Working Paper: CEPR ID: DP15872
Authors: Charles Engel; Ekaterina Kazakova; Mengqi Wang; Nan Xiang
Abstract: We re-examine the time-series evidence for failures of uncovered interest rate parity on short-term deposits for the U.S. dollar versus major currencies of developed countries at short-, medium- and long-horizons. The evidence that interest rate differentials predict foreign exchange risk premiums is fragile. The relationship between interest rates and excess returns is not stable over time and disappears altogether when nominal interest rates are near the zero-lower bound. However, we do find evidence that year-on-year inflation rate differentials consistently predict excess returns – when the U.S. dollar y.o.y. inflation rate has been relatively high, subsequent returns on U.S. deposits tend to be high. We interpret this evidence as being consistent with hypotheses that posit that markets do not fully react initially to predictable changes in future monetary policy. Interestingly, the predictive power of relative y.o.y. inflation only begins in the mid-1980s when central banks began to target inflation more consistently and continues in the post-ZLB period when interest rates lose their primacy as a policy instrument. We address the problems of parameter instability and small-sample bias that plague the conventional Fama (1984) test, while acknowledging these concerns might remain even in our new findings.
Keywords: interest parity; Fama regression; foreign exchange risk premium
JEL Codes: F30; F31; G15
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| interest rate differentials (E43) | foreign exchange risk premiums (F31) |
| nominal interest rates near the zero lower bound (E43) | foreign exchange risk premiums (F31) |
| year-on-year inflation rate differentials (E31) | excess returns (D46) |
| US dollar's year-on-year inflation rate high (E31) | subsequent returns on US deposits high (N22) |
| relative year-on-year inflation targeting by central banks began in the mid-1980s (E31) | predictive power of inflation (E31) |
| shift in economic regime in the 2000s (P19) | slope coefficients in Fama regression exhibit parameter instability (C22) |
| delayed market reactions to monetary policy changes (E52) | UIP puzzle robustness (L15) |