With most fixed-rate mortgages (FRMs), borrowers can choose to pay an upfront fee, or points, in exchange for a more favorable mortgage rate. The purpose of this study is to empirically test two theoretical predictions regarding borrowers’ point choice and their refinancing decisions. One prediction is that borrowers with a shorter expected holding period (higher probability of prepayment) would opt for a loan with fewer points and higher interest rate. The other prediction is that once incurred, points become sunk costs, hence should not affect borrowers’ refinancing decisions in the future. That is, once the difference between the contract rate and the market rate becomes large enough to make it optimal to refinance, the borrower should disregard the points paid at origination. With our data set of 3,899 individual loans between 1996 and 2003, we are able to track a loan from origination till termination and control for observable borrower characteristics as well as macroeconomic variables.
To our knowledge, this is the first empirical test of the rationality of the choices made by borrowers with respect to the points they pay and whether or not these points play a role in their future refinancing decisions. Refinancing loan volume exceeded 2.6 trillion dollars and made up 73.27% of the single-family conventional loan market in 2003. The average points and origination costs for that year were 0.6% of the loan amount, totaling 15.6 billion dollars. According to the 2001 Residential Finance Survey, 27% of all the single-family outstanding first mortgages have positive number of points.
Clearly, understanding the rationality of borrowers' behavior will have significant implications for policy makers as well as pricing of mortgages by financial institutions. Theoretical models in the literature offer different explanations for the borrower’s choice among various rates and points combinations. According to these models, points serve as: a tax shield for borrowers, a way for lenders to charge for the prepayment risk, and, in the case of information asymmetry, a screening instrument to separate borrowers of different prepayment risk types.
Kau and Keenan 1987 use tax benefits to explain the existence of points. They show that if the marginal tax rate of the lender is less than that of the borrower, points can serve as a means for the lender to increase his after-tax yield without increasing the after-tax cost paid by the borrower. Their model suggests that there should be a positive relationship between points and borrower income, since individuals in higher tax brackets have greater desire to deduct points up front. There should also be observed difference between purchase loans and refinance loans because points on purchase loans are tax-deductible in the year they are incurred while points on refinance loans are amortized over the life of the loan.
A number of papers utilize the option theory to study borrowers’ point choice and refinancing decisions. The mortgage can be viewed as a callable coupon bond issued by the homeowner. When the borrower sells the house or refinances, she is exercising the option accompanying the bond and calling it at face value. For this option, the borrower pays a price above that of a non-callable bond. Points can be considered as the price of the prepayment option embedded in fixed rate mortgages. Dunn and Spatt 1988 view a typical fixed-rate loan as an adjustable-rate loan with a protective option to continue the loan at the same rate. Borrowers who are less likely to prepay the loan will find this protective option more valuable, hence will self-select toward loans with high points and low rates. A similar conclusion is reached in Chen and Ling 1989 and Follain et al 1992 who compare the borrower’s expected holding period to the lender’s assumption about the borrowers’ holding period and show that borrowers who expect a shorter duration on their mortgage than the duration assumed by the lender should choose loans with fewer points.
Download
PDF Ebook Do Borrowers Make Rational Choices on Points and Refinancing?
