Center for Research on Economic and Social Theory CREST Working

This paper shows, using a standard CAPM model of security prices in a world market, that even small countries can affect the price of domestically issued risky securities, while large countries can affect the prices of all securities. As a result, countries have the incentive to set tax rates such that in equilibrium investors tend to specialize in domestic securities, and net capital flows between countries are restricted. Each country does this to increase the utility of domestic residents, taking as given the tax policies of other governments, but the net outcome is a reduction in world efficiency and likely a reduction in the utility of all investors. Address. Roger H. Gordon, Department of Economics, University of Michigan, Ann Arbor, MI 48109, (313)-764-6769 Taxation of Asset Income in the Presence of a World Securities Market Roger H. Gordon Hal R. Varian Whenever a country is large enough to be able to affect the international price of a commodity that it trades in, then it will be tempted to set its policy so as to take advantage of this market power, at least so long as it can ignore any threat of retaliation by other countries. This observation forms the basis for a variety of results in the trade literature. For example, when a country can affect the price of its exported goods, then it will find tariffs or direct restrictions on exports attractive. 1 Similarly, if a country is a net demander (supplier) of capital, and faces a nonhorizontal supply (demand) curve, then it may attempt to restrict its net demand (supply). 2 The objective of this paper is to explore characteristics of government tax policy and equilibrium resource allocation when countries are not price-takers in the international market for financial securities. Due to risk aversion, the foreign demand for domestic securities should be downward sloping foreign investors need more attractive terms to induce them to concentrate their portfolios further in any one security.3 Similarly, the supply curve of foreign securities should be upward sloping. The above observations suggest that each country would face the incentive to reduce both its net supply of domestic securities to foreigners and its net demand for foreign securities. When each country sets its policy accordingly, the net result will be restricted international trade in financial securities. Since governments have much more market power than any one firm, it is not surprising that at least large countries should have an incentive to restrict international trade in financial securities. However, we show that each country continues to have market power over the price of the equity of domestic firms even as the number of countries becomes large, so that even small countries have the incentive to restrict foreign ownership of domestic equity. In contrast, only large countries have an incentive to restrict domestic ownership of foreign equity, or to restrict net capital flows. This intervention can take many forms. Direct controls on the outflow of capital is obviously one device. To restrict foreign ownership of domestic equity, a dividend withholding tax on dividends sent to foreigners, or a dividend credit available only to domestic residents, can be used. 4 In addition a corporate tax can be used to restrict the total supply of equity in the domestic firms. One way to restrict inflows of capital is to impose extra fees on multinational entrants to a country. Each of these policies is commonly observed, and each seems to us to be difficult to explain on other grounds. Our results also provide one possible explanation for two empirical puzzles. The first is why individual portfolios are so highly concentrated in domestic securities. From a direct application of standard results in finance, one would expect investors to hold a fraction of the world portfolio We would like to thank participants in a seminar at N.B.E.R., and especially Richard Clarida, for comments on a previous draft. The views expressed in this paper are those of the authors and not necessarily those of N.B.E.R. 1 The most obvious example of such export restrictions is OPEC. 2 See, for example, Jones(1967), Gehrels(1971), Feldstein-Hartman(1979), or Hartman(1985ab), for further discussion. When a country has market power in both the capital market and the commodity market, as in Jones and Gehrels, interaction effects add complications. 3 Helpman-Razin(1978,pp. 141-2) noted in this setting the possible gains from government intervention. 4See Booth(1987) for estimates of the degree to which the dividend tax credit concentrates ownership of Canadian securities among Canadian investors.