Bank information monopoly and trade credit: do only banks have information about small businesses?

Abstract

According to previous studies, a bank can set a higher interest rate for small firms by establishing a lending relationship since information asymmetry limits competition between banks. Therefore, the bank can acquire monopoly rent from small firms. However, if small firms can use trade credit which is another financial source, the bank cannot extract monopoly rent. In this article, we examine whether small firms can use trade credit if the bank sets a higher interest rate. Using panel data of small firms in Japan, our analysis shows that when the interest rate the bank sets is too severe or worsened for the borrower, the ratio of trade payables increases and the bank loses its amount of loans. This result implies that trade creditors alleviate the problems of bank information monopolies in Japan.

Acknowledgement

Credit Risk Database (CRD) data was used with permission from the CRD Association. The views expressed in this article do not necessarily reflect those of the CRD Association. I would like to thank Yoshiro Miwa, Motonari Kurasawa, Fumio Akiyoshi, Kaoru Hosono, Hidehiko Ichimura, Yoshitsugu Kanemoto, Michihiro Kandori, Nobuhiro Kiyotaki, Toshihiro Matsumura, Masaya Sakuragawa, Kuniyoshi Saito, Dan Sasaki, Yasuyuki Sawada, Daisuke Shimizu, Iichiro Uesugi, Peng Xu and Noriyuki Yanagawa for helpful comments and discussions. Seminar participants at the University of Tokyo, Gakushuin University, and the 2003 JEA Annual Meeting at Oita University also provided useful comments.

Notes

¹ Uesugi (2005) also argue that Japanese firms rarely offer early payment discount.

² In recent years, large numbers of empirical studies have investigated the function of trade credit (For example, Petersen and Rajan, 1997; McMillan and Woodruff, 1999; Ng, et al., 1999; and Fisman and Love, 2003). An explosion of papers argues that borrowers use more trade credit when credit from banks is unavailable because of the problem of information asymmetry.

³ In the 90's, Japanese economy had experienced negative macro shocks and the supply of bank loans was decreased (Honda, 2002; Suzuki, 2004). According to our analysis, trade credit might substitute for the decline of bank loans. See also Marotta (1997) for the relationship between macro shocks and trade credit.

⁴ See Bernstein (2002) for detailed information about NSSBF.

⁵ See also Boot (2000) and Gorton and Winton (2003) for useful surveys of relationship lending.

⁶ The stories of Japanese main banks are not one of examples for the relationship banking. See Miwa and Ramseyer (2005a).

⁷ See also Ongena and Smith (2000) for empirical studies about multiple banking relationships.

⁸ Rajan (1992) argues that if the firm borrows from multiple informed banks, the Bertrand competition with informed banks occurs and extracting the rent is impossible.

⁹ Berger, et al . (2005) describe that agency problems between a loan officer and a bank must be solved for small business lending. This paper does not discuss agency problems.

¹⁰ If informed banks form a cartel, they can extract rent from the borrower.

¹¹ According to SMEA (2003), the ratio of the small firms with a single lending relationship, where the number of workers is less than 20, is 25.7% and that ratio for firms with two banks is 31.8%. But firm size in SMEA (2003) is larger than average firm size in this paper, so the number of firms that have a single lending relationship in our dataset is larger.

¹² Biais and Gollier (1997) and Wilner (2000) also analyse a theoretical relationship between the long-term relationship and the price of trade credit.

¹³ The problems of information monopolies by supplier are not serious. According to ‘chusho kigyou no keiei jittai tou nikansuru chousa’ (the research paper about management strategy by SMEs) by Shoko Chukin Bank, the ratio of small firms that purchase from less than 10 supplier is only 12.6%. We cannot acquire data about the ratio of firms that purchase from only one supplier, but we can guess that ratio is tiny. See http://www.shokochukin.go.jp/pdf/cb04other04.pdf (in Japanese).

¹⁴ ‘2–10’ means that if buyers pay within ten days, suppliers make a 2% discount on buyers’ payment. ‘net 30’ means “full payment is due 30 days after the invoice date; after that date, the buyer is in default’ (Ng, Smith, and Smith (1999), p. 1110)

¹⁵ In Japan, the cost of credit from financial company called ‘Nonbank’ is higher than interest rates from bank. The ratio of firms that borrow from ‘Nonbank’ is low, about 3.6% (SMEA, 2003).

¹⁶ See http://www.crd.ne.jp/(in Japanese) for more information about the CRD.

¹⁷ According to White Paper on Small and Medium Enterprises in Japan, “Under the Small and Medium Enterprise Basic Law, the term ‘small and medium enterprises’(SMEs) refers in general to enterprises with capital stock not in excess of 300 million yen or 300 or fewer regular employees, and sole proprietorships with 300 or fewer employees. However, SMEs in the wholesaling industry are defined as enterprises with capital stock not in excess of 100 million yen or 100 or fewer employees, SMEs in the retailing industry are defined as enterprises with capital stock not in excess of 50 million yen or a workforce of 50 or fewer, and SMEs in the service industry are defined as enterprises with capital stock not in excess of 50 million yen or a workforce of 100 or fewer. Small enterprises are defined as enterprises with 20 or fewer employees. In the commercial and service industries, however, they are defined as enterprises with five employees or fewer.

¹⁸ This result does not depend on the sample selection. We can observe the result if the sample is divided by each industries and scales.

¹⁹ Previous papers (for example, Smith, 1987; Petersen and Rajan, 1997) argue that suppliers have an information advantage over banks, and therefore suppliers can offer credit to riskier firms.

²⁰ We add one because our dataset includes firms whose sales or firm age are zero.

²¹ We do not have the data on the prime rate in each bank. Hence, we obtained the short-term prime rate at the end of March from Financial and Economics Statistics Monthly issued by the Bank of Japan.

²² The median number of workers for small firms whose sales are less than 100 million yen in 1997 is 4 workers in this data set.

²³ Results do not change if we do not limit the sample to small firms whose sales are less than 100 million yen.

²⁴ The quick ratio difference is defined as the ratio of current asset minus inventories to current liability in year t + 1 less that ratio in year t.