“The world of finance hails the invention of the wheel over and over again, often in a slightly more unstable version.”
John Kenneth Galbraith
The 30 year fixed rate mortgage loan (FRM) has been a mainstay of the American mortgage market for more than 60 years. This loan has proven to be highly beneficial from the perspective of the borrower. The borrower benefits from the security of a fixed payment and balance amortization combined with the option to refinance should the market mortgage rate decline below the note rate on the existing loan. This option, known as the refinance or prepayment option, makes it difficult for most commercial or savings banks to manage the interest rate risk on the loan.
Consequently, most of these loans over the past 20 years have been securitized and sold into the secondary market. Before that, thrift institutions did hold fixed rate loans in portfolio, with disastrous consequences as interest rates rose sharply in the late 1970s. Through the securitization process, fixed rate loans have been sold to institutional investors (including giant banks) that have the motivation and capability to manage the prepayment risk.
This is still the case today, at least for government or agency loans (loans guaranteed or purchased by the FHA or the Government Sponsored Enterprises, Freddie Mac and Fannie Mae). However, the secondary market for non-agency loans is largely dormant. This means that the only game in town for non-agency loans is the traditional “originate to hold” or “portfolio lending” model. Will portfolio lenders support a robust market for the traditional fixed rate loan?
I suspect the answer is no. History has not been kind to portfolio lenders who hold 30 year fixed rate loans. Recall the carnage of the thrift industry. Most depositors at banks or thrifts do not want to lock up their savings for extended time periods, so the duration of the liability side of most bank or thrift balance sheets is pretty short. In order to properly hedge a fixed rate loan portfolio, banks or thrifts would need to engage in significant risk management programs to reduce the duration and convexity mismatches that are inherent in these products. There are many obstacles to the successful implementation of such programs, not the least of them is accounting rules.
Of course one strategy is to offer the 30 year fixed protection, but charge for the prepayment option. That way, these loans would be made available, but only at a substantial rate or point premium to non-callable alternatives.
Another approach to managing the interest rate risk of fixed rate loans is to issue covered bonds. These bonds are protected by the underlying mortgage loans, which remain on the books of the lending institution (issuing institution). Much of the interest rate risk is passed along to the buyers of the bonds. The credit risk remains with the issuer of the bonds. So far, the covered bond concept has not obtained huge traction in the U.S.
Other loan products are more attractive to the portfolio lender. Most other countries do not rely on the 30-year FRM. Yet many of them have home ownership rates comparable to or higher than the USA. While a variety of mortgage types exist in various countries, the most prevalent loan is a five year rollover. This loan is fixed for five years and then every five years “rolls over” to current market rates. Loans generally contain a yield maintenance provision such that prepayment of a loan in a lower interest rate market entails a make whole payment to the lender.
It is much simpler to manage the interest rate risk of this loan. For example, the lending bank can accept deposits according to the preferred habitat of its deposit customers, and then if need be use interest rate swaps to adjust the effective liability maturity to the effective maturity of the loan portfolio. The borrower assumes more of the interest rate risk.
Role of the GSEs
The GSEs have kept in portfolio large quantities of fixed rate loans. In 1981, Fannie Mae was market value insolvent due to the same rising rate pressures that killed the thrift industry. Subsequently, the GSEs developed the callable debt market as a means of more accurately “match funding” the fixed rate loan portfolio. Given this funding vehicle and low borrowing rates due to their implied government guarantee, GSE portfolio buying pushed FRM rates very low, certainly too low for thrifts or banks to compete.
While the thrifts were generally supportive of the rise of Freddie Mac and Fannie Mae, due to the enhanced liquidity provided by them, the logic of the GSE model was destructive for the thrifts. Benefiting from lower funding costs due to the implied government guarantee and massive leverage, the GSEs were able to offer fixed rate loans on terms more favorable to customers than could the thrifts. So, the thrifts shifted to adjustable rate loans and ceded the fixed rate market to the GSEs.
No one was too worried about this at the time because it looked like the new mortgage banking model (originate to distribute) was more efficient than the old portfolio lending model (originate to hold). Now people are thinking twice about this issue. Was the rise in the shadow banking system more “efficient” only due to government subsidy?
If the GSE portfolio business is wound down over the next few years, that would seem to strongly diminish the demand for fixed rate loans. However, should the guarantee function of the GSEs remain in place, then it is reasonable to expect that sophisticated investors will continue to buy fixed rate MBS. They will be comfortable bearing prepayment risk, provided there is an adequate risk premium built into the yield.
Further, it is likely that the secondary market for non-agency loans will eventually come back to life, with more strict underwriting standards and more comprehensive information available to investors. Thus, in all likelihood, the 30 year fixed rate loan will not disappear.