The great advantage of using panel data over a simple cross-sectional sample is that one can control for the country-specific fixed effects ai. Failure to do so leads to biased estimates if these fixed or latent effects are correlated with the explanatory variables, as is likely to be the case. However, unfortunately the use of panel data also leads to more compli- cations if some or all of the variables in the estimating equation follow a trend over time. Such trending typically implies what econometricians call non-stationarity. One implication of non-stationarity is that the estimated coefficients and their standard errors cannot be trusted. In formal terms, a variable is.