An application of the consumer lending credit scoring model to the health education assistance loan program: Testing its effectiveness in predicting repayment behavior
Abstract
Recognizing that educational loan (specifically HEAL) defaults represent a significant and unnecessary federal expenditure, the purpose of this research is to examine a specific educational loan program within the context of a model that has limited defaults by predicting borrower repayment behavior. Since the public policy/educational lending field has not produced this type of model; and in attempting to limit defaults has either created policies/procedures that are discriminatory, and/or imposed ineffective administrative burdens on the educational institutions, this research turned to an alternative discipline, (i.e. the consumer lending industry) that has minimized defaults, and has done so in a non-discriminatory fashion. The Health Education Assistance Loan (HEAL) program for graduate health professions students represented a feasible avenue for this application because the HEAL program: (1) was originally established as financially self-sustaining, but is currently requiring federal support due to the increasing level of defaults; and (2) as a non-subsidized, variable rate loan, this program more closely resembles consumer financing than other educational loan programs. This research utilized a prototype of the consumer lending credit scoring model, incorporating operational definitions of the borrower's 'character', 'capacity', and 'capital' for the purpose of determining the significance of each in predicting the HEAL borrower's ultimate repayment pattern. That is, using operational definitions of these specified 'c's', and applying such to a group of HEAL borrowers who had graduated from six different allopathic, educational institutions, it was the purpose of this research to identify the characteristics, pertinent to HEAL borrowers, that are associated with belonging to a specified repayment category (i.e. non-delinquent repayment, deferment, or forbearance). This application of the credit scoring model was successful in identifying HEAL borrower characteristics that are significantly associated with repayment category membership (i.e. borrower behavior). Taken in combination, these variables offer a summary description of the borrowers represented in each of the specified categories. The findings of this study offer a framework by which: (1) future research can be conducted (i.e. research that includes defaulted and delinquent borrowers and incorporates the variables that were not available to this study); and (2) the characteristics of those currently in postponed repayment can be identified, which will offer guidance to Health and Human Services in the development of future default reduction policies and procedures.
Subject Area
School administration|Higher education|School finance
Recommended Citation
Cooter, Raelynn, "An application of the consumer lending credit scoring model to the health education assistance loan program: Testing its effectiveness in predicting repayment behavior" (1994). Dissertations available from ProQuest. AAI9427516.
https://repository.upenn.edu/dissertations/AAI9427516