Purchasing Power Parity and Tests of
Spot Exchange-Market Efficiency
Richard J. Sweeney
Recent work on purchasing power parity (PPP) and on the efficiency of spot exchange markets and its implications for exchange-market intervention are examined in this paper. As will be shown, the evidence from the PPP literature provides virtually no support for intervention, despite some claims to the contrary. Work on spot exchange-market efficiency does provide at least tentative support for intervention, though further work is necessary, particularly on implementation.
The PPP studies considered here principally involve various kinds of regressions of exchange rates (or their changes) on relative national price levels (or their changes). The cross section and cross sectiontime series analyses examined can shed no light on the possibility of beneficial intervention. Most of these studies view exchange-rate changes as lagging behind relative national price-level changes, but they cannot distinguish among alternative versions of PPP. It might be plausible, for example, that a country with an inflation trend 5 percent higher than its partner's will experience a 5 percent trend rate of depreciation. If there is a burst of inflation above the given trend, however, will this be (at least partly) reflected in future exchange-rate movements? If the exchange rate will ultimately reflect these price-level developments, perhaps intervention can hasten the movement of the exchange rates. Thus, the important question for intervention is not the relationship in trends in inflation and depreciation but the deviations from these trends. Unfortunately, most empirical work does not allow for discrimination between the two cases. One paper that ex-
Thanks are due to Gottfried Haberler, Charles Pigott, and Thomas D. Willett. Work on this paper was supported, in part, by a grant from the General Electric Foundation.