Coronavirus (COVID-19): Relief options and Additional Resources

The Role of Social Relationships in Financial Intermediation: Empirical Evidence from the United States Small Business Credit Market







United States Small Business Administration
Office of Advocacy
RS 173


The Role of Social Relationships
in Financial Intermediation:

Empirical Evidence from the United States
Small Business Credit Market

by Margaret Jane Miller


1996. 21p. Margaret Jane Miller, University of California, Berkeley CA. under contract no. SBA-8027-OA-93



Small business owners regularly cite inadequate access to credit as one of the most important problems they face. One factor that discourages lending is the lack of information on the creditworthiness of small firms. Although information problems exist in all credit transactions, they are particularly acute in the small business finance market. The cost to a bank for searching out credit information on an individual small business may be large in comparison with the loan size, thus discouraging the bank from offering credit.

The lack of published and standardized financial information on small businesses increases the importance of informal information, such as that which is gathered through personal interaction with the borrowers or third parties. In the case of small businesses there is relatively little third-party information available, so the relationship between the borrower and the lender is exceedingly important.

This research project, part of a Ph.D dissertation, tests the hypothesis that personal relationships between borrowers and lenders affect loan terms and conditions in the U.S. small business credit market.


Scope and Methodology

A theoretical model was developed to describe the role played by relationships in the credit market. The dissertation suggests that the relationship that exists between the borrower and lender is an important factor in the loan quality that can affect the interest rate charged by lenders. A relationship provides frequent opportunities for the lender to monitor and directly communicate with the borrower, thus reducing the transaction costs. Moreover, repeated interaction between the lender and the borrower creates a "reputation" for the borrower -- his or her good name serves as a form of collateral securing the loan.

The theoretical model provides a framework for empirical tests. First, it suggests that a borrower­lender relationship should reduce the interest rate charged in a competitive credit market, other things being equal. Second, it suggests that the distance separating a bank and a given client may affect loan interest rates. Third, the model specifies that different client characteristics, such as equity investment, may affect the interest rate on credit as well as the size of the loan.

Information available in the National Survey of Small Business Finances (NSSBF) was used to empirically test the interest rate model. The NSSBF, commissioned jointly by the U.S. Small Business Administration and the Board of Governors of the Federal Reserve System in 1988, provides data on approximately 3,400 enterprises. Of particular importance for this study, firms were asked numerous questions regarding the nature of their relationships with up to six financial institutions. Direct evidence of the importance of relationships was solicited, including information on the distance between the business and financial institution, the method of communication (in person or by telephone, etc.), and the owners' reasons for choosing their financial institutions.

The variables used to measure the closeness of the borrower-lender relationship were grouped into five categories for a regression analysis: (1) loan characteristics; (2) financial characteristics of the small firm borrower; (3) relevant non­financial characteristics of the small firm; (4) characteristics of the borrower­lender relationship; and (5) information on the macroeconomic environment.



  • The variable indicating a personal relationship between firm owners or managers and bank staff was found to be statistically significant for small firms (at the 95-percent confidence interval). The interest rate paid by small firms that did banking in person or by telephone was reduced by more than three­quarters of a point in several of the statistical estimations.
  • Firms within close proximity to the bank received credit on more favorable terms. All other things being equal, small firms that received loans from banks located within one­half mile of their place of business paid approximately three­quarters of a point less in interest.
  • Some other variables that measured the closeness of the relationship between the bank and business, and which did not perform as well, included funds in a checking or savings account, the total amount of credit outstanding with an institution, and the number of years they have done business together. None of these relationships proved to be significant at the 90-percent confidence level.
  • The empirical results were strongest for smaller firms. When the sample firms in the NSSBF were divided into percentiles in terms of number of employees and total firm assets, the interest rate model performed best for the lowest 75th percentile of firms by number of full­time employees, indicating the bank­business relationship is of greatest value to smaller firms.


Ordering Information

The complete report is available from:

National Technical Information Service
U.S. Department of Commerce
5285 Port Royal Road
Springfield, VA 22161
(703) 487-4650
(703) 487-4639 (TDD)

Order number: PB96 195433

Price codes: A03/$19.50; A01/$10.00 Microf.
*Last Modified 6-11-01