Examining the use of Farm Service Agency guaranteed loans by commercial banks

Latisha A Settlage, Purdue University

Abstract

This dissertation examines the behavior of banks participating in the Farm Service Agency guaranteed loan program between the fiscal years of 1995 and 2003. Guaranteed loans are originated by banks but backed by the federal government. Borrowers must be creditworthy but unable to receive credit at reasonable terms and rates. If the borrower defaults, the federal government reimburses the lender for a portion of lost principal and interest. While banks are not the only commercial lenders eligible to originate guaranteed loans, they are the major users. The study objective is to identify lender and farm financial characteristics that influence the propensity of a bank to use guaranteed loans as well as variables that affect the level of usage of the program. A portfolio selection model is modified to include an asset choice for guaranteed loans. The theoretical model provides insight into the choices of independent variables for the econometric models. Bank asset size, loan-to-asset ratio, agricultural loan-to-total loan ratio, and multi-bank holding affiliation are among the lender variables considered. State-level farm financial variables include debt servicing ratio, debt-to-asset ratio, variability in net farm income, variability in the value of farm assets, as well as others. Bank use of guaranteed loans is considered for all guaranteed borrowers, beginning farmers, and socially disadvantaged farmers. Regression models are estimated for two types of guaranteed loans, farm ownership and operating loans. A double-hurdle specification is employed to account for incidental truncation in the dependent variable. Incidental truncation occurs because loan volume is only observed if a bank has the propensity to use the program. Results indicate that larger banks, banks specializing in agricultural lending, banks with aggressive lending strategies, and banks with a history of significant program participation are more likely to originate guaranteed loans and are predicted to originate higher guaranteed loan volume. Farm financial variables are also found to significantly indicate propensity and intensity of guaranteed loan use. In particular, higher debt servicing ratios and more variability in farm income and asset values are positively related to guaranteed loan use. Rates of return and default for other assets are also significant.

Degree

Ph.D.

Advisors

Dixon, Purdue University.

Subject Area

Agricultural economics|Finance|Banking

COinS