Terms of Trade Effects: Theory and Measurement

Abstract

Foreign trade enables a nation to consume a different mix of goods and services than it produces, so to measure real gross domestic income (GDI) for an open economy, we must deflate by an index of the prices of the things that this income is used to buy, not the price index for GDP. The differences between these two indexes come from the export and import components of GDP, and are measured by the trading gains index. Fisher indexes are a natural way to estimate the conceptual economic indexes of trading gains and real GDI because they are averages of the theoretical upper and lower bounds of the economic indexes. They can be decomposed in a way that permits analyses of the factors driving changes in trading gains, such as changes in the terms of trade and in the relative price of tradables, or changes in the prices of particular commodities. Applying these methods to the United States, we find that trading gains have a median absolute effect on U.S. real GDI of 0.2 percentage points in annual data. The petroleum price shocks that occurred in late 1973 and in 1980 subtracted more than a full percentage point from the annual growth of real GDI, and in the first half of 2008 price increases in petroleum and other imported commodities subtracted 2 percentage points from the annual rate of growth of real GDI, making it negative despite the steady growth of real GDP. On the other hand, with petroleum prices excluded, U.S. terms of trade begin to improve steadily starting in 1995 and the relative price of tradables falls. These effects increase the growth rate of U.S. real GDI by 0.15 percent per year on average.