Is there PPP in the data? : new evidence from a structural DSGE model

There has been a long-running debate in international macroeconomics about whether there is purchasing power parity (PPP) in the data. But while the literature has been dominated by reduced-form evidence focusing on the real exchange rate dynamics including its long-run behaviour, little attention has been paid to establishing whether a full structural model with PPP can empirically explain the data. In this paper we test a DSGE model in which the long-run forces of trade force the exchange rate back to PPP, while shocks to preferences, technology and policy drive it away from PPP in the short run. We do so by using the method of Indirect Inference where we test whether the modelís simulated behaviour can match the reduced-form VARX behaviour found in the data. We Önd, using a sample of the UK and US data observed between 1997 and 2020, that this PPP-constrained model Öts the data well. Thus, even over quite a short recent sample the real exchange rate is being consistently driven back to a PPP equilibrium, in spite of all the shocks and policy pressures creating divergence in this turbulent period, which includes the global Önancial crisis, Brexit and Covid. As the model suggests, such convergence is driven by forces of trade generated by cross-country competition which drives price equivalence as productivity converges globally, as implied by classical trade theory.