“…It was proposed as a method to overcome the sensitivity of results with respect to the set of controlling variables in a regression. Since then, BMA has been applied widely in the empirical growth literature (e.g., Durlauf et al, 2008;Prüfer and Tondl, 2008;Winford and Papageorgiou, 2008;Ciccone and Jarocinski, 2010;CrespoCuaresma et al, 2011) and in other areas of economics (e.g., Koop and Tole, 2003;Tobias and Li, 2004). Recent papers have contributed towards the development of summary measures of the output (Ley and Steel, 2007;Doppelhofer and Weeks, 2009); led to greater understanding of prior assumptions (e.g., Ley and Steel, 2009;; and extended the technique in ways that are relevant to growth regressions such as threshold models (Crespo-Cuaresma and Doppelhofer, 2007), heteroscedasticity (Doppelhofer and Weeks, 2008), endogeneity (Cohen-Cole et al, 2009;Lenkoski et al, 2011;Koop et al, 2011;Karl and Lenkoski, 2012), and panel data models (León-González and Montolio, 2004;Moral-Benito, 2010;2012, Chen et al 2011.…”