We use macroeconomic panel data from 16 countries to estimate a
model that enables us to find evidence for the UIP. Interestingly
enough, we find strong evidence for UIP when we use long-term
interest rate differentials. Moreover, the corresponding
short-term interest rate coefficient turns out to be zero. We
conclude that UIP does not hold in the short-run. Since the static
fixed effects model may be inefficient and biased because of
significant first order autocorrelation, we specify a dynamic
panel data model, which is in turn estimated by GMM. Results from
the dynamic model are in line with what has been found before from the
static model.