References

  • Brixi, Hana Polackova, Christian M. A. Valenduc, and Zhicheng Lee Swift. 2014. Tax Expenditures—Shedding Light on Government Spending through the Tax System. Washington, DC: World Bank.

    • Search Google Scholar
    • Export Citation
  • Burman, Leonard, and Marvin Shaup. 2011. “Tax Expenditures, the Size and Efficiency of Government Spending, and Implications for Budget Reform.” NBER Working Paper 17268, National Bureau of Economic Research, Cambridge, MA.

    • Search Google Scholar
    • Export Citation
  • Coady, David, Ian Parry, Louis Sears, and Baoping Chang. 2015. “How Large Are Global Energy Subsidies?IMF Working Paper 15/105, International Monetary Fund, Washington, DC.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Grote, Martin. 2017. “How to Establish a Tax Policy Unit.” How-to-Note 7, International Monetary Fund, Washington, DC.

  • Heady, Christopher. 2011. “Tax Expenditures: Definitional and Policy Issues.” In Tax Expenditures: State of the Art, edited by L. Philipps, N. W. Brooks, and J. Li. Toronto: Canadian Tax Foundation.

    • Search Google Scholar
    • Export Citation
  • Hutton, Eric. 2017. The Revenue Administration—GAP Analysis Program: Model and Methodology for Value-Added Tax Gap Estimation. IMF Technical Notes and Manuals, International Monetary Fund, Washington, DC.

    • Search Google Scholar
    • Export Citation
  • Inter-American Development Bank (IDB). 2010. “Tax Expenditure Budgets: Concepts and Challenges for Implementation.” IDB Working Paper IDB-WP-131, Washington, DC.

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund (IMF). 2015. “Options for Low-Income Countries’ Effective and Efficient Use of Tax Incentives for Investment Background Document: Tools for the Assessment of Tax Incentives,” Washington, DC.

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund (IMF). 2018. Fiscal Transparency Handbook. Washington, DC.

  • Kassim, Lanre, and Mario Mansour. 2018. “Les rapports sur les dépenses fiscales des payes en développement: une évaluation.” Revue D’Économie du Développement 2018/2 (26): 11367.

    • Search Google Scholar
    • Export Citation
  • Keen, Michael. 2013. “The Anatomy of the VAT.” National Tax Journal 66 (2): 42346.

  • Organisation for Economic Co-operation and Development. 1996. “Tax Expenditures—Recent Experiences.” Paris.

  • Organisation for Economic Co-operation and Development. 2010. “Tax Expenditures in OECD Countries.” Paris.

  • Redonda, Augustin, and Tom Neubig. 2018. Assessing Tax Expenditures Reporting in G20 OECD Economies. Discussion Note 2018/3. Council on Economic Policies, Geneva.

    • Search Google Scholar
    • Export Citation

This note was prepared by Christopher Heady (School of Economics, University of Kent; c.j.heady@kent.ac.uk) and Mario Mansour (Fiscal Affairs Department, International Monetary Fund; mmansour@imf.org).

1

Although the emphasis is on developing and emerging market economies, it is worth noting that not all advanced economies publish tax expenditure reports. Regarding transparency in fiscal management, this how-to note is relevant to all countries; developing and emerging market economies, however, have fewer resources and less capacity to produce tax expenditure reports.

3

A cost-benefit analysis in the case of tax expenditures that target investment can be found in IMF (2015).

4

See, for instance, OECD (1996), Heady (2011), and country examples in Brixi (2004) and Inter-American Development Bank (2010). This definition derives from the practice of tax expenditure accounting, rather than theory, and goes back more than 50 years, when it started in Germany.

5

These are consistent with the levels of practice in the IMF Fiscal Transparency Code (see IMF 2018).

8

The fourth stage in this process, which is the format and content of the tax expenditure report, is discussed in section V. It is not necessary that countries publish an extensive tax expenditure report as part of initial reporting on the cost estimates of tax expenditures.

9

A comprehensive income tax is defined by a rate schedule that applies to an individual’s income from all sources (that is, wages, capital income and capital gains, rental income, business income, gifts, inheritances, and so on).

10

Note that when businesses in the middle of supply chains are exempt from VAT it can mean higher total VAT revenue liability further along the supply chain. A negative tax expenditure can result, because VAT on purchases of exempt items cannot generally end up as an input tax credit.

11

This is not a persuasive argument: high-income households end up reaping most of the benefits of such policies, because they consume more (in absolute terms) than low-income households.

12

In practice, no country has a true comprehensive income tax, in that some incomes are exempt or taxed at lower rates than the normal schedule. The guidance here takes a pragmatic approach and defines the benchmark relative to what a PIT most closely resembles, or what it is intended to do as a policy objective.

13

Most high-income countries, and increasingly other countries, are introducing rules to limit interest expenses and other costs, such as management and service fees, particularly when such costs accrue to nonresident parties related to the payers.

14

This additional objective of excise duty is especially relevant in countries with a broad-based consumption tax. In such cases, the primary objective of raising revenue efficiently should rest with the consumption tax; excise taxes are added to certain items to expand the policy objectives beyond revenue.

15

There are strong reasons to use as a benchmark the rate that reflects the externality, especially given the damages caused by the consumption of certain items, such as energy (see Coady and others 2015).

16

Under the alternative revenue gain method, a tax expenditure is estimated taking into account behavioral changes and the revenue effects on other taxes. While this provides a better approximation of the revenue effect of repealing a certain tax provision, it differs from a pure tax expenditure estimate. For instance, when determining expenditures of ordinary government spending, dynamic behavioral effects on tax revenues are typically ignored in the cost estimate. Estimating dynamic effects requires an understanding of taxpayers’ behavior, including tax evasion and the elasticity of demand and supply of the goods and services/incomes associated with the tax provision, as well as the effects on the revenues raised in other markets that might be affected by the removal of the tax expenditure.

17

Since this change in compliance behavior is also a response to a change in policy, it is difficult to disentangle it from the dynamic effects noted above.

18

In this regard, it is useful for countries to estimate compliance and policy gaps—as in the IMF’s VAT Revenue Administration Gap Analysis Program methodology—as a complement to the cost estimates of tax expenditures (see Hutton 2017; and Keen 2013). Under this methodology, the policy gap is derived assuming full compliance.

19

A similar approach can be followed for individuals. In some countries where PIT operates largely as final withholding on wage income, the public sector is well represented but the private sector is not—because, among other things, compliance is low among the self-employed. In this case, countries might want to rely on a representative sample of private sector employees, and other data sources, to estimate the cost of tax expenditures.

20

Under both (1) and (2), compliance is likely overstated, with the result that the cost of tax expenditures is overestimated. This is also likely the case when data used to estimate the cost of tax expenditures are not from the tax returns or information reported by taxpayers to the tax authorities.

Tax Expenditure Reporting and Its Use in Fiscal Management: A Guide for Developing Economies
Author: International Monetary Fund. Fiscal Affairs Dept.