Time-Varying Mortgage Prepayment Penalties

Posted: 17 May 2002

See all articles by Austin Kelly

Austin Kelly

Federal Housing Finance Agency (FHFA)

V. Carlos Slawson, Jr.

Louisiana State University


Recent empirical findings reveal that prepayment decisions of commercial property owners are slower than predicted by the pure options pricing model (OPM). Borrower decisions appearing slow, however, may be quite rational when prepayment penalties of a time varying nature are incorporated into the OPM. This paper uses a competing risks OPM, adjusted for each of four different categories of prepayment penalties, to analyze borrower prepayment behavior. We find the value of delaying prepayment is often higher for mortgages with declining rate penalties than for mortgages with static rate penalties, frequently requiring a substantially higher interest rate spread to trigger a refinance. Multifamily loan prepayment records reveal the type of prepayment pattern that the adjusted OPM indicate should occur, reducing the gap between empirical findings and theoretical predictions. The results have implications for the specification of regressions fit to historical data, for the pricing of newly originated commercial mortgages, and for pricing in the single-family market where prepayment penalties are reemerging.

Keywords: prepayment penalty, multifamily prepayment, option pricing model, competing risks

Suggested Citation

Kelly, Austin J. and Slawson, Jr., V. Carlos, Time-Varying Mortgage Prepayment Penalties. Available at SSRN: https://ssrn.com/abstract=302392

Austin J. Kelly

Federal Housing Finance Agency (FHFA) ( email )

1700 G St. NW
Washington, DC 20552
United States
202 414-1336 (Phone)

V. Carlos Slawson, Jr. (Contact Author)

Louisiana State University ( email )

E. J. Ourso College of Business
Department of Finance
Baton Rouge, LA 70803-6308
United States
225-578-6291 (Phone)

HOME PAGE: http://www.lsu.edu/business/finance/profile-viewer.php?un=cslawson

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Abstract Views
PlumX Metrics