NATIONAL BUREAU OF ECONOMIC RESEARCH
NATIONAL BUREAU OF ECONOMIC RESEARCH

Tax Effects on Foreign Portfolio Investment


"U.S. equity Foreign Portfolio Investment in the average treaty country rose by over 90 percent relative to U.S. equity holdings in the average non-treaty country, in response to the large relative decrease in the dividend tax rate for corporations in treaty countries."

Because so many other factors come into play, determining how taxes -- and particularly tax reforms -- influence portfolio choice traditionally has proven difficult to identify cleanly. But in Taxes And Portfolio Choice: Evidence from JGTRRA's Treatment of International Dividends (NBER Working Paper No. 13281), co-authors Mihir Desai and Dhammika Dharmapala overcome many of these difficulties by analyzing a tax reform that differentially changed the tax treatment of investments in a manner that was unlikely to produce confounding side issues such as changes in risk assessment or supply side responses. In their study, Desai and Dharmapala not only assess the impact of a particular tax reform but also see lessons for future tax policies.

Ideally, say Desai and Dharmapala, a tax reform with clear consequences for investor aftertax returns and with no effects on supply decisions would most conclusively isolate tax effects. For just this reason they focus on the Jobs and Growth Tax Relief Reconciliation Act (JGTRRA) of 2003 which allowed for dividend tax relief to U.S. investors who owned stock in countries with suitable tax treaties with the United States. The "treaty countries" constitute a treatment group, with equities held in those countries enjoying a reduced U.S. personal tax rate relative to equities held in the control group of non-treaty countries. Furthermore, because the tax reform affected only American investors, it was unlikely to cause a supply response from foreign firms that would affect Americans' portfolio choices.

Desai and Dharmapala based their study on data of outbound U.S. foreign portfolio investment (FPI) from the Treasury International Capital (TIC) reporting system, which cover some 213 countries. Prior to the JGTRRA, dividends were taxed as ordinary income, at a rate of 38.6 percent for taxpayers in the top tax bracket. Under JGTRRA, dividends were taxed at the same rate as capital gains, a reduced maximum rate of 15 percent. This lower rate applied to dividends paid by domestic corporations and by "qualified" foreign corporations. A foreign corporation qualified if it satisfied at least one of several tests. One of those tests was the "Treaty Test," which encompassed corporations resident in countries with which the United States had a tax treaty containing certain information-exchange requirements. Fifty-two countries met this requirement.

Even though JGTRRA's favorable tax treatment of dividends excluded such relatively significant investment destinations as Argentina, Brazil, Malaysia, Singapore, and Taiwan, the reform applied to an extensive subset of foreign corporations. The 52 countries hosted 82 percent of U.S. outbound equity FPI holdings in 2001.

Analyzing the portfolio choices in these countries against a background of data compiled for all countries in the TIC reporting system over a number of years before and after JGTRRA went into effect produces stark results. Desai and Dharmapala find that despite factors that would tend to bias the estimate downward, the positive effect of JGTRRA was quite large. In their baseline specifications, the estimated coefficient of 0.649 implies that U.S. equity FPI holdings in the average treaty country rose by over 90 percent relative to U.S. equity holdings in the average non-treaty country, in response to the large relative decrease in the dividend tax rate for corporations in treaty countries. (The rate for treaty countries fell from 38.6 percent to 15 percent while the rate for non-treaty countries fell to 35 percent.)

The data also show in particular that because of JGTRRA, American investors' holdings of lightly taxed foreign equities increased significantly in a manner consistent with an implied elasticity of asset holdings with respect to the tax rate of approximately -1.6. This elasticity is larger than most estimates of the responsiveness of portfolio shares to tax rates but is consistent with estimates of the sensitivity of FDI to taxes. The researchers stress that their results cannot be explained by a number of other potential alternative hypotheses, such as differential changes to the preferences of American investors, differential changes in investment opportunities, differential time trends in investment in treaty and non-treaty countries, or changes in patterns of tax evasion.

Desai and Dharmapala weigh the possible impacts of these first three factors but conclude that they are negligible, adding that the strong information-exchange provisions in treaty countries are hardly conducive to tax evasion. To the degree that evasion occurs, they say, it is more reasonable that it would occur through investment in tax haven countries with weak or no information-exchange provisions and no withholding taxes, and that the effect of reduced evasion after JGTRRA would appear in the form of transfer of funds from havens to treaty countries.

Desai and Dharmapala say their results suggest that taxes can play a large role in shaping international portfolio choices. The results, moreover, have implications for tax policies aimed at corporate tax integration, which has been widely supported by economists as a means of reducing distortions to firm payout and financing decisions. If dividend relief is not fully extended to foreign dividends, the researchers surmise, corporate tax integration may cause significant distortions in international portfolios resulting in welfare losses. Such effects, say Desai and Dharmapala, could be considerable.

-- Matt Nesvisky


The Digest is not copyrighted and may be reproduced freely with appropriate attribution of source.
 
Publications
Activities
Meetings
Data
People
About

Support
National Bureau of Economic Research, 1050 Massachusetts Ave., Cambridge, MA 02138; 617-868-3900; email: info@nber.org

Contact Us