MORE EVIDENCE ON THE LINK BETWEEN BANK HEALTH AND INVESTMENT IN JAPAN

Among stock-market-listed Japanese firms in 1994-1995, the financial h ealth of the firm's main bank did not significantly affect its investm ent behavior, after controlling for stock market valuation and cash fl ow. However, among the subset of bank-dependent firms, investment was lower by over 50% at firms that have one of the lowest-rated banks as their main bank. Splitting the sample by firm size rather than depende nce on bank finance did not reveal a significant effect of bank health on investment at either large or small firms, suggesting that the com mon practice of using size as a proxy for dependence on bank finance m ay be misleading. Because low-rated banks are smaller and deal with fe wer firms, and because bank-dependent firms themselves tend to be smal ler than non-bank-dependent firms, the aggregate effect on business in vestment in 1994-1995 that I identify is tiny. These results contrast with M. Gibson (1995, J. Bits. 68, 281-308), a similar study which, us ing data for 1991-1992, found a small effect of poor bank health on in vestment for all stock-market-listed Japanese firms and no difference between bank-dependent and non-bank-dependent firms. J. Japan. Int. Ec on., September 1997, 11(3), pp. 296-310. Division of International Fin ance, Federal Reserve Board, Washington, D.C. 20551. (C) 1997 Academic Press.