3. Tax expenditures in Latin America and the Caribbean

The tax system serves as the primary tool by which governments collect funds to finance public spending and investment. It also makes an important contribution to the broad objectives of fiscal policy, such as reducing inequality through income redistribution, supporting vulnerable populations, bolstering economic growth and job creation, boosting private investment, attracting foreign capital, supporting specific economic sectors and regions, and promoting or disincentivising the production or consumption of certain goods and services. Instead of spending public funds to meet these goals, the government may forego revenues through preferential tax treatments, commonly referred to as tax expenditures.

There are a variety of conceptual definitions of tax expenditure. The OECD (2004[1]) refers to tax expenditure as a transfer of public resources by the reduction of tax obligations with respect to a frame of reference or benchmark, rather than by direct public expenditure. The Inter-American Center of Tax Administrations defines tax expenditures as resources forgone by the State due to the existence of incentives or benefits that reduce the direct or indirect tax burden of certain taxpayers in relation to a reference tax system, to achieve certain economic and social policy objectives (CIAT, 2011[2]).

Tax expenditures come in various forms, such as exemptions, deductions, credits, reduced rates, tax deferrals and accelerated depreciation systems. This variety provides policy makers with flexibility in designing policies involving tax expenditures (Table 3.1). The impact of these instruments differs, including in terms of the revenues foregone and changes in the behaviour of beneficiaries.

There are three main approaches to measuring tax expenditures: revenue forgone, revenue gain and outlay equivalent. The revenue forgone approach calculates the revenue loss from a preferential treatment through a partial equilibrium analysis assuming no change in taxpayer behaviour. The revenue gain method estimates the increased revenue that would result from eliminating the preferential treatment and considers the impact on beneficiary behaviour. Finally, the outlay equivalent approach measures the cost of providing the same financial benefit offered by the tax treatment through direct expenditure; that is, the direct spending needed to provide an equivalent net tax benefit to those who receive the preferential treatment being evaluated (Villela, Lemgruber and Jorrat, 2009[5]).

Most countries in the LAC region consider tax expenditures as revenue not received or not collected (Table 3.2). However, some countries define them differently: Brazil refers to them as indirect government expenditures, Nicaragua calls them transfers from the state, and Uruguay defines them as lost revenue (Campos Vázquez, 2022[6]).

Although it is more common for LAC countries to measure tax expenditures as revenue not collected, it is important to acknowledge that the estimates of tax expenditures presented by each country do not necessarily align with the revenue that would be generated if a particular preferential treatment were abolished. The ultimate impact would be contingent on how economic agents respond to the tax changes. Because these estimations are made independently, the total of their results would not necessarily match the hypothetical forgone amount, as eliminating each treatment separately is not equal to eliminating them all at once.

In general, the frame of reference used to identify and measure tax expenditures can be based on either a legal or a theoretical conceptual framework. The legal framework uses the structure established by tax law as a reference, so deviations from the general provisions of tax legislation are considered tax expenditures. The theoretical framework uses a broad tax base, calculated according to a theoretical concept of income, consumption or value-added, as a reference, and any deviation from this base results in forgone revenue. A third, less common, approach is the analogous subsidy approach, whereby tax benefits that are similar to direct subsidies are regarded as tax expenditures.

There is significant variation across countries in the LAC region in terms of what constitutes tax expenditures and the methodologies employed to estimate their size. These differences make it difficult to compare tax expenditures across and within countries over time. Additionally, estimates of individual tax expenditures are not additive; all countries in the LAC region employ the revenue-foregone approach, which estimates each tax expenditure in isolation and does not capture dynamic effects on the behaviour of economic actors or the interaction with other preferential treatments. As such, caution is warranted in the analysis of the potential revenue gains from eliminating certain tax expenditures. These limitations notwithstanding, this section reviews tax expenditures in the region to provide an illustrative analysis of their magnitude and composition.

Revenue losses due to tax expenditures are significant in the LAC region, although there is heterogeneity between countries. In 2021, foregone revenues in Latin America averaged 3.7% of GDP, equivalent to 19% of general government tax revenues (Figure 3.1). While this represents a modest decline with respect to 2015, most of the shift in the regional average is due to large contractions resulting from methodological changes in the estimation of tax expenditures, principally in Chile.1 At the country level, revenues lost to tax expenditures range from 1.4% of GDP in Paraguay to 6.7% of GDP in Honduras. These results have remained relatively stable between 2015 and 2021.

Another key metric for the importance of tax expenditures is to compare them with overall tax revenues. The ratio of tax expenditures to tax revenues is particularly high in countries with a low tax take, irrespective of the level of tax expenditures. In the Dominican Republic, tax expenditures were equivalent to 30% of general government tax revenues while in Brazil they represented 13% of the total, despite having a similar level of tax expenditures as a percentage of GDP.

Given that tax expenditures can be conceived of as public spending through the tax system, it is also relevant to compare their magnitude with public outlays in priority public policy areas (Figure 3.2). Social spending in Latin America is low relative to the OECD, despite an increase during the 2008-2009 global financial crisis and more recently in 2020 as a result of the COVID-19 pandemic. The pandemic laid bare the low level of social spending, with the region’s health and social protection systems unable to respond to pressing needs. Against this backdrop, tax expenditures relative to central government social spending are significant, averaging 33% in Latin America, with some countries registering much higher ratios. Likewise, capital expenditure in the region has suffered during the last decade, emerging as the principal variable of fiscal adjustment. The ratio of tax expenditures to central government capital expenditure exceeds 100% in many countries, reaching a high of 578% in Brazil.

In terms of composition, tax expenditures in the region are highly concentrated in preferential treatments for VAT and income taxes (Figure 3.3). On average, VAT benefits represented 55% of total tax expenditures in 2021, with income taxes accounting for a further 31%. In the case of VAT, a key source of tax revenues in the region, forgone revenues were above 2% of GDP in five countries – Ecuador, Uruguay, Costa Rica, Honduras and Panama – and exceeded 1% of GDP in another seven countries. In the case of income taxes, the cost of these treatments is above 2% of GDP in only three countries: Costa Rica, Ecuador and Honduras. There are also significant differences in the relative weight of tax expenditures for different tax instruments. In Bolivia, Guatemala, Nicaragua, Panama, Paraguay and Peru, VAT-related measures represent more than 60% of overall tax expenditures. In contrast, tax expenditures related to income tax are the most prominent in Brazil, Chile, Costa Rica, Ecuador and Mexico.

VAT expenditures in the region cover a range of goods and services. One of the principal preferential tax treatments for VAT is related to purchases of food or other items in the basic consumption basket, often with the aim of supporting low-income households. These foregone revenues can be substantial, equivalent to 1% of GDP or more in some cases (Figure 3.4).

Beyond the magnitude of these tax expenditures, they also vary considerably in terms of design. In Costa Rica, a reduced rate is applied to the basic consumption basket, which is based on the principal goods purchased by the 30% of the population with the lowest incomes, according to data collected by the National Institute of Statistics and Census (INEC). In Mexico, purchases of foods are zero-rated. In contrast, foods are exempt from VAT in the Dominican Republic.

An important consideration for VAT expenditures for food, especially in an inflationary context, is that the benefits do not accrue equally across society. In Mexico, an estimated 27% of the total revenue foregone through the tax expenditure from the zero rate for purchases of food accrues to the top income quintile, compared with 13% for the lowest quintile (SHCP, 2022[8]).

Preferential VAT treatments are also commonly applied to the consumption of health services and medicines. These measures can be seen as providing relief to households for the high levels of out-of-pocket health expenditure in the region. Special VAT treatments for health services and medicines are considerable in some countries, reaching upwards of 0.8% of GDP in Uruguay.

The design of these preferential tax treatments highlights a strongly country-specific set of policy objectives. Tax exemptions for health services provided by national public health institutions or social security systems are the principal form of health-related tax expenditures in Argentina and Uruguay. In Chile, Colombia and the Dominican Republic, private purchases of health services are exempt from VAT. A similar exemption exists in Ecuador, complemented by an exemption for purchases of medicines.

Tax expenditures for income taxes are common across countries, and they relate predominately to CIT (Figure 3.5). CIT exemptions for free zones are common, leading to significant foregone revenues in El Salvador (0.5% of GDP), Dominican Republic (0.4% of GDP) and Uruguay (0.2% of GDP), for example. A related objective is the promotion of investment, often in strategic sectors. Ecuador provides an eight-year CIT exemption for new productive investments in priority sectors in Quito and Guayaquil, which rises to 15 years for other areas of the country. Uruguay offers a tax credit for investments in fixed assets, among others, that are used in manufacturing or agriculture. Preferential regimes also exist for SMEs, including a reduced CIT rate for SMEs as part of the Propyme programme in Chile.

Foregone revenues due to PIT expenditures can also be significant in some countries. In Mexico, the principal PIT expenditures are exemptions for retirement income, social welfare benefits and various types of special salary payments (such as annual bonus and overtime, among others). In Brazil, the largest foregone revenues resulting from PIT expenditures relate to exemptions for retirement income and special treatments for health expenditure. In El Salvador, deductions for education and health expenditure are among the most important preferential PIT treatments. In contrast, Ecuador provides a blanket deduction for personal expenses (resulting in foregone revenues equivalent to 0.4% of GDP). In Chile, the principal PIT expenditure is an exemption for capital gains from the sale of real estate up to a specified threshold.

Given the large foregone revenues associated with tax expenditures, it is important to fully weigh their potential benefits and costs. Evaluating their effectiveness is key to making informed policy decisions, since these preferential tax treatments are used by governments in lieu of public spending to promote specific policies and objectives, ranging from economic growth and job creation to environmental protection and social welfare.

ECLAC/Oxfam International (2019[9]) recommends evaluating three crucial points when considering the use of a differential tax treatment that leads to forgone revenue: (1) Do tax expenditures result in a more efficient allocation of resources; (2) What effect do they have on the horizontal and vertical equity of the tax system; and (3) Is there transparency and are such tax expenditures included in the budgetary process?

To assess the efficiency and effectiveness of tax expenditures, it is crucial to define the desired objective, identify the impact indicators, and determine the cost associated with each differential treatment. Attention should also be paid to the implications of additional tax system complexity, particularly in terms of higher compliance and enforcement costs. Policy makers should carefully consider the pros and cons of using tax expenditures relative to direct expenditure when making decisions about the allocation of resources. Depending on the objective, direct expenditure may be more efficient.

Tax expenditures, such as exemptions, deductions, credits, reduced rates, deferrals and accelerated depreciation systems, are sometimes used to “correct” certain distortions in the tax system that inhibit desirable economic and social outcomes. However, these treatments may not always be the most efficient method of addressing such distortions and can result in unequal treatment of taxpayers. Some taxpayers may not be able to take advantage of these benefits, leading to horizontal inequity whereby similar taxpayers have different tax burdens. For example, preferential tax treatments for firms and individuals operating in the frontier regions of Mexico can lead to situations where subsidiaries of multinational enterprises pay significantly lower taxes compared to domestic firms in similar industries located elsewhere in the country (Campos Vázquez, 2022[6]).

Additionally, preferential treatments can negatively impact the progressivity and redistributive impact of taxes, leading to vertical inequity. For example, when the preferential treatment is applied to income received from dividends and other financial instruments that is typically concentrated among the highest income groups, it reduces the progressivity of the PIT.

Even though tax treatments may offer certain benefits when compared to other policy tools, their costs, as well as some of their advantages and disadvantages, are often "hidden" from policy makers and the general public. A lack of transparency also exists within the public sector. Since tax expenditures are typically not subject to the same level of control and evaluation as direct expenditure (they are not included in the budget and are usually automatically renewed annually), this lowers the transparency and accountability of fiscal policy, making it harder to determine who benefits from them.

In recent years, there has been increasing interest in the LAC region in evaluating the impact of tax expenditures on various outcomes such as growth, investment and employment. To assess the overall impact of a policy, it is crucial to compare its benefits against the costs, including fiscal costs from lost revenue, as well as the impact on efficiency, equity and transparency.

Many studies in the region are limited to estimating the loss of revenue from tax incentives, with few conducting a comprehensive evaluation of the benefits and costs. In such evaluations, the benefits should include the actual impact of the tax treatment on factors such as increased revenue from new investments and the associated social benefits, while the costs should consider the loss of revenue from investments that would have occurred regardless of the incentives, increased tax administration costs, potential abuse that leads to tax evasion, economic distortions, distributive effects, and their impact on macroeconomic variables (Campos Vázquez, 2022[6]).

Comprehensive studies of tax expenditures are limited in Latin America; Table 3.3 summarises the main findings of the few that exist, which suggest that tax expenditures, and investment incentives in particular, are not necessarily cost-effective in achieving their objectives.

The 2030 Agenda for Sustainable Development, adopted by the United Nations General Assembly in September 2015, sets 17 goals and 169 targets aimed at promoting economic, social, and environmental sustainability for the 193 Member States that adopted it. Implementing the 2030 Agenda for Sustainable Development will be a major challenge for the LAC region; substantial investments are needed to achieve the SDGs, especially given the region’s economic and social disparities. One of the key challenges is domestic resource mobilisation, as the tax take remains insufficient to meet current and future demands on public spending.

Against this backdrop, the role of tax policy in support of the SDGs has taken on renewed importance (ECLAC, 2019[14]; 2020[15]; 2021[16]). Tax expenditures, particularly investment incentives, have emerged as a major policy issue. While tax expenditures can play a role in supporting the achievement of the SDGs, there is significant work to be done to ensure that they are an effective tool for development. For the region, this depends to a large extent on establishing an institutional framework capable of managing the design, implementation, and governance of tax expenditures, as well as ensuring transparency and accountability (Figure 3.6).

This framework depends on the credible measurement and evaluation of tax expenditures, backed by binding legal requirements to ensure their implementation. There are still opportunities in the region to incorporate best practices in the measurement of tax expenditures. The evaluation of tax expenditures significantly lags efforts to quantify tax expenditures and will require a concerted effort to build new capabilities. Regional and international cooperation can play an important role in supporting these efforts. To this end, the OECD has developed the OECD Investment Tax Incentives Database, which allows comparison of tax expenditure policies across developing regions (Box 3.1).

References

[11] Agostini, C. and M. Jorratt (2013), “Política tributaria para mejorar la inversión y el crecimiento en América Latina”, Serie Macroeconomía del Desarrollo (n. 130), Economic Commission for Latin America and the Caribbean, Santiago.

[6] Campos Vázquez, R. (2022), “Measurement of tax expenditures in Latin America”, Project Documents (LC/TS.2022/148), Economic Commission for Latin America and the Caribbean, Santiago.

[19] Celani, A., L. Dressler and T. Hanappi (2022), “Assessing tax relief from targeted investment tax incentives through corporate effective tax rates: Methodology and initial findings for seven Sub-Saharan African countries”, OECD Taxation Working Papers, No. 58, OECD Publishing, Paris, https://doi.org/10.1787/3eaddf88-en.

[17] Celani, A., L. Dressler and M. Wermelinger (2022), “Building an Investment Tax Incentives database: Methodology and initial findings for 36 developing countries”, OECD Working Papers on International Investment, No. 2022/01, OECD Publishing, Paris, https://doi.org/10.1787/62e075a9-en.

[2] CIAT (2011), Manual de buenas prácticas en la medición de los gastos tributarios: una experiencia iberoamericana, Inter-American Center of Tax Administrations, Panama.

[13] CIAT/UN-DESA (2018), Diseño y evaluación de los incentivos tributarios en países en desarrollo: Temas seleccionados y un estudio país, Inter-American Center of Tax Administrations/United Nations Department of Economic and Social Affairs.

[3] DIPRES (2012), Medición y evaluación del gasto tributario, Budget Office of Chile, http://www.dipres.gob.cl/598/articles-94691_doc_pdf.pdf.

[16] ECLAC (2021), Fiscal Panorama of Latin America and the Caribbean, 2021 (LC/PUB.2021/5-P), Economic Commission for Latin America and the Caribbean, Santiago.

[15] ECLAC (2020), Fiscal Panorama of Latin America and the Caribbean, 2020 (LC/PUB.2020/6-P), Economic Commission for Latin America and the Caribbean, Santiago.

[14] ECLAC (2019), Fiscal Panorama of Latin America and the Caribbean, 2019 (LC/PUB.2019/8-P), Economic Commission for Latin America and the Caribbean, Santiago.

[9] ECLAC/Oxfam International (2019), “Los incentivos fiscales a las empresas en América Latina y el Caribe”, Project Documents, (LC/TS.2019/50), Economic Commission for Latin America and the Caribbean/Oxfam International, Santiago.

[20] IMF (2020), “Chile: Technical Assistance Report – Assessment of Tax Expenditures and Corrective Taxes”, IMF Country Report No. 20/305, International Monetary Fund, Washington, D.C., https://www.imf.org/en/Publications/CR/Issues/2020/11/19/Chile-Technical-Assistance-Report-Assessment-of-Tax-Expenditures-and-Corrective-Taxes-49906.

[18] OECD (2022), OECD Investment Tax Incentives Database – 2022 Update: Tax incentives for sustainable development” (brochure), OECD, Paris, https://www.oecd.org/investment/investment-policy/oecd-investment-tax-incentives-database-2022-update-brochure.pdf.

[1] OECD (2004), “Best practice guidelines: off budget and tax expenditures”, paper presented at the 25th Annual Meeting, Madrid, 9–10 June, Organization for Economic Cooperation and Development, Paris.

[7] OECD et al. (2023), Revenue Statistics in Latin America and the Caribbean 2023, OECD Publishing, Paris, https://doi.org/10.1787/a7640683-en.

[8] SHCP (2022), Renuncias recaudatorias 2022, Secretariat of Finance and Public Credit of Mexico, Mexico City.

[4] SHCP (2021), Renuncias recaudatorias 2021, Secretariat of Finance and Public Credit of Mexico, Mexico City.

[21] SII (2021), Informe de Gasto Tributario: 2019 a 2021, Servicio de Impuestos Internos, https://www.sii.cl/aprenda_sobre_impuestos/estudios/gasto_tributario_2019_2021.pdf.

[5] Villela, L., A. Lemgruber and M. Jorrat (2009), “Tax expenditure budgets: concepts and challenges for implementation”, IDB Working Paper series, No. 131, Inter-American Development Bank, Washington, D.C..

[12] World Bank (2017), “Hacia un sistema tributario más eficiente: Evaluación de la eficiencia fiscal, análisis de costos y beneficios de los gastos fiscales y estudio sobre el empleo informal y sus repercusiones en el sistema tributario”, Santo Domingo.

[10] World Bank (2012), “El Gasto Tributario en Colombia: Una propuesta de evaluación integral y sistémica de este instrumento de política pública”, Bogotá.

Note

← 1. In Chile, the 2020 edition of the tax expenditure report (covering the years 2019-2021) incorporated a revised benchmark tax system, greater coverage of preferential treatments, and methodological improvements to better estimate some tax expenditures, especially as related to the income tax, which were in line with recommendations provided by the IMF and the OECD at the behest of the Ministry of Finance of Chile (SII, 2021[21]; IMF, 2020[20]).

Metadata, Legal and Rights

This document, as well as any data and map included herein, are without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries and to the name of any territory, city or area. Extracts from publications may be subject to additional disclaimers, which are set out in the complete version of the publication, available at the link provided.

© OECD 2023

The use of this work, whether digital or print, is governed by the Terms and Conditions to be found at https://www.oecd.org/termsandconditions.