Testing for output convergence: a re-examination

This paper investigates output convergence for the G7 countriesusing panel time-series techniques. We consider both the nullhypotheses of no convergence and convergence. It is shown thatinferences on output convergence depend on which one of thetwo null hypotheses is considered. Further, the no convergenceresults reported in previous studies using the time-series definitionmay be attributed to the low power of the test procedures beingused. Our results also highlight some potential problems oninterpreting results from some typical panel unit root and stationaritytests.