Imperfect Exchange Rate Pass-through: Empirical Evidence and Monetary Policy Implications

We construct a small open economy (SOE) DSGE model interacting with the rest of the world (ROW). We depart from the standard SOE model along several dimensions. Firstly, we nest two different pricing paradigms: local currency pricing (LCP) alongside producer currency pricing (PCP). Second, the production function incorporates capital and intermediate inputs produced domestically and abroad. Finally, international asset markets are incomplete. Using US and Canadian data, we explore the empirical evidence for PCP vs LCP pricing paradigms through a Bayesian estimation likelihood race and a comparison with the second moments of the data. We then examine the implications of these two paradigms for the conduct of monetary policy using optimized Taylor-type inertial interest rate rules with a zero lower bound constraint. The main results are: first, in a likelihood race LCP easily beats PCP and fits reasonably the second moments of the data; second, whereas for the closed economy ROW the price-level rule closely mimics the optimized general inflation-output rule, for the SOE the corresponding result requires a nominal income rule.