## 1. The GIMF Model, Public Investment Efficiency and the Evaluation of Fiscal Reforms ## Source details **Canonical URL:** [1. The GIMF Model, Public Investment Efficiency and the Evaluation of Fiscal Reforms](https://www.imf.org/-/media/files/publications/selected-issues-papers/2025/english/sipea2025143.pdf) ## Other formats - [Markdown version](/-/media/files/publications/selected-issues-papers/2025/english/sipea2025143.pdf.md) - [Structured JSON version](/-/media/files/publications/selected-issues-papers/2025/english/sipea2025143.pdf.json) --- ### Introduction and key facts - Around 180 active tax expenditures in Uruguay. - Estimated revenue foregone equivalent to about 6 percent of GDP in 2021. - Ten measures account for nearly half of all tax expenditures. - Most affected taxes: Value-Added Tax (IVA) and Corporate Income Tax (IRAE). - Paper objective: analyze Uruguay's tax incentives, benchmark cross-country, and simulate macroeconomic effects using the Global Integrated Monetary and Fiscal (GIMF) model. ### Composition, trends, and distribution - Uruguay exhibits a complex tax expenditure system with many measures in place for decades, limiting fiscal flexibility and potential reallocation to public investment, infrastructure, and social programs. - Growth in tax expenditures over time is consistent with patterns in many high-income countries. - Main stated policy intent of measures: promote private investment, sectoral development, and reduce household vulnerabilities to income shocks. - Distribution and concentration: - Expenditures approximately equally distributed between individuals and firms, with little variation across years. - Very small fraction corresponds to non-residents. - Sectoral concentration: heavily concentrated in investment-related activities, followed by health, industry, energy, and housing. - The most affected taxes: IVA and IRAE — both represent more than 60 percent of the total revenue foregone. - Form of measures: Exemptions account for about 60 percent of revenue foregone; tax credits represent around 20 percent. - Concentration of measures: just ten measures account for nearly half of all tax expenditures. - Property tax expenditures in the form of exemptions represent about 150 percent of the property tax collected. - For IRAE, exemptions and tax credits are roughly equally split. For IVA, reduced rates and tax credits are also relevant. - Historical cost evolution: - Estimated fiscal cost: 6 percent of GDP in 2024. - Historical comparison: estimated at 4 percent of GDP (or 19 percent of tax revenues) in 2005. - Persistence: - Many measures are longstanding—some dating back to the 1970s and 1980s—highlighting large inertia within the tax system. ### International perspective (GTED comparisons and stylized statistics) - GTED-based cross-country comparisons (2021) — selected indicators (Uruguay; LAC; Emerging & developing; Advanced economies): - Total foregone revenue as percent of tax revenues: Uruguay 32.3; LAC 28.7; Emerging & developing 17.6; Advanced economies 21.9. - Total foregone revenue as percent of GDP: Uruguay 6.0; LAC 4.1; Emerging & developing 3.1; Advanced economies 5.5. - HHI index of measures: Uruguay 421.9; LAC 1342.2; Emerging & developing 1993.3; Advanced economies 1437.2. - Number of measures: Uruguay 186.0; LAC 103.6; Emerging & developing 91.7; Advanced economies 142.3. - Type of tax (shares, percent) in Uruguay: - Goods and services: 48.2 - Income: 30.8 - Property: 21.0 - Type of tax expenditure (shares, percent) in Uruguay: - Exemption: 59.5 - Reduced rate: 13.3 - Tax credits, rebates and refunds: 11.6 - Deduction: 3.6 - Type of beneficiary (shares, percent) in Uruguay: - Businesses: 39.7 - Households: 31.3 ### Empirical findings from cross-country analysis - Regression evidence (OLS, country-year 2000-2023) on revenue foregone (RF, log): - GDP per capita (log): coefficient 0.273***. - Sound macroeconomics: coefficient 0.122***. - Free trade: coefficient 0.153*** (in one specification). - Legal/property rights: coefficient 0.0730** (in one specification). - Government efficiency: coefficient -0.0352 (not statistically significant). - Interpretation: Countries with higher GDP per capita and stronger legal, macroeconomic, and trade environments tend to forgo more revenue through tax expenditures. Government size shows no robust significant relationship. ### International comparison and empirical residuals (Uruguay) - Residuals from an OLS regression indicate Uruguay (URY) lies systematically above predicted values given governance, institutional quality, and economic indicators. - Positive residuals imply Uruguay's actual tax expenditures exceed what is typical for countries with similar profiles, suggesting room to streamline or rationalize existing tax incentives. - Regression framework note: One observation = a tax measure in a country and year; total observations = 98,905. Regression includes year fixed effects, country fixed effects, and region fixed effects. Robust standard errors reported; significance levels: *** p<0.01, ** p<0.05, * p<0.1. ### Types of tax expenditures (definitions and examples) - Exoneraciones (exemptions): incomes exempt from both corporate and personal income tax; transactions of goods released from tax obligations. - Alicuotas reducidas (reduced rates): reduced rate in the Value Added Tax. - Deducciones Especiales (special deductions): increased deductions in corporate income (e.g., expenses in R&D, staff training). - Creditos (credits): tax credits for investment promotions in corporate income tax. - Regimenes simplificados (simplified regimes): fixed-sum or turnover based taxes for microenterprises. ### Illustrative tax reform assumptions and model setup - Illustrative reform assumed to increase tax revenues by 0.4 percent of GDP. - Alternative summary in conclusion: a reform could increase total government revenues-to-GDP by about 0.35 percentage points over the period. - In the baseline simulation, the reform’s burden is assumed to fall on consumption in the theoretical model (GIMF). - Behavioral response of taxpayers and elasticities of demand for consumption and investment are crucial and should be calibrated ex-ante by tax authorities. - Assumed use of additional revenues: translated into an increase in productive public expenditure, preserving the fiscal balance. - Public investment efficiency parameter (휉휉) in GIMF calibrated to 80 percent. ### GIMF simulation results — macroeconomic dynamics and key statistics - Reform directs about 0.4 percent of GDP to productive public expenditure; government absorption is "over 2 percent higher." - Cumulative real GDP impact: a cumulative increase of 0.5 percent compared to baseline by 2037. - Government debt-to-GDP improves by 0.4 percentage points by 2037. - Output, investment, and consumption: - Private investment initially falls due to the reduction in tax expenditures but gradually rises as productivity increases. - Household consumption decreases initially (higher taxes) and then recovers over the medium term due to higher labor income from increased productive capacity. - External sector and exchange rates: - Nominal Effective Exchange Rate (NEER) depreciates consistently—by about -1. 2 percent lower compared to baseline in 2037. - Real Effective Exchange Rate (REER) initially appreciates but then stabilizes. - Real exports increase, with an increase of about 0.8 percent above baseline by 2037. - Real imports: initially rise but start declining from 2028 onwards. - Imports of investment goods increase significantly, reaching almost 3 percent above baseline by 2037. - Current account deviation improves over time, turning positive from 2028 onwards. - Prices and monetary policy: - Headline inflation spikes on impact due to the one-off change in the VAT, but is rapidly contained. - Monetary policy rate (and real interest rate) increases only slightly in the short term. - Inflation experiences a slight initial increase but stabilizes over time due to monetary policy response and a disinflationary productivity shock. ### Trade-offs, calibration needs, and alternative fiscal uses - Alternative use of additional revenues: maintain public expenditure levels to reduce the fiscal deficit. - Importance of ex-ante calibration: behavioral reactions of households and firms can produce short-term economic losses depending on reform design; these must be weighed against long-term benefits. - Policy design must account for heterogeneity in responses across consumption and investment, and for which specific incentives are removed. ### Policy recommendations and evaluation framework - Conduct regular cost-benefit analysis of tax incentives, using detailed micro-level data and robust econometric methods, to assess behavioral effects and whether measures meet objectives. - Consider phasing out obsolete or inefficient measures and redirecting savings toward high-priority areas such as education, infrastructure, and innovation to boost private investment, productivity, and long-term growth. - Develop a comprehensive benchmark tax system to improve transparency and provide a consistent framework for evaluating tax expenditures. - Leverage international and regional experience to better understand behavioral responses to incentives. - Recognize implementation obstacles due to high inertia in tax incentives; reforms should be carefully designed and communicated. - Systematic review and modernization: revise and modernize tax incentives to eliminate ineffective or obsolete tax expenditures and direct fiscal resources toward policies that generate meaningful economic impact. - Data-driven assessment: conduct cost-benefit analysis grounded on detailed micro-level data and robust econometric methods to identify measures that drive productive investment versus those that primarily benefit specific sectors. - Define a clear benchmark tax system to provide greater transparency and consistency in evaluating tax expenditures. - Outcome-based evaluation: measure incentive impacts on relevant outcomes such as job creation, innovation, or regional development, and regularly evaluate effectiveness and phase out measures that do not contribute to sustainable and inclusive economic growth. - Learning from recent reforms: recent reform initiatives in other countries could provide lessons for Uruguay. *Source: sipea2025143 - 1. The GIMF Model, Public Investment Efficiency and the Evaluation of Fiscal Reforms 14* ### 1. The GIMF Model, Public Investment Efficiency and the Evaluation of Fiscal Reforms 14 ### 1. The GIMF Model, Public Investment Efficiency and the Evaluation of Fiscal Reforms 14 ### Introduction and key facts - Around 180 active tax expenditures in Uruguay. - Estimated revenue foregone equivalent to about 6 percent of GDP in 2021, ranking third in Latin America. - Ten measures account for nearly half of all tax expenditures. - Most affected taxes: Value-Added Tax (IVA) and Corporate Income Tax (IRAE). - Paper objective: analyze Uruguay's tax incentives, benchmark cross-country, and simulate macroeconomic effects using the Global Integrated Monetary and Fiscal (GIMF) model. ### Composition, trends, and distribution - Uruguay exhibits a complex tax expenditure system with many measures in place for decades; this limits fiscal flexibility and potential reallocation to public investment, infrastructure, and social programs. - Growth in tax expenditures over time is consistent with patterns in many high-income countries. - Main stated policy intent of measures: promote private investment, sectoral development, and reduce household vulnerabilities to income shocks. ### International perspective (GTED comparisons and stylized statistics) - GTED-based cross-country comparisons (2021) — selected indicators (Uruguay; LAC; Emerging & developing; Advanced economies): - Total foregone revenue as percent of tax revenues: Uruguay 32.3; LAC 28.7; Emerging & developing 17.6; Advanced economies 21.9. - Total foregone revenue as percent of GDP: Uruguay 6.0; LAC 4.1; Emerging & developing 3.1; Advanced economies 5.5. - HHI index of measures: Uruguay 421.9; LAC 1342.2; Emerging & developing 1993.3; Advanced economies 1437.2. - Number of measures: Uruguay 186.0; LAC 103.6; Emerging & developing 91.7; Advanced economies 142.3. - Type of tax (shares, percent) in Uruguay: - Goods and services: 48.2 - Income: 30.8 - Property: 21.0 - Type of tax expenditure (shares, percent) in Uruguay: - Exemption: 59.5 - Reduced rate: 13.3 - Tax credits, rebates and refunds: 11.6 - Deduction: 3.6 - Type of beneficiary (shares, percent) in Uruguay: - Businesses: 39.7 - Households: 31.3 ### Empirical findings from cross-country analysis - Regression evidence (OLS, country-year 2000-2023) on revenue foregone (RF, log) highlights: - GDP per capita (log): coefficient 0.273*** (statistically significant across specifications). - Sound macroeconomics: coefficient 0.122***. - Free trade: coefficient 0.153*** (in one specification). - Legal/property rights: coefficient 0.0730** in one specification. - Government efficiency: coefficient -0.0352 (not statistically significant). - Interpretation: Countries with higher GDP per capita and stronger legal, macroeconomic, and trade environments tend to forgo more revenue through tax expenditures. Government size shows no robust significant relationship. ### Macroeconomic simulation using GIMF and policy-relevant quantitative results - Simulated reform: reducing cost-inefficient tax incentives and redirecting resources to productive public expenditure. - Quantitative simulation finding: a reform that increases tax revenues by 0.4 percentage points of GDP could lead to a long-term increase in annual real GDP growth of about 0.5 percent. - Policy trade-off emphasized: design reforms to balance fiscal sustainability and economic competitiveness so tax incentives support strategic development goals without undue fiscal pressure. ### Types of tax expenditures (definitions and examples) - Exoneraciones (exemptions): incomes exempt from both corporate and personal income tax; transactions of goods released from tax obligations. - Alicuotas reducidas (reduced rates): reduced rate in the Value Added Tax. - Deducciones Especiales (special deductions): increased deductions in corporate income (e.g., expenses in R&D, staff training). - Creditos (credits): tax credits for investment promotions in corporate income tax. - Regimenes simplificados (simplified regimes): fixed-sum or turnover based taxes for microenterprises. ### Policy recommendations and institutional measures discussed - Conduct regular cost-benefit analysis of tax incentives, using detailed micro-level data and robust econometric methods, to assess behavioral effects and whether measures meet objectives. - Consider phasing out obsolete or inefficient measures and redirecting savings toward high-priority areas such as education, infrastructure, and innovation to boost private investment, productivity, and long-term growth. - Develop a comprehensive benchmark tax system to improve transparency and provide a consistent framework for evaluating tax expenditures. - Leverage international and regional experience to better understand behavioral responses to incentives. - Recognize implementation obstacles due to high inertia in tax incentives; reforms should be carefully designed and communicated. *Source: sipea2025143 - 1. The GIMF Model, Public Investment Efficiency and the Evaluation of Fiscal Reforms 14* ### 11. Empirical evidence suggests that the level of tax expenditures in Uruguay may lay ### 11. Empirical evidence suggests that the level of tax expenditures in Uruguay may lay ### International comparison and empirical residuals - Residuals from an OLS regression indicate Uruguay (URY) lies systematically above predicted values given governance, institutional quality, and economic indicators. - Positive residuals imply Uruguay's actual tax expenditures exceed what is typical for countries with similar profiles, suggesting room to streamline or rationalize existing tax incentives. - Data sources referenced: GTED database on tax expenditures (2000-2023), Fraser Institute database on economic freedom, and IMF staff calculations. ### Regional patterns and tax-measure level findings - Tax expenditures vary significantly across regions; Latin America and the Caribbean (LAC) exhibits: - Higher reliance on exemptions and non-traditional beneficiaries. - Lower revenue foregone for taxes on goods and services and other taxes relative to advanced economies. - Higher revenue foregone for tax exemptions and deductions, and more substantial tax benefits aimed at "other" beneficiaries (distinct from businesses or households). - Regression framework note: One observation = a tax measure in a country and year; total observations = 98,905. Regression includes year fixed effects, country fixed effects, and region fixed effects. Robust standard errors reported; significance levels: *** p<0.01, ** p<0.05, * p<0.1. ### Tax expenditures in Uruguay: scope and composition - Number of active measures: approximately 180 active measures. - Estimated fiscal cost: 6 percent of GDP in 2024. - Historical comparison: estimated at 4 percent of GDP (or 19 percent of tax revenues) in 2005. - Distribution and concentration: - Expenditures approximately equally distributed between individuals and firms, with little variation across years. - Very small fraction corresponds to non-residents. - Sectoral concentration: heavily concentrated in investment-related activities, followed by health, industry, energy, and housing. - The most affected taxes: Value-Added Tax (IVA) and Corporate Income Tax (IRAE) — both represent more than 60 percent of the total revenue foregone. - Form of measures: Exemptions account for about 60 percent of revenue foregone; tax credits represent around 20 percent. - Concentration of measures: just ten measures account for nearly half of all tax expenditures. - Property tax expenditures in the form of exemptions represent about 150 percent of the property tax collected. - For IRAE, exemptions and tax credits are roughly equally split. For IVA, reduced rates and tax credits are also relevant. ### Persistence, fiscal significance, and policy implications - Many measures are longstanding—some dating back to the 1970s and 1980s—highlighting large inertia within the tax system. - The cost of tax expenditures has gradually increased over the years, becoming increasingly significant as a share of public revenue. - These features raise questions about fiscal sustainability and policy efficiency and point to potential reform targets: construction, healthcare, housing, and energy sectors. ### Illustrative tax reform assumptions and model setup - Illustrative reform assumed to increase tax revenues by 0.4 percent of GDP. - Alternative summary in conclusion: a reform could increase total government revenues-to-GDP by about 0.35 percentage points over the period. - In the baseline simulation, the reform’s burden is assumed to fall on consumption in the theoretical model (GIMF). - Behavioral response of taxpayers and elasticities of demand for consumption and investment are crucial and should be calibrated ex-ante by tax authorities. - Assumed use of additional revenues: translated into an increase in productive public expenditure, preserving the fiscal balance. - Public investment efficiency parameter (휉휉) in GIMF calibrated to 80 percent, given the good quality of public investment management in Uruguay. - Sources for model and simulations: GIMF (Global Integrated Monetary and Fiscal model) and IMF staff simulations. ### GIMF simulation results — macroeconomic dynamics - Public finances and growth: - Reform directs about 0.4 percent of GDP to productive public expenditure; government absorption is "over 2 percent higher." - Cumulative real GDP impact: a cumulative increase of 0.5 percent compared to baseline by 2037. - Government debt-to-GDP improves by 0.4 percentage points by 2037. - Simulations show elimination early on of a small initial government deficit, stabilizing at near-zero levels while government spending as a share of GDP increases. - Output, investment, and consumption: - Private investment initially falls due to the reduction in tax expenditures but gradually rises as productivity increases. - Household consumption decreases initially (higher taxes) and then recovers over the medium term due to higher labor income from increased productive capacity. - External sector and exchange rates: - Nominal Effective Exchange Rate (NEER) depreciates consistently—by about -1. 2 percent lower compared to baseline in 2037. - Real Effective Exchange Rate (REER) initially appreciates but then stabilizes. - Real exports increase, with an increase of about 0.8 percent above baseline by 2037. - Real imports: initially rise but start declining from 2028 onwards. - Imports of investment goods increase significantly, reaching almost 3 percent above baseline by 2037. - Current account deviation improves over time, turning positive from 2028 onwards. - Prices and monetary policy: - Headline inflation spikes on impact due to the one-off change in the VAT, but is rapidly contained. - Monetary policy rate (and real interest rate) increases only slightly in the short term. - Inflation experiences a slight initial increase but stabilizes over time due to monetary policy response and a disinflationary productivity shock. ### Trade-offs, calibration needs, and alternative fiscal uses - Alternative use of additional revenues: maintain public expenditure levels to reduce the fiscal deficit. - Importance of ex-ante calibration: behavioral reactions of households and firms can produce short-term economic losses depending on reform design; these must be weighed against long-term benefits. - Policy design must account for heterogeneity in responses across consumption and investment, and for which specific incentives are removed. ### Policy recommendations and evaluation framework - Systematic review and modernization: - Revise and modernize tax incentives to eliminate ineffective or obsolete tax expenditures and direct fiscal resources toward policies that generate meaningful economic impact (World Bank, 2024). - Data-driven assessment: - Conduct cost-benefit analysis grounded on detailed micro-level data and robust econometric methods to identify measures that drive productive investment versus those that primarily benefit specific sectors (World Bank, 2020; IMF, 2022). - Define a clear benchmark tax system: - A benchmark tax system should be defined to provide greater transparency and consistency in evaluating tax expenditures and serve as a standardized reference for international comparison (IMF, 2019b). - Outcome-based evaluation: - Measure incentive impacts on relevant outcomes such as job creation, innovation, or regional development, and regularly evaluate effectiveness and phase out measures that do not contribute to sustainable and inclusive economic growth (World Bank, 2024). - Learning from recent reforms: - Recent reform initiatives in other countries (example references to Chile reforms in 2022 and 2024 in the source) could provide lessons for Uruguay. *Source: IMF staff chapter on Uruguay tax expenditures and GIMF simulations (GTED database on tax expenditures (2000-2023), budget execution reports (Rendición de Cuentas) 2014–2024, IMF staff calculations, and cited World Bank and IMF literature).* --- _Source: https://www.imf.org/-/media/files/publications/selected-issues-papers/2025/english/sipea2025143.pdf_