Variance of Firm Performance and Leverage of Small Businesses

Abstract

We investigate the relationship between leverage and firm performance using small business data from Japan by estimating the effects of leverage on both average firm performance and the variance of firm performance. We find that leverage has a negative effect on average firm performance and a positive effect on the variance of firm performance. This suggests that the problem of moral hazard is severe for highly leveraged firms. However, when highly leveraged firms have sufficient collateral assets, the effects of leverage are positive for average performance, but negative for the variance of performance. This implies that when small firms have sufficient collateral assets, highly leveraged businesses are better performers.

Notes

13. See Freixas and Rochet (2008) regarding a simple model of a credit market with moral hazard.

14. Cerqueiro, Degryse, and Ongena (2011) use the regression model with multiplicative heteroscedasticity to examine the determinants of loan rate setting by banks.

15. According to Frank and Goyal (2003), leverage is determined by cash flows, the annual change in fixed assets (as a proxy for investment), working capital, and long‐term debt.

16. In contrast, as Lang, Ofek, and Stulz (1996) argue, firms with strong investment opportunities may choose lower leverage because of the cost of high leverage. If correct, the performance of highly leveraged firms will be lower on average.

17. See http://www.crd-office.net/CRD/english/ (last access date: January 2014) for information about the CRD.

18. According to the 2005 White Paper on Small and Medium Enterprises in Japan, “[U]nder the Small and Medium Enterprise Basic Law, the term small and medium enterprise (SME) refers in general to enterprises with capital stock of not in excess of 300 million or 300 or fewer regular employees, and sole proprietorships with 300 or fewer employees.”…“Small enterprises are defined as enterprises with 20 or fewer employees.”(p. viii).

19. According to Sakai, Uesugi, and Watanabe (2010), the CRD covered about 60% of small incorporated firms in Japan in 2001. As we limit our analysis to firms operating for more than five consecutive years in the CRD, the sample size in our analysis is smaller than the full sample collected.

20. Under this definition, the maximum (minimum) leverage for observations for which the high leverage dummy equals zero (one) is 1.0928 (0.8649).

21. The value at the third quartile for the ratio of a firm's tangible fixed assets to total assets is 0.5371.

22. The value at the third quartile of the annual growth rate in tangible fixed assets is 0.0095.

23. See Ch. 10 in Wooldridge (2010) for a detailed discussion.

24. The results are similar if we use the high leverage dummy.

25. We do not employ the square of firm age because the estimated coefficient for this variable is not statistically significant.

26. Ono, Uesugi, and Yasuda (2013) show that firms are more likely to default when banks offer credit guaranteed loans. This suggests that banks offer guaranteed loans to risky (including highly leveraged) firms.