“…We experiment with many potential measures of elites (see e.g., Angeles, 2007;Angeles and Neanidis, 2009). Using variables that capture the presence of an influential elite as in Barth, Caprio, and Levine (2006) and Morck, Yavuz, and Yeung (2011) we find that the correlations between these measures and the supervisory power index are very low and statistically insignificant. 11 In a previous version of this paper we also used a panel VAR approach, in the fashion of Love and Zichinno (2006) and Holtz-Eakin, Newey, and Rosen (1988) to identify the isolated impact of bank regulations on inequality.…”