Working Paper: NBER ID: w15624
Authors: Paul R. Bergin; Reuven Glick; Jyhlin Wu
Abstract: This paper reconciles the persistence of aggregate real exchange rates with the faster adjustment of international relative prices in microeconomic data. Panel estimation of an error correction model using a micro data set uncovers new stylized facts regarding this puzzle. First, adjustment to purchasing power parity deviations in aggregated data is not just a slower version of adjustment to the law of one price in microeconomic data, as arbitrage occurs in different markets, in response to distinct macroeconomic and microeconomic shocks. Second, when half-lives are estimated conditional on macro shocks, micro relative prices exhibit just as much persistence as aggregate real exchange rates. These results challenge theories of real exchange rate persistence based on sticky prices and on heterogeneity across goods, and support an explanation based on the presence of distinct macro and microeconomic shocks.
Keywords: Purchasing Power Parity; Real Exchange Rates; Micro Prices; Macroeconomic Shocks
JEL Codes: F4
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Adjustment to purchasing power parity (PPP) deviations in aggregated data (F31) | Distinct mechanisms of adjustment (F32) |
| Local goods prices actively adjust to restore the law of one price in disaggregated data (F16) | Adjustment in aggregated data occurs through nominal exchange rates (F31) |
| Halflives estimated conditional on macroeconomic shocks (C41) | Micro relative prices exhibit persistence similar to that of aggregate real exchange rates (F31) |
| Responses to micro and macro shocks (E39) | Conventional estimates of speed of adjustment may be biased (C51) |
| Heterogeneity among goods (F12) | Persistence puzzle cannot be solely explained by aggregation bias (C41) |
| Significant portion of heterogeneity in adjustment speeds (C22) | Associated with responses to macroeconomic shocks (E39) |
| Evidence challenges traditional sticky price models (C54) | Firms adjust more to idiosyncratic shocks specific to their industry rather than to common macroeconomic shocks (L16) |
| Need for a model that incorporates roles of both macro and micro shocks (E13) | Explaining dynamics of real exchange rates and price adjustments (F31) |