Aggregate loan quality assessment in the Farm Credit System

Changes in loan quality affect loan pricing, credit policy, and the capital structure of the Farm Credit System (FCS). Improving the assessment and anticipation of changes in loan quality is a continual challenge. This research shows that loan quality assessment through aggregate credit scoring models that analyze farm sector financial information is a valuable addition to risk management in the FCS.Aggregate models of PCA and FLB loan quality were developed, using Ordinary Least Squares in a pooled cross section time series framework, through analysis of loan quality and farm sector financial information for the St. Louis Farm Credit District. Collateral values, change in farmland values, and government farm policy are significant factors in the PCA model. Liquidity, change in farmland values, and off-farm income are significant in the FLB model. The models explain much of the variation in loan quality over time and are robust to various validation tests. The estimation process was hampered by multicollinearity which must be guarded against in using aggregate financial data.The research further demonstrates, through a loan pricing model, how the aggregate loan quality models can enhance risk management in a forward planning process. Insights from the research also reveal that loan quality can change rapidly and cannot be accurately predicted in advance; no clear early warning indicators were found.The aggregate credit scoring model can be a useful analytical tool for evaluating a loan portfolio and can provide new insights into the broader lending environment of the Farm Credit System.U of I OnlyETDs are only available to UIUC Users without author permissio