Long-run purchasing power parity revisited: a Monte Carlo simulation

Abstract

The existence of long-run purchasing power parity (PPP) implies that a cointegration vector of nominal exchange rate, domestic price, and foreign price is expected regardless of using the Engle-Granger two-step method or Johansen maximum likelihood approach. However, this paper has found conflicting results: the Engle-Granger technique tends to reject the long-run PPP hypothesis whereas the Johansen method is generally supportive of long-run PPP. Via Monte Carlo simulations, the present paper finds that the Johansen approach has a bias toward supporting long-run PPP especially under the circumstances in which the assumption of normally or/and independently and identically distributed disturbance terms is violated.