TOWARDS A RECONCILIATION OF THE EMPIRICAL-EVIDENCE ON THE MONETARY APPROACH TO EXCHANGE-RATE DETERMINATION

The monetary approach to exchange rate determination (MAER) has served as a theoretical workhorse in open economy macroeconomics, yet empirical evidence concerning its validity is mixed: tests based on structural forms of the model are typically negative, while cross-equation restrictions tests based on nonstructural representations are typically favorable. We seek a reconciliation of these results by investigating the small-sample performance of cross-equation restrictions tests. Our investigation indicates that the tests have surprisingly low power in detecting nontrivial departures from the model, thus we conclude that something is amiss with standard versions of the MAER.