations. Taking adequate time to get this right is crucial. Unprepared administrations could face significant revenue leakages, to the detriment of public support for the VAT. At the same time, if taxpayers have difficulty complying with changes, pressures for reversal of reform could become intense (Gillis, 1989; and Thirsk, 1997).

Personal Income Taxation

India has already reformed key elements of its PIT regime. In the 2005/06 budget, thresholds were increased dramatically, rates were lowered modestly, and a variety of savings-related exemptions were consolidated into one deduction. The government is now considering modalities for taxing withdrawals from small savings funds.5 Small savings incentives cost the government in the range of ¼ percent of GDP annually in forgone income tax revenue (Ministry of Finance, 2002a). However, the FRBMA road map suggested grandfathering many existing savings schemes, which would limit near-term revenue gains.

There remain important base-broadening measures, however, that the Indian authorities could take. These would not be easy—well-organized vested interests would need to be confronted—but could yield some ¾ percent of GDP a year:

  • Tightening the tax treatment of charities. Charities are generally exempt from taxes when they perform activities of social value that are not for profit. At present in India, however, trusts carry on many activities that are for profit, and surveys have suggested that business income may represent 50–60 percent of their total income. This leads to revenue losses of almost 0.2 percent of GDP a year and distorts competition and horizontal equity (Bagchi and others, 2005).

  • Subjecting agricultural income to taxation.6 Besides creating horizontal equity problems, the exemption has led to significant evasion. High administrative and compliance costs could provide a rationale for some special treatment, but the present high income tax threshold in India (see below), which would eliminate all small and many medium-sized farmers from the tax net, already accommodates this concern.7 The exemption is estimated to cost ¼ percent of GDP a year (Bagchi and others, 2005).

  • Eliminating the tax deductibility of mortgage interest. This deduction raises issues of vertical equity—high-income taxpayers who face higher marginal tax rates receive a larger benefit—and distorts investment incentives.8 Its elimination could bring an additional 0.2 percent of GDP a year in revenue (Ministry of Finance, 2002a).

Table 8.3.

India and Other Economies: Structure of the Personal Income Tax System

article image
Sources: PricewaterhouseCoopers (2003b); country authorities; and IMF staff estimates.

In percent of per capita GDP.

For a married taxpayer with two children earning the per capita income and taking advantage of all savings, dependent, and education deductions.

India’s high income tax threshold would also restrain the growth of revenues over the medium term, but addressing this is not part of the current reform strategy. Compared to PIT systems in more mature emerging market countries, India’s threshold for income taxation is very high relative to per capita income, even adjusting for different deductions available (Table 8.3).9 The amount of tax revenue forgone may be significant: perhaps 1⅓ percent of GDP.10 The PIT threshold has just been raised in India, breaking a long decline over the past 40 years relative to average income (Figure 8.2). This issue is considered in more detail in the section “The Income Tax Threshold” below.

Figure 8.2.
Figure 8.2.

Evolution of the Personal Income Threshold

(Ratio to per capita income)

Sources: Ministry of Finance (2002a); and IMF staff estimates.

Corporate Income Taxation

Many aspects of India’s corporate tax regime are now in line with international practice, but low revenue productivity signals too narrow a base. India’s corporate tax rate remains high, but it has done a good deal of late to bring its other CIT provisions in line with international practice (Table 8.4). However, even with these changes, the productivity of the corporate tax lags more mature emerging market countries. This indicates significant leakage through exemptions, and India does maintain extensive tax holidays, including for export-related activities, specific sectors, and regions. The direct cost of holidays for regions, exports, and the construction sector alone has been estimated at some ½ percent of GDP (Bagchi and others, 2005), and there is little evidence that these holidays have generated significant investment or employment.11 One result of excessive leakage in India has been a proliferation of other corporate taxes (e.g., the minimum tax and wealth tax), which complicate the tax system and raise administrative and compliance costs. Recent practice in more mature emerging market countries has been to reduce or eliminate tax holidays.12

Table 8.4.

India and Other Economies: Structure of the Corporate Income Tax System

article image
Sources: PricewaterhouseCoopers (2003b); country authorities; and IMF staff estimates.

Including surcharges.

Number of years; “Nlim” denotes no limit; “cap” indicates that the amount in any one year is capped.

Machinery and equipment, under straight line method.

The need to broaden the corporate tax base has long been recognized in India. The Kelkar Report (Ministry of Finance, 2002a) and FRBMA road map (Ministry of Finance, 2004) advocated the removal of tax holidays and discussed options including upfront elimination, a rapid two- to three-year phaseout, and sunset clauses with no new entrants. Time consistency would argue in favor of the first option: a phase-out would allow vested interests to lobby for continuation. The government has been pursuing the third route, but some setbacks are evident: in the 2005/06 budget, exemptions covering research and development facilities in specific sectors and investments in Jammu and Kashmir were extended by two years, and new tax benefits have been offered to corporates operating within special economic zones. A key issue, considered in the section “Creating an Environment for Reform,” is how to create an environment in which base-broadening measures would stand more of a chance of success.

The Income Tax Threshold

Elaborating a strategy to increase the PIT threshold over the medium term should be a focus for the Indian authorities. International experience suggests that having well-specified tax reform plans is important: sudden political or economic events often provide an unexpected impetus to reform.13 Doing nothing would allow inflation to erode the threshold (so-called “bracket creep”), but this would require 25 years to bring the threshold into line with the average in more mature emerging market countries. Moreover, without an explicit strategy, there is a significant risk that the threshold will eventually be raised again, leaving the problem in place even as India develops into a more mature emerging market.

Several arguments have been put forward in India in favor of a high threshold:14

  • Tax administration constraints. Due to resource constraints and organizational shortcomings the tax administration is unable to challenge effectively taxpayers’ declarations, allowing them to underdeclare income. They tend to do so at levels just above the threshold, and thus a higher threshold may even enhance revenues.

  • High compliance costs for taxpayers. These raise the social cost of extracting resources from the private sector. Chattopadhyay and Das Gupta (2002) have estimated compliance costs in India to be well above those in developed countries.

  • Social considerations. There is no formal social security system in India, so that a higher level of exemption may be needed to allow Indians to self-insure.

  • The high level of indirect taxation relative to comparable developing countries. A higher threshold may ensure that the effective rate of tax on labor (i.e., including consumption taxes) is reasonable in India for low income earners.

The arguments in favor of a high threshold have merit, but are less compelling from a medium-term perspective. Tax administration reforms, for instance the expansion of the taxpayer identification number (TDS system), computerization, introduction of a large taxpayer unit, and expanded collection of third party information, should strengthen tax administration. The information technology revolution should at the same time rapidly reduce compliance costs for taxpayers. In addition, the government has prioritized the enhancement of social spending over the medium term, and is introducing elements of a safety net. Finally, in the medium term, India will join the ranks of more mature emerging markets, and compared to them, its level of goods and services taxation lags.

Improvements in tax administration and better targeted social spending could open the door to a change in the threshold. As administrative and compliance costs fall and the government becomes more effective at redistributing income, political opposition could lessen. A formal social security system funded by a payroll tax is not necessarily needed but, if India does follow the lead of more mature emerging markets and introduces one, the payroll tax could promote better compliance, given the benefit motivation for payroll tax payment and the administrative synergies in payroll tax and PIT collection.

Creating an Environment for Reform

International experience suggests that successful tax reform, in particular, base broadening, requires focused efforts to build and maintain public support. In this regard, improvements in simplicity and horizontal equity (fairness) proved to be strong selling points in generating public support for tax reform; vertical equity and economic efficiency—while important from a theoretical perspective—did not (Thirsk, 1997). Moreover, tax base broadening also seems to be associated with successful revenue-based fiscal consolidation episodes, perhaps due to improved efficiency and macroeconomic outcomes (Tsibouris and others, 2006).15

To promote a broader tax base, many emerging market countries have become more transparent about the revenue forgone from various exemptions, that is, tax expenditures. Brazil, Mexico, Turkey, and Ukraine report tax expenditures and Russia and the Philippines have partial reporting.16 Indeed, in some cases this seems to have had a direct impact on subsequent policy (e.g., Ukraine).17 The Kelkar commission and FRBMA road map recognized the importance of transparency about tax expenditures, and the Ministry of Finance has been working on estimates for major items. Box 8.1 discusses the key issues to consider in this process.

Maintaining a broad tax base also requires careful administration. Budget process controls are important and are reasonable in India: the Ministry of Finance is required to vet and cost all new tax expenditure proposals and can propose alternative modalities for delivering support (e.g., on-budget subsidies). Moreover, the FRBMA in India effectively requires that new measures with a cost be compensated elsewhere. However, the system has failed in practice to stem the flow of exemptions, and recent legislation on special economic zones evolved to include extensive tax holiday provisions. An option to exert greater control would be to make all steps of the process more transparent, including what would now be internal Ministry of Finance deliberations on alternative subsidy mechanisms.

Enhancing Transparency About Tax Expenditures

Improving transparency about tax expenditures requires consideration of several issues (Craig and Allan, 2001):

  • Definitions. The two most commonly used approaches are the conceptual and reference law approaches. In the former, tax expenditures are defined relative to a pure theoretical baseline (e.g., a single rate VAT). In the latter, only exemptions relative to the existing tax law are considered (e.g., lower rate of VAT would not be considered a tax expenditure). The conceptual approach produces the widest accounting, but should be tempered to reflect administrative feasibility.

  • Measurement. The standard approach is to focus only on the direct reduction of tax liability and to avoid assumptions about a behavioral response. Estimation can be done using survey or other data, but over time tax forms and filing requirements can be adapted to permit more exact measurement. A de minimis rule—the exclusion of small items—can be used to reduce the administrative burden of measurement.

  • Publication. The budget documents should ideally include information covering the past two years, plus the projection, and should distinguish any new initiatives. The documents should spell out the estimation methodology and discuss the risks to revenues, and budget implementation, from misestimation.

  • Ex post assessment and audit. Standard compliance audit is appropriate, but value-for-money audit is crucial. The Auditor General should ask whether the instrument achieved the policy goal. If this is not feasible, specialized studies can be undertaken, as has been done in India for tax holidays in the northeastern region. Any assessment should be published.

To protect CIT revenues, India may also need to address risks from harmful international tax competition. As India enters into bilateral and regional free trade arrangements, and companies are able to supply the Indian domestic market from other locations, the various jurisdictions may compete for mobile tax bases.18 Tax coordination can help to reduce the extent of competition. The least intrusive form would be a nonbinding code of conduct (as in the European Union). More developed forms would involve agreement on tax floors and on acceptable incentives (see Easson, 2004).


There is ample room for further revenue gains in India. The introduction of a GST, reduction of income tax thresholds, extension of income tax to the agricultural sector, and elimination of corporate tax exemptions would go a long way to raising India’s revenue ratio over the medium term, to achieve revenue collection performance on par with more mature emerging market countries (Table 8.5).

Table 8.5.

Summary of Key Revenue Reforms

article image
Source: (Bagchi and others, 2005); Ministry of Finance (2002a and 2002b); and IMF staff estimates.

Yield at existing income tax threshold.

Including agriculture.

Moving difficult reforms forward in India requires some thought about strategy. Fiscal federal agreements would be needed to secure a full GST and agricultural income taxation, while PIT reform would likely gain traction with improvements in the social safety net and tax administration reform. Corporate tax reforms may benefit from improvements in transparency, and from efforts to ensure appropriate tax coordination.


  • Aggarwal, Aradhna, 2004, “Export Processing Zones in India: Analysis of the Export Performance,” ICRIER Working Paper No. 148 (New Delhi: Indian Council for Research on International Economic Relations).

    • Search Google Scholar
    • Export Citation
  • Alesina, Alberto, and R. Perotti, 1996, “Fiscal Adjustments in OECD Countries: Composition and Macroeconomic Effects,” NBER Working Paper No. 5730 (Cambridge, Massachusetts: National Bureau of Economic Research).

    • Search Google Scholar
    • Export Citation
  • Bagchi, Amaresh, R.K. Rao, and B. Sen, 2005, “Raising the Tax-Ratio by Reining in the ‘Tax Breaks’” (New Delhi: National Institute for Public Finance).

    • Search Google Scholar
    • Export Citation
  • Bird, Richard, and P.P. Gendron, 2005, “VAT Revisited: A New Look at the Value Added Tax in Developing and Transition Countries,” paper presented at the USAID Workshop for Practitioners on Tax on May 4 (Washington: U.S. Agency for International Development).

    • Search Google Scholar
    • Export Citation
  • Chattopadhyay, Saumen, and A. Das Gupta, 2002, “The Compliance Cost of the Personal Income Tax and Its Determinants” (New Delhi: National Institute for Public Finance).

    • Search Google Scholar
    • Export Citation
  • Craig, Jon, and William Allan, 2001, “Fiscal Transparency, Tax Expenditures, and Budget Processes: An International Perspective,” Proceedings of the Ninety-Fourth Annual Conference 2000, pp. 25864 (Washington: National Tax Association).

    • Search Google Scholar
    • Export Citation
  • Easson, Alex, 2004, “Harmful Tax Competition: An Evaluation of the OECD Initiative,” Tax Notes International (June 7), pp. 103777.

    • Search Google Scholar
    • Export Citation
  • Gillis, Malcolm, ed., 1989, Tax Reform in Developing Countries (Durham, North Carolina: Duke University Press).

  • Government of India, 2004, Report of the Twelfth Finance Commission (New Delhi).

  • Gupta, Sanjeev, B. Clements, E. Baldacci, and C. Mulas-Granados, 2004, “The Persistence of Fiscal Adjustments in Developing Countries,” Applied Economics Letters, Vol. 11, No. 4, pp. 20912.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • International Monetary Fund, Reports on Observance of Standards and Codes, various reports (Washington).

  • Khan, Mahmood, 2001, “Agricultural Taxation in Developing Countries: A Survey of Issues and Policy,” Agricultural Economics, Vol. 24, No. 3, pp. 31528.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ministry of Finance, 2002a, Report of the Task Force on Direct Taxes (New Delhi: Government of India).

  • Ministry of Finance, 2002b, Report of the Task Force on Indirect Taxes (New Delhi: Government of India).

  • Ministry of Finance, 2004, Report of the Task Force on Implementation of the Fiscal Responsibility and Budget Management Act, 2003 (New Delhi: Government of India).

    • Search Google Scholar
    • Export Citation
  • Ministry of Finance, 2005, Budget 2005–06 (New Delhi: Government of India).

  • PricewaterhouseCoopers, 2003a, Corporate Taxes 2004–05: Worldwide Summaries (Toronto: John Wiley & Sons Canada Ltd.).

  • PricewaterhouseCoopers, 2003b, Individual Taxes 2004–05: Worldwide Summaries (Toronto: John Wiley & Sons Canada Ltd.).

  • Thirsk, Wayne, ed., 1997, Tax Reform in Developing Countries (Washington: World Bank).

  • Tsibouris, George C., Mark A. Horton, Mark J. Flanagan, and Wojciech S. Maliszewski, 2006, Experience with Large Fiscal Adjustments, IMF Occasional Paper No. 246 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Zee, Howell, J. Stotsky, and E. Ley, 2002, “Tax Incentives for Business Investment: A Primer for Policymakers in Developing Countries,” World Development, Vol. 30, No. 9, pp. 14971516.

    • Crossref
    • Search Google Scholar
    • Export Citation

One possibility is that expenditure cutbacks are less reversible, for example if canceling programs permanently weakens the lobby groups promoting them.


This approach can be characterized as a dual VAT. An alternative approach in a fiscal federal system is the CVAT, which imposes a creditable tax on interstate trade to minimize opportunities for cross-border fraud. See Bird and Gendron (2005) for a discussion of the merits and demerits of each of these approaches.


There is also a need to review remaining excises on energy products, alcohol, and tobacco.


The Report of the Task Force on Implementation of the FRBMA (Ministry of Finance, 2004) foresaw compliance-driven revenue improvements of about ½ percent of GDP through the service tax alone.


Deposits, withdrawals, and interest earned are exempt from taxation; international practice has moved toward exemptions for deposits and interest only (Ministry of Finance, 2004).


See Bagchi and others, (2005) for a discussion of the arguments in favor of an agricultural exemption, and why they fail in the Indian context.


The vast majority of developing countries do not exempt agricultural income from taxation (Khan, 2001). However, difficulties in measuring income in the agricultural sector have led to widespread use of presumptive methods of taxation.


See Ministry of Finance (2002a) for a discussion of the pros and cons of the deduction. Among countries surveyed in the Kelkar report, one-half of high-income countries and one-quarter of emerging market and low-income countries had such a provision.


This is also before accounting for the fact that social considerations are partly built into the threshold in India: it is 25 percent higher for women, and 50 percent higher for retirees. Such a design is uncommon, although South Africa also applies a higher threshold for the aged.


Calculated using 2003 data and assuming a threshold of 40 percent of per capita income, 35 million current taxpayers, and 70 million new nonagricultural taxpayers with an average income of Rs 20,000 and savings deductions of Rs 4,000. Calculations also include agriculture, beyond the value of ending the exemption, and are made using data from Bagchi and others (2005). The gains should be understood as relevant for the medium term.


See Bagchi and others (2005). Aggarwal (2004) considers export zones and finds that exports per employment unit declined sharply after a period, as the incentives could not compensate for poor governance and infrastructure in the zones.


See Zee, Stotsky, and Ley (2002) for a discussion of the merits and demerits of various types of investment incentives. Income tax holidays are considered to be among the worst. South Africa’s practice—budget subsidies—is an example of a more transparent approach.


See, for instance, Ministry of Finance (2004).


See Chapter 9 for a discussion of the growth-enhancing impact that a tax base broadening and rate reduction reform could have.


China, Colombia, and South Africa do not report tax expenditures. Beyond emerging markets, all G-7 countries except Japan report tax expenditures (although Italy’s reporting is partial). Among higher income Asian economies, Korea also reports tax expenditures. Source: IMF Report on Observance of Standards and Codes, various reports.


The Ukrainian authorities began publishing tax expenditure estimates in 2002, and efforts were made to broaden the tax base. Steady successes were followed by sweeping reform in 2005, when a new government sought resources to fund social initiatives.


In India, before 2001, states competed for investment via incentives and lower tax rates. An agreement in 2001, which set floor rates of sales tax and eliminated some incentives, was widely seen as mitigating the problem. See Twelfth Finance Commission report (Government of India, 2004).