The early empirical literature pertaining to monetary and portfolio balance models has been surveyed extensively (Frankel and Rose, 1995; Sarno and Taylor, 2002; Taylor, 1995).
In the wake of the collapse of the Bretton Woods system two major strands of models dominated the literature: the monetary and portfolio balance approaches. Both approaches focused on stocks of outside assets—money in the former and both money and bonds in the latter. While this perspective is completely natural from today's vantage point, it is easy to forget how much this approach differed from the older flow perspective of the Mundell-Fleming model. This approach stressed current account and capital account flows as the determinants of exchange rates, and was largely superseded in the 1970s. Consequently, I do not discuss this approach in detail.
Although the monetary and portfolio approaches share a focus upon stocks, they differ in their views of the substitutability of capital. In practice, the difference amounts to whether uncovered interest rate parity holds, or whether the forward rate differs from the expected future spot rate by an exchange risk premium. Of the two approaches, by far the most common approach in the nominal exchange-rate literature has been the monetary model.
The monetary approach views the exchange rate as the relative price of currencies, when that relative price depends on the relative demands and supplies of ...