Credit risk and the demand for agricultural loans

This paper uses the lender-borrower relationship to provide insight into the empirical estimation of loan demand/contract curves for agricultural loans. Loan demand is shown to be determine partly by lenders' willingness to provide debt. The implicit solution to the loan contract curve in the lender - borrower relationship is derived from the cumulative probability distribution function of loan losses, which is the same measure used as the dependent variable in credit scoring models. Consequently, empirical estimation of loan demand can be obtained from credit scoring models. This paper presents the theory and then provides loan demand estimates and elasticities using Farm Credit Corporation cross-sectional and time-series data. Empirical estimates indicate the possibility of a backward-bending loan demand curve, which may indicate some credit rationing in agriculture.