ArticlesVol. 1, No. 2

Dual Income Taxation and Developing Countries

Richard M. Bird1,

Eric M. Zolt2

1 Professor Emeritus of Economics, University of Toronto,
2 Michael H. Schill Professor of Law, UCLA School of Law
Citation:1 Colum. J. Tax L. 174

Abstract

The dual income tax combines a progressive tax on labor income and a lower flat rate tax on income from capital. Denmark, Finland, Norway, and Sweden adopted dual income taxes to address a set of tax challenges that arose in the late 1980s and early 1990s. Although developing countries face much different economic, political, and tax environments from the Nordic countries, the dual income tax may be the right solution to the different set of challenges facing many developing countries. Providing separate tax rates for labor and capital income allows countries greater flexibility in addressing tax competition while retaining progressive tax rates for labor income. A dual income tax regime may also allow developing countries to rationalize the taxation of income from active business operations under the personal and corporate tax systems and the taxation of passive investment income under the personal tax system. Developing countries could also use the move to a dual income tax system as an opportunity to make broader reforms in their personal, corporate, and payroll tax systems. Finally, recent tax reforms in Russia, Ukraine, and several countries in Central and Eastern Europe have led to flat tax regimes that generally apply a single tax rate to all types of income above some zero0bracket amount. We contend that a dual income tax may provide policymakers in developing countries with an attractive alternative that addresses tax competition concerns while maintaining a progressive tax on labor income.

 

 

  • I. INTRODUCTION
  • II. DUAL INCOME TAX SYSTEMS IN DEVELOPED COUNTRIES
    • A. The Nordic Pioneers
    • B. Design of Dual Income Tax Systems
    • C. Arbitrage Opportunities
    • D. Reforms and Proposals Outside the Nordic Countries
  • III. DUAL INCOME TAX IN DEVELOPING COUNTRIES
    • A. Tax Systems in Developing Countries
    • B. Design Choices
      • 1. Personal income tax system
      • 2. Relationship of Personal Income Tax Rates and Corporate Tax Rates under a Dual Income Tax
      • 3. Taxation of Normal and Excess Returns
    • C. Advantages and Disadvantages of Dual Income Tax Regimes in Developing Countries
  • IV. DUAL INCOME TAX VERSUS FLAT TAX REGIMES
    • A. Recent Flat Tax Reforms
    • B. Choosing Between a Dual Income Tax and Flat Tax System
  • V. CONCLUSION

I. INTRODUCTION

Businesses often seek to find new applications for existing products. TNT was originally designed as a yellow dye. Listerine, a popular mouthwash, was used as an antiseptic for surgery. Viagra was initially designed to treat hypertension. The dual income tax, a combination of a progressive tax on labor income and a lower flat rate tax on income from capital, was adopted in the Nordic countries to address a set of tax challenges that arose in the late 1980s. Although developing countries face much different economic, political, and tax environments from the Nordic countries, the dual income tax may be the right solution to the different set of challenges facing many developing countries.

Almost all income tax regimes combine a personal income tax and a corporate income tax. The personal income tax generally applies to wage income, as well as different types of income from capital such as dividends, interest, rents, royalties, and profits from sole proprietorships and partnerships. The corporate income tax generally applies to profits on entities operating in corporate form. The dual income tax seeks to tax wages and labor income attributable to sole proprietorships and partnerships at progressive tax rates and tax capital income at a flat rate under either the personal or corporate tax systems.

Transplanting legal regimes, like using prescription drugs for purposes other than those for which they were designed, may result in undesirable and unintended consequences.[1] Nonetheless, dual income taxation may be an effective tool to improve income taxation in developing countries. Global competition provides strong incentives for countries to reduce tax rates, especially developing countries that are more dependent on foreign capital inflows. Providing separate tax rates for labor and capital income allows countries greater flexibility in addressing tax competition and greater opportunity to retain progressive tax rates for labor income. A dual income tax regime may also allow developing countries to rationalize the taxation of income from active business operations under the personal and corporate tax systems and the taxation of passive investment income under the personal tax system. Developing countries could also use the move to a dual income tax system as an opportunity to make broader reforms in their personal, corporate, and payroll tax systems.

The dual income tax approach rejects the long-held ideal of a progressive global personal income tax (a “comprehensive income tax”).[2] To some, this may seem a step in the wrong direction. For decades, tax policy advisors in both developed and developing countries focused on a comprehensive income tax as an essential keystone to any modern tax system. Although there have always been those who questioned the conventional wisdom, the comprehensive income tax was long accepted as the ideal towards which all income tax systems should strive.[3]

In reality, however, the personal income tax system in most developing countries (as well as most developed countries) has never been global or very progressive.[4] The so-called comprehensive income tax system is often a set of incomplete and sometimes inconsistent rules applicable to different types of income. Developing countries face serious challenges in improving the taxation of income, whether from labor, capital, or a combination of labor and capital.[5] The result is that personal income taxes play a very limited role in developing countries.[6] Many developing countries subject income from capital to relatively light effective tax burdens under the personal income tax system through policy design, poor tax administration, or both. However, the way in which countries achieve this low tax burden is often so cumbersome and inefficient that countries reap both the costs of taxing capital (economic distortions, compliance, and enforcement costs) and the costs of failing to do so (inequity, as well as administrative problems from tax arbitrage).

Developing countries also face serious challenges in implementing corporate income tax systems. Low-income developing countries are more likely than high-income developing countries or developed countries to adopt tax incentives that reduce the corporate tax base.[7] Tax holidays, taxfree zones, and other tax incentives erode the potential corporate income tax base and may, sometimes deliberately, create advantages for foreign over domestic firms. While all countries face difficulties in taxing the agricultural sector and small businesses, developing countries face greater challenges both because of the lack of administrative capacity to tax these sectors effectively and because the costs of taxes and other forms of government regulation may outweigh the benefits from operating in a formal economy. This encourages businesses to stay in the informal sector and discourages modernization and growth.[8]

One way to improve tax regimes in developing countries is to abandon taxing income from capital and move to a personal consumption tax of the types that have been proposed over the last few decades.[9] We believe good reasons exist to retain both personal and corporate income taxes. Although the personal income tax in most developing countries is limited in its ability to redistribute income, it can and should play an important role both in generating revenues and in helping to establish a more politically cohesive and stable state.[10] Despite its flaws the personal income tax is probably the only significantly progressive element found in most developing countries’ tax systems.[11] In addition, the corporate income tax plays a much greater role in raising revenue in developing countries than in developed countries and, in particular, succeeds in capturing at least some location-specific rents.[12]

Although we focus here on the dual income tax in the developing country context, we note that many of the same advantages and disadvantages of moving from a nominal comprehensive income tax to a dual income tax apply to the United States and other developed countries. The U.S. tax system, like other tax systems, already subjects income from capital to tax treatment different from the treatment of income from labor,[13] including lower tax rates applicable to dividends and capital gains.[14] Global competition provides incentive to reform tax systems and tax competition concerns in the U.S. have prompted legislative proposals to reduce corporate income tax rates significantly below tax rates under the personal income tax system.[15] The U.S. Department of Treasury has previously given serious consideration to forms of dual income taxation. As discussed in Section II.D, in 1992, the Treasury Department included as part of its report on the integration of the personal and corporate tax system a prototype called the “Comprehensive Business Income Tax” that is a form of dual income taxation.[16] We believe that a move to a dual income tax regime in the United States, as for developing countries, would be an improvement over the current tax system.[17]

We begin in Section II with a brief review of the Nordic experience with dual income tax regimes and several parallel reforms that have been proposed or implemented in other developed countries, including the United States.[18] Section III then reviews the different ways in which developing countries might implement a dual income tax regime and considers briefly the possible advantages and disadvantages of the different approaches. Substantial differences exist in the political, economic and tax environments both between and among developed and developing countries. These differences influence both the form and desirability of separating the taxation of labor and capital income.

Section IV compares the dual income tax with flat tax proposals of the type that have been adopted in several Eastern European countries and the countries of the former Soviet Union. We argue that the dual income tax, properly conceived, can combine many of the advantages of flat tax regimes with the traditional virtues of a personal income tax system that has explicitly progressive tax rates. All major tax policy or tax administration reforms carry economic and political risks, and the risk-reward calculus will differ from country to country. Nonetheless, the move to a dual income tax may facilitate a number of useful broader tax reforms in developing countries, including reducing or eliminating corporate tax incentives, rationalizing the taxation of portfolio income of domestic and foreign investors, adopting withholding regimes that may increase tax compliance and tax revenue, coordinating presumptive tax regimes for small and medium businesses with the corporate tax system, and integrating payroll or social security tax systems with the personal income tax systems. Section V concludes.

II. DUAL INCOME TAX SYSTEMS IN DEVELOPED COUNTRIES

A. The Nordic Pioneers

Denmark, Finland, Norway and Sweden were among the first countries to explicitly reject the comprehensive income tax model and adopt separate tax regimes for taxing income from labor and income from capital. While Denmark was the first country to adopt a dual income tax, it has subsequently moved to a hybrid comprehensive income tax and a dual income tax system.[19] Table 1 describes the dual tax regimes in those countries as of 2008.

Table 1. Dual Income Tax Regimes (2008)20

Table 1. Dual Income Tax Regimes (2008)[20]

Several factors contributed to the decisions to adopt dual income tax regimes in the Nordic countries.[21] In the 1980s, these countries were concerned that domestic investors were transferring portfolio investments (such as bank deposits, stocks, and corporate and government bonds) outside their countries. The highest marginal personal income tax rates in the mid-1980s ranged from 66% in Norway to 87% in Sweden.[22] Given the high tax rates that individuals were facing on portfolio income in the Nordic countries, strong incentives existed for domestic investors to move their investments offshore to avoid domestic taxation.[23] The globalization of capital markets made it easier for individual investors to establish and maintain investment and bank accounts outside their home country.[24]

At the same time, relatively high inflation rates resulted in substantial effective tax rates on capital income.[25] Even at modest rates of inflation, subjecting nominal interest income to tax rates over 50% results in very high effective tax rates on real interest income. Tax systems in most countries do not provide for adjustments for inflation.[26] The combination of high tax rates, relatively effective tax administration, and no inflation adjustments made the problems more acute in the Nordic countries.[27] Applying a low flat rate to capital income provided a rough adjustment for inflation.[28]

The then-existing tax regimes in the Nordic countries also provided substantial tax preferences for capital investments.[29] As in other countries, these preferences often took the form of current deductibility of many expenses while taxes on the income associated with the expenses were deferred. In particular, the full deductibility of interest on debt incurred to finance tax-favored assets resulted in negative effective tax rates on capital income.[30] Taxpayers used deductions associated with capital income to reduce their tax liability on labor income.[31] By establishing separate tax regimes for income from labor and capital, the Nordic countries were able to stop the erosion of the tax base for labor income.[32] Lowering the tax rate on capital income made it possible for the Nordic countries to broaden the capital income tax base.[33]

Finally, the economies of the Nordic countries faced difficult challenges during this time period. In particular, substantial concerns existed about high levels of unemployment. While Nordic levels of unemployment were low by international standards, the levels of unemployment in the late 1980s and early 1990s were high by historical standards (and would continue to increase).[34] These concerns helped provide political support for reducing the tax burden on income from capital as a means to increase economic activity.[35]

In addition to the challenges facing the Nordic countries, tax reforms throughout the world in the mid-to-late 1980s had resulted in significant rate cuts for individual and corporate income taxes.[36] While the Nordic countries increased taxes on consumption during this period, large social programs still required substantial revenues from income taxes to fund government operations.[37] Essentially, these countries faced two choices in reforming their income tax systems: reduce tax rates for all income and generate insufficient income tax revenue to support social programs, or bifurcate the tax regime and apply a lower rate to income from capital while maintaining higher progressive tax rates on labor income. The Nordic countries chose the latter approach.

Steffen Ganghof has set out a taxonomy that nicely demonstrates the competing considerations a country may face in designing an income tax system.[38] Ganghof begins with three domestic policy goals that governments may consider in designing an income tax regime. The first goal is tax progressivity.[39] This goal addresses vertical equity concerns by requiring high-income recipients to pay a higher proportion of their income in tax.[40] The second goal is comprehensiveness, which requires equal treatment of income from capital and income from labor.[41] This addresses horizontal equity concerns, and along with the first goal, results in taxing individuals in accordance with “ability to pay” principles.[42] The third goal is symmetry, which exists when all types of capital income are subject to the same tax regime.[43] This goal addresses concerns of allocative efficiency as well as tax-induced distortions in investment and savings decisions.[44]

In a closed economy, a comprehensive income tax can theoretically satisfy these three objectives. At least in its ideal form, the comprehensive income tax achieves all three goals by taxing income from all sources equally, in a progressive manner according to a taxpayer’s ability to pay, so that a taxpayer faces the same tax regime for different types of capital income.[45] With the competitive pressures in an open economy, however, all three goals are rarely met. When international competitiveness is added to the three domestic tax policy goals, governments may no longer be able to satisfy all goals, particularly as cross-border mobility of capital varies both between types of capital and types of countries.[46] In Ganghof’s terminology, governments face an income tax quadrilemma: their only choice is which goal to sacrifice: progressivity, comprehensiveness, symmetry, or competitiveness.[47]

Depending on economic and political factors, countries face different considerations in determining which objective to sacrifice. The Nordic countries sacrificed comprehensiveness by providing different treatment for capital and labor income.[48] Other countries have chosen at various times to sacrifice progressivity, symmetry, or competitiveness.[49] Because of the different tax environments facing the Nordic countries and developing countries, many of the specific considerations that prompted the Nordic countries to sacrifice comprehensiveness may not apply in developing countries. Nonetheless, as we argue below, developing countries may be able to improve their tax system substantially by sacrificing comprehensiveness and adopting separate tax regimes for capital and labor income.

B. Design of Dual Income Tax Systems

As Table 1 shows, the dual income tax systems in Finland, Norway, and Sweden provide for a progressive income tax rate schedule applicable to labor income and a flat tax rate on capital income.[50] Under this approach, an individual’s tax liability depends not only on total income, but also on the split between labor and capital income.[51] The tax rate on capital income is at or near the lowest positive rate for labor income and the highest marginal tax rate on labor income is about 15–25% higher than the tax rate on capital income.[52] However, as discussed below, many variant designs are also possible. For instance, in the so-called “pure” version of the dual income tax, the tax rate on capital income is aligned with the corporate tax rate.[53] Income from capital taxable under the dual income tax system generally includes business profits, dividends, interest, rents and some types of royalties. In principle, policymakers could also choose to expand the scope of dual income tax systems to include such items as imputed rent on owner-occupied housing and returns on pension savings (as well as other forms of current tax-favored savings) in capital income.[54]

The tax systems differ in how they divide income from labor and capital for purposes of determining tax liability. Finland and Sweden generally applied a strict schedular approach, placing labor income and capital income in separate baskets. Finland and Sweden adopted separate rate schedules for labor and capital income, limited the use of capital losses to offset labor income, and provided for any personal allowances to be applied only to labor income.[55] In contrast, before the 2004 tax reforms, Norway provided for a “general” tax base that was taxed at a single flat rate and qualified for personal allowances.[56] Both capital and labor incomes were included in the general tax base, but only labor income was included in the “personal” tax base.[57] The personal tax base was subject to progressive tax rates, but contained a substantial zero-bracket amount that resulted in only relatively high levels of labor income being subject to the higher progressive tax rates. This version of a dual income tax is equivalent to a flat tax on all income with a surcharge on labor income above a certain threshold.

C. Arbitrage Opportunities

The move from a comprehensive income tax to a dual income tax regime generally presents tax planning (or tax evasion) opportunities. Lower tax rates on income from capital than on income from labor create a strong incentive to characterize income from labor as income from capital. This incentive exists in many tax systems, especially those that provide for favorable tax treatment for capital gains or subject labor income to payroll or social security taxes that substantially increase the effective marginal tax rates for labor income but leave unchanged the tax rates for income from capital.[58]

The major challenge in the Nordic dual tax systems is to limit a taxpayer’s ability to convert labor income from self-employment (such as sole proprietorships or partnerships) or from wages of owner-employees of closely-held corporations into income from capital. The dual income tax regimes of the four countries reflect different approaches to prevent taxpayers from transforming labor income into capital income.

One approach provides for an imputed return to the firm’s business assets by multiplying the value of the assets by an assumed rate of return on capital (for instance, the interest rate on government debt plus some risk premium).[59] The asset base could be the firm’s gross assets, in which case the firm’s financial liabilities are not deducted from the asset base.[60] Alternatively, countries could determine the imputed return of the net assets and thus provide a deduction for business liabilities (as well as adjusting net profits for interest deductions).[61] Profits in excess of this imputed return are deemed to be returns from labor and are subject to taxation as labor income whether or not distributed to the owner.[62]

Taxpayers may also seek to avoid higher tax rates on labor income in cases where owner-employees own a large percentage of the shares of closely-held corporations.[63] Such owner-employees will seek to minimize their wage compensation and extract returns in the form of lower taxed dividends and capital gains. To limit this tax strategy, countries could tax normal returns to capital as income from capital, and then tax excess returns to capital under the progressive tax rates applicable to labor income.[64] Alternatively, countries could choose to require closely-held businesses to be taxed as flow-through entities, regardless of legal form of organization.[65]

The rules designed to prevent re-characterization of income generated their own set of distortions. For example, the regime that determines capital income with reference to the amount of capital invested encourages taxpayers to stuff significant assets into their corporations that are related to business operations. Those rules that only apply to closely held corporations with a percentage of active owners encourage taxpayers to add passive owners to avoid application of these rules.[66] The larger the spreads between the flat tax rate applicable to capital income, the corporate tax rates, and the progressive tax rates applicable to labor income, the greater the arbitrage opportunities. While the Nordic and other countries face significant challenges to prevent taxpayers from disguising labor income as capital income, as we discuss later, different considerations may apply in developing countries.

D. Reforms and Proposals Outside the Nordic Countries

Over the years, several countries outside the Nordic region have considered or adopted different types of dual income tax systems.[67] Even under a comprehensive income tax, tax regimes frequently establish different effective taxes on different types of income.[68] Important elements of dual income tax regimes already exist in many countries. For example, many countries have adopted final withholding taxes on interest and dividends at rates below the top marginal rates under the personal income tax system.[69] Elements of dual income tax regimes are also present in regimes that provide rate preferences or exclusions for capital gains.

Different tax regimes for income from capital and income from labor have been discussed extensively in many countries, including the United States. In the U.S., a version of a dual income tax was proposed by the U.S. Department of Treasury in 1992, the Comprehensive Business Income Tax (CBIT).[70] The “classical” corporate income tax system distorts choices between operating in non-corporate and corporate form, choosing between debt finance and equity finance, and retaining rather than distributing corporate profits.[71] The U.S. Treasury was concerned about the range of tax treatment applicable to income from entities operating in corporate form.[72] In some cases, corporate income was subject to double taxation under the personal and corporate tax systems.[73] In other cases, however, income attributed to corporate entities was subject to zero tax (for example, interest income paid to tax-exempt entities or portfolio interest paid to foreign holders) or three or more layers of tax (for example, for multiple levels of corporate holdings not subject to consolidation).[74] The 1992 Treasury Report set forth various proposals to reduce distortions and rationalize the treatment of business income.[75] The policy aim was to design a tax regime that subjected business income to a single layer of tax.[76]

In the U.S., as well as many other countries, taxpayers can choose among different types of legal forms in which to conduct business operations.[77] While there may be valid business and legal reasons to allow taxpayers to choose among different legal forms, differential tax treatment of those different legal forms is much less compelling.[78] An additional objective of the CBIT proposal was to design a tax regime that would be applicable to business income, without regard to the legal form chosen by an individual or groups of individuals to conduct operations.[79]

Finally, the CBIT proposal addressed the concern of differential tax treatment of debt and equity capital, both at the corporate level, where interest paid on debt was deductible and dividends paid on common shares were not deductible, and at the investor level, where tax-exempt and taxfavored investors paid little or no tax on interest and dividends received.[80] Holders of debt and equity have similar claims to the cash flow of the business activity, subject to different competing rights as to risk, return, duration and control.[81] While good reasons may exist for treating debt and equity capital differently for economic and legal purposes, differential tax treatment is, again, much less compelling.

The easiest ways to eliminate the distinction between debt and equity in the corporate tax regime are either to provide deductions for dividends paid or to disallow interest deductions. The CBIT proposal eliminated the usual deduction for interest payments in order to impose tax liability on the share of business income attributable to debt capital.[82] This was the most controversial aspect of the proposal partly because it resulted in clear winners and losers among corporate taxpayers and investors.[83] Alternatively, the CBIT proposal could have included a deduction for interest at the corporate level and a tax on the investor level collected through final withholding (at the corporate tax rate).[84]

Simply put, CBIT was designed as a flat tax on business income, collected at the source (with no further taxation at the investor level).[85] CBIT is thus in effect a form of dual income tax that imposes a flat tax on income from businesses (regardless of form or label) while maintaining a progressive tax on labor income.

III. DUAL INCOME TAX IN DEVELOPING COUNTRIES

A. Tax Systems in Developing Countries

The tax environment in most developing countries differs substantially from that in developed countries. Taxing patterns differ. Tax burdens as a percentage of GDP are roughly 18% for developing countries, about half the tax-to-GDP ratio of developed countries.[86] Developing countries rely relatively more on taxes on consumption rather than on income as compared to developed countries, with VATs and excise taxes providing a substantial portion of tax revenues.[87] Whether measured as a percentage of GDP or a percentage of overall tax revenue, personal income taxes play a much smaller role in developing countries than in developed countries.[88] While the revenue from corporate income taxes in developing countries varies substantially by region, corporate income taxes provide a major source of revenue in many countries, and generally, the poorer the country, the greater the proportion of total income taxes from corporate taxes.[89]

Many developing countries have a large traditional agricultural sector and a significant informal (shadow) economy, both operating largely outside the formal tax system.[90] No country has managed to tax either of these sectors effectively.[91] These hard-to-tax sectors constitute a much higher portion of total economic activity in developing countries than in developed countries.[92] As a result, the tax base that tax authorities can potentially reach is relatively small in many developing countries.

Depending on the legal form of business operations, active business income could be taxed under either the personal income tax or corporate tax systems. Alternatively, a country could adopt a business or an enterprise tax that applied to all forms of business—corporations, partnerships, and sole proprietorships.[93] Developing countries have also adopted presumptive tax regimes that seek to tax business activity by reference to factors other than income.[94]

The sources of revenue under the personal income tax system vary substantially among countries. While good comparative information about the revenue composition of personal income taxes in developing countries is not available, it is not uncommon for over 90% of personal income tax revenue in developing countries to come from wage withholding in the formal sector.[95] It follows that personal income tax systems in developing countries raise relatively small amounts of revenue from either active business operations or from “passive” investment income such as dividends and interest.

Developing countries often provide for deductions and exemptions for different types of savings under the personal income tax.[96] In several countries, individuals can deduct life insurance premiums and pension contributions and are not taxed on most types of interest and dividend income.[97] But even when countries include different types of portfolio income in the tax base, the tax administration often lacks capacity to tax this income effectively.[98] Those countries that succeed in collecting tax revenue generally do so by using final withholding tax regimes, often at tax rates substantially below rates applicable to other types of income.[99]

Both developed and developing countries face substantial challenges in taxing portfolio income that crosses national borders.[100] First, consider the challenges in taxing residents on their portfolio income earned outside the country. The “international tax compromise” resulting from the League of Nations model treaty generally allocates active business income to the country where it is earned (the source jurisdiction) and portfolio income to the country from which the capital is supplied (the residence jurisdiction).[101] But while countries have the right to tax their residents on their portfolio income earned outside the country, several countries choose to tax their residents only on income earned in their country.[102] In the absence of effective information exchange or withholding regimes, even those countries that tax their residents on their world-wide income have limited ability to tax their residents’ foreign-source income effectively.[103] The use of tax havens and bank secrecy laws make it difficult for taxing authorities to track the investments of their residents.[104] Developed countries such as the United States, Germany, and Japan face substantial challenges in taxing this foreign-source income; developing countries with less administrative capacity face even greater difficulties.[105]

Second, consider the challenges to taxing foreign investors on their portfolio income from local investments. While many source countries impose withholding taxes on portfolio income paid by domestic persons to foreign investors, both developed and developing countries collect little revenue from portfolio investments held by non-residents.[106] Exceptions to withholding requirements often apply to interest income. In 1984, Congress repealed the U.S. withholding tax on portfolio interest income held by non-U.S. residents.[107] The failure of the U.S. to tax foreign investors on income earned on U.S. debt obligations makes it difficult for all countries to impose taxes on portfolio interest, especially developing countries. With respect to other types of portfolio income, many countries have entered into tax treaties that trade away the right to collect income on passive investments held by foreign investors, presumably hoping to attract more foreign investment.[108]

Developing countries vary greatly in the tax wedge applicable to labor income. Differences in relative tax rates under the personal income tax system partly explain the variations, but most of the difference is attributable to payroll or social security taxes imposed on workers and employers to fund unemployment, medical care, or pension benefits.[109] These taxes on labor (and associated benefits) are especially high in many Eastern and Central European countries, countries of the former Soviet Union, and several Latin American countries.[110] To the extent that these payroll or social security taxes exceed the value of expected benefits, they constitute an additional tax on labor income.

Corporate tax systems in developing countries vary greatly in scope and design. Many developing countries receive a large percentage of their corporate tax revenues from a relatively small number of taxpayers, whereas other countries have a more diversified tax base.[111] Countries differ both as to the level of corporate income tax rates and the relationship of the corporate rate to personal income tax rates. As Table 2 indicates, while some regional convergence does exist, there is no easy explanation for either the variation in the level of tax rates or the differences between personal and corporate rates in these countries.

Table 2. Relative Personal and Corporate Income Tax Rates in Selected Developing Countries112

Table 2. Relative Personal and Corporate Income Tax Rates in
Selected Developing Countries[112]

Spending patterns also differ. The aggregate levels of taxes and expenditures are smaller in developing countries than in developed countries.[113] Expenditure programs to alleviate poverty or to reduce inequality are much smaller in developing countries than programs in most developed countries.[114] Nonetheless, much as in the Nordic countries in the 1980s, developing countries require tax revenues to support government programs and also face challenges in maintaining or increasing income tax revenues from increasing global tax competition.[115]

B. Design Choices

As in the Nordic cases, several design options exist for dual income taxes. The considerations in choosing tax rates and bases differ from the Nordic countries and also vary substantially among developing countries.

1. Personal income tax system.

The first set of design options under the personal income tax system involves choices as to the level of tax rates and the relationship of tax rates for income from capital and income from labor. The options for any particular country are strongly influenced by the tax rates of the existing tax regime, as well as the tax rates of other countries in the region.[116]

Choices as to the level of tax rates include both the positive marginal tax rates and the zero-bracket amounts (also referred to as the tax free minimum or basic allowance). Marginal income tax rates under personal income tax systems have declined substantially in recent decades.[117] The dispersion of rates among countries has also narrowed considerably, particularly within regions.[118] As markets for both labor and capital become more integrated, policymakers face greater constraints on their ability to set tax rates for labor and capital income.[119] Such constraints are clearly much greater for countries in the developing world than they are for larger, more developed countries (like United States and Germany) that may to some extent be able to tax agglomeration (location) rents without unduly discouraging investment.[120] The most common location rents found in developing countries are those in the natural resource sector or those related to government grants of exclusive rights to provide goods or services.[121]

For developing countries with relatively high top marginal rates under their personal income tax systems (35% or greater), the move to a dual income tax regime will likely mean substantial reductions in tax rates (which may allow policymakers the opportunity to expand the tax base).[122] However, the amount of room for tax rate reductions depends on the distribution of taxable income and the relative amounts of tax revenue collected from taxpayers in the highest marginal tax brackets.[123] Even countries with moderate or comparatively low marginal tax rates will likely face pressure to reduce tax rates as part of any major tax reform.

Under the existing tax regimes, countries have taken different approaches to the size of the zero-bracket amount. The choice of the zerobracket amount influences the proportion of the population subject to tax liability and the level of progressivity of the personal income tax, especially for low-income individuals.[124] The choice also affects the relationship between the regular income tax system and presumptive tax regimes that apply to taxpayers who are either operating in the informal sector or are considered too small to be worth including in the regular tax system.

Many developing countries have different types of presumptive tax regimes that seek to tax micro and small businesses by reference to factors other than income, such as sales, number of employees, or size of establishment.[125] The aim of such presumptive tax regimes is to reduce administrative and compliance costs and to educate and develop taxpayers to where they can enter the regular income tax system. When presumptive tax regimes are in place, introducing a dual income tax which taxes corporate (or large enterprise) profits and personal capital income (or small enterprise profits) at the same rate reduces tax differences based on size of enterprise. Another partial solution would be to adopt different tax rates on active business income and passive income from capital, with the former close to the higher labor income rate and the latter closer to the lowest rate on labor income.[126]

Policymakers also need to consider whether to adopt a zero-bracket amount for capital income under a personal income tax system. On equity grounds, providing tax relief for small investors may make sense. However, countries with final withholding regimes for different types of capital income are likely to find it administratively challenging to have a sufficiently good reporting system to provide for refunds for small investors.

The major policy decision in adopting a dual income tax regime is setting the relative tax rates for capital and labor income. While the Nordic countries set the rate for capital income at (or near) the lowest tax rate applicable to labor income, different considerations may apply in developing countries. In the pre-dual income tax regime of the Nordic countries, many types of capital income were nominally subject to the high marginal tax rates applicable to income from labor, although these rates were applied very unevenly.[127] By contrast, in developing countries where almost all tax revenue is from wage withholding in the formal sector, applying (and effectively, collecting) any tax on either active business income (from businesses not in corporate form) or passive income would likely be a tax increase. For countries that apply final withholding taxes under their current personal income tax regimes on passive income such as dividends, interest, and some types of rents and royalties, the existing tax rates on this income from capital are likely substantially below the rates applicable to income from labor.[128] Box 1 sets forth some rate alternatives under a dual income tax for a personal income tax system.

Box 1. Rate Alternatives under Dual Income Tax for Personal Income Tax Systems

Box 1. Rate Alternatives under Dual Income Tax for Personal
Income Tax Systems

Consider the potential consequences from changing the relationship between tax rates for capital and labor. Assuming, for purposes of this discussion, reasonable levels of tax compliance, setting the tax rate on capital at the highest rate applicable to labor income would reverse the tax incentives present in Nordic countries by encouraging taxpayers to characterize capital income as labor income. However, if compliance levels were low, it is unlikely that many potential evaders would have to engage in such strategy. If they want to evade taxes, they can do so more easily by simply under-reporting income. Setting the tax rate on capital at the middle of the labor income tax range would also create some interesting incentives. Taxpayers whose labor income would be taxed at the higher end of the progressive tax rate schedule may seek to characterize part of their income from labor as income from (lower-taxed) capital. On the other hand, taxpayers whose labor income would be taxed at the lower end of the progressive tax rate schedule may try to characterize part of their income from capital as income from labor, but not enough to move the labor income to a higher bracket.[129]

The trade-off is between achieving additional progressivity from taxing labor income and the potential distortions and arbitrage opportunities from the different tax rates applicable to income from labor and capital. While the gap in the Nordic countries between the flat tax rate applicable to income from capital and the top rate under the progressive schedule for labor income was (and continues to be) substantial,[130] the difference between the tax rates for capital and labor income in many developing countries would likely be much smaller, primarily because of the difficulties developing countries would face in imposing such high tax rates on labor income. The smaller the gap between capital and labor income tax rates, the less incentive taxpayers have to disguise the form of their income thereby lowering the potential revenue losses from such strategies.

Another design consideration is how broadly to define the capital income component of a dual income tax under a personal income tax system. Some forms of capital income are simple to define in principle: interest income, dividends, royalties, and rents. In almost all countries, it should be possible to tax the first three of these items at a flat rate under a final withholding regime. The application of final withholding taxes to these payments from firms, whether to domestic or foreign recipients, would improve efficiency by increasing the symmetry of tax treatment of capital income and simplify administration. For many countries, adopting final withholding regimes would both expand the tax base and likely improve the overall progressivity of the tax system.[131]

Assuming taxes on dividends are collected under a final withholding regime (or alternatively, dividends are excluded from income), the question of whether and how to tax capital gains on corporate shares at the individual level remains. One alternative is simply to exempt these capital gains from taxation. If the dual income tax system provided for full integration of distributed earnings and interest payments, any remaining gains on corporate shares would presumably largely reflect inflation, and exemption would be appropriate.[132] A second alternative is to tax capital gains at the individual level. If such gains were taxed when realized, individual taxpayers should be allowed to adjust the cost basis of their shares to reflect any taxed, but undistributed, earnings at the corporate level. A third alternative is to tax capital gains only for shares held for less than a specific period of time, for example, one year. This approach would prevent taxpayers from engaging in schemes to avoid tax on labor income by selling their shares to reduce or evade tax liability. Unfortunately none of these solutions is perfect, or simple to implement.

An interesting design alternative maintains the progressive rate structure for labor income and provides for two different regimes for taxing income from capital: a relatively low flat tax rate on portfolio investment income and a higher flat tax rate on income related to active businesses. Uruguay adopted a version of this dual income tax system (or perhaps more accurately, a tripartite tax system) in July 2007.[133] Under the Uruguayan dual income tax, labor income is taxed at rates ranging from 10% to 25%.[134] Portfolio income is taxed at a 12% rate through a final withholding regime.[135] Self-employed persons can choose between being taxed under the business tax regime (at a 25% tax rate after deducting business expenses) or paying under the labor income tax regime at the 25% rate (with allowance for personal deductions and presumed expenses of 30% of receipts, an effective rate of 17.5%).[136]

2. Relationship of Personal Income Tax Rates and Corporate Tax Rates under a Dual Income Tax.

Once the tax rates are set under the personal income tax system and the line is somehow drawn between capital and labor income, the next issue is the relationship of the personal income tax rate to the corporate income tax rate. As set forth in Box 2, several alternatives exist.

Box 2. Relationship of Tax Rate on Capital under the Personal Tax Regime and the Corporate Tax Rate

Box 2. Relationship of Tax Rate on Capital under the Personal Tax
Regime and the Corporate Tax Rate

Under the “pure” Nordic dual income tax system, the corporate tax rate is set at the tax rate for capital income (and the lowest positive rate on labor income).[137] The income shifting resulting from lower corporate income tax rates has, in several European countries, shown up primarily in the form of lower personal income tax revenues as more small businesses have incorporated to take advantage of the tax differential.[138] For businesses in developing countries that are actually paying tax under the personal income tax system, a lower corporate tax rate would encourage businesses to incorporate. If greater use of the corporate form were to occur in developing countries, this would have the beneficial effect of increasing the size of the formal sector in those countries and expanding the potential tax base reachable with the available administrative resources.

A second alternative sets the corporate tax rate higher than the personal tax rates for either income from labor or income from capital. To the extent corporate tax revenues are derived primarily from large corporations with substantial foreign ownership, the corporate income tax acts as a final withholding tax on the income derived in the country. If the corporate tax applies primarily to businesses engaged in extracting natural resources or other activities that generate location-specific excess returns, then this alternative might raise revenue without substantial economic distortions.[139] However, setting the corporate tax rate higher than the personal tax rate would discourage small businesses from incorporating, and depending on the corporate tax rate chosen compared to tax rates in other countries, may discourage foreign investment.[140]

A third alternative sets the corporate tax rate below the personal tax rates either for income from labor or income from capital. For those countries where income tax revenues are derived almost completely from wage withholding, this alternative achieves many of the same results as the Nordic dual income tax systems. It provides for a lower tax rate on income in corporate form as a means of addressing tax competition concerns. Low corporate taxes could encourage domestic and foreign investment, while progressive tax rates could be maintained for labor income. It would also, however, create incentives for those with substantial income from labor to incorporate to reduce tax liability.

3. Taxation of Normal and Excess Returns.

Economists distinguish between the normal return to capital and rents or excess returns. For debt capital, the normal rate of return is the market rate of interest on debt in the relevant risk class. For equity capital, the normal rate of return is determined by the market rate on stocks with similar risk characteristics.[141] An extensive economic literature argues against imposing tax on the “normal” rate of return on capital, especially in small open economies in which both the elasticity of supply of capital and the desire to increase the capital stock are likely quite high.[142]

Box 3 sets forth several alternative regimes for taxing entities on the returns from debt and equity investments. The first alternative is the conventional corporate income tax which allows the deduction of interest but taxes the full returns to corporate equity without distinguishing between normal and excess returns. The second alternative seeks to tax all returns at the entity level either by disallowing a deduction for interest or by imposing a final withholding tax on dividends and interest at the entity level at the same rate as the corporate tax. The CBIT proposal discussed earlier is an example of this approach. The third alternative seeks to tax only excess returns. In recent years, several alternatives for excluding the normal return on capital from the tax base have been proposed under a variety of names and labels, such as cash flow taxation and Allowance for Corporate Equity (ACE).[143] A dual income tax system with a relatively low but positive tax rate on capital income might be an appropriate compromise between the theoretically optimal zero tax rate on capital in a small open economy and the political economy arguments that might justify high tax rates on capital income.[144] In addition, in at least some instances, countries may wish to consider imposing a separate cash-flow tax for firms that earn returns well in excess of the normal return on capital (for example, because they are able to capture location rents from access to limited natural resources).[145] The fourth alternative provides entities with a deduction for an imputed rate of return on both debt and equity capital and imposes taxes at the individual level on interest, dividends, and capital gains. The final alternative taxes the normal returns to capital at a low flat rate but taxes excess returns at the same progressive marginal tax rates applicable to labor income. Norway has adopted a variation of this approach with respect to income of closelyheld corporations.[146]

Box 3. Treatment of ormal and Excess Returns under Dual Income Tax Systems

Box 3. Treatment of Normal and Excess Returns under Dual
Income Tax Systems

Choosing among the alternatives requires policymakers to balance several competing objectives. Reducing economic distortions will provide support (particularly in small economies) for exempting the normal returns from capital.[147] However, if countries seek to raise the same amount of revenue from corporate taxes, a smaller tax base under a corporate tax regime that exempts normal returns will require higher tax rates. For example, if policymakers seek to raise $100 million in tax revenue, they could raise that amount by choosing to tax only the excess returns at the entity level (alternative 3) and have tax rates of, say, 35%, or tax full returns of debt and equity at the entity level (alternative 2) and have tax rates of 20%.

Finally, with the reduction of personal and corporate tax rates, the need for specific provisions to reduce the double taxation of distributed corporate income (and the double taxation associated with capital gains attributable to previously taxed corporate income) may be much reduced in developing countries. In addition, the potential double taxation also serves to reduce incentives for owner-managers of closely-held corporations to disguise labor income as capital income.[148] With respect to capital gains on corporate shares, if the corporate tax rate is set equal to the capital income tax rate under the dual income tax, then not only can dividends be exempt from withholding taxes but in effect the proportion of capital gains attributable to increased corporate earnings is also taxed at the same rate.[149] Although it is difficult to think of an administratively feasible way to tax the component of real capital gains on shares attributable to goodwill, even this limited leveling out of the tax arbitrage playing field would be a substantial improvement in most developing countries.

C. Advantages and Disadvantages of Dual Income Tax Regimes in Developing Countries

Moving to a dual income tax regime offers both advantages and disadvantages as compared to the current tax regimes in developing countries. As tax systems and tax environments differ greatly among countries, so do the relative advantages and disadvantages, as well as the relative costs and benefits, from undertaking major tax reforms. Moreover, in some instances similar advantages could be achieved by reforms other than moving to a dual income tax regime.

The dual income tax provides greater flexibility to developing countries that wish to reduce tax rates by allowing them to lower tax rates on capital income while maintaining (or adopting smaller reductions for) tax rates on labor income. As discussed in the next section, the ability to set separate rates for labor and capital income is also an important factor in choosing between dual income taxes and flat taxes. Adopting separate tax rates for labor and capital income may have both revenue and progressivity implications. Countries can respond to pressure to reduce tax rates by lowering rates on capital income while maintaining, and perhaps increasing, the current tax revenue generated from labor income. The progressivity implications of moving to a dual income tax are less clear-cut and depend greatly on the distribution of tax burdens for labor and capital income under the current tax system.

In addition, a move to a dual income tax regime could help rationalize the taxation of income from capital, whether income from passive investments or active business operations. A single rate would provide more symmetric and less distorting treatment of income from passive investments. It might also allow countries to conform the relative tax treatment of passive investment income received by resident and nonresident investors. The move to a dual income tax regime could result in significant administrative gains from increased use of information reporting as well as increased use of provisional and, especially, final withholding.[150] The move to a dual income tax could also help rationalize the tax treatment of income from active businesses under the personal or corporate tax regimes. Adopting the same tax rate for income from active business under the personal tax system and under the corporate tax system will eliminate tax-induced distortion as to choice of business form.

A dual income tax regime with relatively low tax rates on business income and relatively low corporate tax rates may represent an improvement over tax regimes that combine high tax rates on capital with special tax incentives, especially those incentives designed to attract foreign investment. Whether this move to a dual income tax would lower costs of capital and increase domestic and foreign investment is a difficult empirical question and likely varies depending on country-specific factors.[151] It depends partly on how successful past tax incentives have been in attracting investment in developing countries. Although the evidence is not entirely clear, such incentives have been more successful in increasing the inflow of foreign direct investment to middle-income countries or those with significant natural resources than to low-income countries without significant natural resources, which have made most use of these incentives.[152] It is also not clear whether tax incentives generate additional new investment or new ownership of existing investments.[153] Although foreign capital inflows appear to respond to tax incentives, what may be changing is more the ownership of capital than the amount of real investment.[154] The relative advantage of a dual income tax over current regimes that pair high tax rates with high tax incentives also depends on how successful a low flat rate tax on capital would be in increasing the level of investment as compared to an existing regime with high tax rates and generous tax incentives. While supporters of low flat tax rates claim low taxes have increased investment and increased tax revenue in several countries, it is difficult to isolate the effect of tax changes on the costs of capital from the effect of other tax and fiscal reforms.[155]

Removing tax preferences and exclusions and simply moving to a lower statutory rate would simplify tax regimes for both individual and corporate taxes, thus reducing both administrative and compliance costs. The latter point is particularly important because such costs appear to create a substantial fiscal barrier to the formalization of many economic enterprises in developing countries.[156] The move to a dual income tax regime could also provide a platform on which to incorporate other related reforms, such as coordinating presumptive tax regimes for small businesses with the regular corporate tax regime and integrating the personal income tax system with the payroll or social security tax regimes. Thus, adopting a dual income tax regime may provide the opportunity to coordinate with existing presumptive tax regimes and facilitate moving to a regular tax system.[157]

For countries with substantial payroll taxes, the dual income tax provides an opportunity to impose a single tax on labor income that would be both more transparent and easier to administer. For that matter, simply integrating the administration of payroll taxes with labor income taxes would substantially simplify and improve both systems in many countries.[158]

The move to a dual income tax also has costs and disadvantages. Any major tax reform imposes substantial administrative costs on both taxpayers and tax authorities. Depending on the current tax systems in developing countries, the move to a dual income tax could substantially raise the cost of debt capital, especially if countries adopt more effective final withholding regimes. Differences in tax rates on labor and capital income, as well as any differences between tax rates on capital income under the personal tax system and corporate tax rates, would also present incentives for taxpayers to structure the type and form of income to reduce tax liability. However, this problem is likely of less concern in developing countries, because of the smaller difference in the effective marginal tax rates for capital income, labor income and corporate income rate than exists in developed countries.[159]

Finally, it is important not to compare the dual income tax to some ideal tax regime with high levels of tax compliance, but rather to the current imperfect tax regimes and compliance rates in the particular country.[160] Difficulties exist in estimating the amount of any revenue or progressivity gains or losses from the move to a dual income tax without examining the details of the proposed reforms and the effectiveness of the existing tax regimes. Revenue gains from adopting a dual income tax would likely be greater in those countries where the prior tax treatment of income from capital was ineffective and where the change to the dual income tax might reduce the size of the informal economy by reducing tax barriers to formalization. Similarly, in developing countries that have relatively ineffective taxation of capital income, a dual income tax may even produce some progressivity gains, particularly if reforms succeed in reducing taxes on income of low wage earners and increasing taxes on income of higher wage earners and capital income.

IV. DUAL INCOME TAX VERSUS FLAT TAX REGIMES

If policymakers in developing countries wish to retain an income tax, there are competing tax reform options. Choices include trying to improve the comprehensive income tax model, adopting a dual income tax, or adopting a flat rate tax.[161] The choice between a dual income tax and a flat tax appears relatively straightforward: the dual income tax provides for a single tax rate only on capital income whereas the flat tax applies a single rate to all types of income. The differences between dual income tax and flat tax systems, however, may be greater or less depending on several factors, such as the zero bracket amount and rate structure under the personal income tax, the tax rate differential between the personal income tax system and the corporate tax system, the integration of the personal and corporate tax systems, and the presence of payroll or social security tax on labor income.

A. Recent Flat Tax Reforms

Several recent tax reforms, particularly in Central and Eastern Europe and the countries of the former Soviet Union, have followed a flat tax approach. Table 3 provides information about tax systems in these countries both before and after flat tax reforms.

Table 3. Flat Tax Regimes162

Table 3. Flat Tax Regimes[162]

Almost all of these flat tax reforms are income tax systems that apply a single tax rate above a zero-bracket amount.[163] Though clumped together under the flat tax rubric, these reforms vary substantially across countries and over time. They differ in choice of the tax rate as compared to the range of tax rates before the reform, the types of income subject to the single tax rate, the size of the zero-bracket amount, the treatment of interest, dividends, and capital gains, and the relationship of the personal income tax rate and corporate tax rates before and after the reform. They also differ with respect to interaction with other taxes on labor, such as social security or payroll taxes.

These flat tax reforms are not the consumption-style flat taxes of the type that dominated recent tax discussion in the United States.164 These reforms retain income as the tax base, but eliminate multiple positive marginal tax rates with respect to certain types of income.[165] The first wave of flat tax reforms (Estonia 1994; Lithuania 1994; and Latvia 1995) generally provided a moderately high single tax rate as compared to the previously existing tax rates in those countries and as compared to international tax rates.[166] The reforms applied only to the personal income tax systems and left the separate corporate tax regime (with often higher corporate tax rates) unchanged. In contrast, the second wave of flat tax reforms (for example, Russia 2001; Kazakhstan 2007; and Bulgaria 2008) provided for a flat rate substantially below previous personal income tax rates and also low relative to international tax rates.

As set forth in Table 4, existing flat tax regimes have taken different approaches as to both the level of tax rates and the relationship of the flat tax rates under the personal tax system and the tax rates under the corporate tax system.

Table 4. Relative Personal and Corporate Income Tax Rates under Flat Tax Regimes

Table 4. Relative Personal and Corporate Income Tax Rates under
Flat Tax Regimes

Flat tax systems also differ as to the extent to which they integrate the personal and corporate tax systems. Most flat tax systems retain the double tax on distributed corporate income (often with final withholding regimes), although some exclude tax on further distributions.

B. Choosing Between a Dual Income Tax and Flat Tax System

Comparisons between the dual income tax and the flat tax model depend on the specific details of the proposals as well as the specific circumstances of particular countries. For dual income tax systems with a tax rate on capital income set at the lowest marginal tax rate on labor income, the dual income tax can be designed as a flat tax with a surcharge on high-income wage earners. In other words, the difference between the two approaches is essentially what is gained or lost by having a progressive rate schedule for labor income combined with a flat tax rate on capital income.

Flat tax and dual income tax systems share several advantages over the current tax systems in many developing countries. As discussed in the previous section, a single tax rate on income from passive investments under the personal income tax system will generally provide more symmetric and less distorting treatment of this type of capital income, reducing wasteful tax-induced distortions in investment and saving decisions. A single tax rate would also be a simpler and more effective way to tax capital income and would facilitate use of withholding regimes. Combining either a flat tax or a dual income tax reform under the personal income tax system with an effective corporate income tax reform may result in a more rational approach for taxing active business income. Where the move to either a flat tax or a dual income tax results in reforms that lower tax rates and broaden tax bases, existing capital should be used more efficiently and the flow of capital into the economy should be increased.[168] Both reforms may in some circumstances also expand revenues and make the tax system more progressive with less administrative strain and effort than required under a comprehensive income tax approach. This administrative advantage is not a minor consideration in developing countries in which administrative capacity is a severe constraint on tax policy.[169]

Flat tax regimes have advantages over dual income tax regimes. For example, if a tax system really has only a single rate on all income, it would be simpler to explain, to comply, and to enforce than other systems.[170] If the single rate under the personal income tax system is also aligned with the corporate rate, distortions would be reduced and the opportunity for arbitrage would be less than under a dual income tax.

On the other hand, dual income tax regimes have several advantages over flat tax regimes. By decoupling the taxation of wage and capital income tax, policymakers gain an additional degree of freedom to address global tax competition concerns or domestic political concerns from various interest groups. It may also be a bit easier for policymakers to resist the endless clamor for special concessions and hence avoid complicating either component of the income tax system by various incentive provisions.

Perhaps the strongest argument for choosing a dual income tax over a flat tax is to maintain at least a modest degree of explicit progressivity in the tax system. Market-driven developing countries generate both growth and inequality. All countries face challenges in maintaining high growth rates without being derailed by political and social tensions created by growing inequality. For those who believe that an essential element in sustaining growth and political stability is to provide some check on the ability of those with great private incomes and wealth to influence political outcomes, a sustainable democratic tax system must have some explicit way of taxing the rich more than the poor.[171] Since personal wealth taxes seem unlikely to become any more popular anywhere than they are now, the combination of a more effective flat rate taxation of capital income, a high proportion of which accrues to the rich, and a moderately progressive tax on wage income, which has in recent years become increasingly unequal in many countries, may be an essential ingredient in any sustainable fiscal solution. This is exactly the combination provided by a dual income tax.

In most developing countries sustainable growth also probably requires some expansion of the state to replace the social safety net formerly provided for much of the population by extended rural families.[172] An increasingly educated population and expanded infrastructure are also needed. All this may call for larger government programs supported primarily by higher tax revenues. At least modest progressivity in the taxation of labor income may be necessary to provide adequate revenues to fund government programs that are likely to be an essential component of the growth and economic and political sustainability of most countries. The dual income tax may thus both look better and work better than the flat tax in the context of many developing countries precisely because it includes a more explicit progressive element in the form of a progressive rate structure rather than simply a zero-bracket. In the social and political context of such countries it is not enough simply to tax the not-so-poor at a higher rate than the poor; as a rule, one must also tax the rich more than the merely well-off and more than the average worker as well.[173]

Although it is difficult to estimate the revenue and progressivity effects of moving to a dual income tax or flat tax regime, the introduction of an explicit dual income tax approach may thus strengthen rather than weaken the role of the personal income tax and the overall tax system in developing countries. An additional consideration of significance in some more decentralized countries may be that adopting a dual income tax system would make it simpler for sub-national governments to piggyback on the labor portion of the tax.[174]

V. CONCLUSION

Moving to a dual income tax system offers both advantages and disadvantages as compared to current tax regimes in developing countries. A dual income tax provides countries with greater flexibility to address tax competition while maintaining a progressive tax regime for labor income. A flat tax rate on capital may provide an opportunity both to rationalize the taxation of different types of passive income from capital and to eliminate many of the tax preferences contained in the personal and corporate tax systems. Generally, the move to a dual income tax system provides a uniform lower capital tax rate which, if set at the same level as the corporate income tax, could reduce the apparently fatal attraction that draws many developing countries towards tax incentives that often erode the tax base, provide incentives for rent-seeking and corruption, and provide greater benefits to foreign, rather than, domestic investors. At the same time, retaining a progressive tax on labor income will provide some progressivity in countries where economic changes result in greater inequality. The combination of these two features provides an opportunity to improve tax policy and tax administration substantially in many developing countries.

Dual income tax systems have worked relatively well in the Nordic countries. If developing countries seek to follow this approach, they have many options to choose from with respect to the details of a proposed dual income tax regime. The flat personal tax rate on income from capital could either follow the Nordic model (set at or near the lowest positive tax rate for labor income), or fall at the middle or top of the progressive rate schedule for labor income. To improve compliance, countries could provide for final withholding for dividends and interest payments from corporations. Ideally, final withholding would be expanded to cover other payors.[175] Developing countries could also choose to follow the Uruguayan model, and provide for a flat tax rate on passive income from capital at the lowest positive tax rate on labor income and a flat tax rate from capital from sole proprietorships and closely held corporations at the middle or highest rates.[176] Alternatively, developing countries could elect to adopt different tax treatment for normal and excess returns under dual income tax regimes, either exempting normal returns from tax or taxing the normal returns at a preferential low flat tax rate and taxing excess returns at the progressive marginal tax rates applicable to labor income.

While the dual income tax provides for a single rate on income from capital, at least initially, the flat tax adopted in Eastern European countries and the countries of the former Soviet Union provided for a single tax rate on income from labor.[177] These regimes, however, generally do not integrate payroll taxes with the income tax. This results in considerably higher taxes on labor than on capital income. Moving to an explicit dual income tax in which the labor income tax includes the payroll tax would provide an opportunity to rationalize the generally erratic distributional effects of the existing personal income tax and payroll tax on labor income. To the extent that the flat tax regimes increase the coverage of income from capital under the personal income tax and income under the corporate tax, these tax regimes also may achieve the goals of symmetry and comprehensiveness, and depending on the rate adopted, competitiveness. Although it is an empirical question, it seems likely that in the context of most developing countries a dual income tax would be more progressive than a flat tax regime at comparable tax rates.

Developing countries face different and difficult challenges in reforming their tax system; different environments, different objectives, capacities and policy trade-offs all come into play. Given these differences, no one approach can resolve the tax challenges facing developing countries. Nonetheless, the dual income tax approach appears to provide an exceptionally promising basis on which to construct more rational, sustainable, productive, and perhaps even more progressive tax systems in many developing countries.


  1. Legal sociologists sometimes argue that “legal transplants” are virtually impossible. Pierre Legrand, What “Legal Transplants”?, in ADAPTING LEGAL CULTURE 55 (David Nelken & Johannes Feest eds., 2001).
  2. Tax systems are considered global if they include income from all sources in a common tax base. Tax systems are progressive if higher marginal income tax rates apply to greater amounts of taxable income. The general recommendations of tax advisors in recent decades for a “broad-based, low rate” approach to tax reform in developing countries do not represent a shift away from the comprehensive income tax approach. Richard M. Bird, The BBLR Approach to Tax Reform in Emerging Countries, in PUBLIC ECONOMICS: THEORY AND POLICY (M. Govinda Rao & Mihir Rakshit eds., forthcoming 2010).
  3. Distinguished academics such as Stanley Surrey and Richard Musgrave spread the word to developing countries to help modernize and improve the design and implementation of tax systems. See, e.g., Stanley S. Surrey & Oliver Oldman, Report of Preliminary Survey of the Tax System of Argentina, 16 PUBLIC FINANCE 155 (1961); RICHARD A. MUSGRAVE & MALCOLM GILLIS, FISCAL REFORM FOR COLOMBIA: FINAL REPORT AND STAFF PAPERS OF THE COLOMBIAN COMMISSION ON TAX REFORM (1971); RICHARD A. MUSGRAVE, FISCAL REFORM IN BOLIVIA: FINAL REPORT OF THE BOLIVIAN MISSION ON TAX REFORM (1981).
  4. In all countries, a large portion of private savings is tax-favored. Low or no taxes apply to savings in owner-occupied housing, pension and other retirement savings. See, e.g., Edward J. McCaffery, Tax Policy under a Hybrid Income-Consumption Tax, 70 TEX. L. REV. 1145 (1992). Weak tax administration in developing countries accentuates both the equity and efficiency problems that arise from such differential treatment. Richard M. Bird & Eric M. Zolt, Tax Policy in Emerging Countries, 26 ENV’T & PLAN C: GOV’T & POL’Y 73 (2008).
  5. Vito Tanzi & Howell H. Zee, Tax Policy for Emerging Markets: Developing Countries (Int’l Monetary Fund, Working Paper No. 00/35, 2000).
  6. Richard M. Bird & Eric M. Zolt, Redistribution via Taxation: The Limited Role of the Personal Income Tax in Developing Countries, 52 UCLA L. REV. 1627 (2005).
  7. Michael Keen & Alejandro Simone, Tax Policy in Developing Countries: Some Lessons from the 1990s, and Some Challenges Ahead, in HELPING COUNTRIES DEVELOP: THE ROLE OF FISCAL POLICY 302, 327–32 (Sanjeev Gupta, Ben Clements & Gabriela Inchauste eds., Int’l Monetary Fund 2004).
  8. Friedrich Schneider & Dominik H. Enste, Shadow Economies around the World: Size, Causes, and Consequences (Int’l Monetary Fund, Working Paper No. 00/26, 2000).
  9. For a useful review of such proposals that have been made in the U.S., see Charles McLure & George Zodrow, Consumption-Based Direct Taxes: A Guided Tour of the Amusement Park, 63 FINANZARCHIV: PUB. FIN. ANALYSIS 285 (2007).
  10. CAROLYN WEBBER & AARON WILDAVSKY, A HISTORY OF TAXATION AND EXPENDITURE IN THE WESTERN WORLD 526 (1986) note that the extent to which a nation’s finances rely on the taxation of income has historically been considered “a mirror of democracy” in the sense that it symbolized the commitment to social justice. More recently, the important political dimension of taxation as a means of building state capacity in developing countries has been emphasized. See generally TAXATION AND STATE-BUILDING IN DEVELOPING COUNTRIES (Deborah Brautigam, Odd-Helge Fjeldstad & Mick Moore eds., 2008).
  11. Ke-young Chu, Hamid Reza Davoodi & Sanjeev Gupta, Income Distribution and Tax and Government Social Spending Policies in Developing Countries 35 (UNU/WIDER Working Papers No. 214, 2000).
  12. Peter Birch Sørensen, Can Capital Income Taxes Survive? And Should They?, 53 CESIFO ECON. STUD. 172, 181 (2007).
  13. See generally Eric M. Zolt, The Uneasy Case for Uniform Taxation, 16 VA. TAX REV. 39 (1996).
  14. Certain dividends are subject to tax as net capital gains under I.R.C. § 1(h)(11) and capital gains qualify for lower tax rates. I.R.C. § 1(h) (2010).
  15. For example, Senators Ron Wyden (D-Ore) and Judd Gregg (R-N.H.) have proposed legislation to reduce corporate tax rates from 35% to 24%. Bipartisan Tax Fairness and Simplification Act of 2010, S. ___, 111th Cong. § 2 (2010).
  16. See infra notes 70–85 and accompanying text. DEPARTMENT OF THE TREASURY, INTEGRATION OF THE INDIVIDUAL AND CORPORATE TAX SYSTEMS—TAXING BUSINESS INCOME ONCE 106 (1992) [hereinafter United States Treasury]. One of the co-authors of this Article, Eric Zolt, while serving as Deputy Tax Legislative Counsel, Office of Tax Policy, was one of the primary authors of the Treasury Report.
  17. See also Edward D. Kleinbard, An American Dual Income Tax: $ordic Precedents, 5 NW. J. L. & SOC. POL’Y 41 (2010) (arguing that an implementable dual income tax would likely be superior on equity and efficiency grounds to the current tax system).
  18. For reviews of dual income systems in developed countries, see Sijbren Cnossen, Taxing Capital Income in the $ordic Countries: A Model for the European Union?, in TAXING CAPITAL INCOME IN THE EUROPEAN UNION: ISSUES AND OPTIONS FOR REFORM 180, 211 (Sijbren Cnossen ed., 2000); Bernd Genser & Andreas Reutter, Fiscal Policy in Action: Moving Toward Dual Income Taxation in Europe, 63 FINANZARCHIV: PUB. FIN. ANALYSIS 436 (2007); Kleinbard, supra note 17; Peter Birch Sørensen, Dual Income Taxation: Why and How?, 61 FINANZARCHIV 559 (2005) [hereinafter Sørensen (2005)]; Peter Birch Sørensen, From the Global Income Tax to the Dual Income Tax: Recent Tax Reforms in the $ordic Countries, 1 INT’L TAX & PUB. FIN. 57 (1994) [hereinafter Sørensen, From the Global Income Tax]. Earlier discussions of the dual income tax in developing countries may be found in Robin W. Boadway, Income Tax Reform for a Globalized World: The Case for a Dual Income Tax, 16 J. ASIAN ECON. 910 (2005) and Alberto Barreix & Jerónimo Roca, Strengthening a Fiscal Pillar: The Uruguayan Dual Income Tax, 92 CEPAL REV. 121 (2007).
  19. Sørensen, From the Global Income Tax, supra note 18, at 60–61.
  20. The data for this table are from Sørensen, From the Global Income Tax, supra note 18, at 59, IBFD, GLOBAL INDIVIDUAL TAX HANDBOOK (2008), and IBFD, GLOBAL CORPORATE TAX HANDBOOK (2008).
  21. Steffen Ganghof provides an excellent review of the political considerations in Denmark, Finland, Norway, and Sweden in reforming their tax systems to adopt dual income taxes. STEFFEN GANGHOF, THE POLITICS OF INCOME TAXATION: A COMPARATIVE ANALYSIS 77–111 (2006).
  22. For example, Sweden’s top marginal rate under the personal income tax system was 87% in 1979, 65% in 1990, and 51% in 1991. Sven Larson, The Swedish Tax System: Significant Features and Lessons for Policymakers, 43 TAX NOTES INT’L 395, 396 (2006). Sweden was not alone in reducing tax rates. Peter, Buttrick, and Duncan note that the GDPweighted average top statutory marginal personal income tax rate in the 108 countries for which they have data for the entire period was 62% in 1981, fell to 43% in 1991 and continued to decline to only 36% in 2005. Klara Sabirianova Peter, Steve Buttrick & Denvil Duncan, Global Reform of Personal Income Taxation, 1981–2005: Evidence from 189 Countries (IZA Discussion Paper No. 42228, 2009).
  23. Sørensen, From the Global Income Tax, supra note 18; Peter Birch Sørensen, The $ordic Dual Income Tax: Principles, Practices, and Relevance for Canada, 55 CAN. TAX J. 557, 565–66 (2007) [hereinafter Sørensen, The $ordic Dual Income Tax].
  24. Sørensen, From the Global Income Tax, supra note 18; Sørensen, The $ordic Dual Income Tax, supra note 23.
  25. Sørensen, From the Global Income Tax, supra note 18, at 62–64; Sørensen, The $ordic Dual Income Tax, supra note 23, at 11.
  26. Victor Thuronyi, Adjusting for Inflation, in TAX LAW DESIGN AND DRAFTING 434 (Victor Thuronyi ed., 1998).
  27. Sørensen, From the Global Income Tax, supra note 18; Sørensen, The $ordic Dual Income Tax, supra note 23.
  28. Sørensen, From the Global Income Tax, supra note 18; Sørensen, The $ordic Dual Income Tax, supra note 23.
  29. Sørensen, The $ordic Dual Income Tax, supra note 23, at 10.
  30. Sørensen contends that before introducing dual income tax regimes, Norway and Sweden likely had negative tax revenue on capital income under the personal income tax system. Id. Similar results have been found in other developed countries. For a discussion of revenue from taxation of capital income in the United States, see Roger Gordon & Joel Slemrod, Do We Collect Any Revenue from Taxing Capital Income?, 2 TAX POL. & THE ECON. 89 (1988) and Roger Gordon, Laura Kalambodidis & Joel Slemrod, Do We Now Collect Any Revenue from Taxing Capital Income?, 88 J. PUB. ECON. 981 (2004); for a discussion of revenue from taxation of capital income in Germany, see Johannes Becker & Clemens Fuest, Does Germany Collect Revenue from Taxing Capital Income? (CESifo Working Paper No. 1489, 2003).
  31. Cnossen, supra note 18, at 191.
  32. The U.S. took a different approach to protect the labor tax base in its 1986 reform by adopting loss rules that prevented taxpayers from sheltering labor income from losses associated with passive investments. I.R.C § 469 (2010).
  33. Sørensen, The $ordic Dual Income Tax, supra note 23, at 10.
  34. After a decade with an unemployment rate averaging only 2.5%, unemployment in Norway began to rise sharply in 1989 to a peak of 6.0% in 1993, followed by gradual decline to 3.2% by the end of the 1990s. Unemployment in Sweden was even lower in the 1980s, but began to rise sharply in 1991, peaking at 10.0% in 1997. OECD Statistics, LFS by sex and age – indicators, http://stats.oecd.org/Index.aspx?DatasetCode=LFS_SEXAGE_I_R (select “Unemployment rate” in Series; then select “All persons” in Sex; then select “Total” in Age; then click “Time & Frequency” hyperlink and select 1979 to 1999; then click “View Data”) (last visited May. 14, 2010).
  35. For a discussion of the politics of tax reform in Sweden, see Sven Steinmo, Globalization and Taxation: Challenges to the Swedish Welfare State, 3 COMP. POL. STUD. 839 (2002).
  36. Peter, Buttrick & Duncan, supra note 22, at 48 (on individual income taxes) and Michael Devereux, Ben Lockwood & Michela Redoano, Do Countries Compete Over Corporate Tax Rates?, 92 J. OF PUB. ECON. 1210, 1222 (2008) (on corporate income tax rates).
  37. For information on changes in reliance on consumption taxes in the Nordic countries, see OECD, CONSUMPTION TAX TRENDS, 2008 EDITION Tables 3.1, 3.2, at 41–42 (2009). For an interesting detailed appraisal of the historic evolution of the Nordic tax financing model, see PETER LINDERT, GROWING PUBLIC: SOCIAL SPENDING AND ECONOMIC GROWTH SINCE THE EIGHTEENTH CENTURY 235–45, 267–95 (2003).
  38. Steffen Ganghof, Global Markets, $ational Tax Systems, and Domestic Politics: Rebalancing Efficiency and Equity in Open States’ Income Taxation 7–14 (Max Planck Institute Discussion Paper, 2001) [hereinafter Ganghof (2001)]; Steffen Ganghof, The Politics of (Income) Tax Structure (Yale Conference on Distributive Politics, Working Paper, 2005).
  39. Ganghof (2001), supra note 38, at 8.
  40. Id.
  41. Id. at 8–9.
  42. Id.
  43. Id. at 9.
  44. Id.
  45. Id. at 9–11.
  46. Sørensen, supra note 12, at 181–84.
  47. See Ganghof (2001), supra note 38, at 14–25. 48.
  48. Id. at 19–23.
  49. Denmark chose to sacrifice symmetry also by imposing differential tax rates on capital income. Australia also sacrificed symmetry, but in a different way, by electing to cut only the corporate income tax rate. New Zealand chose instead to cut all income tax rates, thus sacrificing progressivity. Ganghof (2001), supra note 38, at 14–25. See also Steffen Ganghof and Richard Eccleston, Globalization and the Dilemmas of Income Taxation in Australia, 39 AUST. J. POL. SCI. 519, 530 (2006). In contrast, Germany, until its most recent reforms, sacrificed competitiveness in order to retain the other three objectives.
  50. Good descriptions of the Nordic dual income tax systems may be found in Genser and Reutter, supra note 18, and Sørensen, From the Global Income Tax, supra note 18.
  51. Sørensen, From the Global Income Tax, supra note 18, at 61.
  52. In the initial dual income tax reforms, only Norway aligned the personal tax rate on capital income with the lowest positive tax rate on labor income. In Finland, the rate on capital income was slightly higher than the lowest positive labor tax rate and in Sweden the personal income tax rate on capital income was slightly lower than the lowest positive labor tax rate. Denmark kept progressive taxes on both labor and capital income, with lower rates on the latter.
  53. A truly “pure” version would provide for full integration of personal and corporate level taxes to avoid double taxation of corporate income. For example, Sørensen and Johnson propose combining a dual income tax with an allowance for corporate equity to eliminate double taxation. Peter Birch Sørensen & Shane Matthew Johnson, Taxing Capital Income—Options for Reform in Australia 100 (Paper prepared for Australia’s Future Tax System Conference June 2009), available at http://taxreview.treasury.gov.au/content/html/conference/downloads/attachment_08_Draft_P eter_Birch_Sorensen_paper.pdf (last visited May 14, 2010). But integration is not a necessary consequence of aligning the capital income and corporate income tax rates. Norway historically provided the greatest degree of integration. Until 2005, Norway eliminated the double taxation of equity income by providing an imputation system for distributed dividends and allowed shareholders to increase the basis of their corporate shares by their pro-rata amount of the retained profits that was subject to corporate tax. In the Norwegian tax system this was known by its Norwegian acronym RISK. Vidar Christiansen, $orwegian Income Tax Reforms, 2 CESIFO DICE REPORT, Autumn 2004, at 9, 10. A similar version was proposed by the U.S. Treasury Department as a deemed dividend reinvestment plan (DRIP) system. United States Treasury, supra note 16, at 106.
  54. Robin W. Boadway, The Dual Income Tax System—An Overview, 2 CESIFO DICE REPORT, Autumn 2004, at 3, 7. For such a tax proposal for Australia, see Sørensen and Johnson, supra note 53. A dual income tax could include returns on consumer durables and returns to human capital accumulation, although no proposal has yet gone that far.
  55. Cnossen, supra note 18, at 183–86.
  56. Kleinbard, supra note 17, at 54–55.
  57. Id.
  58. Sørensen, supra note 12, at 213.
  59. For example, the Norwegian approach uses the interest rate on 5-year government bonds plus a risk premium of 4%. If the imputed rate of return equals the interest on business debt, it will not matter whether the calculation is based on gross assets or net assets (excluding liabilities). If the rates differ, taxpayers under a net asset regime may have incentives to adjust their borrowings to maximize the amount of income from capital.
  60. Sørensen, From the Global Income Tax, supra note 18, at 73–75.
  61. Sørensen notes that Norway adopted a version of the gross asset approach, Finland and Sweden adopted versions of the net asset approach, and Denmark adopted a regime that allows taxpayers to choose between variations of the two approaches. Sørensen (2005), supra note 18, at 573.
  62. A variation of this approach uses the costs of shares as the base for computing the return to capital and then treats any excess returns as labor income.
  63. Sørensen, From the Global Income Tax, supra note 18, at 75–76; Kleinbard, supra note 17, at 64–67.
  64. In part because of the difficulties in taxing closely held businesses, Norway changed its dual income tax regime to tax the normal returns to capital at a lower flat rate, but tax excess returns of closely held corporations at the higher progressive tax rates applicable to labor income. Sørensen (2005), supra note 18, at 575–79; Peter Birch Sørensen, $eutral Taxation of Shareholder Income, 12 INT’L. TAX & PUB. FIN. 777 (2005).
  65. Sørensen notes that the mandatory income-splitting features of the Norwegian tax system worked relatively well when applied to self-employed taxpayers, but were less effective when applied to active owners of small corporations. Sørensen, supra note 12, at 214–17. These employee-shareholders could avoid the mandatory income-splitting rules by increasing the percentage of shares owned by passive investors to avoid treating part of the income they received from the corporation as income from labor. Beginning in 2006, active owners of small businesses in Norway were taxed at the lower capital income tax rates on the imputed normal rate of return on the value of their shareholdings, but at the higher effective tax rate applicable to labor income for amounts of realized income (dividends and capital gains) in excess of the normal rate of return. The business sector in all countries consists of very heterogeneous firms. Consequently, no one right system to distinguish labor and capital income exists, even in the most technologically and administratively advanced countries.
  66. Sørensen, From the Global Income Tax, supra note 18, at 75–76; Kleinbard, supra note 17, at 66–67.
  67. See, e.g., OECD, OECD TAX POLICY STUDIES NO. 13, FUNDAMENTAL REFORM OF PERSONAL INCOME TAX, (2006) (providing a brief review of dual and “semi-dual” income tax systems in OECD countries); Barreix and Roca, supra note 18 (reviewing dual income tax systems in Latin America and Spain).
  68. Zolt, supra note 13.
  69. In Europe, for instance, such countries include Austria, Belgium, Italy, Portugal, the Czech Republic, Lithuania, and Poland. Genser & Reutter, supra note 18, at 448.
  70. United States Treasury, supra note 16, at 39–60.
  71. Investment decisions with respect to industry, asset mix, location, risk-taking, and timing may be influenced by variations in effective tax rates. Inter-temporal decisions, like inter-sectoral decisions, are also affected by taxes on capital income, with the result that private savings are diminished. Moreover, the complexity of corporate taxes may impose significant costs and barriers to the expansion of new and small firms, while uncertainty as to the precise tax implications of various corporate decisions may act as a general deterrent to investment. For a recent comprehensive review of the distortions arising from typical corporate income taxes see Alan J. Auerbach, Michael P. Devereux & Helen Simpson, Taxing Corporate Income, Paper prepared for the Mirrlees Report (March 2008).
  72. United States Treasury, supra note 16, at 3–14.
  73. Corporate income is taxed first when earned at the corporate level and then again upon distribution to shareholders under the personal income tax. Corporate income may also be subject to double taxation when shareholders recognize capital gains attributable to previously taxed, but not distributed, corporate income.
  74. United States Treasury, supra note 16, at 12.
  75. Id. at 15–60.
  76. Id. at 12–14.
  77. See, e.g., Ellen P. Aprill & Sanford Holo, Choice of Entity: Considerations and Consequences, in 61 MAJOR TAX PLANNING Ch. 5 (USC Inst. Fed. Tax’n 2009).
  78. William A. Klein & Eric M. Zolt, Business Form, Limited Liability, and Tax Regimes: Lurching Toward a Coherent Outcome?, 66 U. COLO. L. REV. 1001, 1010–17 (1995).
  79. Many countries treat small businesses differently for various reasons of varying persuasiveness. For a recent discussion of taxing small businesses in developing countries, see WORLD BANK GROUP, DESIGNING A TAX SYSTEM FOR MICRO AND SMALL BUSINESSES: GUIDE FOR PRACTITIONERS (2007).
  80. United States Treasury, supra note 16, at 67–69.
  81. G. Mitu Gulati, William A. Klein & Eric M. Zolt, Connected Contracts, 47 UCLA L. REV. 887, 941–43 (2000).
  82. The U.S. Treasury Department has generally required full deductibility of interest as a requirement for a foreign country’s income tax system to qualify as a tax on “net income” for tax credit purposes. See Charles E. McLure, Jr. & George M. Zodrow, The Economic Case for Foreign Tax Credits for Cash Flow Taxes, 51 NAT’L TAX J. 1 (1998).
  83. The move to the CBIT regime from the current regime would benefit those entities that rely relatively more on equity finance and penalize those firms that rely relatively more on debt finance.
  84. This approach was problematic in the United States because of the current tax regime applicable to the large holdings of tax-exempt and non-resident investors in the U.S. corporate debt markets. This would likely be of less concern in developing countries. For a discussion of CBIT with final withholding on interest, see Cnossen, supra note 18, at 211.
  85. As we discuss briefly later, it is difficult to determine how best to treat capital gains under any income tax regime, whether in developed or developing countries. See infra notes 148–149 and accompanying text.
  86. Roy Bahl & Richard M. Bird, Tax Policy in Developing Countries: Looking Back—and Forward, 61 NAT’L TAX J. 279 (2008); Tanzi & Zee, supra note 5, at 8.
  87. Bird & Zolt, supra note 6, at 1630; Tanzi & Zee, supra note 5, at 9–15.
  88. Bird & Zolt, supra note 6, at 1653–60.
  89. Roger Gordon & Wei Li, Tax Structure in Developing Countries: Many Puzzles and A Possible Explanation 4 (Nat’l Bureau of Econ. Research, Working Paper, No. 11267, 2005).
  90. Schneider & Enste, supra note 8.
  91. TAXING THE HARD-TO-TAX: LESSONS FROM THEORY AND PRACTICE (James Alm, Jorge Martinez-Vazquez & Sally Wallace, eds., 2004).
  92. For estimates of the size of the informal sector in developing countries, see James Alm, Jorge Martinez-Vazquez, & Friedrich Schneider, “Sizing” the Problem of the Hard-to- Tax, in TAXING THE HARD-TO-TAX: LESSONS FROM THEORY AND PRACTICE 11, 24 (James Alm, Jorge Martinez-Vazquez & Sally Wallace eds., 2004). The size of the informal economy may itself be a function of the design and implementation of the tax system. For example, the high social insurance tax rates levied by some countries create an incentive for a large informal economy by discouraging employers from reporting the extent of employment and encouraging the under-reporting of wages. Jan Rutkowski, Taxation of Labor, in FISCAL POLICY AND ECONOMIC GROWTH: LESSONS FOR EASTERN EUROPE AND CENTRAL ASIA 281 (Cheryl Grey, Tracey Lane & Aristomene Varoudakis eds., 2007). The resulting lower tax revenues often lead governments to raise tax rates still further, thus exacerbating incentives to evade taxes.
  93. For example, Bolivia adopted an enterprise tax that applied to all businesses regardless of legal form. MUSGRAVE, supra note 3, at 320.
  94. Several Latin American countries such as Mexico and Colombia have imposed a presumptive tax on assets as a form of alternative minimum tax. For further discussion, see Vito Tanzi & Efraim Sadka, A Tax on Gross Assets of Enterprises as a Form of Presumptive Taxation (IMF Working Paper, No. 92/16, 1992). Such systems are also found in some developed countries. For example, France taxes small farmers presumptively on the basis of a presumed return on land, and Israel uses a similar system, in part based on presumed return on assets, more widely. Victor Thuronyi, Presumptive Taxation of the Hard-to-Tax, in TAXING THE HARD-TO-TAX: LESSONS FROM THEORY AND PRACTICE 101, 111–12. 118–19 (James Alm, Jorge Martinez-Vazquez & Sally Wallace eds., 2004).
  95. Bird & Zolt, supra note 6, at 1658–59.
  96. Ved. P. Gandhi, Relevance of Supply-Side Tax Policy to Developing Countries: A Summary, in SUPPLY-SIDE TAX POLICY: ITS RELEVANCE TO DEVELOPING COUNTRIES 9, 24– 25 (Ved P. Gandhi et al. eds., 1987).
  97. Jitendra R. Modi et al., Statistical Tables, in SUPPLY-SIDE TAX POLICY: ITS RELEVANCE TO DEVELOPING COUNTRIES 337, Table A10 at 366–67 (Ved P. Gandhi et al. eds., 1987) [hereinafter Modi et al., Statistical Tables].
  98. For extensive discussion of the weakness of tax administration in most developing countries, see, e.g., IMPROVING TAX ADMINISTRATION IN DEVELOPING COUNTRIES (Richard M. Bird & Milka Casanegra, eds., 1992) and ARINDAM DAS-GUPTA & DILIP MOOKHERJEE, INCENTIVES AND INSTITUTIONAL REFORM IN TAX ENFORCEMENT: AN ANALYSIS OF DEVELOPING COUNTRY EXPERIENCE (1998).
  99. Tanzi & Zee, supra note 5, at 18–19; Modi et al., Statistical Tables, supra note 97, Table A11, at 368.
  100. Michael J. Graetz & Itai Grinberg, Taxing International Portfolio Income, 56 TAX L. REV. 537 (2003); Reuven S. Avi-Yonah, Globalization, Tax Competition, and the Fiscal Crisis of the Welfare State, 113 HARV. L. REV. 1573 (2000).
  101. See generally, Reuven S. Avi-Yonah, The Structure of International Taxation: A Proposal for Simplification, 74 TEX. L. REV. 1301 (1996).
  102. HUGH J. AULT & BRIAN J. ARNOLD, COMPARATIVE INCOME TAXATION: A STRUCTURAL ANALYSIS 372–75 (2d. ed 2004).
  103. Avi-Yonah, supra note 100, at 1584.
  104. Id. at 1576.
  105. Id. at 1581–86.
  106. For a review of the challenges to taxing non-residents in the Philippines see Richard D. Pomp, The Experience of the Philippines in Taxing its $on-Resident Citizens, in INCOME TAXATION AND INTERNATIONAL PERSONAL MOBILITY 43 (J. Bhagwati & J. Wilson eds., 1989).
  107. Deficit Reduction Act of 1984, Pub. L. No. 98-369, §127(a) (codified as amended at I.R.C. § 871 (h) (1994)).
  108. For a recent review of the effectiveness of tax incentives in attracting foreign investment, see Alexander Klemm, Causes, Benefits, and Risks of Business Tax Incentives (IMF Working Paper, No. 09/21, 2009) and for an empirical analysis, see Alexander Klemm & Stefan Van Parys, Empirical Evidence on the Effects of Tax Incentives (IMF Working Paper, No. 09/136, 2009).
  109. See World Bank, Paying Taxes 2010, app. Table 1.4 (2010) for the estimated size of social security and labor taxes paid by employers in most developing countries.
  110. For discussion, see Rutkowski, supra note 92, at 281; James Alm & Hugo López- Castaño, Payroll Taxes in Colombia, in FISCAL REFORM IN COLOMBIA 195 (Richard M. Bird et al. eds., 2005); John Norregaard & Tehmina S. Kahn, Tax Policy: Recent Trends and Coming Challenges 12–15 (Int’l Monetary Fund, Working Paper No. 07/274, 2007).
  111. In many developing countries, as little as one percent of taxpayers generate over 70% of corporate tax revenues. See Katherine Baer, Olivier P. Benon & Juan A. Toro Rivera, Improving Large Taxpayers’ Compliance: A Review of Country Experience (IMF Occasional Paper 215, 2002). This marked concentration of the tax base has, in many countries, led to the creation of Large Taxpayer Units (LTUs) dedicated specifically to servicing and monitoring such taxpayers.
  112. The data are from Ernst & Young, Global Executive (2009) [hereinafter EY Global Executive] and Ernst & Young, The 2009 Worldwide Corporate Tax Guide (2009) [hereinafter EY Worldwide Corp. Tax Guide].
  113. Robin Burgess & Nicholas Stern, Taxation and Development, 31 J. OF ECON. LIT. 762, 764–75 (1993).
  114. Chu, Davoodi & Gupta, supra note 11, at 34–37.
  115. Avi-Yonah, supra note 100, at 1640–41. See also Martin Ravallion, Do Poorer Countries Have Less Capacity for Redistribution? (World Bank Policy Research Paper 5046, September 2009). Ravallion estimates that guaranteeing poverty-line income ($1.25/day) to the poorest in those countries with annual consumption per capita of less than $2,000 would require marginal tax rates on the non-poor (those above the “poverty line”) of over 100% to eliminate the poverty gap, while for high-income developing countries (those with per capita consumption over $4,000), this goal could be achieved with an average rate of only 1%. Id. at 20.
  116. Norregaard & Kahn, supra note 110, at 3–6.
  117. Peter, Buttrick & Duncan, supra note 22, at 2.
  118. In Latin America, for example, the average tax rate on corporations fell from 41%in 1985 to 29% in 2003, and the top rate on personal income from 51% to 28%. Eduardo Lora & Mauricio Cárdenas, La Reforma de las instituciones fiscales en América Latina 34 (Inter-American Dev. Bank Res. Dept., Documento de Trabajo No. 559, 2006).
  119. See generally Richard M. Bird & Charles E. McLure Jr., The Personal Income Tax in an Interdependent World, in THE PERSONAL INCOME TAX: PHOENIX FROM THE ASHES? 235 (S. Cnossen & R.M. Bird, eds., 1990).
  120. See Sørensen & Johnson, supra note 53, at 25.
  121. Anwar Shah and John Whalley argue persuasively that such “rents” are generally much more important in developing than developed countries. Anwar Shah & John Whalley, Tax Incidence Analysis of Developing Countries: An Alternative View, 5 WORLD BANK ECON. REV. 532 (1990).
  122. Many of the flat tax reforms discussed in the next section provided for an expansion of the tax base to help minimize revenue losses from rate reductions. See infra notes 165–67 and accompanying text.
  123. In a few developing countries, such as Papua New Guinea, almost all personal income tax revenues were collected from expatriates who were subject to the top rate of the personal income tax rate schedule. Richard M. Bird, Taxation in Papua New Guinea: Backwards to the Future?, 17 WORLD DEV. 1145, 1150–51 (1989). Such countries may have little to gain by lowering this rate and much revenue to lose.
  124. Countries vary greatly in the proportion of individual income tax payers to total population, with low-income countries having much lower participation rates as compared to medium-income and high-income countries. Modi et al., Statistical Tables, supra note 97, Table A5, at 360. While in many countries the choice of the zero-bracket amount is driven largely by revenue concerns, several other important design elements may also arise. For example, raising the zero-bracket amount would relieve a substantial number of existing taxpayers from the obligation to file income tax declarations. This was an important consideration in Uruguay’s move to a dual income tax. Barreix & Roca, supra note 18, at 131.
  125. For some examples, see Richard M. Bird & Sally Wallace, Is It Really So Hard to Tax the Hard-to-Tax? The Context and Role of Presumptive Taxes, in TAXING THE HARD-TOTAX: LESSONS FROM THEORY AND PRACTICE 121, 135–36 (James Alm et al. eds., 2004).
  126. This approach was adopted in Uruguay. Barreix & Roca, supra note 18.
  127. Sørensen, From the Global Income Tax, supra note 18, at 60.
  128. Several countries in Latin America impose schedular taxes on some types of capital income. These schedular capital taxes generally apply different tax rates to different types of capital income. See, e.g., Barreix & Roca, supra note 18, at 129 (tax treatment of capital income in Uruguay).
  129. The existence, level, and structure of payroll taxes will also influence incentives to avoid characterization as labor income.
  130. See supra notes 50–52 and accompanying text.
  131. Some adjustments in tax treaties would likely be required. One reason for imposing similar final withholding tax regimes on payments to foreign and domestic taxpayers is to reduce incentives for domestic taxpayers to use foreign entities to avoid withholding tax liabilities. Policymakers could also extend final withholding taxes to returns from pensions. Sørensen & Johnson, supra note 53. However, doing so would raise the difficult question of whether such returns constitute capital income or deferred labor income.
  132. George R. Zodrow, Corporate Income Taxation in Canada, 56 CAN. TAX J. 392, 461 (2009). Alternatively, Sørensen & Johnson, supra note 53, at 101, 121–26 propose a system of accrual taxation for gains in financial assets. Successful implementation of such a system, however, is likely beyond the capacity of most developing and developed countries.
  133. Barreix & Roca, supra note 18, at 122.
  134. Unlike the prior tax regime in Uruguay, the dual income tax regime imposed taxes on professional services, interest, rent, and capital gains. A substantial zero bracket excludes 60% of the population from the tax system. Id. at 131.
  135. Id. at 130.
  136. Id.
  137. Sørensen, supra note 12, at 212–13.
  138. Ruud de Mooij & Gaetan Nicodème, How Corporate Tax Competition Reduces Personal Tax Revenue, 6 CESIFO DICE REPORT, Spring 2008, at 27.
  139. As Keen and Mansour note, however, even in the case of natural resources countries may compete owing to scarcity of skills needed for resource extraction. Michael Keen & Mario Mansour, Revenue Mobilization in Sub-Saharan Africa: Challenges from Globalization 11 (Int’l Monetary Fund, Working Paper No. 09/157, 2009).
  140. Klemm and Van Parys emphasize that even if (as their study suggests) countries may sometimes succeed through fiscal incentives in attracting foreign investment, there is no persuasive evidence that the result is increased total investment and growth. Klemm & Van Parys, supra note 108, at 21.
  141. Sørensen, supra note 12, at 176.
  142. Id. at 175; Roger H. Gordon, Can Capital Income Taxes Survive in Open Economies?, 47 J. FIN. 1159 (1992).
  143. For a recent review of such proposals in the context of a developed country, see Sørensen & Johnson, supra note 53, at 79. See also Klemm, supra note 108, at 9 (considering similar arguments in a developing country context).
  144. George R. Zodrow, Corporate Income Taxation in Canada, 56 CAN. TAX J. 392, 460 (2008). On the arguments for taxing capital income, see, e.g., John Head & Richard Krever, Taxing Capital Income, 22 AUSTRL. TAX F. 83 (2007).
  145. See ROSE GARNAUT & ANTHONY CLUNIES-ROSS, TAXATION OF MINERAL RENTS (Oxford 1983) for an example of a resource tax proposal.
  146. Sørensen, supra note 12, at 214–17.
  147. See Joel Slemrod, Carl Hansen & Roger Procter, The Seesaw Principle in International Tax Policy, 65 J. PUB. ECON. 163 (1997).
  148. Kleinbard, supra note 17, at 48.
  149. Dividend exemption is not an essential element of a dual income tax. If full dividend integration is not desired, countries could subject dividends to taxation under the personal income tax regime, or provide for partial imputation by providing credit for taxes paid at the entity level.
  150. Whether taxpayers’ compliance costs would also be reduced depends on the obligations imposed on taxpayers and other parties to report and withhold, and on the specific types of income included in the dual income tax regime.
  151. Liam P. Ebrill, Income Taxes and Investment: Some Empirical Relationships for Developing Countries, in SUPPLY-SIDE TAX POLICY: ITS RELEVANCE TO DEVELOPING COUNTRIES 115, 133–34 (Ved P. Gandhi et al. eds., 1987).
  152. Klemm & Van Parys, supra note 108, at 18 find (for a wider sample of developing countries) that foreign direct investment responds positively to increased tax incentives, and in particular, tax holidays.
  153. The available data for developing countries do not separate foreign capital inflows into real and financial flows. It is therefore not clear whether the tax incentives generate additional new investment or new ownership of existing investments. Because neither increased total real investment nor economic growth appears to be associated with the increased foreign direct investment inflows, it may be that either such financial transfers dominate or higher foreign investment inflows ‘crowd out’ domestically-financed investment. Klemm & Van Parys, supra note 108, at 15.
  154. Id. at 19–20. This result is not inconsistent with the meta-study of corporate tax elasticities in developed countries as reported by Ruud A. de Mooij and Sjef Ederveen in Corporate Tax Elasticities: A Reader’s Guide to Empirical Findings, 24 OXFORD REV. ECON. POL’Y. 680, 686 (2008). This study finds by far the highest and most significant responses of international corporate capital flows reflect profit shifting in response to changes in statutory rates, but that there is also a significant effect on real capital inflows when effective marginal tax rates on capital are lowered, as by incentives. Id. at 694–95. However, as Klemm & Van Parys, supra note 108, at 20, note that such positive effects of tax incentives on real investment (and growth) appear to be conditional on the existence of precisely those institutional and financial aspects of absorptive capacity that are generally missing in developing countries. See, for example, the extensive analysis in WORLD BANK, DOING BUSINESS 2010: REFORMING THROUGH DIFFICULT TIMES (2009) [hereinafter WORLD BANK].
  155. Compare Daniel J. Mitchell, Eastern Europe’s Flat Tax Revolution, 37 TAX NOTES INT’L 989 (2005)(flat tax reforms have generated impressive results, including faster growth, more jobs, and increased competitiveness) with Clifford G. Gaddy & William G. Gale, Demythologizing the Russian Flat Tax, 37 TAX NOTES INT’L 983, 988 (2005)(difficult to establish causal link between flat tax reforms and economic growth and higher tax revenue because of other concurrent factors that influence growth and tax compliance).
  156. WORLD BANK, supra note 154, at 9.
  157. The economic and administrative importance of providing an incentive for firms to move from a presumptive to the regular income tax regime is emphasized by Bird & Wallace, supra note 125, at 143–44.
  158. Peter Barrand, Stanford Ross & Graham Harrison, Integrating a Unified Revenue Administration for Tax and Social Contribution Collections: Experiences of Central and Eastern European Countries 33 (IMF, Working Paper No. 04/237, 2004).
  159. The rate differentials will likely be smaller in developing countries because it is unlikely that developing countries could successfully adopt and enforce personal income tax rates comparable to the high tax rates found in Nordic countries.
  160. For a similar conclusion with respect to adopting a dual income tax in the United States, see Kleinbard, supra note 17, at 45.
  161. See generally Michael Keen, Yitae Kim & Ricardo Varsano, The “Flat Tax(es)”: Principles and Experience, 15 INT’L TAX PUB. FIN. 712, 715 (2008). Earlier flat tax experiments took place in such diverse places as Hong Kong, and Jamaica. See Richard Cullen & Antonietta Wong, Globalization and the Hong Kong Revenue Regime (Paper presented at symposium in honor of Alex Easson Queen’s University Law School, February 2008); James Alm et al., A Program for Reform, in THE JAMAICAN TAX REFORM 153, 157–58 (Roy Bahl ed., 1991).
  162. See Keen, Kim & Varsano, supra note 161; EY Global Executive, supra note 112; EY Worldwide Corp. Tax Guide, supra note 112.
  163. See Brian Jenn, Flat Taxes: Past, Present, and Future, 43 TAX NOTES INT’L 995 (2006); Pablo Saavedra, Flat Income Tax Reforms, in FISCAL POLICY AND ECONOMIC GROWTH: LESSONS FOR EASTERN EUROPE AND CENTRAL ASIA 253 (Cheryl Grey, Tracey Lane & Aristomene Varoudakis eds., 2007).
  164. For a survey of such proposals, see McLure & Zodrow, supra note 9.
  165. With the exception of Georgia, these reforms retain a zero-bracket amount, so they provide for some progressivity in the tax system.
  166. Keen, Kim & Varsano, supra note 161, at 715–18.
  167. The data for this table are from EY Global Executive, supra note 112 and EY Worldwide Corp. Tax Guide, supra note 112.
  168. Studies looking at the revenue and progressivity consequences of moving to a flat tax regime emphasize the difficulty of isolating the effects of flat tax reforms from other concurrent tax and fiscal reforms. Keen, Kim & Varsano, supra note 161, at 741.
  169. Richard M. Bird, Administrative Dimensions of Tax Reform, 10 ASIA-PAC. TAX BULL. 134 (2004).
  170. Keen, Kim & Varsano, supra note 161, at 732–36.
  171. Different countries may of course choose very different balances of tax levels and tax progressivity: see, for instance, the two quite different approaches to this issue in Lindert, supra note 37, and Alberto Alesina & George-Marios Angeletos, Fairness and Redistribution: U.S. versus Europe (Nat’l Bureau of Econ. Research, Working Paper No. 9502, 2003).
  172. For an historical example, see Burbidge’s discussion of the development of social security in Canada. JOHN BURBIDGE, SOCIAL SECURITY IN CANADA: AN ECONOMIC APPRAISAL (1987).
  173. On the importance of the role of tax systems in increasing state legitimacy and hence capacity to govern effectively, see Brautigram et al. eds., supra note 10. The key role improved governance plays in increasing tax capacity is empirically demonstrated in Richard M. Bird, Jorge Martinez-Vazquez & Benno Torgler, Tax Effort in Developing Countries and High Income Countries: The Impact of Corruption, Voice and Accountability, 38 ECON. ANAL. & POL’Y 55 (2008).
  174. Boadway, supra note 18, at 926.
  175. For an early proposal along these lines, see Richard M. Bird & Oliver Oldman, The Transition to a Global Income Tax: A Comparative Analysis, 31 BULL. INT’L FISC. DOC.439, 442 (1977).
  176. See supra notes 133–36 and accompanying text.
  177. See supra notes 162–67 and accompanying text.