Journal of Real Estate Finance and Economics, 23:3, 375±378, 2001
# 2001 Kluwer Academic Publishers. Manufactured in The Netherlands.
Comment: Secondary and Primary Mortgage
DAVID J. PEARL
Of®ce of Federal Housing Enterprise Oversight, 1700 G. St. N.W. Washington, DC 20552
``A New Spin on the Jumbo/Conforming Loan Rate Differential,'' by Brent W. Ambrose,
Richard Buttimer, and Thomas Thibodeau, and ``Credit Scoring and Mortgage
Securitization: Implications for Mortgage Rates and Credit Availability,'' by Andrea
Heuson, Wayne Passmore, and Roger Sparks, bear on two important areas for housing
®nance practitionersÐthe extent to which homeowners bene®t from the implicit federal
guarantee of securities issued by Fannie Mae and Freddie Mac (here referred to as ``the
enterprises'') and the evaluation and pricing of mortgage credit risk.
These comments re¯ect the discussants' business school training and extensive
experience in the mortgage ®nance industry and in enterprise regulation. The comments
focus on these areas from two vantage pointsÐthat of a safety and soundness regulator and
that of an industry participant concerned with national housing ®nance policy. Credit risk
is a key concern for regulators because credit losses erode capital and capital protects
against losses under economic stress. Companies in the business of mortgage-credit-risk
management, along with safety and soundness regulators, use credit-risk models to price
risk and to evaluate the extent of credit losses that might result under adverse economic
conditions. Ambrose et al. implicitly suggest an increased emphasis on house-price
volatility as a factor in this type of analysis. (Factors typically include loan-to-value
ratioÐcurrent and contemporaneousÐloan age, the spread between the mortgage coupon
and current mortgage rates, burnout, and more recently, consumer credit scores.) In basing
its analysis on a variant of the Black-Scholes model, Ambrose et al. also raise the question
of what types of models best evaluate or price credit risk. Heuson et al. raise the subject of
credit scores as measures of mortgage-credit risk.
From the standpoint of national housing ®nance policy, the articles question the
effectiveness of government-sponsored enterprises in reducing the cost of homeowner-
ship. Both suggest that not all the bene®ts of an implied federal guarantee of the
enterprises' securities are passed on to homeowners in the form of lower interest rates.
Ambrose et al. suggest that the differential between conforming and nonconforming
mortgage rates may be attributed, in part, to a credit-risk differential rather than to the
pass-through of bene®ts of the implicit guarantee alone. Heuson et al. suggest that
borrowers don't bene®t from the liquidity premium created by enterprise securitization.
The practical applicability of research depends on the extent to which it takes the
realities of the marketplace into consideration. In some respects, this discussant sees those