FDI Return Differentials: An Explanation Based on Offshore Profit Shifting

Abstract Beginning in the mid-1990s, U.S. multinational enterprises began to rapidly shift some profit that would otherwise be earned in the United States to their offshore affiliates located in low-tax jurisdictions. While much as been written about the implications of offshore profit shifting for the U.S. tax base, little is known about the effects of profit shifting on measurement in the domestic economy. Using confidential firm-level data from the Bureau of Economic Analysis, we develop a methodology to measure the profit shifted offshore by U.S. multinational enterprises. In 1990, profit shifting was less than one percent of U.S. private business sector value added. By 2008, profit shifting was about 2.5 percent of U.S. private business sector value added. Reattributing these profits back to the U.S. economy increases U.S. GDP. How does this affect measurement of the U.S economy? We show that accounting for the rapid increase in profit shifting adds to GDP and productivity growth rates for 1994–2008, partially mitigating the productivity slowdown found in official statistics. Most of the profit shifting lies in R&D intensive firms and industries. Our adjusted measure of GDP in R&D intensive industries is as much as 8 percent larger than that found in official statistics. Profit shifting matters for measurements beyond just headline GDP. For 2014, we find that adjusting for profit shifting roughly halves the U.S. trade deficit because adjusting for profit shifting doubles the trade surplus in services (our adjustment does not change the goods trade balance). The profit we reattribute to the United States is capital income, which adds 1.9 percentage points to the decline in the net labor share. The profit reattributed to the United States decreases the income earned by U.S. multinational enterprises on their foreign direct investments (and raises their domestic earnings). As a result, our adjustment reduces the rate of return on U.S. direct investment abroad from 7.3 percent to 3.0 percent, roughly eliminating the difference between the rate of return earned by U.S. multinationals on their foreign investments and the rate of return earned by foreign multinationals on their investments in the United States. In addition, the adjusted data show an even steeper decline in the US labor share in the last 20 years than what was previously documented in earlier work using unadjusted data. Finally, reattributing profits also has implications for the share of corporate profits earned by private business relative to C-corporations: it mitigates the decline in the share earned by the latter relative to what was found in unadjusted data.