We show that previous results suggesting that government ownership of banks has a negative effect on economic growth are not robust to adding more 'fundamental' determinants of economic growth, such as institutions. We also present regression results from a more recent period (1995-2007) which suggest that, if anything, government ownership of banks has been associated with higher long run growth rates, even after controlling for institutions and other variables suggested by the growth literature. Drawing on the current global financial crisis, we provide a conceptual framework which explains why under certain circumstances government owned banks could have a greater effect on economic growth than privately-owned banks.