NOTE NUMBER 330 67806 viewpoint PUBLIC POLICY FOR THE PRIVATE SECTOR FINANCIAL AND PRIVATE SECTOR DEVELOPMENT VICE PRESIDENCY DECEMBER 2011 Reforming Business Taxes Miriam Bruhn What Is the Effect on Private Sector Development? Miriam Bruhn (mbruhn@ Tax rate s an d th e admi n i strati ve c osts of tax c ompl i an c e are ke y worldbank.org) is an economist in the World c on c e rn s of bu si n e ss. S tu di e s wi th i n an d ac ross c ou n tri e s su gge st th at Bank’s Development l owe ri n g c orporate tax rate s c an i n c re ase i n ve stme n t, re du c e tax Research Group. e vasi on by formal fi rms, promote th e c re ati on of formal fi rms, an d This Note was written as u l ti mate l y rai se sal e s an d G DP. Th e se be n e fi ts, h owe ve r, n e e d to be part of the Investment bal an c e d agai n st oth e r obj e c ti ve s of th e tax re gi me . A l th ou gh l e ss Climate Impact Project, a joint effort of the World i s kn own abou t th e e ffe c ts of re du c i n g c ompl i an c e c osts, e vi de n c e Bank Group’s Investment su gge sts th at th i s too c an l e ad to more formal fi rms an d h i gh e r sal e s. Climate Department, IFC’s Investment Climate Busi- The tax regime is one of the most prominent tend to face disproportionately high compliance ness Line, and the World aspects of a country’s business environment. In costs.1 Bank’s Development Re- many countries most formal firms are required to Corporate tax rates have been almost univer- search Group, in collabora- file and pay taxes repeatedly throughout the year sally reduced over the past decade as a result of tion with IFC’s Develop- ment Impact Department, and must dedicate substantial staff time to the the competition for increasingly mobile capital. the Development Impact process. Thus it is unsurprising that in the World Yet important differences remain across coun- Evaluation Initiative, the Bank Enterprise Surveys businesses consistently tries.2 The World Bank’s Doing Business report, FPD Chief Economist’s rank tax rates and the tax administrative process which collects data on statutory tax rates around Office, and the Global among the most important constraints they face the world, finds significant variation, with rates Indicators and Analysis (figure 1). Tax policy and administration are a ranging from 0 percent in Moldova to 40 percent THE WORLD BANK GROUP Unit. The project is funded key part of a country’s private sector development in Chad in 2010 (World Bank 2011). by the U.S. Agency for strategy. They are also often a political minefield, But statutory tax rates do not necessarily rep- International Develop- subject to conflicting objectives. resent the taxes that firms actually need to pay. ment and the World The regulatory burden of taxation has been The reason is that deductions, depreciation, and Bank Group’s Investment increasingly highlighted in surveys on doing busi- other factors influence how much of corporate Climate Department. ness, and country case studies suggest that high income is taxable. A more meaningful measure compliance costs can contribute to the decision of corporate taxes is the effective tax rate, which of businesses to operate informally—that is, to measures the actual taxes paid (after taking into not register with the tax authority at all (see, account deductions, depreciation, and other for example, Thiessen 2003). This is particularly factors) as a percentage of profits. Djankov and relevant for small and medium-size firms, which others (2010) report both statutory and effec- REfORMING BUSINESS TAxES WhAT IS ThE EFFECT ON PRIVATE SECTOR DEVELOPMENT? Figure Share of firms identifying issue as a major constraint Lower tax rates, higher investment While both tax rates and tax administration are 1 Percent key concerns of the business community, the lack 50 of comparable information on administrative 40 structures and, more important, administrative 30 and taxpayer compliance costs constrains cross- 20 country studies of tax administration. 10 The tax component of the World Bank’s Doing 0 Business survey is one of the few attempts to col- lect standardized information on tax compliance s io n ty ns e on ce s ts g ion ls ns ate or im sin kil ici ur tio an tio ati pt tat tit Cr Co rs ctr en xr ula fin ula str rru pe costs, though only for a single hypothetical firm. or bo Lic Ele Ta ini om sp Co eg eg to La dm an rr er ss rc Using this database, Djankov and others (2010) Tr bo ce ad xa cto La Ac Tr Ta se examine the relationship between the complexity al m or Inf of paying taxes and investment. They find that Note: Data cover 128 countries and are the latest available for each country. neither the number of hours needed to comply Source: Author’s calculations based on data from World Bank Enterprise Surveys in 2005–10. with taxes nor the number of tax payments, as captured in the survey, is significantly correlated tive tax rates for 85 countries.3 Effective tax rates with investment across countries (table 1). But tend to be lower (the average statutory tax rate given the regressive nature of compliance costs, is 29 percent, and the average effective tax rate their effect can be expected to be more impor- 17.4 percent), but they still vary widely across tant for firms that are smaller than the case study countries—from 6.6 percent in Mongolia to 39.9 firm used by the Doing Business survey. Thus percent in Bolivia.4 more research is needed to study the relation- Measuring the effects of tax reforms on eco- ship between tax simplification and investment nomic outcomes such as investment is challeng- across countries. There is no within-country study ing. It involves extensive data requirements, examining this relationship. nonuniform reporting practices, identification Most empirical papers on the private sector problems, and a wide range of imperfect mea- effect of tax reforms focus on the relationship sures and methodologies to overcome these between corporate income tax rates and invest- challenges. Further complicating the task is ment across countries. For example, one recent that investment decisions usually are not based study finds that a decrease in the statutory cor- only on corporate income tax. Value added tax, porate tax rate of 10 percentage points is associ- depreciation, dividend tax, and capital taxes all ated with an increase in foreign direct investment play an important part as well, particularly for equivalent to between 0.33 and 0.45 percent of small domestic firms. Indeed, one reason why GDP (Klemm and Van Parys 2009). The same countries have lowered corporate income tax study finds no significant relationship between rates during the past decade is that they have statutory tax rates and domestic investment. also reduced tax incentives and have put more Two cross-country studies using effective emphasis on value added tax instead, raising the tax rates find stronger results. They show that rates and widening the coverage. a decrease in the effective corporate tax rate of 10 With these challenges in mind, this Note is percentage points is associated with an increase in aimed at providing an initial overview of the foreign direct investment equivalent to between empirical evidence on the effects of business tax 1.6 and 2.1 percent of GDP (Djankov and oth- reforms that change the tax rate or administrative ers 2010; Van Parys and James forthcoming).6 processes (particularly filing requirements and Similarly, one shows that a decrease in the effec- the time required to pay taxes) on investment, tive corporate tax rate of 10 percentage points tax evasion by formal firms, formal firm creation, is associated with an increase in domestic invest- and economic performance.5 Tax reforms clearly ment equivalent to 2 percent of GDP (Djankov have implications for other economic outcomes and others 2010).7 as well, particularly government revenue. But the A potential limitation of cross-country studies focus here is on the effect of these reforms on is that the estimated correlations between tax private sector development. rates and investment are not necessarily causal, since they can be driven by other country char- of 2000 increased tax evasion by formal firms by acteristics that are hard to control for (see also 10.6 percent of sales (table 2). OECD 2007). Within-country studies can often A micro-level study successfully measures tax account for these factors, however, and can evasion by calculating the reporting gap between therefore identify a more reliable, causal rela- administrative data on exports from Hong Kong tionship. Two within-country studies examine SAR, China, to China and administrative data on the effect of effective tax rates on investment. imports by China from Hong Kong SAR, China An analysis of the U.S. Tax Reform Act of 1986 (Fisman and Wei 2004). The study finds that a shows that for sole proprietors a 10 percentage 10 percentage point increase in the tax rate on point decrease in the marginal tax rate led to a exports, defined as the import tariff plus value 20 percent increase in investment (Carroll and added tax rate for each product, raises the report- others 1998). In contrast, a within-country study ing gap between exports and imports by 30 percent. in India finds that the Finance Act of 2000, which There are no studies that measure the effect of included a reduction in tax benefits for export compliance costs on tax evasion by formal firms. income (increasing the effective tax rate from 0 to 18 percent), appears to have had no effect Lower and simpler taxes, more formal on firms’ investment (James 2010). But further firm creation investigation suggests that firms were able to Another important outcome of tax reforms is avoid paying higher taxes after the reform by formal firm creation, encompassing both the overstating expenses. formation of new firms and the formalization of previously informal firms. A cross-country study Higher tax rates, more tax evasion shows that a 10 percentage point decrease in the by formal firms effective corporate tax rate is associated with an Estimating the effect of compliance costs and tax increase in the total number of registered busi- rates on tax evasion by formal firms is difficult nesses of 2 per 100 people of working age (Djankov because tax evasion is hard to measure empiri- and others 2010; table 3).8 The same study finds cally. Using detailed data from tax auditors to no significant correlation between the number capture tax evasion, the within-country study in of hours needed to comply with taxes and formal India (James 2010) shows that the increase in firm creation across countries. But a 10 percent effective tax rates ensuing from the Finance Act decrease in the number of tax payments is associ- Table Effect of tax reforms on investment 1 Increase in domestic Increase in foreign Country Study Reform investment direct investment United States Carroll and others 1998a Tax Reform Act of 1986: 10 percentage 20 percent point decrease in marginal tax rate India James 2010 Finance Act of 2000 (marginal None effective tax rates)b Cross-country Djankov and others 2010 10 percentage point decrease in effective 2 percent of GDP 2 percent of GDP corporate income tax ratec 10 percent decrease in number of None hours needed to comply with taxes 10 percent decrease in number of None tax payments Cross-country Klemm and Van Parys 10 percentage point decrease in statutory None 0.33–0.45 percent of GDP 2009 corporate income tax rated Cross-country Van Parys and James 10 percentage point decrease in effective 1.6–2.1 percent of GDP forthcoming corporate income tax rateb a. Examines the effect of sole proprietors’ personal income taxes on their capital investment decisions. The measure of income taxes used is the marginal federal individual income tax rate, which accounts for both the statutory rate schedule and implicit tax rates that arise from special features of the tax code. b. Refers to the marginal effective tax rate on capital, defined as the amount of taxes paid as a percentage of the pretax return on investments that are marginal (that is, just sufficient to cover financing and tax costs). c. Refers to the first-year effective tax rate, defined as the actual first-year corporate income tax liability relative to pretax earnings, taking into account all available deductions. d. Tax data are from the PricewaterhouseCoopers worldwide summaries of statutory corporate tax rates. Table Effect of tax reforms on tax evasion by formal firms 2 Country Study Reform Increase in tax evasion China Fisman and Wei 2004 10 percentage point increase in tax rate 30 percent (in gap between on exports from Hong Kong SAR, China, reported exports and imports) to Chinaa India James 2010 Finance Act of 2000 (marginal 10.6 percent of sales effective tax rates) a. Tax rate is import tariff plus value added tax rate. 4 ated with an increase in business registrations of from tax reforms are also reflected in higher 1.6 per 100 people of working age. sales and GDP. In the context of the U.S. Tax The study by Djankov and others (2010) does Reform Act of 1986, Carroll and others (2001) not disentangle whether the increase in business find that a 10 percentage point decrease in the registrations is due to the formalization of previ- marginal tax rate for sole proprietors led to an ously unregistered firms or to new firm creation. But increase in firm sales of about 15 percent (table within-country evidence from Brazil suggests that 4). Similarly, Fisman and Svensson (2007) show tax reforms can provide an incentive for informal that a 10 percentage point decrease in the effec- firms to register. Fajnzylber, Maloney, and Montes- tive corporate tax rate in Uganda is associated Rojas (forthcoming) show that the introduction with a 15 percentage point increase in annual of the “SIMPLES� tax regime—which reduced the sales growth. The introduction of the SIMPLES tax rate by up to 8 percent of annual revenue for tax regime in Brazil increased firm revenues by both micro and small firms and consolidated six 37 percent (Fajnzylber, Maloney, and Montes- separate federal tax and social security payments Rojas forthcoming). into a single monthly payment—increased the Only two papers examine how corporate tax share of micro firms that are registered with the rates affect GDP growth across countries. The tax authorities by 7.2 percentage points.9 first, using statutory tax rates, finds no robust Since the SIMPLES reform in Brazil included correlation between these rates and GDP growth both a simplification of payments and a reduction (Klemm and Van Parys 2009). The second uses in tax rates, the effects cannot be clearly attrib- statutory top corporate tax rates and finds that uted to either change. There is no within-country a 10 percentage point decrease in these rates study that estimates the effects of simplifying pay- increases GDP per capita growth by 1.82 percent- ments and reducing tax rates separately. age points (Lee and Gordon 2005). The results of the second paper may be more reliable, since this Improved economic performance paper includes a larger set of countries and uses Several studies suggest that the increases in an instrumental variables strategy to capture the investment and formal firm creation resulting causal effect of corporate tax rates on growth. Table Effect of tax reforms on formal firm creation 3 Country Study Reform Increase in formal firms Brazil Fajnzylber, Maloney, Introduction of SIMPLES 7.2 percentage pointsa and Montes-Rojas forthcoming Cross-country Djankov and others 10 percentage point decrease in 2 per 100 people of working age 2010 effective corporate income tax rateb 10 percent decrease in number of None hours needed to comply with taxes 10 percent decrease in number 1.6 per 100 people of working age of tax payments a. Refers to the share of micro firms (urban self-employed and firms with at most five paid employees, excluding domestic workers) registered with the tax authorities. b. Refers to the first-year effective tax rate, defined as the actual first-year corporate income tax liability relative to pretax earnings, taking into account all available deductions. Table Effect of tax reforms on economic performance 4 Increase in sales Country Study Reform (or GDP growth) United States Carroll and others Tax Reform Act of 1986: 10 percentage About 15 percent 2001a point decrease in marginal tax rate Brazil Fajnzylber, Maloney, Introduction of SIMPLES 37 percent and Montes-Rojas forthcoming Uganda Fisman and Svensson 10 percentage point decrease in effective 15 percentage pointsc 5 2007 corporate income tax rateb Cross-country Klemm and Van Parys 10 percentage point decrease in statutory None (GDP growth) 2009 corporate income tax rate Cross-country Lee and Gordon 10 percentage point decrease in statutory 1.82 percentage points (GDP per 2005 top corporate tax rated capita growth) a. Examines the effect of sole proprietors’ personal income taxes on the sales growth of their enterprises. The measure of income taxes used is the marginal federal individual income tax rate, which accounts for both the statutory rate schedule and implicit tax rates that arise from special features of the tax code. b. Refers to firms’ reported tax payment (all types of taxes) as a share of sales. c. Refers to sales growth, not the level of sales. d. Rates are from the World Tax Database of the Office of Tax Policy Research at the University of Michigan. Conclusion 4. Moldova and Chad, the countries with the lowest Both within- and cross-country studies suggest and highest statutory tax rates according to the World that lowering corporate tax rates can increase Bank’s Doing Business 2012 (2011), are not in the sam- investment, reduce tax evasion by formal firms, ple of countries covered by Djankov and others (2010). promote the creation of formal firms, and ulti- 5. A related literature examines the effects on invest- mately raise sales and GDP. These benefits, ment of tax incentives such as tax holidays and exemp- however, need to be balanced against other tions from import duties and consumption taxes on raw objectives of the overall tax regime. Less is materials and inputs (for an overview of this literature, known about the effects of reducing compliance see Zee, Stotsky, and Ley 2002). This literature sug- costs, largely because of a lack of comparable gests that tax incentives can stimulate investment (for information. The few completed papers on this example, Van Parys and James 2010). But a country’s topic provide suggestive evidence that simplify- overall economic characteristics may be more impor- ing taxes can increase formal firm creation and tant for the success or failure of industries than any tax firms’ sales. But more work, particularly at the incentive package (Zee, Stotsky, and Ley 2002). Several within-country level, is needed in this area to papers find that even if tax incentives stimulate invest- allow firm conclusions. ment, they are not cost-effective, because they entail a revenue loss that is larger than the investment they cre- ate (Chai and Goyal 2008; Zee, Stotsky, and Ley 2002). Notes 6. See OECD (2007) and Mooij and Ederveen (2008) The author would like to thank Jacqueline Coolidge, for an overview of the earlier literature on corporate tax Sebastian S. James, Michael Keen, Jan Loeprick, Marialisa rates and foreign direct investment. Motta, Massimiliano Santini, Richard Stern, and Uma 7. Using statutory rather than effective tax rates, Subramanian for their valuable inputs and comments. Djankov and others (2010) find no significant effect 1. These costs usually include the cost of preparing on domestic investment and a slightly smaller effect on and paying taxes (above and beyond normal business foreign direct investment. accounting) for profit tax, value added or sales tax, and 8. Using statutory rather than effective tax rates, Djank- payroll taxes (see also Engelschalk 2007 and Interna- ov and others (2010) find a slightly smaller but positive tional Tax Dialogue 2007). and significant effect on the number of registered busi- 2. For an overview, see Deloitte (2011). nesses per 100 people of working age. 3. The effective tax rates are calculated for a standard- 9. This corresponds to a 55 percent increase in the ized domestic enterprise operating in each country. share of micro firms registered with the tax authorities, REfORMING BUSINESS TAxES WhAT IS ThE EFFECT ON PRIVATE SECTOR DEVELOPMENT? though from a low base (from only 13 percent before the International Tax Dialogue Conference, Buenos the reform to 20.2 percent after the reform). Aires, October 17–19. James, Sebastian S. 2010. “The Effect of Tax Incentives References on Investment: Analysis of Exporters in India.� Carroll, Robert, Douglas Holtz-Eakin, Mark Rider, and World Bank, Washington, DC. Harvey S. Rosen. 1998. “Entrepreneurs, Income Taxes, Klemm, Alexander, and Stefan Van Parys. 2009. “Empir- viewpoint and Investment.� NBER Working Paper 6374, National ical Evidence on the Effects of Tax Incentives.� IMF Bureau of Economic Research, Cambridge, MA. Working Paper WP/09/136, International Monetary is an open forum to ———. 2001. “Personal Income Taxes and the Growth Fund, Washington, DC. encourage dissemination of of Small Firms.� In Tax Policy and the Economy, vol. Lee, Young, and Roger H. Gordon. 2005. “Tax public policy innovations 15, ed. James M. Poterba, 124–48. Cambridge, MA: Structure and Economic Growth.� Journal of Public for private sector–led and MIT Press. Economics 89: 1027–43. market-based solutions for Chai, Jingqing, and Rishi Goyal. 2008. “Tax Conces- Mooij, Ruud A. de, and Sjef Ederveen. 2008. “Corporate development. The views sions and Foreign Direct Investment in the Eastern Tax Elasticities: A Reader’s Guide to Empirical Find- published are those of the Caribbean Currency Union.� IMF Working Paper ings.� Oxford Review of Economic Policy 24 (4): 680–97. authors and should not be WP/08/257, International Monetary Fund, Wash- OECD (Organisation for Economic Co-operation and attributed to the World ington, DC. Development). 2007. Tax Effects on Foreign Direct Bank or any other affiliated Deloitte. 2011. “Global Tax Rates 2011.� http://www Investment: Recent Evidence and Policy Analysis. OECD organizations. Nor do any .deloitte.com/. Tax Policy Studies, no. 17. Paris: OECD. of the conclusions represent Djankov, Simeon, Tim Ganser, Caralee McLiesh, Rita Thiessen, Ulrich. 2003. “The Impact of Fiscal Policy and official policy of the World Ramalho, and Andrei Shleifer. 2010. “The Effect of Deregulation on Shadow Economies in Transition Bank or of its Executive Corporate Taxes on Investment and Entrepreneur- Countries: The Case of Ukraine.� Public Choice 114 Directors or the countries ship.� American Economic Journal: Macroeconomics 2 (3/4): 295–318. they represent. (3): 31–64. Van Parys, Stefan, and Sebastian S. James. 2010. “The Ef- Engelschalk, Michael. 2007. Designing a Tax System for fectiveness of Tax Incentives in Attracting Investment: To order additional copies Micro and Small Businesses: Guide for Practitioners. Panel Data Evidence from the CFA Franc Zone.� contact Ryan Hahn, Washington, DC: International Finance Corpora- International Tax and Public Finance 17 (4): 400–29. managing editor, tion. ———. Forthcoming. “Why Tax Incentives May Be an Room F 4P-252A, Fajnzylber, Pablo, William F. Maloney, and Gabriel V. Ineffective Tool to Encouraging Investment? The The World Bank, Montes-Rojas. Forthcoming. “Does Formality Im- 1818 H Street, NW, Role of Investment Climate.� World Economy. Washington, DC 20433. prove Micro-Firm Performance? Evidence from the World Bank. 2011. Doing Business 2012: Doing Business Brazilian SIMPLES Program.� Journal of Development in a More Transparent World. Washington, DC: World Telephone: Economics. Bank. 001 202 473 4103 Fisman, Raymond, and Jakob Svensson. 2007. “Are Cor- Zee, Howell H., Janet G. Stotsky, and Eduardo Ley. Fax: ruption and Taxation Really Harmful to Growth? 2002. “Tax Incentives for Business Investment: A 001 202 522 3480 Firm-Level Evidence.� Journal of Development Econom- Primer for Policy Makers in Developing Countries.� Email: ics 83 (1): 63–75. World Development 30 (9): 1497–516. rhahn@worldbank.org Fisman, Raymond, and Shang-Jin Wei. 2004. “Tax Rates and Tax Evasion: Evidence from ‘Missing Imports’ Produced by Carol Siegel in China.� Journal of Political Economy 112 (2): 471–96. Printed on recycled paper International Tax Dialogue. 2007. “Taxation of Small and Medium Enterprises.� Background paper for This Note is available online: http://www.worldbank.org/fpd/publicpolicyjournal