September 24, 2014, Revised September 27, 2014

For the last 24 years, rates have been in the single-digits, supporting a widespread presumption in the mortgage industry that this is the normal state of affairs. Yet the rate was 10% or higher during most of the 11 year period 1979-1990. Given this history and the current outlook for the economy and monetary policy, it is prudent to consider the consequences of a return to double-digits, and to take whatever measures are available to soften their impact.  The need for preventative measures is particularly acute in the reverse mortgage market.

The focus of this article is the transition from the current 4-5% market to a 10% market. Problems associated with the higher rate level -- living with a 10% rate rather than a 4-5% rate – will be discussed in another article.

Transition to a 10% Forward Mortgage Market

The transition problem arises out of the impact of rate increases on mortgage applicants who have loans in process that have not yet been locked – the final rate has not yet been set. These applicants have made decisions to proceed based on one set of rates, and have incurred expenses in the process, perhaps including a non-refundable purchase deposit, only to find themselves forced either to pay a rate higher than they had expected, or to drop the deal and count their losses.  

Every time there is a rate spike, mortgage applicants who are not locked must either pay up or give up. The shorter the period in which the rate increase is concentrated, the larger the fall-out and losses.

Unlocked applicants caught by a rate spike fall into four groups. The first group do not yet meet lock requirements because their applications have not been fully processed; perhaps their appraisals have not yet been delivered, or problems may have emerged in connection with their credit. A second group meets all the requirements to lock but needs a longer lock period than the 60-90 days customarily available, usually because they have a house under construction. The third group consists of “floaters” who could lock but choose not to because they believe interest rates are going down, or at least rates will be stable and they will avoid the lock fee. A fourth group believes themselves to be locked but aren’t because the lenders who locked them did not hedge their own positions and will be unable to honor their locks when the interest rate spikes.

There are no statistical data on the number of unlocked applicants in the forward market, but my rough guess is that it is less than 10% of all applicants. Further, the great majority of borrowers are aware that their rate is not final until it is locked. Unless rate increases in the future are more concentrated in short periods, the fall-out and losses from future rate spikes should be similar to those we have had in the past. Those who are caught will be seriously hurt but the market will go on as before. 

Transition to a 10% Reverse Mortgage Market

The impact of rising rates on the HECM reverse mortgage market could be much worse, perhaps devastating. This market has never been tested by rising rates, since it was just getting started when the period of double-digit rates was ending in the early 90s.

The HECM market is much more vulnerable than the forward market, for two reasons. First, most reverse mortgage applicants on any given day are unlocked. In part, this is due to the longer processing period, which among other things reflects the legal requirement that every borrower must be counseled. But the major factor is that locks are not available for periods longer than 2 weeks. Loans are not contractually locked until shortly before closing, and therefore locks provide very little rate protection.

The second reason the HECM market is more vulnerable than the forward market to rate increases is that most forward borrowers are aware of their exposure to a rate increase if they aren’t locked while most reverse mortgage borrowers are not. Reverse mortgage merchandising is primarily focused on the amounts seniors can draw rather than the rate. Most lenders don’t show rates on their web sites. While close scrutiny of the Good Faith Estimate that the applicant receives shortly after applying would reveal that the rate is not locked, the disclosure is oblique and few seniors grasp it.  

In addition, some (perhaps most) HECM lenders voluntarily lock the rate that prevailed when the applicant was first told how much could be drawn, even though this might be higher or lower than the market rate on the lock day. This avoids having to explain to the client early on that the rate won’t be known for sure until shortly before they close. The practice of providing voluntary but non-contractual rate locks is facilitated by the relatively large markups that prevail in this industry.

This practice works only so long as rate increases are small enough to be accommodated by lenders through tolerable reductions in their markups. If the transition from 5% to 10% is slow and steady, all will be well but expecting that is wishful thinking. It is very unlikely that lenders can absorb a rate increase larger than 1%. Yet there were at least four occasions during the years 1979-81 when the increase exceeded 1% within one month, and three additional periods when the rate rose by more than 1% over two months.

The worst case is that a rate spike too large to be covered by a reduction in lender markups would confront reverse mortgage applicants in process with large and unexpected rate increases. They will all claim with good reason that nobody warned them that this might happen. The media will portray the episode as the exploitation of senior homeowners by greedy lenders, and resolutions will be introduced in the Congress to fold up the HECM program.

The best case is that nothing of the sort would happen because of another unique feature of the HECM market. The amount that can be drawn on a HECM is based not on the HECM rate but on an “expected” rate, and this rate is locked by HUD for at least 120 days beginning with the date the application is submitted. Hence, if a rate spike occurs, borrowers who close within that period will be able to draw the amounts they had been promised earlier. While their HECM rate might go from 5% to 10%, it wouldn’t reduce their draw amount.

The best case scenario is based on the premise that borrowers are only concerned with the draw amounts, which is certainly true for some. The borrower who intends to hold a credit line unused for 20 years will be indifferent, because the initial line will not be reduced by the rise in rates while future growth of the line will be at the higher rate. But such borrowers are rare. Most HECM borrower are concerned with how rapidly their home equity declines, those with short time horizons particularly so, and they will be adversely affected. Assuming the worst is the prudent course of action.

One way to prevent a bad scenario is to develop and deploy longer lock periods, at least comparable to those available in the forward market. But that will require an industry effort that includes the secondary market players along with GNMA. It could take years even if the process were to start tomorrow.

A second approach, which can be adopted by individual lenders, is to provide an early disclosure that makes crystal clear that applicants assume a rate risk associated with market volatility, but that volatility works in both directions and they will benefit if rates decline. I have developed such a disclosure that is available on request. Lenders who provide voluntary but non-binding locks at application would instead lock at the price prevailing on the lock day – just as it is done in the forward market.

To make the message fully credible, however, the applicant must be able to monitor their rate at any time prior to the lock day. While only a few lenders offer this feature to their clients today, the technology required is relatively simple and readily available.

Implementation of this approach will not prevent a rate spike from torpedoing deals and making unlocked applicants unhappy. However, it will clarify that market changes work in both directions, that the applicants who were caught by the spike had been warned about the risk, and that they had not been abused by the lender.