Deteriorating Bank Health and Lending in Japan: Evidence from Unlisted Companies under Financial Distress

Abstract

When a borrower faces a hold-up problem, deteriorating bank health might reduce a borrower's credit availability. However, a bank with an impaired balance-sheet might attempt to ‘gamble for resurrection’ and hence might increase risky lending to zombie firms. The purpose of this paper is to investigate what impacts weakened financial conditions of banks had on loans outstanding to medium size firms in Japan. Estimating lending functions, we examine the determinants of lending to unlisted Japanese companies in the late 1990s and the early 2000s. We find that two alternative measures of the bank health, regulatory capital adequacy ratios and ratios of non-performing loans (NPLs), had opposite impacts on lending. In the case of regulatory capital adequacy ratios, its deterioration had a perverse impact on the bank's lending. The deteriorating NPL ratios, however, increased lending to troubled firms to keep otherwise economically bankrupt firms alive.

Key Words:

• Bank–firm relationship
• credit crunch
• misallocation
• unlisted firms

JEL CLASSIFICATIONS:

• G21
• G31
• G32

Acknowledgements

Earlier versions of this paper were presented at the Bank of Japan (Research and Statistics Department) and at the Third Financial Markets Asia-Pacific Conference in Sydney. We would like to thank the participants for their helpful comments. Views expressed in this paper are those of authors and do not necessarily reflect those of the Bank of Japan or the Research and Statistics Department.

Notes

1. This issue is similar to the debate on the problem of ‘soft budget constraints’ discussed by Dewatripont & Maskin (1995) and Berglöf & Roland (1995).

2. Ahearne & Shinada (2005) also reached similar conclusions.

3. Authors such as Ito & Sasaki (2002) and Woo (2003) showed that deteriorating capital adequacy ratios of Japanese banks caused ‘capital crunch’ in the 1990s.

4. Some of the problems appeared as a Japan premium (see, for example, Batten & Covrig, 2004).

5. We regarded the data as outliers when the growth rate of the outstanding loan exceeds 200 percent, the absolute value of the normalized profit is greater than 20, and the absolute value of the debt–asset ratio exceeds 10.

6. Many companies close their books in March, but not all the companies covered by the analysis did so. Data are, thus, arranged on the basis of fiscal year when books were closed.

7. In equation (1), the threshold is equal to –γ/(2δ). Estimating similar equations, SKS showed that the threshold was less than 0.5 for listed firms in Japan.