One size does not fit all... Panel data : Bayesian model averaging and data poolability

We show in this paper why researchers ought to pay particular attention to the issues of model uncertainty, data poolability, and country heterogeneity in their panel data applications. We focus on the identification of robust determinants of current account balances (CABs). Applying Bayesian Model Averaging, we adopt a flexible modelling approach to highlight that (i) many determinants have limited relevance when accounting for model uncertainty; (ii) OECD and non-OECD countries cannot be pooled together when accounting for slope heterogeneity; iii) cross-sectional and time-series relationships can diverge. Using a novel approach which allows us to control for country fixed effects in a cross-sectional setting, we demonstrate that this last finding is the outcome of substantial unobserved country heterogeneity and not diverging responses across time horizons. Our methodologies offer direct solutions to the issues raised in this paper and have broad applications.