We propose an IV panel unit root test robust to nonstationary error volatility. Its finite-sample performance is convincing even for many units and strong cross-correlation. An application to GDP prices illustrates the inferential impact of nonstationary volatility.

The puzzling Monte Carlo finding that the size distortion of meta-analytic panel unit root tests increases with the number of panel series is explained as the cumulative effect of arbitrarily small size distortions in the time series tests composing the panel test.

We investigate the OLS-based estimator s2 of the disturbance variance in an error component linear panel regression model when the disturbances are homoskedastic, but spatially correlated. Although consistent (Song and Lee, Econ. Lett. 2008), s2 can be arbitrarily biased towards zero in finite...