PFR Fundamentals: Tax Productivity June 2025 World Bank Economic Policy Global Department Fiscal Policy © 2025 The World Bank 1818 H Street NW, Washington DC 20433 Telephone: 202-473-1000; Internet: www.worldbank.org Some rights reserved This work is a product of The World Bank. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy, completeness, or currency of the data included in this work and does not assume responsibility for any errors, omissions, or discrepancies in the information, or liability with respect to the use of or failure to use the information, methods, processes, or conclusions set forth. The boundaries, colors, denominations, links/footnotes and other information shown in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. The citation of works authored by others does not mean the World Bank endorses the views expressed by those authors or the content of their works. Nothing herein shall constitute or be construed or considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. Rights and Permissions The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Attribution—Please cite the work as follows: “Bradley Larson, PFR Fundamentals: Tax Productivity. Washington, D.C. World Bank Group.” Any queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@worldbank.org. i Contents About the Note and Acknowledgments........................................................................................................ 1 Introduction .................................................................................................................................................. 2 Methodology................................................................................................................................................. 3 VAT C-efficiency ........................................................................................................................................ 3 CIT Productivity ......................................................................................................................................... 4 PIT Productivity ......................................................................................................................................... 4 Data Sources ................................................................................................................................................. 6 Tax productivity estimates ............................................................................................................................ 7 Interpretation of estimates......................................................................................................................... 11 Country applications ................................................................................................................................... 11 Conclusion ................................................................................................................................................... 13 References .................................................................................................................................................. 14 Annex: High, and potentially problematic, tax productivity estimates ...................................................... 15 ii About the Note and Acknowledgments PFR Fundamentals is a series of analytical and how-to notes prepared by the Fiscal Policy Unit to assist task teams in preparing and implementing Public Finance Reviews (PFRs). This note was prepared by Bradley Larson. Overall guidance was provided by Emilia Skrok. Tuan Minh Le, Alaistair Thomas, and Eric Lacey provided helpful comments and inputs. The peer reviewers were Cristina Savescu, Barbara Cunha, Rafael Chelles Barroso, and Desislava Enikova Nikolova. 1 Introduction The ability of governments to raise tax revenues to finance public services and promote economic development varies due to a wide variety of economic, institutional, and political factors. Many of these variables are difficult or impossible to observe, especially in data-poor environments. Nonetheless, it is helpful to assess a country’s tax performance against a common standard to reveal scope for improvement against nationally defined goals, past performance, or the performance of peer countries. Tax productivity indicators measure how well a tax system generates revenue relative to the size of the tax base and a standard tax rate. VAT C-efficiency is calculated for Value Added Tax (VAT), and is a relatively well-established technique.1 CIT productivity and PIT productivity are calculated for Corporate Income Tax (CIT) and Personal Income Tax (PIT), respectively.2 There is limited academic literature on CIT productivity and PIT productivity, but they have been used extensively within the World Bank and by other providers of technical assistance to developing countries, including the International Monetary Fund, regional development banks, and bilateral donors.3 The nomenclature for these indicators is not strictly applied, and recent Public Finance Reviews (PFRs) have used the terms “efficiency” and “productivity” interchangeably. This note uses “tax productivity” to refer collectively to all of the indicators. It uses the term “VAT C-efficiency” to refer to the indicator calculated for VAT, since that specific term is well established in the literature, especially at the IMF (it is referred to as the “VAT revenue ratio” by the OECD). Finally, this note uses “CIT productivity” and “PIT productivity” to refer to these less-well established indicators to avoid additional confusions that might arise from using the term “efficiency”. Tax productivity indicators are simple, easy to calculate metrics that provide valuable insights into the performance of most countries’ main revenue sources. In each case, a value of 1 (or 100 percent) indicates that actual tax revenue is equal to the theoretical maximum. Deviations from that ideal— indicated by values less than 1 (or 100 percent)—can be due to myriad causes, including: 1) a narrow tax base—due to tax incentives, including exemptions and reduced rates, as well as the structure of the economy (informality and hard-to-tax sectors); (2) low tax compliance—driven by the complexity of the tax system, weak tax administration, and informality; and (3) measurement error. Interpretation of tax productivity scores therefore requires analysis and understanding of the respective tax systems. Alternative—or complementary—measures include a simple calculation of tax revenues as a share of GDP or a more advanced measure of tax performance that goes beyond the size of the tax base and tax rates to control for observable economic, institutional, and demographic characteristics such as tax potential,4, tax gaps, and effective tax rates.5 Measures of tax productivity can be used to answer three questions: 1 Ebrill et. al, 2001; Keen, 2013. 2 See, Gallagher, Mark. 2005. Benchmarking Tax Systems. Article in Public Administration and Development, May 2005. 3 The World Bank has applied various estimation of tax productivity and VAT C-efficiency for the preparation of Public Finance Reviews (PFRs) and other major tax diagnostic analysis (e.g., CEM; tax policy reviews). 4 “PFR Fundamentals: Tax Potential and Tax Effort” details the methodology, data requirements, and application of the more advanced approach. 5 “PFR Fundamentals: Measuring tax burden on capital investment” (forthcoming) details the methodology, data requirements, and application for income taxes. 2 • What is the overall efficiency/productivity of the major taxes (VAT, CIT, and PIT) of a country, given their respective tax bases and statutory rates? • How does a country’s performance on these measures change over time? • How does a country’s performance compare with its peers? The calculation of tax productivity is just a first step in analyzing the performance of the tax system. It is recommended that the results are coupled with additional data or contextual knowledge about the country’s tax policy and tax administration. This can help explain fiscal outcomes or inform more detailed policy recommendations. VAT C-efficiency, CIT productivity, and PIT productivity indicators are calculated by the Fiscal Policy unit using the methodology and data described below. Users can find the results in the PFR Resource center, as a bulk download or charted in the PFR Data Visualization Tool. The next section of this note details the methods used to calculate tax productivity indicators. Following that is a brief section on data sources. Next, the calculated estimates are presented along with brief guidance on their interpretation. The final section shows how tax productivity metrics were applied in select PFRs. Methodology The formulas below are used to calculate the standard indicators available on the PFR resource center. Each consists of three basic elements: the revenue raised, a measure of the tax base, and the relevant tax rate. Variations are given for each indicator in case users have better data available for their given country. However, users should exercise caution, since using different formulas for different countries would make cross-country comparisons misleading. VAT C-efficiency VAT C-efficiency measures how well the value-added tax system raises revenue relative to a country’s consumption expenditure and the standard VAT rate, using the formula: – = × In the formulation above, Final Consumption Expenditure (FCE) provides a more precise measure of the tax base than gross domestic product (GDP)—the tax base used due to data limitations to calculate the standard indicators for CIT and PIT.6 6 Ebrill, Keen & Perry, 2001. FCE is the sum of household final consumption expenditure (private consumption) and general government final consumption expenditure (general government consumption). Final consumption expenditure is defined further below. General government final consumption expenditure (formerly general government consumption) includes all government current expenditures for purchases of goods and services (including compensation of employees). It also includes most expenditures on national defense and security, but excludes government military expenditures that are part of government capital formation. Data can be found in the World Bank’s World Development Indicators database. 3 Alternatively, users can calculate VAT C-efficiency using Household Final Consumption Expenditure (HFCE),7 which is usually the primary component of FCE, as the tax base if data are available. This can be preferable when government consumption is not subject to VAT (which is common in developing countries). However, it is important to use the same definition of the tax base for all countries if cross- country comparisons are made. In the few cases where data on final consumption expenditure is not available—and VAT C-efficiency for a given country therefore does not appear in the Fiscal Unit’s established dataset—users can calculate “VAT productivity” using GDP as the tax base.8 Although not as precise as using consumption expenditure, GDP is a fairly good proxy for the VAT tax base, since final consumption is generally the largest component of GDP. However, there is still some variation across countries in final consumption’s share of GDP, particularly when comparing fast-growing emerging economies—where investments are a larger share of GDP—against slower-growing developed economies. CIT Productivity CIT productivity measures how well the CIT system generates revenue relative to the size of the economy and the standard tax rate, using the formula: = × CIT rates are usually flat, making them a good standard for benchmarking CIT productivity. However, caution should be applied if a country provides exceptions for select industries, offers simplified regime for small enterprises, or has a CIT tax base that is disproportionately composed of only a few companies. GDP is used here as a proxy for the tax base in calculating CIT productivity because it is widely available. A more accurate method would be to use the operating surplus (or corporate profits) component of GDP, which better reflects the actual CIT base. This measure is not widely available for developing countries—and the Fiscal Unit therefore does not calculate CIT productivity using it—but users can do so if data are available for their case. Using operating surplus as the tax base would yield higher estimates of CIT productivity all else equal. However, it is important to use the same definition of the tax base for all countries if cross-country comparisons are made. PIT Productivity 7 HFCE excludes the government consumption but continues to include consumption by Non-Profit Institutions Serving Households (NPISHs) in the World Bank’s common measure. Specifically, household and NPISHs final consumption expenditure (formerly private consumption) is the market value of all goods and services, including durable products (such as cars, washing machines, and home computers), purchased by households. It excludes purchases of dwellings but includes imputed rent for owner-occupied dwellings. It also includes payments and fees to governments to obtain permits and licenses. This indicator includes the expenditures of nonprofit institutions serving households even when reported separately by the country. Data can be found in the World Bank’s World Development Indicators database. 8 Here the VAT indicator is referred to as “productivity” to distinguish it from the standard C -efficiency ratio, which requires a stricter definition of the tax base. 4 PIT productivity measures how well the PIT system generates revenue relative to the size of the economy and the standard tax rate, using the formula: = × ℎℎ Given its composition, caution should be exercised in the use and interpretation of PIT productivity. Most PIT systems use a progressive rate schedule, which makes it difficult to properly control for the impact of the differing PIT statutory rates across different income brackets. The highest marginal PIT rate is commonly used as a proxy for the effective PIT rate. However, it can obscure important differences in the progressivity of PIT systems in different countries, making cross-country comparison misleading. This shortcoming could be addressed by calculating the effective PIT rate as the quotient of gross average income and PIT collection across all income brackets (Box 1). However, that would require administrative data from the PIT systems of each country and is therefore not feasible for a global dataset. GDP is again used here as a proxy for the tax base, given data availability. A more accurate method would be to use the wage bill rather than GDP, particularly in developing countries. Again, this measure is not widely available for developing countries—and the Fiscal Unit therefore does not calculate PIT productivity using it—but users can do so if data are available for their case. The wage bill would be preferable since GDP includes income not subject to PIT, such as capital income, agricultural income, and earnings from the informal sector. The wage bill, by focusing on taxable income, provides a more accurate representation of the tax base. Data on the wage bill in developing countries can often be found in national statistics bureau reports. Using the wage bill as the tax base would yield higher estimates of PIT productivity all else equal. However, it is important to use the same definition of the tax base for all countries if cross-country comparisons are made. 5 Box 1. The Effective PIT Rate The effective PIT rate can be calculated as the quotient of total gross income and total PIT collection (Table 1) (the other columns of the table are shown for context). In this example, the overall effective PIT rate is 21.4 percent, which would be a better representation of the statutory tax rate than the highest marginal PIT rate of 27 percent. However, collecting this data on an annual basis from the tax authorities in every country is not feasible, which necessitates the use of the highest marginal PIT rate as a proxy for the PIT statutory rate in the global dataset. Table 1. Calculation of the effective PIT rate in a hypothetical progressive PIT system Income Nominal Gross Taxable Effective Bracket Rate Gross Income Income PIT Collection Rate <1000 0% 300,000 270,000 - 0.0% 1,000-10,000 5% 200,000,000 180,000,000 9,000,000 4.5% 10,001-30,000 20% 500,000,000 450,000,000 90,000,000 18.0% 30,001+ 30% 900,000,000 810,000,000 243,000,000 27.0% Total 1,600,300,000 1,440,270,000 342,000,000 21.4% All else equal, a lower effective tax rate would yield a higher efficiency/productivity measure, since the tax rate appears in the denominator of each formula. Data Sources Data used in the Fiscal Policy unit’s calculations of VAT C-efficiency, CIT productivity, and PIT productivity come from the World Bank,9 IMF,10 Organization for Economic Cooperation and Development (OECD),11 KPMG,12 and PwC (Table 2).13 In the more than 80 percent of cases, the last 9 One variable comes from the World Bank’s World Development Indicators: Final consumption expenditure (% of GDP) (indicator code: NE.CON.TOTL.ZS). Three variables come from the World Bank Fiscal Unit’s World Bank Fiscal Database (WBFD): VAT revenue (current LCU); CIT revenue (current LCU); and PIT revenue (current LCU). The WBFD is compiled from numerous global fiscal datasets, including the World Bank’s MFMOD, the IMF’s Government Finance Statistics, and the IMF’s World Economic Outlook. 10 One variable comes from the IMF’s World Economic Indicators: GDP, current prices (indicator code: NGDP). 11 Three variables, representing OECD countries, come from the OECD: standard VAT rate, CIT rate, and highest marginal PIT rate. These variables represent rates prevailing on January 1 of the given year. They are collected using a web scraper from the OECD Data Explorer (https://data-explorer.oecd.org/). 12 Three variables, representing non-OECD countries for the period 2000–2020, come from KPMG: standard VAT rate (https://kpmg.com/bs/en/home/services/tax/tax-tools-and-resources/tax-rates-online/indirect-tax-rates- table.html), CIT rate (https://kpmg.com/dk/en/home/insights/2016/11/tax-rates-online/corporate-tax-rates- table.html), and highest marginal PIT rate (https://kpmg.com/vg/en/home/services/tax1/tax-tools-and- resources/tax-rates-online/individual-income-tax-rates-table.html). 13 Three variables, representing non-OECD countries for the period 2021–2024, come from PwC: standard VAT rate, CIT rate, and highest marginal PIT rate. They are collected using a web scraper from PwC’s Quick Charts websites for VAT (https://taxsummaries.pwc.com/quick-charts/value-added-tax-vat-rates), CIT (https://taxsummaries.pwc.com/quick-charts/corporate-income-tax-cit-rates), and PIT 6 observed value for the tax base is in 2023, with most of the remainder in 2022. In cases where the observed tax rates data are older than revenue and GDP data, the most recent tax rate is carried forward to create a fuller data panel. Table 2. Data sources for VAT C-efficiency, CIT productivity, and PIT productivity Coverage Estimate Variable Data source Countries Years VAT C-efficiency VAT revenue World Bank 137 1990–2024 Final consumption expenditure World Bank 187 1960–2023 Standard VAT rate KPMG 133 2000–2020 Standard VAT rate PwC 97 2021–2024 Standard VAT rate OECD 38 2003–2024 CIT productivity CIT revenue World Bank 163 1990–2024 GDP IMF 214 1990–2024 CIT rate KPMG 133 2000–2020 CIT rate PwC 102 2021–2024 CIT rate OECD 38 2003–2024 PIT productivity PIT revenue World Bank 161 1990–2024 GDP IMF 214 1990–2024 Highest marginal PIT rate KPMG 133 2000–2020 Highest marginal PIT rate PwC 95 2021–2024 Highest marginal PIT rate OECD 38 2003–2024 Note: Coverage reflects the maximum number of countries and years for each variable; data are not necessarily available for all enumerated countries in all years of the stated coverage. Tax productivity estimates The number of countries for which VAT C-efficiency, CIT productivity, and PIT productivity can be estimated has increased steadily since observations began, reaching a critical mass in 2005 and peaking in 2020 ( Figure 1, panel a). Before 2005, the number of reporting countries was too small to be representative. Although retained in the dataset, they are excluded from the results presented here. Since 2020, country coverage has declined due to lags in reporting of tax revenue. In each year, the number of CIT productivity estimates has exceeded the number of PIT productivity estimates, which in turn has exceeded the number VAT C-efficiency estimates. This pattern is likely to continue in the future, as tax is generally easier to collect from corporations than individuals, and the use of sales tax in some countries means that VAT C-efficiency is not applicable. Over the entire period, 2005–2023, CIT productivity and PIT productivity estimates were both significantly lower than estimates for VAT C-efficiency ( Figure 1, panel b). These patterns conform to expectations, given the methodology of the estimates described above—particularly the measure of the tax base. The kernel density of PIT productivity estimates is bimodal, and VAT C-efficiency nearly so, suggesting distinct clusters of countries or time periods within the full distribution, which is substantiated by the patterns observed across income (https://taxsummaries.pwc.com/quick-charts/personal-income-tax-pit-rates), respectively. PwC’s tax rates data is current as of the last review of the respective country and may not be updated every year. 7 groups and regions below. Point estimates generally the broader distribution, with the averages, range, and standard deviation of VAT C-efficiency estimates exceeding that of CIT productivity and PIT productivity estimates ( Table 3). CIT productivity has improved over time, in contrast with VAT C-efficiency and PIT productivity ( Figure 1, panel c). It also appears that the median country responds to international macroeconomic shocks. Median scores for each the indicators declined following the Great Financial Crisis (2007–2009). Similarly, VAT C-efficiency and CIT productivity—but not PIT productivity—declined during the COVID-19 pandemic (2020). However, the latter pattern is less clear, and there is a lot of noise in each series. Figure 1. Country coverage, density, and change in median value of estimates, by indicator (2005– 2023) Source: World Bank staff estimates using data from World Bank, IMF, KPMG, PwC, and OECD. Note: Estimates greater than or equal to 1 are excluded. Observations prior to 2005 are excluded. Table 3. Summary statistics, by indicator VAT C- CIT PIT efficiency Productivity Productivity Minimum 0.044 0.002 0.001 Maximum 0.995 0.831 0.530 Median 0.497 0.116 0.117 Mean 0.501 0.137 0.144 Standard 0.155 0.092 0.096 deviation Source: World Bank staff estimates using data from World Bank, IMF, KPMG, PwC, and OECD. Note: Estimates greater than or equal to 1 are excluded. Observations prior to 2005 are excluded. 8 It is technically possible to get a value above 1 (or 100 percent) for any of the indicators, which may reflect either positive structural changes or timing issues in the annual data. In the first case, for example, if a large part of the informal sector becomes formal—due to policy changes or improved enforcement—actual tax collections may exceed expectations in a given year. This can happen because the denominator (e.g., GDP) might not change significantly, failing to fully capture informal activity. In the second case, taxes collected from previous periods (such as arrears or audit outcomes), or one-off revenues like oil royalties, fines, or major corporate payments may temporarily boost collections beyond what is expected from the current year’s economic base. This is especially true for VAT if a country issues no or delayed refunds, the tax base reported is gross VAT, or if there are too many exemptions. However, values above 0.9 (or 90 percent) should be very rare given all of the reasons tax collection can fall short. In total, 25 such observations were observed out of 5,776 calculated estimates—23 of which were for VAT (Annex 1). Although these estimates are retained in the dataset, values equal to or greater than of 1 (or 100 percent) are excluded from the descriptive statistics and charts presented here, and users should exercise caution if they are encountered in their own analysis. Tax productivity estimates vary by tax type, income group, and region (Figure 2). As expected, VAT C- efficiency was significantly greater than CIT productivity and PIT productivity overall. VAT C-efficiency and PIT productivity have shown no clear improvement over time, in contrast to CIT productivity, which improved over time in most income groups and regions. As expected, tax productivity measures are positively correlated with income per capita (as a proxy for administrative capacity), with the high income group (HIC) outperforming the upper-middle income (UMIC), lower-middle income (LMIC), and low income (LIC) groups. Countries in the Europe and Central Asia (ECA) region generally perform better than countries in other developing-country regions for each tax type, with the exception of CIT productivity for some years. Countries in the Latin America and Caribbean (LAC) and Middle East and North Africa (MNA) regions generally perform better than countries the South Asia (SAR) and Sub- Saharan Africa (SSA) regions. CIT productivity briefly surged in MNA and was especially low in the LIC SAR, and SSA groups. PIT productivity was especially high in the HIC and ECA groups compared to others, likely largely due to stronger tax administration, a large formal sector, and higher voluntary compliance rates.14 14 Pessino and Fenochietto (2013) in their IMF Working Paper “Understanding Countries’ Tax Effort” find that tax effort and productivity are significantly correlated with income level, institutional quality, and the size of the informal sector. High-income countries tend to score much higher due to stronger institutions and lower informality. Keen and Simone (2004) in “Tax Policy in Developing Countries: Some Lessons from the 1990s” (IMF) note that developed countries are more successful at income taxation because of broader coverage, better enforcement, and more accurate income reporting. 9 Figure 2. VAT C-efficiency, CIT productivity, and PIT productivity estimates, by income group and region (2005–2023) Source: World Bank staff estimates using data from World Bank, IMF, KPMG, PwC, and OECD. Note: Regional averages comprise only developing countries (high-income countries are excluded). Estimates greater than or equal to 1 are excluded. Observations prior to 2005 are excluded. 10 Interpretation of estimates Relatively higher estimates on any of the tax productivity metrics imply a better functioning tax system relative to other years within a country or other countries with similar tax systems. Conversely, a relatively lower estimate implies greater administration, compliance, or policy gaps. As such, tax productivity metrics indicate where there might be a problem and when a country is making progress or slipping. They cannot themselves identify the specific gaps, source of problems, or potentially useful policy reform—but they do suggest where to look. Tax productivity metrics can also be used to estimate the potential impact of a change in tax rates. A value of 0.5, for example, implies that a 1 percentage point increase in the relevant tax rate would increase tax revenue by 0.5 percent of GDP, ceteris paribus. The usefulness of tax productivity indicators varies by context, and particular caution should be applied with PIT productivity (Table 4). Tax revenue collected is typically the most well-defined variable in the formulas used to calculate the indicators. Uncertainty arises in the tax rates and tax base. Table 4. Suitability of tax productivity measures to different applications Application VAT CIT PIT Indication of inefficiencies/gaps Good Fair Poor Measuring progress within countries over time Good Good Fair Benchmarking performance against peers Good Good Poor VAT C-efficiency is generally the most reliable of the indicators, as the tax base is well defined and highly relevant to the tax type and the VAT rate is generally consistent across consumers. CIT productivity holds a middle ground. CIT rates are usually flat, although some countries provide exceptions for select industries or preferential treatment for small and micro firms. However, if the country has a handful of companies that pay a disproportionate share of CIT, their performance can obscure the performance of small taxpayers. PIT productivity is the most problematic given the widespread use of progressive tax rates and tax allowances, as well GDP being far from an ideal proxy for the tax base (i.e., earnings). In countries with a progressive personal income tax system that is working exactly as intended, the use of the top tax rate to calculate PIT productivity can give a false impression of low productivity, simply because less tax revenue is collected from taxpayers with a lower effective tax rate. Even for comparisons over time within a country, PIT productivity can be imprecise, as GDP includes untaxed income. Structural changes (such as shifts toward agriculture or increased informality) or rising non- wage income may distort the indicator by shifting the base away from taxable earnings. Country applications VAT C-efficiency, CIT productivity, and PIT productivity indicators are commonly cited in World Bank PFRs. They are most effective when coupled with a detailed analysis of the relevant tax systems that diagnoses the cause of the observed gaps and offers targeted reforms to improve performance.15 15 Efficiency and productivity scores, explanations, and reforms were taken directly from cited PFRs. In most cases, more recent data are now available. 11 In Albania, VAT C-efficiency was low compared to regional peers but on par with the European Union average. VAT was the main source of the country’s tax expenditures, amounting to almost 5 percent of GDP. CIT productivity was low and PIT productivity was especially low compared to peers. Bunching of reported incomes around the minimum wage suggested that many taxpayers were underreporting their incomes to reduce their tax liability. The PFR argued that better monitoring and enforcement could improve PIT productivity.16 In Cabo Verde, CIT productivity and PIT productivity were lower than most peer countries. The authors argued that tax reforms were critical for fiscal sustainability, and highlighted efforts related to tax administration, tax expenditures and incentives, and regional tariffs. VAT C-efficiency, on the other hand, was higher than in most peer countries. The authors argued that reducing the number of exempt and zero-rated goods could improve revenue mobilization from VAT and reduce the cost of enforcement, further increasing VAT C-efficiency.17 In Georgia, VAT C-efficiency was higher than its peers prior to the COVID-19 pandemic. The authors argued that lower VAT revenue relative to high-performing peers was likely due to a smaller share of private consumption in GDP or a lower statutory rate and that the decline observed during the pandemic could be explained by an increase in the output gap. The authors concluded that CIT productivity and PIT productivity were not relevant for Georgia, and the measures could be misleading, due to the unique design of its CIT system (Distributed Profit Tax) and the flat-rate PIT regime.18 In Peru, VAT C-efficiency was somewhat higher than the regional average, but still low enough to indicate that than less than half of potential VAT revenue was being collected. Possible causes of the gap included exemptions, a large informal sector, and compliance and enforcement challenges. CIT and PIT productivity were low compared to peer countries, possibly due to the high exemption threshold for the PIT, onerous tax expenditures, a large informal sector, and tax evasion.19 In Sri Lanka, CIT productivity was low. Although the CIT statutory rate was on par with its peers, the country collected only half as much revenue as a share of GDP. The authors argued this was due to tax expenditures and policies that created opportunities for tax avoidance, including preferential tax regimes, special economic zones, and profit shifting.20 The PFR did not directly address VAT C-efficiency or PIT productivity. In Zimbabwe, VAT C-efficiency was low and had fallen over the last decade. However, only about one- quarter of the gap was due to shortcomings in tax administration; removing VAT exemptions and taxing zero-rated goods at the standard tax rate would increase VAT revenue by an estimated 4.1 percent of 16 World Bank, Albania Public Finance Review: Enhancing Fiscal Sustainability for Resilience and Human Development, Working Draft, May 2025, p. 17. 17 World Bank, Cabo Verde Public Finance Review: Enhancing Fiscal Sustainability in the Face of Shocks, March 2025, pp. 30, 42–43. 18 World Bank, Georgia Public Finance Review: Fiscal Policy for Inclusive Growth , June 2024, pp. 69–70. 19 World Bank, Peru Public Finance Review: Building Trust: Peru’s Path Forward , Working Draft, April 2025, pp. 28– 29. 20 World Bank, Sri Lanka Public Finance Review: Towards a Balanced Fiscal Adjustment, Working Draft, May 2025, p. 35. 12 GDP. CIT efficiency was low due to a combination of tax incentives, a large informal sector, and compliance challenges.21 A review of Public Finance Reviews (PFRs) that incorporate efficiency and productivity analysis shows that while VAT C-efficiency, CIT productivity, and PIT productivity indicators serve as useful benchmarks, their real value lies in uncovering underlying structural and policy issues within each tax system. These indicators can reveal areas of weakness—such as widespread exemptions, weak enforcement, or a narrow tax base—and help identify targeted, evidence-based reforms. Conclusion VAT C-efficiency, CIT productivity, and PIT productivity measure how well the respective tax systems generate revenue relative to the size of their tax base and a standard tax rate. By doing so, these simple to calculate metrics can help reveal scope for improvement against nationally defined goals, past performance, or the performance of peer countries. These indicators have been used extensively within the World Bank, especially in PFRs, and by other providers of technical assistance to developing countries. Users should be wary of very large estimates and should always couple estimates with qualitative analysis of the respective tax systems. The estimates can point to problems with tax administration or compliance, but country expertise is necessary to explain why they arise and how they can be addressed. Country teams are encouraged to use these indicators as a starting point, while also improving the precision of estimates by employing better proxies for tax bases—particularly for CIT and PIT—and by using more accurate measures of tax rates, such as effective tax rates or weighted average rates, rather than relying solely on top nominal rates. Furthermore, these analyses should be supplemented with additional tools where available, such as tax potential models, marginal tax rate analysis, and tax gap assessments, to provide a more comprehensive and policy-relevant picture of tax system performance. 21 World Bank, Zimbabwe Public Finance Review: Anchoring Macroeconomic Stability through Fiscal Policy, February 2025, pp. xiv, 53, 57–60 13 References Ebrill, L.; Keen, M.; Bodin, JP; and Perry, V. (2001). The Modern VAT. International Monetary Fund. https://www.elibrary.imf.org/display/book/9781589060265/9781589060265.xml Gallagher, Mark. 2005. Benchmarking Tax Systems. Article in Public Administration and Development, May 2005 Keen, M. and Simone, A. (2004). Tax policy in developing countries: Some lessons from the 1990s and some challenges ahead. Helping Countries Develop: The role of Fiscal Policy. Keen, M. (2013). The Anatomy of the VAT. International Monetary Fund. https://www.imf.org/external/pubs/ft/wp/2013/wp13111.pdf International Monetary Fund (IMF). (2023). Government Finance Statistics. https://data.imf.org/?sk=a0867067-d23c-4ebc-ad23-d3b015045405 KPMG. (2023a). Corporate tax rates table. KPMG. (2023b). Personal tax rates table. KPMG. (2023c). Indirect tax rates table. Pessino and Fenochietto. (2013). Understanding Countries’ Tax Effort. IMF Working Paper. https://www.imf.org/external/pubs/ft/wp/2013/wp13244.pdf. PwC. (2023a). Corporate income tax (CIT) rates. PwC. (2023b). Personal income tax (PIT) rates. PwC. (2023c). Value-added tax (VAT) rates. Ueda, J. (2018). Estimating the Corporate Income Tax Gap: The RA-GAP Methodology. International Monetary Fund. https://www.imf.org/en/Publications/TNM/Issues/2018/09/12/Estimating-the- Corporate-Income-Tax-Gap-The-RA-GAP-Methodology-45890 World Bank. (2023). World Bank DataBank, GDP (current LCU). https://data.worldbank.org/indicator/NY.GDP.MKTP.CN World Bank, Albania Public Finance Review: Enhancing Fiscal Sustainability for Resilience and Human Development, Working Draft, May 2025. World Bank, Cabo Verde Public Finance Review: Enhancing Fiscal Sustainability in the Face of Shocks, March 2025. World Bank, Georgia Public Finance Review: Fiscal Policy for Inclusive Growth, June 2024. World Bank, Peru Public Finance Review: Building Trust: Peru’s Path Forward, Working Draft, April 2025. World Bank, Sri Lanka Public Finance Review: Towards a Balanced Fiscal Adjustment, Working Draft, May 2025. World Bank, Zimbabwe Public Finance Review: Anchoring Macroeconomic Stability through Fiscal Policy, February 2025. 14 Annex: High, and potentially problematic, tax productivity estimates VAT C-efficiency, CIT productivity, and PIT productivity estimates equal to or above 0.9 (or 90 percent) are rare. In total, 25 such observations were observed in the calculated estimates (Table 5). The estimates are preserved in the dataset, and users should exercise caution if they are encountered. Table 5. High, and potentially problematic, tax productivity estimates Tax base Efficiency/ Country Tax Tax rate Tax revenue (million Productivity Country code Year type (%) (million LCU) LCU) Estimate Bahamas, The BHS 2016 VAT 7.5 628 8,926 0.94 Bahamas, The BHS 2018 VAT 7.5 681 9,875 0.92 Bahamas, The BHS 2022 VAT 12.0 1,136 10,422 0.91 Bahamas, The BHS 2023 VAT 12.0 1,252 11,530 0.90 Bolivia BOL 2012 VAT 13.0 16,299 136,516 0.92 Bolivia BOL 2013 VAT 13.0 19,033 156,833 0.93 China CHN 2007 VAT 17.0 2,241,211 14,013,877 0.94 China CHN 2008 VAT 17.0 2,689,368 16,094,309 0.98 China CHN 2009 VAT 17.0 2,830,804 17,708,705 0.94 China CHN 2010 VAT 17.0 3,477,717 20,329,620 1.01 China CHN 2011 VAT 17.0 4,141,158 24,712,501 0.99 China CHN 2012 VAT 17.0 4,566,370 28,030,693 0.96 Fiji FJI 2016 VAT 9.0 740 8,547 0.96 Japan JPN 2014 VAT 5.0 19,135,400 402,378,279 0.95 Luxembourg LUX 2012 VAT 15.0 3,196 23,547 0.90 Luxembourg LUX 2013 VAT 15.0 3,449 24,361 0.94 Luxembourg LUX 2014 VAT 15.0 3,770 25,259 1.00 Mongolia MNG 2018 VAT 10.0 2,195,907 23,019,101 0.95 Mongolia MNG 2019 VAT 10.0 2,486,268 26,627,539 0.93 Mongolia MNG 2021 VAT 10.0 2,837,710 29,580,844 0.96 Mongolia MNG 2022 VAT 10.0 3,946,188 35,422,937 1.11 Mongolia MNG 2023 VAT 10.0 4,773,366 40,524,049 1.18 Zimbabwe ZWE 2017 VAT 15.0 0.44 3.13 0.93 Barbados BRB 2020 CIT 5.5 613 10,337 1.08 Montenegro MNE 2022 PIT 1.0 83 5,924 1.40 15