tax competition, benefit taxes, and fiscal federalism - National Tax ...

National Tax Journal Vol 51 no. 3 (September 1998) pp. 579-86

TAX COMPETITION, BENEFIT TAXES, AND FISCAL FEDERALISM

TAX COMPETITION, BENEFIT TAXES, AND FISCAL FEDERALISM TIMOTHY J. GOODSPEED

*

Abstract - As the world becomes more globalized and resources become more internationally mobile, the issue of tax competition is moving to the forefront of debates surrounding national tax systems. Those who have studied taxation in a federal system of government will recognize that this debate is not new. This paper presents the conceptual issues that have arisen concerning tax competition in the fiscal federalism literature, and explores the reasons why the empirical measurement of the effects of tax competition, particularly corporate tax competition, is so difficult. Given the complicated nature of the problem and the theoretical ambiguity on the consequences of tax competition, it should perhaps not be surprising for the empirical literature to find that competition sometimes increases and sometimes decreases tax rates.

is moving to the forefront of debates surrounding national tax systems. Those who have studied taxation in a federal system of government will recognize that this debate is not new. Federal systems of government inherently involve issues concerning the taxation of mobile tax bases. This is fortuitous, for we have a good deal of literature to rely upon in thinking about the consequences of tax competition in the global economy. Despite this knowledge, the consequences of tax competition remain ambiguous and difficult to measure empirically. This paper presents the conceptual issues that have arisen concerning tax competition in the fiscal federalism literature, and explores the reasons why the empirical measurement of the effects of tax competition, particularly corporate tax competition, is so difficult. I begin in the next section by discussing two conceptual issues. First, I address the question of whether tax competition is good or bad. The answer depends on the type of tax used to finance public expenditures. Second, I ask whether we can say that tax competition leads to too little or too much taxation on various tax bases, and I distinguish between vertical and horizontal tax competition.

INTRODUCTION As the world becomes more globalized and resources become more internationally mobile, the issue of tax competition *

Department of Economics, Hunter College and CUNY Graduate Center, New York, NY 10021.

579

NATIONAL TAX JOURNAL VOL. LI NO. 3

The third section of the paper discusses the reasons that empirical measurement of the effect of tax competition is so difficult. I outline four reasons. First, tax systems are complex and tend to be opaque. Second, there are many factors besides taxes, particularly spending decisions and other location factors, that influence the location decisions of tax bases. Third, it is difficult to know the benefits derived by taxed factors from public spending. Fourth, there are often several governmental policies at work that have the same aim and this makes it difficult to discern the effect of tax competition in isolation. Given these complications, and the theoretical ambiguity on the consequences of tax competition, it should perhaps not be surprising for the empirical literature to find that competition sometimes increases and sometimes decreases tax rates.

suggests, free migration implies that firms and individuals will sort themselves among jurisdictions to obtain their most preferred tax-expenditure package in the same way that individuals shop in the private market. Moreover, the resulting allocation of resources will be Pareto efficient so that no reallocation can be made that will make one economic agent better off without making another worse off. However, the Tiebout efficiency result obtains only if taxes are benefit taxes. Benefit taxes reflect social marginal cost and therefore lead consumers and firms to choose jurisdictions efficiently. If taxes do not reflect benefits, however, Oates (1972) suggests that externalities are created so that tax prices diverge from social marginal cost. This creates incentives for inefficient location decisions. But, as Hamilton (1976) has argued, even nonbenefit taxes may be converted to benefit taxes. For instance, Hamilton (1976) shows that, given certain zoning restrictions (in particular, forcing houses to meet some minimum value), property taxes are converted into benefit taxes. Moreover, appropriate tax base restrictions will convert any local tax to a benefit tax. For instance, restricting residence on the basis of a minimum level of income will convert an income tax into a head tax. Further, even if nonbenefit taxes are used, the inefficiency created may be small, as suggested by the general equilibrium simulation results of Goodspeed (1989).

THE CONCEPTUAL ISSUES Is Tax Competition Good or Bad? Competition in the private sector is normally associated with an efficient allocation of resources. Tax competition results when the tax system of one government entity affects the tax system of a second governmental entity, usually through an effect on the second entity’s tax revenues. Whether tax competition in the public sector is efficient depends on the type of tax used. One key result in the study of tax competition is based on the analysis of Tiebout (1956); one modern discussion is contained in Oates and Schwab (1988). This result indicates that horizontal tax competition can result in an efficient allocation of resources if the taxes used are benefit taxes. If taxes are commensurate with benefits, Tiebout

It is also useful to note that, in the efficient Tiebout equilibrium, tax rates are not equal across jurisdictions. Differing tax rates simply reflect differing demands for public services. Consequently, one cannot gauge the level and effects of tax competition simply by analyzing differences in tax rates.

580


Does Tax Competition Lead to Too Little or Too Much Taxation?

level of public services than is optimal. These are the basic results of Zodrow and Mieszkowski (1986) and Wildasin (1989), for instance.1 This suggests that tax competition results in taxes that are lower than otherwise. It also suggests a less equitable tax system if capital taxes are viewed as a redistributive element of the system.

In discussing whether tax competition leads to too much or too little taxation, it is useful to distinguish between horizontal and vertical competition. Horizontal tax competition refers to two (or more) governments at the same level. An example of horizontal tax competition would be two local (or two national) governments that tax a mobile base; the tax rate set by any one of the jurisdictions then affects the tax revenues of the other jurisdiction. Vertical tax competition refers to two (or more) governments at different levels. An example of vertical tax competition would result if the central and state governments tax the same base; again, the tax rate set by one of the jurisdictions affects the tax revenues of the other jurisdiction.

The international taxation literature goes even further in suggesting that capital tax rates may be competed to zero. This is the result of Bond and Samuelson (1989) and Razin and Sadka (1991), for instance. Capital tax rates get competed to zero in these models because they ignore public goods (or equivalently assume that capital receives no benefits from public goods). Gordon (1992) finds that, if one country acts as a Stackelberg leader, positive tax rates on capital emerge as an equilibrium solution, even ignoring the benefits of public goods.

Much of the literature has centered on horizontal tax competition. If benefit taxes are not used, tax competition may lead to nonoptimal levels of taxation; this is referred to as “destructive” tax competition by McGuire (1991). Since redistributive taxes are by definition nonbenefit taxes, one focus of research has been on whether redistributive taxes will be eliminated by tax competition. But the literature also suggests that horizontal tax competition can result in either too little or too much taxation; these I will refer to as “destructive downward” and “destructive upward” tax competition.

The fact that tax rates are set within a political system means that whether tax rates are “too high” or “too low” may not even be theoretically easy to determine, however. Generally, as shown in Goodspeed (1995), there are two efficiency conditions that must be satisfied simultaneously—factors need to be efficiently located across jurisdictions and public goods need to be provided efficiently. Goodspeed (forthcoming) shows that, if the political system is such that inefficient levels of public goods are chosen, the efficiency condition for the optimal allocation of factors across jurisdictions changes. In essence, benefit taxation will no longer be (second-best) efficient, and taxes that differ from benefit taxes may actually improve matters.

Destructive downward tax competition has been the focus of work on small open economies that rely exclusively on capital taxes. In this framework, it is generally found that capital taxation leads to (1) a distortion in the allocation of capital, (2) capital taxes that are lower than otherwise, and (3) a lower

The small empirical literature on capital tax competition does not come to any

581


strong conclusions. Some empirical studies, such as Weiner (1996) and Grubert (1997), that address corporate tax competition do not address the issue of competition per se, but rather the related but different question of whether capital responds to tax differences. While this is valuable information, it does not tell us much about tax competition, since we would expect capital responses regardless of whether capital taxes represent benefit taxes.

expenditures, which implies a reduction in the tax rate. A second response is to attempt to maintain its revenues; whether this will require an increase or decrease in the tax rate is ambiguous. (However, Goodspeed (1998) shows that this requires an increase in the tax rate if the elasticity between the tax rate and the tax base is constant and if an increase in the tax rate results in an increase in tax revenues.) A third response results from the fact that an increase in the tax rate by one level of government affects the deadweight loss of the tax for the other level of government. Essentially, the tax becomes more expensive relative to the other taxes, and the response is to lower the tax rate.

Destructive upward tax competition can result when a government possesses market power. The typical scenario here is the case of “tax exporting,” in which a government takes advantage of some unique characteristic, such as the fact that a place is a tourist or financial center. The government may be able to take advantage of such a characteristic and place relatively high taxes on those that rely on or consume the unique characteristic, but may not consume much in the way of public services that the tax dollars support. In this case, taxes on these factors may be too high; those that consume public services are able to free ride on payments of those who do not consume public services.

The theoretical direction of the response is ambiguous in sign, so Besley and Rosen (1997) and Goodspeed (1998) turn to empirical tests. But their empirical tests come to contrary conclusions! Besley and Rosen (1997) find that higher U.S. federal gasoline and cigarette taxes lead to higher state taxes on these products. Goodspeed (1998) finds that increases in higher level government income taxes lead to decreases in lower level government income taxes.

New research on vertical tax competition also comes to somewhat contrary results on whether competition between higher and lower levels of government will lead to higher or lower tax levels. Consider two levels of government that tax the same base to raise revenue. Will an increase in the tax rate on this base by one level of government increase or decrease the tax rate imposed by the other level of government? Research by Besley and Rosen (1997) and Goodspeed (1998) indicates three ways that a government may respond when another level of government infringes on the common tax base by raising its tax rate. One response is to lower

WHY IS EMPIRICAL MEASUREMENT OF THE EFFECTS OF TAX COMPETITION SO DIFFICULT? Thus far, we have seen that it is difficult to find a consensus in discussions of tax competition. Both empirical and theoretical explanations find that in some cases tax competition is good and in others it is bad; in some cases, tax competition leads to tax rates that are higher than otherwise; in other cases, it leads to tax rates that are lower than otherwise. Why is the effect of tax competition so difficult to measure? I suggest four 582


reasons. First, tax systems are complex and tend to be opaque. Second, there are many factors besides taxes, particularly spending decisions and other location factors, that influence the location decision of tax bases. Third, without knowing the benefits derived by factors from public spending, it is difficult to say how closely a given tax system mimics benefit taxation, and without this information, it is difficult to assess whether tax competition is good or bad. Fourth, there are often several governmental policies at work that have the same aim; this makes it difficult to discern the effect of tax competition in isolation.

something more in taxes to have access to a good transportation system rather than relying on dirt roads and outdated ports to ship its product. To illustrate the consequences of not considering benefits as well as taxes, consider two concrete examples of attempts to tax capital. In the late 1980s, Switzerland imposed a 30 percent tax on interest and West Germany attempted to impose a withholding tax on interest income. Since Switzerland provides a lot of benefits in its bank secrecy laws, one would expect there to be some willingness to pay this tax in exchange for these benefits. On the other hand, the attempt by the former West Germany to impose a withholding tax on interest income failed. This was an attempt to tax a highly mobile resource that got very little direct benefit in return for its tax payments. Predictably, West Germany soon found that it had to remove the tax or lose this tax base. If we look at the case of Switzerland without considering benefits, we would conclude that taxes on capital have little effect on the location of capital. In contrast, if we look at the case of West Germany in isolation, we would conclude that taxes on capital have a large impact on the location of capital.

One reason that the effect of tax competition has been difficult to measure is that corporate tax systems tend to be opaque. One must consider the definition of the tax base, including such things as depreciation allowances and investment tax credits. The use of effective tax rates goes some way toward solving this problem, although effective tax rates ignore important problems such as transfer pricing. More fundamentally, the Tiebout literature suggests that differences in effective tax rates are not necessarily inefficient. Such differences may simply reflect differences in demand for publicly provided goods and services, or other characteristics that are willingly paid for by companies (or individuals). Presumably, for a given amount of taxation, more spending on infrastructure and schools makes a location more attractive. Surely a company is willing to pay something in taxes to have access to a well-trained local labor force. The effect of increased public spending on infrastructure on private sector output has been debated in the literature. (See, for instance, Holtz-Eakin, 1994.) But presumably a company is willing to pay

The theoretical argument for including public spending and other benefits specific to location as well as taxes is clear. Moreover, Wilson (1993), in a survey of multinational companies, finds public spending to be an important factor in location decisions. Yet studies of the effect of taxes on investment location often exclude public spending and other locational factors as an explanatory variable. It should be noted that a well-specified empirical model is not an easy task, as it 583


involves general equilibrium changes in factor prices. An empirical method used in the fiscal federalism literature, first considered in Oates (1969), is to measure the extent to which taxes and public service levels are capitalized into housing values in local jurisdictions. There is no reason in general to restrict the analysis to capitalization in the housing market, however. The methodology can be generalized to reflect international factor movements and changes in wages, for instance.

neously. Consider, for instance, policies aimed at encouraging investment in Ireland. Ireland assesses companies a reduced 10 percent tax rate (in lieu of the standard 38 percent), a tax holiday, for manufacturing and traded services as well as financial services based in Ireland. The European Union complains that this is unfair tax competition. If a country that practices worldwide taxation were to tax this residual, any incentive created by the tax holiday would be removed. Yet, many Irish tax treaties with members of the European Union that use worldwide taxation contain tax sparing provisions, which exempt the difference between the standard and tax holiday rates from taxation by the treaty partner. Other countries that use the exemption method also aid the tax holiday since they exempt foreign source income from tax.

A third reason that the effect of tax competition is difficult to measure is our lack of knowledge about the benefits that capital receives from public spending.2 The West German predicament showcases some of the problems in attempting to assess tax competition. It is difficult to ascertain whether the competition that led to the downfall of the withholding tax is good or bad. On the one hand, it suggests that governments will not tax mobile resources that do not benefit from services provided by the government; this is efficient, though it may impinge on the redistributive capability of taxation in the face of competition. On the other hand, perhaps public services are now underprovided within Germany, precisely because capital is mobile and benefit taxes are unavailable; the threat of exit has caused an insufficient level of tax to be levied on capital. In either case, since most of escaping capital found its way into foreign bank accounts, the owners of which were unknown to the German government, it seems likely that sharing of tax information between countries could go a long way toward stemming some types of capital flows.

Of course, tax sparing itself raises several questions. For instance, if one wants to stimulate investment, tax sparing is probably not the best method. In order to benefit from tax sparing, a company must have profits. Tax sparing therefore does not subsidize start-up ventures that initially incur losses. Second, tax sparing opens another avenue for possible transfer pricing abuse; moreover, some suggest that the Irish government turns a blind eye to some practices that inflate Irish source profits. In any case, a logical starting point for European Union countries that practice worldwide taxation would be a renegotiation of tax treaties to abolish tax sparing provisions.

A fourth reason that the effect of tax competition is difficult to measure is that many governmental policies having the same aim are at work simulta-

A third policy is that the European Union hands out grants to countries with lower than average incomes. Much of this aid goes to Ireland, and the aim, 584


encouraging investment in Ireland, is the same as both the tax holiday and tax sparing. One might argue that one of these policies is better than another at stimulating investment. For instance, direct aid may be preferable in that it avoids the transfer-pricing incentives of tax holidays and tax sparing. The drawbacks of this type of policy to stimulate investment are that a bureaucracy needs to be set up to administer the grants and bureaucrats rather than the private market determine which projects get financed. In any case, since all the policies exist, empirical attempts to get a handle on the effect of tax competition need to take into account the fact that all these policies are influencing investment location simultaneously.

outline four reasons. First, tax systems are complex and tend to be opaque. Second, there are many factors besides taxes, particularly spending decisions and other location factors, that influence the location decision of tax bases. Third, it is difficult to know the benefits derived by taxed factors from public spending. Fourth, there are often several governmental policies at work that have the same aim and this makes it difficult to discern the effect of tax competition in isolation. Given these complications, and the theoretical ambiguity on the consequences of tax competition, it should perhaps not be surprising that empirical studies, such as those of Besley and Rosen (1997) and Goodspeed (1998), show differing results for different taxes. It may be that tax competition, both theoretically and empirically, has different consequences in different situations.

Conclusions

ENDNOTES 1

Increasing globalization and greater ease of movement of tax bases has moved the issue of tax competition to the forefront of debates surrounding the interrelationship of national tax systems. Many of the issues involved in the taxation of mobile resources have been addressed in the study of taxation in a federal system of government. This paper begins by discussing the conceptual issues regarding tax competion that have arisen in the fiscal federalism literature. Two major themes are addressed: whether tax competition is good or bad and whether tax competition leads to lower or higher tax rates in equilibrium. The theoretical results can go either way, and empirical results, particularly for vertical tax competition, also find contrasting results.

2

However, as Wilson (1987) notes, it is also possible for public goods to be overprovided. Oakland and Testa (1996) attempt to infer the level of business taxes that should be imposed based on the benefits that companies derive from public spending. Their work is testimony to the difficulties and ambiguities involved. For instance, they do not assign any benefits to companies from educational spending; yet, a well-trained labor force surely yields some benefit to corporations.

REFERENCES Besley, Timothy J., and Harvey S. Rosen. “Vertical Externalities in Tax Setting: Evidence from Gasoline and Cigarette.” Princeton Center for Economic Policy Studies Working Paper No. 46. Princeton: CEPR, 1997. Bond, E. W., and L. Samuelson. “Strategic Behaviour and the Rules for International Taxation of Capital.” Economic Journal 99 No. 398 (December, 1989): 1099–111. Goodspeed, Timothy J. “A Re-Examination of the Use of Ability to Pay Taxes by Local Governments.” Journal of Public Economics 38 No. 3 (April, 1989): 319–42.

The paper then discusses the reasons that empirical measurement of the effect of tax competition is so difficult. I

Goodspeed, Timothy J. “Local Income Taxation: An Externality, Pigouvian Solution, and 585


Public Policies.” Regional Science and Urban Economics 25 No. 3 (June, 1995): 279–96.

Values: An Empirical Study of Tax Capitalization and the Tiebout Hypothesis.” Journal of Political Economy 77 No. 6 (November–December, 1969): 957–71.

Goodspeed, Timothy J. “Tax Structure in a Federation.” Hunter College. Unpublished paper, 1998.

Oates, Wallace E. Fiscal Federalism. Princeton: Harcourt Brace Jovanovich, Inc., 1972.

Goodspeed, Timothy J. “On the Importance of Public Choice in Migration Models.” Economics Letters 5 (forthcoming, 1998).

Oates, Wallace E., and Robert M. Schwab. “Economic Competition Among Jurisdictions: Efficiency Enhancing or Distortion Inducing?” Journal of Public Economics 35 (1988): 333– 54.

Gordon, Roger. “Can Capital Income Taxes Survive in Open Economies?” Journal of Finance 47 (1992): 1159–80.

Razin, Assaf, and Efraim Sadka. “International Tax Competition and the Gains from Tax Harmonization.” Economics Letters 37 (1991): 69–76.

Grubert, Harry. “Tax Planning by Companies and Tax Competition by Governments: Is There Evidence of Changes in Behavior?” Paper presented at NBER Conference on International Taxation, Cambridge, MA, November 14–15, 1997.

Tiebout, Charles. “A Pure Theory of Local Government Expenditures.” Journal of Political Economy 64 No. 5 (October, 1956): 416–24.

Hamilton, Bruce. “Capitalization of Intrajurisdictional Differences in Local Tax Prices.” American Economic Review 66 No. 5 (December, 1976): 743–53.

Weiner, Joann. “Estimates of How the Unitary Tax Affects Business Investment.” U.S. Department of the Treasury. Unpublished paper, 1996.

Holtz-Eakin, Douglas. “Public-Sector Capital and the Productivity Puzzle.” Review of Economics and Statistics 76 No. 1 (February, 1994): 12–21.

Wildasin, David. “Interjurisdictional Capital Mobility: Fiscal Externality and a Corrective Subsidy.” Journal of Urban Economics 25 No. 2 (March, 1989): 193–212.

McGuire, Therese. “Federal Aid to States and Localities and the Appropriate Competitive Framework.” In Competition Among States and Local Governments, edited by Daphne Kenyon and John Kincaid, 153–64. Washington, D.C.: The Urban Institute Press, 1991.

Wilson, G. Peter. “The Role of Taxes in Location and Sourcing Decisions.” In Studies in International Taxation, edited by Alberto Giovannini, R. Glenn Hubbard, and Joel Slemrod. Chicago: University of Chicago Press, 1993. Wilson, John D. “Trade, Capital Mobility, and Tax Competition.” Journal of Political Economy 95 No. 4 (August, 1987): 835–56.

Oakland, William, and William Testa. “StateLocal Business Taxation and the Benefits Principle.” Federal Reserve Bank of Chicago Economic Perspectives 20 No. 1 (January– February, 1996): 2–19.

Zodrow, George, and Peter Miezskowski. “Pigou, Tiebout, Property Taxation, and the Underprovision of Local Public Goods.” Journal of Urban Economics 19 No. 3 (May, 1986): 356– 70.

Oates, Wallace E. “The Effects of Property Taxes and Local Public Spending on Property

586