We show how agency problems between lenders (principals) and third-party originators (TPO; agents) imply that TPO-originated loans are more likely to default than similar retail-originated loans. The nature of the agency problem is that TPOs are compensated for writing loans, but are not completely held accountable for the subsequent performance of those loans. Using a hazard model with jointly estimated competing risks and unobserved heterogeneity, we find empirical support for the TPO/default prediction using individual fixed-rate subprime loans with first liens secured by residential real estate originated between January 1, 1996, and December 31, 1998. We find that apparently equal loans (similar ability to pay, option incentives and term) can have unequal default probabilities. We also find that, initially, the agency-cost risk was not priced. At first, the market did not recognize the higher channel risk, since TPO and retail loans received similar interest rates even though the TPO loans were more likely to default. We also show that this inefficiency was short-lived. As the difference in default rates became apparent, interest rates on TPO loans rose about 50 basis points above otherwise similar retail loans. Copyright 2002 American Real Estate and Urban Economics Association.

No abstract

No abstract

Supply equations for five output groups and demand equations for four input groups in ten regions of the United States are estimated and evaluated. The econometric estimation is conducted for complete regional product supply and input demand systems subject to competitive theory. The results document the extreme diversity of production relationships within the United States. They clearly indicate the unequal effects of changes in economic conditions and government policies on major production regions.

A Markov chain is a natural probability model for accounts receivable. For example, accounts that are 'current' this month have a probability of moving next month into 'current', 'delinquent' or 'paid-off' states. If the transition matrix of the Markov chain were known, forecasts could be formed for future months for each state. This paper applies a Markov chain model to subprime loans that appear neither homogeneous nor stationary. Innovative estimation methods for the transition matrix are proposed. Bayes and empirical Bayes estimators are derived where the population is divided into segments or subpopulations whose transition matrices differ in some, but not all entries. Loan-level models for key transition matrix entries can be constructed where loan-level covariates capture the non-stationarity of the transition matrix. Prediction is illustrated on a $7 billion portfolio of subprime fixed first mortgages and the forecasts show good agreement with actual balances in the delinquency states.