Can regulatory interventions alleviate financial crises? If so, which ones work? We draw inferences from the Japanese banking crisis of the 1990s using a hand-gathered database of bank loans gathered from original sources. Our results indicate that whereas risk-based capital infusions in Japan (similar to those following the 2009 Supervisory Capital Assessment Program (stress tests) in the US) were successful in stimulating aggregate lending by Japanese banks, earlier blanket infusions (comparable to the 2008 Troubled Asset Relief Program (TARP) in the US) were not effective. Moreover, changes in accounting rules in Japan that revalued bank assets (similar to the relaxation of mark-to-market requirements for banks in the US) did not increase aggregate Japanese bank lending, but rather reallocated it. Capital constraints during the crisis also induced many Japanese banks to close their overseas branches and switch their charters from international to domestic. This endogenous charter switch reversed the process of foreign direct investment (FDI) for many Japanese banks. Therefore we use the Japanese banking crisis as a natural experiment to test FDI theories and find empirical support for the relative access hypothesis, but not for the industrial organization approach or for the relative wealth hypothesis.
ASJC Scopus subject areas