A Taxonomic Model for the Risk-Rating of Residential Mortgages

Abstract

A competitive economic system relies on its financial institutions and markets to allocate credit within an economy. For an efficient allocation of credit flows, the participants in financial markets need to have adequate information concerning the risk and return of competing investment opportunities. In the secondary mortgage market, two governmental instrumentalities-the Federal National Mortgage Association (FNMA) and the Federal Home Loan Mortgage Corporation (FHLMC)-provide assessments of the yields on alternative mortgage investments.' However, the information on the default risk of conventional mortgages offered in the secondary market has been rather limited. At the present time, a prospective investor must subjectively analyze the likelihood of default of each individual mortgage available in the market. Similar to the rating system in the bond market, it would be advantageous for the mortgage market to have a procedure for providing information on the relative default risk of a residential mortgage. An An innovation that would enhance the efficiency of the secondary mortgage market is the introduction of a methodology for the risk-rating of residential mortgages. This study develops a quantitative procedure for assessing the relative levels of default risk inherent in residential mortgages at origination. Through the application of a series of multivariate techniques, a taxonomic model is derived for the assignment of a conventional mortgage to one of six risk classes. The model is then utilized in the formation of a riskrating system for the secondary mortgage market.