Purpose -The purpose of this paper is to estimate a three-equation model of US commercial bank usage of the Farm Service Agency's (FSA) guaranteed operating loan and interest assistance programs. Also, to identify the key farm and banking variables that affect the decision to use loan guarantees and the volume of loans with interest assistance. Design/methodology/approach -A triple hurdle, three-equation system is estimated to model three decisions: to participate in the FSA operating loan program; whether to use interest assistance given the decision to participate in the operating loan program; and then the degree of participation in the interest assistance program. Statistical selection is modeled. Data on almost all commercial banks in the USA from 1995 to 2003 are used in the estimation sample. Findings -Statistical selection is statistically significant so selection must be included in the models. Variables reflecting state-level characteristics such as farm debt servicing ratio, individual bank loan-to-asset ratio, bank size and the general guaranteed loan and interest assistance environment are significant in all three equations. Intensity of interest assistance use varies markedly across states. Originality/value -The interest assistance program has high subsidy costs and is an important source of support for financially marginal farmers. Scant prior research has investigated this program. The present study also shows that modeling interest assistance usage must be embedded in a larger model to give a complete specification.
PurposeThe purpose of this paper is to examine the performance of the agricultural banking industry using both traditional and risk‐adjusted non‐parametric efficiency measurement techniques. In addition to computing efficiency scores, the risk preference structure of the agricultural banking industry is examined.Design/methodology/approachThe paper used data envelopment analysis (DEA) to examine the efficiency of agricultural banks in the year 2001. Standard cost efficiency is computed and compared to both profit and risk‐adjusted profit efficiency scores. The risk‐adjustment is a modification of traditional DEA wherein firm preferences are represented via a mean‐variance criterion. The risk‐adjusted technique also provides estimates of firm level risk aversion.FindingsResults from the traditional approach that does not account for risk indicate a low degree of efficiency in the banking industry, while the risk‐adjusted approach indicates banks are much more efficient. On average, 77 percent of the inefficiency identified by the standard DEA formulation is actually attributable to risk averse behavior by the firm. In addition, most banks appear to be substantially risk averse.Research limitations/implicationsThe risk‐adjusted DEA technique used in this study should be applied to other, diverse data sets to examine its performance in a broader context.Practical implicationsResults from this study support the idea that traditional DEA methods may mischaracterize the level of efficiency in the data if agents are risk averse. In addition, the paper outlines a practical method for deriving firm level risk aversion coefficients.Originality/valueThis paper sheds light on the agricultural banking industry and illustrates the power of a new efficiency and risk analysis technique.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.