Taking these findings as a starting point, it seems doubtful whether sluggish prices offer an explanation
for slow adjustment to PPP. The results from the multivariate linear model established in Section
3.2 are compatible with results previously derived in the literature. This is in particular valid with
respect to the argument raised by Rogoff (1996) who states that the convergence speed to PPP is too
low to be fully explained by stickiness in nominal prices in the tradition of Dornbusch (1976). At first
glance, the smaller half-lives of shocks established by this study therefore seem to be more consistent
with standard macroeconomic models based on nominal rigidities. However, the overall results
suggest that the nominal exchange rate carries most of the adjustment burden and the speed of price
adjustment is not a main factor in the context of PPP. Hence, the argument of price stickiness might be
invalidated not by low convergence of real exchange rates but by the minor contribution of prices to
adjustment. A related argument has been raised by Cheung et al. (2004). A major task for theoretical
models is to incorporate the nonlinear adjustment of real exchange rates while keeping an eye on the
main role of nominal exchanges.