An empirical investigation of lending to small businesses

by Min, Kyung-Seol

Abstract (Summary)
ABSTRACT MIN, KYUNG-SEOL, An Empirical Investigation of Lending to Small Businesses. (Under the direction of Douglas K. Pearce and Karlyn Mitchell.) This dissertation analyzes lenders' two important decisions (loan approval decision and loan rate decision) in their loan evaluation procedures using the SSBF (Survey of Small Business Finances) data sets. First, I examine what factors play important roles in the determination of loan interest rates. In this test procedure, following Vickery (2007), I split the data sets into fixed-rate and variable-rate loans to find systematic differences in the loan rate determination between these two loan types. The regression results of the loan rate model are different from general expectations in many respects. Not many independent variables have significant coefficient estimates. There is little consistency among the three regression results (2003, 1998, and 1993 SSBF data sets) both in fixed-rate and in variable-rate loans. Wide use of credit scoring is analyzed to be one explanation of the disappointing results. Lenders evaluate the credit risk of loan applicants with their own credit scoring system, and they mainly use the credit scores in deciding whether to approve the loan, and for the approved loans, the loan rates do not vary much according to their credit scores. Next, the test results of the credit rationing theory by Stiglitz and Weiss (1981) show that in the higher market interest rate period lenders' credit standards get also higher, so they ration credit more than in the lower market interest rate period. Finally, in the analysis of ethnic discrimination, Asians and Hispanics were less discriminated against in 2003 than in 1998, but African-Americans still higher loan denial rate than white applicants. The enhanced objectivity in lenders' loan evaluation procedures by using the credit scoring models might be associated with the decreased ethnic discrimination, but more collection of the SSBF data sets and analyses using these data sets are needed to support this explanation.
Bibliographical Information:

Advisor:Lee Craig; John Lapp; Douglas Pearce; Karlyn Mitchell

School:North Carolina State University

School Location:USA - North Carolina

Source Type:Master's Thesis



Date of Publication:11/07/2008

© 2009 All Rights Reserved.