Back To a ‘Normal’ Market?

We’re starting to see higher risk coming into the market, notes Jonathan Glowacki of Milliman, as part of an analysis of the present and immediate future of the industry.

A recent MBA article noted that we’re seeing the lowest rate of mortgage delinquencies in 18 years, coupled with  strong employment.

That’s true, says Jonathan Glowacki, principal and consulting actuary for Milliman, among the premier firms in the risk management field — but there are some extenuating circumstances behind that low rate. And he expects to see it tick up a bit in the near future.

By and large, Glowacki sees risk coming back into the market, though not necessarily at a level to be overly worried about. “We are starting to know that we’re reverting to a more normal market, with a little bit higher risk coming in the door.”

Glowacki spoke with Mortgage Media’s SA Ibrahim about his take on the market as it stands now and in the immediate future, regarding mortgage delinquencies, home prices, the nature of today’s investors, and any red flags or clouds he sees on the horizon.


What’s behind the low delinquency rate?

Glowacki and his colleagues play close attention to mortgage delinquency and how the market is evolving — many of Milliman’s clients are involved in taking mortgage risk through reinsurance, credit risk transfer (CRT) exposures and guarantees through government agencies. He referred to the MBA report showing the delinquency rate at about 4 percent, down from 5 percent a year prior.

According to Glowacki, there are a couple of factors behind this:

  1. Natural disasters in late 2017 into 2018 (a time frame that includes hurricanes Harvey, Irma, Maria, Florence and Michael, as well as the California wildfires) resulted in a lot of mortgages becoming delinquent but then quickly getting paid off — thus not resulting in a credit event to any of the investors or the mortgage collateral guarantors.
  2. We’re coming down from a very high peak in the number of delinquencies that occurred during the mortgage crisis of the 2000s. “What we’re seeing is that those loans from the 2005/2008 origination years had very high delinquency rates — and we’re 10 years on now, and a lot of those loans have either paid off going through the foreclosure process, or just made it through,” Glowacki said. “And so the delinquency rates from that pool of loans is really starting to come down, and that’s putting some downward pressure too on where we are today.”
  3. Origination from 2010/11 through 2015 has been underwritten in a “very strong environment,” Glowacki said: “The credit quality of those loans is extremely high, the quality of loan production, and that’s really the diligence that lenders are placing on loans before originating them, has been very strong, and we’ve observed very low delinquency trends on originations.”


The home-price factor

Home prices tends to drive the difference between a low-default and high-default market, Glowacki noted. Referring to the scenario immediately preceding large crises in the mortgage market — the mid-2000s crisis, the 1990s S&L debacle, etc. — “there was a strong correlation in default incidents and in negative home price appreciation, or depreciation.”

While agreeing with Ibrahim that unemployment rates also are a significant factor — “I think in between the margins, unemployment does tend to drive a lot of delinquency rates” — Glowacki added that borrowers tend to sell the property and avoid foreclosure if they truly can’t make their payments. “When that default goes to resolution, whether or not it’s going to end in a foreclosure, or whether or not it’s going to be sold, and the property’s worth more than the loan amount, you tend to have the property be sold by the borrower.” he noted.

“Where we are now is we’ve seen very strong home price growth, and a lot of that has been driven by the tighter underwriting, cleaning out the borrowers — and then obviously people starting in and buying up homes after the crisis at pretty good prices, not discounts, but values for the homes, and then those have grown over time. So we’ve seen a lot of appreciation over the last couple of years and most of it, actually, has been generally in line.”

That raises the issue of affordability, of course. Glowacki noted that there are many areas where there is a disconnect, with home prices slightly above wage growth, a result of monetary policy and the low interest rate environment. That scenario isn’t exclusive to housing, he added: “A lot of assets are more expensive today than they were two or three years ago, and housing just happens to be one of them.”

It’s a widespread enough issue over enough types of assets that it generally doesn’t ring any bells, except in regions where the difference is significant, where wage growth far trails home-price growth. Among the larger disconnects: Colorado, which has seen large growth in economic activity since 2012, translating in home prices doubling or more between 2012 and 2018. Wages haven’t kept up, he said. Colorado is the region that sticks out in Milliman’s algorithms as having the most risk in the current environment, Glowacki noted: Other areas, like San Francisco, have experienced similar economic-activity growth but have some underlying demographic trends that serve as mitigating factors— there’s not a lot of new homes being built in San Francisco, he noted. North Dakota and Texas also had big price hikes driven by economic activity (in oil/natural gas production), but not as notable.


Red flags

Having noted a strong underwriting quality in the 2012-2015 period, Glowacki said it hasn’t remained at that level: “We have started to notice some credit drift in the types, the quality of borrowers in 2017-18.” The FICO scores are still relatively high; the average loan-to-value (LTV) ratios are still pretty good, thanks in part to policies implemented by FHA and the GSEs, such as limiting cash-out refis to 80-85 percent LTV. But “where we have seen a little bit opening up on the credit box, really, is debt-to-income ratios.”

In 2014, around 15 percent of loans had a DTI ratio above 45 percent, in which 45-plus percent of gross pay would go toward mortgage. “That’s a pretty large amount, to say half of my income goes to pay just where I stay and where I sleep, and the other half of it has to pay for all your other expenses,” Glowacki said. In 2018, the percentage of loans with a 45-plus DTI ratio had nearly doubled, to around 30 percent. That metric doesn’t show everything, he cautioned; there are some offsetting factors.

“However, what I can say is that we also do watch the development of delinquencies by origination year,” he said. “What I mean by that is how many loans are delinquent after four quarters of origination.” And that, he said, is about 150 percent higher for 2016 or 2018 loans than for 2013 — though he noted that starts from a low base, since 2013 had some of the lowest delinquency rates going back as far as 1990.

“We are starting to see those seasoning curves rise a little bit, and so we do note that there is risk coming back into the market,” Glowacki said. “It’s not at a point where we’re going to say ‘stay away from the market’ or that it’s gonna be a risky investment. It’s not there, by any means. However, we are starting to know that we are reverting back to a more normal market, with a little bit higher risk coming in the door.”


Speculative buying?

When asked by Ibrahim about the impact of speculative homebuyers betting on home prices continuing to go up with plans for re-sale, Glowacki said he doesn’t see that as a major factor. Granted, he noted, a lot of those types of loans are funded outside of agencies, or are portfolio loans at banks that still don’t have much transparency into that market.

But really, “I still see a lot of the demand for housing is going to be driven by purchase mortgages, particularly a lot of first-time home buyers, coming into the market for the first time” — many in the millennial demographic, who stayed out of the market for a while partly due to the mid-2000s crisis (and partly due to student loans). “There has been good research saying millennials are slower at buying homes than their counterparts of prior decades or generations. They are starting to come into the market,” Glowacki said.


Who are today’s investors?

So who, Ibrahim asked, are the investors today? How are they different from before, and what’s happening to the risk from the point the loan is originated?

“We’re seeing a little bit of activity in the non-agency market, so those are going to be non-conforming loans that are originated by different lenders, packaged in securities, and sold without a guarantee or without any wrap from Freddie or Fannie,” Glowacki said. “That market was a very large portion of the overall originations pre-crisis and post-crisis. It essentially went to zero. It’s come back a little bit, but predominantly in the form of jumbo prime loans. We’re seeing a little bit of talk of what’s being called expanded credit, but I think overall origination and volume is still light, maybe a couple billion dollars here or there for the year, but nothing as big as we saw pre-crisis.”

What has been grabbing attention, he said, was over $2 trillion of face amount being distributed to the market through credit risk transfer (CRT) securities. That’s taken two different routes, execution-wise, he noted: Participants buying a mortgage bond, collateralized by mortgages, and taking some of the credit risk; or participants pursuing reinsurance, in which capital gets invested in various types of “low-frequency, high-severity exposure” — natural catastrophe risk. Earthquakes, hurricanes, floods.

And these investors are very interested in the value at risk (VAR) levels and their probable maximum loss, Glowacki said: “They really put their foot down when it comes to the type of collateral that goes into their securities, and how much they need to offset the risk that they’re taking.”


Clouds on the horizon?

Credit risk has always been cyclical, Ibrahim noted, and it’s been a while without a down cycle. He asked where Glowacki sees the next cycle happening, and what clouds he may be seeing on the horizon.

“A big driver of what we call mortgage losses is going to be home prices, and we are seeing some pockets, as I mentioned, of a disconnect between changes in income and home prices,” Glowacki said. “This indicates a hot housing market which at some point … goes in a pendulum and is going to swing the other way.

“Now I don’t think it’s going to be near to the extent as the last crisis we saw,” he added, explaining, “Generally we’re going to see a continuation of the pendulum swinging the other way — a little bit looser credit, a little bit more affordability, products are going to be developed. Expanded credit, for what that’s worth, is going to continue to grow over the next couple of years, and that’s gonna result in — just the fact that you’re issuing high-risk credit at origination — increased delinquencies down the road.” It won’t happen this year, he said, but he forecasts that in the next one to two years that the 4 percent delinquency rate MBA reported is going to start ticking back up. “That’s because the loans that are performing so well now are gonna season, and we’re gonna transition into the newer books that are being originated today,” those of a somewhat lower quality.

He doesn’t anticipate a major impact in the housing market, though it won’t go unnoticed.

“It doesn’t seem to have hit the excesses of a bubble territory, if you will, but I do think that we’ll see some losses starting to accrue and develop over the next maybe two or three, four, five years in the mortgage market,” he said. “I believe it would be more of a mild recession, and the important part with that is who’s taking the mortgage risk, and right now it’s a relatively new community; a lot of reinsurers and investors that are new to this type of exposure — and how they’re going to react when they start to experience or see a rise in delinquencies and losses. That’s an unknown.”

It’s not the only wild card: “There’s also a very big unknown that obviously we haven’t talked about, and that’s what happens with Freddie and Fannie. That’s anyone’s guess. I really don’t have any opinion on that one.”


Line in the sand?

With that pendulum swinging back, Ibrahim noted that defaults are likely to rise — maybe a tiny move from lowest to very low but it will likely still be reported in the popular press as a huge hike. But does Glowacki see a line in the sand where defaults start approaching a certain level and people should start getting worried?

Not necessarily, Glowacki responded. Partly due to overreaction to price shifts: “We really think that there’s going to be a knee-jerk reaction when we start seeing an increase in delinquencies and losses. That would be people buying the bonds of CRT, or things like that. So I think there’s going to be a big shift in prices, and overreaction to prices. And we actually saw that in 2018, early 2018, with the hurricanes. There was a big reaction that the hurricanes are going to see losses to investors in mortgages, and the prices of those bonds fell pretty dramatically. They quickly recovered once it was realized that they’re not going to result in a loss and things are still going pretty well.”

Then there’s the intrinsic value — which he thinks will be fine in whatever the next cycle ends up being — “but where you would be a little bit concerned and where the industry would start to get concerned is when they start eating into the layers in which investors are participating. A lot of the deals, both non-agency and the CRT deals, have a certain level of credit enhancement at origination when they come out, and I think once that credit enhancement starts getting eaten away and goes through the equity tranches of the deal and maybe into some of the more subordinate, then you might get a little bit of a reaction from the industry in terms of: What do I do with my capital? Do I stay in this for the long run, or do I pull out and call it a day? At which case, that obviously then has its own implications in terms of funding and paying for the mortgages.”

It can become a self-repeating cycle, depending on how much and how quickly the losses come and grow, Glowacki said, adding that he doesn’t really see that happening in the next U.S. recession. “But I do think that we’ll have a change in the players and the attitudes after they start to get some losses.”


‘This is my home’

Even during the last mortgage crisis, Ibrahim noted, there were a lot of struggling home borrowers who were delinquent, but it never translated to losses – because they ultimately came out of delinquency. Many struggling borrowers still managed to pay their mortgages through the worst of the crisis. By and large, he asked, will people generally be loyal to their home?

It comes down to the value of the collateral, and to life circumstances, Glowacki responded. Most people who bought a home continued to pay their mortgage during the crisis. Many defaulted — say 30 percent of a pool — but they were the high-risk collateral, the non-agency types. That’s still some seven out of 10 still paying their mortgages. And there would be even fewer defaults on those loans through the GSEs or FHA, he said.

“The majority of people, I think, still have the mantra that ‘This is my home, and I was gonna pay $1,000 a month to live here, and if it’s worth less, I’m still going to pay $1,000 a month to live here,” he said. Major life events like divorce, long-term unemployment or serious illness can throw a wrench into that, of course.

“The stigma of going through foreclosure and walking away certainly probably is a little bit looser post-crisis than it was pre-crisis, but I think for the majority of borrowers, if you can afford to stay in your home, I think you generally will,” Glowacki said. “It’s the investors and it’s the ones that don’t have a choice that would just be more willing to walk away and leave the keys on the table.”