Publication date: March 7, 2014
Perhaps the most important decision a mortgage borrower makes is the type of mortgage: whether fixed-rate (or FRM), or adjustable-rate (ARM). If it is fixed, they decide the term; if it is adjustable, they decide the initial period until the rate adjusts.
The available terms on FRMs today are 10, 15, 20, 25 and 30 years. The available fixed-rate periods on ARMs are 1, 3, 5, 7 and 10 years. All ARMs have terms of 30 years.
Having choices is a strength of the system, because the different mortgage types can serve borrowers with different needs. The system is weak, however, in helping borrowers make good decisions.
• Poor decision capacity
Mortgages tend to be complicated, and consumers are exposed to them only once or a few times during a lifetime. They have no opportunity to enhance their knowledge through trial and error, which is largely how they learn about other products and services.
While some potential mortgage borrowers, recognizing the importance of the decision, will put in the time required to become knowledgeable, most prefer to depend largely on the advice of others.
• Poor advice
In selecting a mortgage, most borrowers get their advice from their loan provider, a loan officer or mortgage broker. While LPs can no longer steer borrowers to the products that pay them a larger commission — such steering is illegal under regulations issued after the financial crisis — eliminating biased advice does not assure good advice.
LPs are rewarded for their sales skills, not for the quality of their advice or their ability to exposit difficult ideas to borrowers. They tend to avoid extended discussions of different loan types and options because that slows down the process, and can derail it altogether. LPs earn no more if the borrower selects the right type of mortgage than if they select the wrong type. What matters is that the borrower is satisfied with the LP’s advice, and the loan closes.
• Poor disclosures
One of the alleged purposes of mandatory disclosures is to help borrowers compare different types of mortgages. The cornerstone of such efforts is the annual percentage rate or APR, which is supposed to be an objective measure of mortgage cost that allows borrowers to make unbiased comparisons.
But it is nothing of the sort. The APR does not include all mortgage costs, and it is calculated in every case over the term of the mortgage, even though very few mortgages run to term.
It is fortunate that because most borrowers don’t understand the APR, few try to use it.
• A new approach
I recently developed an alternative and much simpler approach based on the following rule: the best mortgage type is the one that has the lowest total cost over the period that is the borrower’s best guess as to how long they will have the mortgage — provided that the borrower can afford the payment.
If she cannot afford the payment, the best mortgage is the one having the second lowest total cost. And so on.
The cost measure that is used includes upfront and monthly costs, lost interest on those costs, less tax savings and balance reduction. On ARMs, costs are calculated on the assumption that interest rates increase to the maximum extent permitted by the ARM contracts.
In the current state of the economy, that appears to be the most plausible assumption, subject to change in future years.
The results based on market rates as of Feb. 28, 2015 are shown in the accompanying box.
For example, a borrower who expects to be in his house 11 to 15 years will minimize costs with a 15-year FRM, but if the payment on the 15 is unaffordable, the next best choice is the 20-year FRM, and if that payment is unaffordable, third best is the 10/1 ARM.
The results may be affected by changes in rate relationships, such as the rate on ARMs falling relative to the rate on FRMs, but these don’t occur very frequently.
My intention is to keep this table up to date on my website.
The writer is Professor of Finance Emeritus at the Wharton School of the University of Pennsylvania. Comments and questions can be left at mtgprofessor.com