Behavior of Reverse Mortgage Borrowers - A case study

Tuesday, 16 September 2008 16:14

Articles - Industry Stats

User Rating: / 0
PoorBest 

    We analyzed borrower behavior and characteristics for all reverse mortgages we have on file. The results may indicate what kind of borrower behavior other originators can expect.

Borrower Characteristics

    We compiled a random sample of data on 212 loans originated since 2005. Of these, 9 were Cash Accounts and 2 were Homekeepers, leaving 201, or 95%, as HECMs. Only 2 of the HECMs were refinances. 90% were in Massachusetts, 10% in New Hampshire.

• Single borrower = 118 (55.6%)
• Couple borrower = 94 (44.4%)
• Total individual borrowers = 306

    The average age of all borrowers was 75.4 years and ranged from 62 to 98. Within couples, the youngest averaged 73.2 years and the oldest 76.4 years. Six couples had an age difference of ten or more years with Couple #212 being 29 years apart (he 91, she 62). This is interesting because the HECM calculations make no distinction for gender and are based strictly off the youngest borrower’s age. For example, our youngest couple, aged 64 and 63 respectively, have a far greater joint life expectancy than Couple #212.

    We had no cases of two borrowers who weren’t a couple (e.g. no senior plus adult child borrowers) and only one case of three or more borrowers on a single loan. Fifteen of the loans had a trust involved (7%) and eight (4%) had a life estate involved. About 5% involved a repair set-aside.

Cash Accounts

    Of our 9 Cash Accounts, 7 were jumbos averaging $1,134,000 appraised value and ranging from $2 million to $600,000. Two others had values in the HECM range but did not qualify for a HECM. Three of our 9 cash accounts were for multi-family residences. (Of the HECMS, 30 or 15% were for 3- or 4--family residences, (none for 2-family), which are fairly common in Massachusetts.)

HECM Appraised Values & Maximum Claim Amounts

    For our HECM loans, the average appraised value was $358,923. The highest value was $1,565,000 and lowest was $140,000. Because the average value is so high, many loans were limited by the county limits. (Of course, the limits varied by county and even changed from 2005 through 2008.)

    For 123 of our 201 HECM loans, the appraisal was below the applicable limit. But, for 78 (or 39%) of the loans the appraisal exceeded the limit. These naturally tended to be in the higher cost Boston suburban areas. (The average single family assessment for all of Massachusetts in 2007 was $404,000.) In our loans, above-limit appraised values were:

For those loans where the appraised value exceeded the limit, $9,065,000 of value was wasted, or an average of $127,676 per loan. We were able to use this analysis to determine if our entire historic portfolio had been generated using the new origination fee structure dictated in S. 2331, the Housing Reform Bill, we would have lost on average about $900 per loan, or a revenue reduction of about 15-20%, a serious issue since the new legislation had no effect on our costs.

Borrower Behavior at Closing

Our borrowers clearly displayed three tendencies at closing:

• About 22% selected the term or tenure option.
• Paying off liens at closing is very common – 79% did so.
• Not taking all of the available money and leaving something in the line of credit.

Term and Tenure

    We had 38 borrowers (or 19% of the HECMs) choose the tenure option. An additional 6 borrowers chose the term option, with term length ranging from 5 to 15 years. The average monthly payment across all term and tenure borrowers was $914.63, and ranged from $160 to $5000. (In this latter case the 86-year old borrower chose a term of 55 months, taking no lump sum and effectively exhausting the line of credit.) The highest tenure payment was $1595.41 to an 83-year old.
    About 25% of our term/tenure borrowers chose a specific, rounded, monthly amount and had the line of credit calculated accordingly.  But 75% worked in reverse, first selecting a round amount for line of credit and/or lump sum and having the term/tenure amount be a calculated, usually odd, amount. For example, borrower #146 chose a lump sum of $25,000, a line of credit of $100,000 and was left with a monthly tenure payment of $862.46.

    Tenure or term borrowers tended to be a little older, averaging 79.3 years, or 3 years older than the average. Due to the higher age, only 35% were couples– as opposed to 45% for all borrowers. Our oldest tenure borrower was 98.

Of the net principal limit, term and tenure borrowers (on average) allocated:

• 5.3% for lump sum payments
• 15% for lien payoffs
• 27.2% for the line of credit
• 52.5% for the NPV of the term or tenure payments themselves

12 of these 44 borrowers (or 27%) left themselves with no line of credit.  Only 14 had no liens to be paid off.

Lump Sum

    The lump sum is an area where borrowers have clear choices. Our borrowers chose lump sum amounts as follows:

The highest lump sums generally occurred with Cash Accounts; the largest was $325,000. The highest HECM lump sum was $294, 317, which was 100% of the net principal amount. We had only two borrowers (or 1%) who neither paid liens off, elected term/tenure, or wanted a line of credit and the lump sum effectively equaled the net principal limit. The average lump sum was almost $27,000 but was only $22,432 excluding the cash accounts.

Lien Payoffs

    Only 21% of our loans had no lien payoff. The remaining 79% of borrowers paid off an average of $76,650. Most money went for first mortgages but an unknown share went for tax or other liens. The high was $526,393 (from a cash account) and the low was a nominal $5. About 28% of the payoffs were above $100,000.

Line of Credit Selection

    The line of credit absorbs whatever net principal limit doesn’t go to lump sum, lien payoffs, and term/tenure NPV. (Net principal limit is calculated after closing costs and servicing set-aside are deducted from the principal limit.) 87% of our loans had a line of credit. The exceptions either had large liens or large lump sums and displayed a high proportion of term/tenure loans. Excluding cash accounts, our largest HECM LOC was $263,200 to the same 98-year old mentioned above who had a tenure payment. Our lowest LOC was $415. The median LOC was $100,000 and the average was $92,243.

    HECM borrowers who did not choose a term or tenure plan appear to act reasonably conservatively. Their net principal limit was divided as:

This is not the whole story. Lien payoffs are mandatory and of course is a major reason for seeking an RM. What about beyond that? Of the monies left after lien payoffs at closing, a mere 20% is actually withdrawn at closing. 80% is reserved for future use.

    Of course, borrowers have no choice, in a sense, as to the size of the line of credit. They have to get it and the line per se costs nothing. All they can do is not access it if they don’t want it. But, it seems clear most borrowers are not so drowning in debt or burdened by other expenses that they need to use most of the money at closing.

Borrower Behavior after Closing

Borrower behavior after closing involves these questions:

1. When do borrowers withdraw how much from their line of credit?
2. Do borrowers ever change their plan?
3. To what extend do borrowers pre-pay?
4. Last, what do they do with the money?

    Our project allowed us to shed some light on these questions, even if definitive answers require more work. In particular, the fungible nature of money makes identifying RM expenditures difficult.

    We were able to review a sample of HECM borrowers whose line of credit ranged from 20 to 29 months old. Conclusions were:

• 84% had made at least one withdrawal from their line of credit, but 16% had not; failure to do so was not related to how old the line was.
• Of 118 separate withdrawal events, 101 (or 85.6%) were $10,000 or less.
• 17 withdrawals (or 14.4%) were greater than $10,000; the largest being $50,000. (In addition, one borrower made two $30,000 withdrawals on the same day.)
• The average withdrawal amount was $6844 and the median was $5000.
• On average, there were about 4 withdrawals per person. The most active person in our sample took 13 withdrawals in 19 months. The average time between withdrawals was 3.8 months and the median was 2 months.

One withdrawal pattern is for the borrower to take out a large sum of money (greater than $20,000) within a month or two of closing and then not touch the LOC for an extended period of time. Apparently, they wish to minimize transaction hassle.
A second pattern involves the borrower not making any withdrawals when the LOC becomes available, perhaps because the money they received in a lump sum from the closing is sufficient to provide for them for several months. Yet, after a year or so, the borrower makes their first withdrawal, and then more frequent withdrawals follow.

    A third pattern is a borrower with no or very limited withdrawals - less than 10% of availability. The logic here is straightforward—the borrower doesn’t currently need cash, or wants to maximize the line growth.

    For our sample, the average original LOC was $106,474. Subsequent line growth raised the current line to an average of $121,069. Yet, the average total amount withdrawn from the line (including accrued interest) was $28,066, meaning only 23.2% of the available money had been withdrawn. As a distribution,

1. About 10% had less than 20% of their LOC left after about 2 years. They tended to have lower original lines, averaging about $30,000.
2. Another 10% had between 20% and 50% of their line left
3. About 37% had between 50% and 90% left
4. About 43% had over 90% left, half of whom had 100% left.

Two examples:

1. Borrower #8 (85 years old), had an original LOC of about $212,000, left untouched for the first 18 months, but then began making $1000 withdrawals on alternate months.
2. Borrower #32 (74 years old) started with a line of $24,000 and drew almost all of it the first month.

    Most borrowers aren’t totally rigorous in withdrawing money, i.e. $5,000 every other month. They tend to be fairly erratic, both in frequency and amount. Otherwise, they could have opted for a tenure/term plan.

    The majority of withdrawals (85.6%) are in even figures (e.g. rounded to the nearest 500), suggesting general usage for goods and services, not anything specific. The remaining 14.4% of the withdrawals are very specific amounts (e.g. $6,404), suggesting the borrower was paying one or more specific bills, such as taxes.

    Borrowers with a tenure plan tended to utilize the line just as much as borrowers without one. Approximately 2/3’s of tenure plan participants make frequent withdrawals, about equal to those without tenure plans.

    A key issue is whether borrowers will exhaust their LOC before mortgage termination. The average age of the borrowers in our data set was 78.6, with a weighted female-male life expectancy of about 10 years. Given 23% of the lines had been withdrawn in about 24 months, the average borrower will use up the line in about 8 years, two years short of termination.

    But, there is a clear dividing line between those spending down the line quickly and those not doing so. We estimate that approximately 50% will exhaust the line prior to termination and will therefore have a balance close to or exceeding the original principal limit.

Changes

Out of 31 observed borrowers, we picked up:

• 1 borrower changed their plan: adding a $3000 monthly term payment 7 months after closing
• An 85-year old borrower repaid $50,000 about 2 years after closing, effectively doubling their line of credit.

    While these are individual stories, our estimation would be that repayment and plan changes occur in no more than 5% of accounts.

Spending the money

    Our respondents indicated general areas where the RM money had gone. In order, the areas were (multiple answers allowed):

Evaluation

    Borrowers’ opinions of their RM are partially a function of their understanding of the RM. We recorded the following:

We could find no particular correlation between satisfaction level and line usage or other financial metrics. However, our sample was small.

Summary

    Lien payoffs and line of credit usage remain the most valued components of a reverse mortgage. However, frequent line of credit withdrawals, and the fact that they are, on average, so much larger than typical term/tenure payments, mean up to half of all borrowers will exhaust the line prior to termination. In those cases, they might have been more financially prudent to have selected the tenure option.

    Our analysis had insufficient data to prove (or disprove) a hypothesis that those who take a larger lump sum at closing tend over time to withdraw more from the line than those who don’t. In other words, either financial necessity or standard of living increases incent them to “prime the pump”, so to speak, toward extra spending.

    The diversity of borrower behavior is clear and makes a strong argument for additional reverse mortgage products. The HECM today is a “one size fits all” arrangement. It is flexible but most borrowers don’t use the flexibility. The low limits mean substantial borrowing power is lost and certain higher home value segments likely see less utility in the product.

    The withdrawal patterns imply up to half of borrowers will exhaust their lines prior to mortgage termination. This could have a serious impact on their life style, once they are accustomed to the line. However, we could not analyze this issue from our database of current borrowers.

    The general satisfaction of borrowers is also evident, as is the variety of expenditure purpose. Practical, but not “day-to-day” expenses seem to take priority. Items that could be deferred, especially repairs and taxes, are way ahead of some items that get more attention in the media, such as annuities or long-term care.

    The extremely heavy emphasis on paying off liens is not new news, of course, but does serve to make the point that, for many borrowers, a reverse mortgage is mainly a conversion of a forward mortgage. The main benefit is simply elimination of payments. Others have already proposed convertible forward mortgages and our research would seem to support that idea.

    All in all, reverse mortgages have a valued role to fill in the future. But the product needs expansion and diversification to completely meet the variegated needs and behavioral patterns of borrowers.

Diogo is a private investor in reverse mortgage companies and other financial services. His previous work experience includes TowerGroup, Ernst & Young, McKinsey, and Bank One. He was educated at MIT and Harvard Business School.








John Brodrick is the CEO and founder of Your Home for Life and brings 20+ years experience in the mortgage business. A graduate of Barrington College, John is a past president of the Massachusetts Mortgage Association.