Working Paper: NBER ID: w2249
Authors: Jeffrey A. Frankel; Richard Meese
Abstract: "Unnecessary variation" is defined as variation not attributable to variation in fundamentals. In the absence of a good model of macroeconomic fundamentals, the question "are exchange rates excessively variable?" cannot be answered by comparing the variance of the actual exchange rate to the variance of a set of fundamentals. This paper notes the failure of regression equations to explain exchange rate movements even using contemporaneous macroeconomic variables. It notes as well the statistical rejections of the unbiasedness of the forward exchange rate as a predictor of the spot rate. It then argues that, given these results, there is not much to be learned from the variance-bounds tests and bubbles tests. The paper also discusses recent results on variation in the exchange risk premiums arising from variation in conditional variances, both as a source of the bias in the forward rate tests and as a source of variation in the spot rate. It finishes with a discussion of whether speculators' expectations are stabilizing or destabilizing, as measured by survey data. The paper concludes that it is possible that exchange rates have been excessively variable -- as, for example, when there are speculative bubbles -- but that if policy-makers try systematically to exploit their credibility in order to stabilize exchange rates, they may see their current credibility vanish.
Keywords: No keywords provided
JEL Codes: No JEL codes provided
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| monetary disturbances (E39) | exchange rate fluctuations (F31) |
| investor expectations (D84) | exchange rate movements (F31) |
| elastic expectations (H30) | exchange rate fluctuations (F31) |
| inelastic expectations (D84) | exchange rate fluctuations (F31) |
| speculative bubbles (D84) | exchange rate variability (F31) |