다수은행 거래관계 기업의 차입집중에 관한 실증분석

It is generally observed that firms maintain relationships with multiple banks. Although most studies investigating the phenomenon of multiple banking relationships assume more or less equal borrowing weights among the banks,in practice creditor concentration or asymmetry in borrowing occurs, bywhich we mean that a firm tends to concentrate its borrowing on one mainbank and to fill the remaining loan necessity from other arm’s-length lenders. A firm can benefit from establishing a close, long-term relationship witha single bank. For example, studies report evidence that a close relationshipimproves accessibility to funds and reduces the cost of capital and collateralrequirements (Petersen and Rajan, 1994; Berger and Udell, 1995). However,the downside of a close relationship with a single bank is the tendency thata firm may find itself in a disadvantageous position for loan bargaining withthe bank, which has acquired monopolistic information concerning thefirm’s true credit worthiness. This is called the‘hold-up problem’, which canbe a significant constraint, at least for informationally-opaque firms. According to von Thadden (1992), multiple banking relationships may bea solution to the hold-up problem because they create competition amongbanks, and thus prevent one bank from extracting the monopolistic rentwhen offering the loan. Multiple banking relationships are reported to reducebankruptcy probability and the costs resulting from actual bankruptcy, andthus limit a firm’s potential financial distress. Finally, multiple relationshipsmay reduce a firm’s liquidity problems by spreading the risks of liquidityproblem of banks themselves. However, the negotiations for loan renewals liquidating value of the firm may be reduced due to the possible failure of coordination among banks(Bolton and Scharfstein, 1996). Considering the above discussion, we assert that creditor concentration (or asymmetric borrowing)with multiple banking relationships allows a firm to overcome the hold-up problem, as well as the possible failure in coordination among many banks. In other words, creditor concentration is a strategic solution for a firm to strike a balance between a single long-term relationship and multiple bank relationships. This paper investigates the influence of variables related to firm, bank, and market characteristics on the degree of creditor concentration in the Korean bank loan market. We use a data set from firms registered on the KISVALUE system of the NICE Information Service from 2006 to 2011. We use a Heckman selection model, and three proxies for measuring the degree of creditor concentration are included in the analysis: the Herfindahl-Hirschman Index (HHI), the largest financing share (CR1), and the Share of Inequality Index (SII). In the first stage of analysis, we estimate a probit for the probability of multiple banking relationships. In the second stage, we estimate the determinantsof the degree of concentration, which is conditional on borrowing from multiple lenders. In this study, we find evidence of stronger creditor concentration for firms with more opaque information, more liquidation value of assets, and lower leverage. In addition, creditor concentration intensifies when a firm is more profitable and less risky. Regarding main banks’characteristics, we find that variables, such as bank size, capital adequacy, and bank ROA, do not significant explain creditor concentration. However, the number of employees to total loans, which can be viewed as a proxy for monitoring costs, is negatively related to creditor concentration. Market conditions are found to influence creditor concentration: there is evidence of stronger creditor concentration during boom times, which seems to be related to the tendency of firms to disperse loans to avoid the liquidity problem during recessions. In addition, a weaker creditor concentration is found in more competitive banking environments, during which banks offer better loan conditions, resulting in more dispersed loans among banks. We divide the sample into SMEs and large firms to examine asymmetric bank borrowing patterns. We find that SMEs display highly concentrated borrowing for the firms with more opaque information, higher liquidation value of assets, lower leverage, more profitable business, and better credit rating. These results are similar to those from the previous analysis of the full sample. However, the relations are weak for large firms, which have weak incentives to make strategic decisions on borrowing allocations because they have various alternative financing sources instead of bank loans, such as issuing stocks or bonds. Next, we analyze whether there is any difference in firms’ borrowing behavior during the financial crisis. The impact of firm and market characteristics on the degree of concentration in borrowing during the financial crisis is weaker than that during the normal period. We conjecture that the firms' strategic decisions are limited by the banks because of their strict lending attitude during the financial crisis. Taken together, these results indicate that creditor concentration is attributed to firms' strategic decisions on balancing the holdup problem of relationship lending with the coordination failure of multiple lending. In particular, SMEs behave more strategically relative to large firms. In this way, creditor concentration allows the borrower to retain some of the benefits of long-term relationship lending as well as to enjoy the insurance effect against liquidity risks originating from the relationship lender.