Why April Reverse Mortgage Numbers Don’t Tell the Full Story

Reverse mortgage endorsement volume for April as reported by the Department of Housing and Urban Development earlier this week indicated a month-over-month decline of roughly 45% — to just over 1,600 loans. But the numbers aren’t reflective of what’s happening in the market, several industry sources have shared with RMD, particularly at a time when many originators are seeing renewed interest among prospects as a result of the COVID-19 pandemic.

The difference could be due to a few different factors including longer endorsement timelines both on the part of lenders and the Department of Housing and Urban Development, industry participants say.

“The endorsements shown for our company, as I suspect for most if not all lenders, do not reflect the actual volume of loans currently being closed,” says All Reverse Mortgage CEO Mike Branson in an email to RMD.


All Reverse, a direct-endorsement lender based on Orange, Calif., which ranked No. 14 on the list of top reverse mortgage lenders by volume based on year-to-date loan counts tracked by Reverse Market Insight, saw only 35% of its loans closed in April as reported in the data. The company saw a similar but less severe discrepancy in the previous months as well, with February numbers off by around 25% and March numbers off by around 30%.

“We can only speculate why the endorsements are lagging but since they only reflect approximately 35% of our actual closings and we didn’t drop down to the bottom of the rankings for the top 100, we can see that it is an industry-wide issue at the present time,” Branson says. “I have not seen a confirmation of this in the form of a written announcement from HUD, but it appears that HUD has gotten behind in this area due to the COVID situation.”

Despite HUD’s ongoing endorsement efforts, a backlog of loans due to remote-working transitions and adoption of electronic case binder submissions to HUD is possible, the agency says. During the COVID-19 pandemic, HUD began allowing digital delivery of reverse mortgage case binders for the duration of the national state of emergency.

“It’s possible that some of the drop in volume is due to remote work transitions during the end of March and beginning of April, and the transition away from paper-based processes,” writes a senior HUD official in an email to RMD. “However, we continue to endorse case binders, including HECM files, through our new electronic process implemented through our April 6 Mortgagee Letter 2020-07. We are working closely with HECM lenders to ensure any backlogs of paper file submissions are being addressed.”

There’s also some volume decline to be expected as the industry continues to originate non-HECM reverse mortgages with several lenders having introduced new proprietary products and competitive product updates in recent months and years.

While proprietary volume is not reported by lenders, reverse mortgage technology company ReverseVision says it is an important part of the overall picture of the reverse mortgage origination landscape.

“While it may be noted that cndorsement volumes for FHA loans may have slowed, this may be more in terms of administrative challenges due to the global pandemic, and to the mix of Reverse Mortgages that are HECM vs. Private Loan products (PLP),” says Joe Langner, ReverseVision president. “The endorsement numbers do not include any representation for Private Loan Products and as they gain more popularity in the market there should a natural adjustment to the HECM endorsements…At ReverseVision we actually have seen solid growth in new proposals and applications with our customer base, in part due to the value of a reverse mortgage during times like these [with] lower interest rates, and the growth of the private loan products in the market.”

Companies featured in this article:

, ,

Join the Conversation (2)

see all

This is a professional community. Please use discretion when posting a comment.

  • One exec states: “’The endorsements shown for our company, as I suspect for most if not all lenders, do not reflect the actual volume of loans currently being closed….’” In further comments, he confirms this lack of knowledge of the HECM endorsement process. Most endorsed HECMs are closed within 50 days of receiving a case number assignment; yet it takes the average endorsed HECM 4 months to go from case number assignment to endorsement. HUD does not report on closed HECMs, neither does RMI; both report HECM endorsements.

    The endorsement process cannot begin until the related HECM has been closed. There are a small percentage of closed HECMs which are not endorsed due to undetected human errors in the HECM closing process but that are discovered in the endorsement process. BUT again it needs to be emphasized that such errors are very uncommon even though they are encountered each fiscal year. As to why the exec has not encountered the phenomenon of the gap between closing and endorsement before is not stated in the article.

    Even though HUD admits they have experienced Covid-19 related problems, the HUD response sounds rather optimistic when you consider that the size of the HECM endorsement volume for last month is so low that it has not been seen in 18 years. Could it be that the paper files for the closed HECMs that are waiting endorsements have yet to start the endorsement process? That seems likely. There is also little doubt that the switch to remote electronic endorsement processing has not been all that smooth or that all problems have been worked out. The HUD spokesperson did not mention how they are accounting for the files that are now being remotely processed.

    So let us look at the case number assignments for November 1, 2019 through January 31, 2020 (the group of three consecutive months of applications with case number assigned most likely to impact the endorsement volume for April 2020). Looking at just the difference between November 2019 and November 2018 case number assignments, November 2019 had 680 more. December 2019 had 975 more case number assignments than December 2018 and finally January 2020 had 723 more case number assignments than January 2019. So each of those three months had a minimum of 680 more case numbers being assigned in fiscal 2020 than in fiscal 2019. All that means is that the HECM endorsement count for last month should have been at least a few hundred more than the amount for April 2019 (of 2,901 endorsements).

    While I have no idea why proprietary reverse mortgages were brought into the conversation, it seems the overall tone of the article is on point since few can readily explain an almost 50% drop in HECM endorsements reported for April 2020 than what was expected.

  • Another way to analyze the situation is examining the pull through rate. Here are the modified and annualized pull through rates (taking into account a four month lag from case number assignment to endorsement) for the following fiscal years (ending September 30):

    2012 — 63.9%
    2013 — 64.3%
    2014 — 63.1%
    2015 — 59.5%
    2016 — 63.7%
    2017 — 67.7%
    2018 — 63.8% and
    2019 — 62.3%

    The average pull through rate for these 8 years is 63.5%. Amazingly the standard deviation over that 8 year period was 2.1%. Six fiscal years had a standard deviation of less than 1 from the average, one month had a standard deviation from the average of less than 2 but greater than 1, and the only other month had a standard deviation of 2. The lowest the pull through rate should be for the fiscal year 2020 is about 59.3%.

    If 1) we do the same thing but for that same time period by months rather than by fiscal years and 2) the pull through rate each month is based on a modified annualized determination using a trailing 12 month methodology, the average pull through rate is 64.2% with a standard deviation of 2.9%. What is surprising is that 71% of the 96 months have a standard deviation from the average of less than 1, followed by 23% of the months having a standard deviation from the average of over 1 but less than 2, followed by 5% of the months having a standard deviation from the average of more than 2 but less than 3 and the only other month had a standard deviation from the average in excess of 3 but was less than 4. So we should have expected that about 61.3% of the December 2019 applications receiving case number assignments in that month would end up endorsed for a total of 2,631 HECM endorsements for April 2020. Of course, with an actual endorsement count of just 1,601, most of us were caught off guard as that was the lowest HECM endorsement cout for any April in 18 years..

    Now we have to start looking to see if there were any unusual problems with the applications that received case number assignments in December 2019. While unknown, it is unlikely the problem lies there. Was there a new condition that would erode the decision of applicants to get the HECM during those four months. While unknown, it is unlikely that is the source.

    Based on the level of information we have at our disposal right now, it seems there is NO reason for the HECM count for April 2020 to be that low unless it is related to the impact of Covid-19 either at the lender level or at HUD. For NOW, it seems to be Covid-19 related and principally involving HUD.

    In my first comment, I was trying to follow the arguments made in the article. The second is more of a statical analysis of the situation, trying to see if we could find a reasonable pull through rate to estimate what the April 2020 HECM endorsement count should have been if the Covid-19 impact were not involved.

string(104) "https://reversemortgagedaily.com/2020/05/08/why-april-reverse-mortgage-numbers-dont-tell-the-full-story/"

Share your opinion

[wpli_login_link redirect="https://reversemortgagedaily.com/2020/05/08/why-april-reverse-mortgage-numbers-dont-tell-the-full-story/"]