Purchasing Power Parity and the Dollar
In terms of purchasing power parity, the dollar seems a tad undervalued these days, but that does not mean it will soon appreciate. Exchange rates can deviate from their purchasing-power-parity levels for long periods. What's more, the necessary adjustment can come through prices, not exchange rates.
People value money for what it buys, and, given the opportunity, they will use the national currency that offers them the greatest purchasing power. If, for example, goods are cheaper in Mexico than in the United States, Americans will trade U.S. dollars for Mexican pesos and buy Mexican goods. Such cross-border arbitrage should affect both exchange rates and prices so as to promote parity among the purchasing powers of the world's currencies. This idea - purchasing power parity - is fundamental to many economic models of exchange-rate behavior and to some descriptions of the dollar's equilibrium value. Observers who complain that the dollar is overvalued or undervalued often do so with reference to the dollar's purchasing-power-parity value.
One way to get a quick fix on the dollar's purchasing-power-parity value is to look at a real exchange rate. Real exchange rates mathematically combine nominal exchange rates - the kind you can find in a newspaper - and price indexes - like consumer price indexes. A rising dollar real exchange rate indicates that American goods, when expressed in a common currency, are becoming more expensive than foreign goods, or that the United States is losing its competitive edge.
Using a real exchange rate to judge whether the dollar is overvalued or undervalued, however, requires some reference point at which purchasing power parity holds. Such a point should also be consistent with a global balance-of-payments configuration that is sustainable. Good luck finding that! If, however, we assume that purchasing power parity holds in the long term - a belief that many economists hold - and if we assume that our sample is lengthy enough to reasonably represent the long term, then we might define purchasing power parity in terms of the mean or median of our exchange-rate data. By this method, the dollar now seems undervalued by roughly 11 percent against a trade-weighted average of developed and developing countries' currencies, and by approximately 7 percent against a trade-weighted average of major countries' currencies.
As our data demonstrate, real exchange rates can show large and persistent deviations from levels consistent with purchasing power parity. Because of the way economists construct real exchange rates, if the prices of some goods change relative to others, the real exchange rate will deviate from its purchasing-power-parity value even if arbitrage is complete for each and every individual good. Global economic shocks can induce relative-price changes, as can productivity differentials between traded- and nontraded-goods sectors in specific countries. Consequently, real exchange rates can deviate from their purchasing-power-parity levels as long as relative-price trends persist.
While the dollar currently seems undervalued relative to its purchasing-power-parity level, dollar exchange rates need not quickly - or ever - appreciate. Instead, the dollar could return to purchasing power parity if prices in the United States rise faster than prices abroad. Inflation expectations in the United States are well behaved and do not suggest fears of rising future inflation. Let's hope the exchange market does not see something that the rest of us are missing.