Tyler Cowen meditates on the topic of exchange rates.

I am headed back from Poland to France, for one more day in Europe. I cannot help but wince at the especially high prices in France, compared to the United States. You may recall my mention of the five dollar chocolate bar; at home I could get the same for less than three dollars.

So a microeconomist might conclude that the U.S. dollar should rise over time. Arbitrage will cause people to buy more in the United States, helping out the dollar.

…Let’s bring the macroeconomist into the picture. He tells us that the United States has unsustainably high trade and budget deficits. The only way to clear these deficits, he says, is for the dollar to fall at least thirty percent.

The first point I would make is that it is very dangerous to extrapolate the relative cost of goods and services in two countries on the basis of one product. I am sure that there are goods that Tyler could find that are much cheaper in France than they are here.

He is calculating one real exchange rate, which is the price of a candy bar in euros in France, converted to dollars, relative to the price of a candy bar in the U.S. in dollars. If goods were costlessly tradable and prices were flexible, then the two prices would have to be equal and the real exchange rate would always be one–one candy bar for one candy bar. However, in the real world there are trading costs, and there are many goods and services, which makes it very difficult to measure “the” real exchange rate.

In a previous life, I helped to popularize some international cost of living comparisons as part of the Salary Calculator. Those comparisons are very tenuous and controversial. There is local lifestyle bias, so that an Italian might say that Japan is expensive based on the price of pizza, while a Japanese might say that Italy is expensive because of the cost of sushi. There are all sorts of tax issues, regulatory issues, and retailing customs that affect relative prices. What Cowen calls the “microeconomic” view of exchange rates is sort of a straw man, in that it assumes away all of these factors and instead predicts that prices will move in the direction that they would if international goods markets were perfectly integrated.

The model of exchange rate determination that was popular when I was in graduate school was based on the view that asset markets adjust quickly, while goods prices adjust slowly. So when investors decide they want more dollar-denominated assets, the value of the dollar goes up. In the goods market, the price of Toyotas stays steady in yen, and since a dollar can buy more yen, our dollars can buy more Toyotas.

The mirror image of strong demand for dollar-denominated assets is a U.S. trade deficit. What Cowen calls the “macroeconomic” view is simply a forecast that at some point foreign investors will want to stop increasing their dollar-denominated holdings, and when they do so the dollar will fall. I find this a compelling forecast, but someone else could take the view that the appetite for dollar-denominated holdings might increase going forward.

For Discussion. Which relative prices in the economy adjust particularly quickly, and which prices adjust particularly slowly?