Small business finance has traditionally been a local and close-knit affair. Small firms, whose informational opacity precludes them access to public capital markets, seek out local bank lenders whose geographic proximity allows them to observe and accumulate the “soft,” non-quantitative information necessary to assess these firms’ creditworthiness (Meyer 1998, Stein 2002, Scott 2004). This relationship based approach remains the method used by many, if not most, community banks to underwrite their small business loans today.

But some exceptions to these tight location-based credit relationships began to emerge during the 1990s, when the geographic distance between small business borrowers and their commercial bank lenders began to increase and some banks began using credit scoring models and “hard,” quantitative information to assess small business loan applications. Increasing geographic distances between small business borrowers and their bank lenders has been documented in a number of recent studies (e.g., Cyrnak and Hannan 2000; Degryse and Ongena 2002; Petersen and Rajan 2002; Wolken and Rohde 2002; Brevoort and Hannan 2004).

In some cases the magnitudes have been substantial. For example, in 2001 the median borrower-lender distance for business loans originated with backing by the Small Business Administration (SBA) was approaching 20 miles, more than triple the distances observed in the mid-1980s (this study, see Table 1 below). The dissemination of small-business credit scoring technology has also been rapid. First implemented in the mid-1990s, by 1997 over half of large U.S. commercial banking companies were using credit scores to assess at least some of their small business loan applications (Mester 1997; Akhavein, Frame, and White 2005).

The apparent decline in the importance of borrower-lender proximity and in person relationships for small business lending has potential implications for the supply and quality of small business credits as well as the strategies of banks that extend those loans. All else equal, greater geographic distance between informationally opaque firms and their bank lenders should increase the cost of lending and reduce the supply of loans e.g., bankers will make fewer in person visits because of high travel expenses, resulting in less accurate information, poorer credit assessments, and higher rates of loan default. Arguably, the implementation of small business credit scoring models could either dampen or exacerbate these outcomes.

If the predominant effect of credit scoring is to improve the quality of banks’ information about borrower creditworthiness, or provide banks with better decision making frameworks based on quantifiable rather than qualitative financial information, then credit scoring can potentially mitigate the adverse information costs associated with borrower-lender distance. But if the predominant effect of credit scoring is to reduce banks’ production costs e.g., by eliminating expensive on site visits, reducing loan analysis time, or generating scale economies associated with automated lending processes (Mester 1997, Rossi 1998) then credit scoring will increase the profitability of the marginal loan application without any mitigation of distance-related information costs.

By changing the optimal tradeoffs among information quality, customer service, loan production costs, and bank scale, these developments have affected banks’ existing competitive advantages and may determine whether banks engage exclusively in either relationship lending and transactional lending, or will do both kinds of lending, going forward (Boot and Thakor 2000; DeYoung, Hunter, and Udell 2004).

Download
PDF Ebook Borrower-Lender Distance, Credit Scoring, and the Performance of Small Business Loans