Neat review post by Edward Glaeser assessing the evidence. He finds it inconclusive, but largely because he seems to be focusing on deltas rather than absolutes - it's true that increases in inequality prior to the crash were not that stark, and that the link between reckless borrowing and increases in inequality is not robust.
However, the biggest financial collapses were in countries with high inequality: US, UK, Ireland, Greece, Spain... whilst egalitarian Sweden is doing OK. Of course, there may not be a direct causal link - it could just as easily be that inequality and financial recklessness are both the result of something else: like an obsession with market liberalism, for example.
Incidentally, this seems to be a problem with the economists' preference for fixed-effects models. The stable cross-national differences that seem to account for outcomes end up getting missed.