The evolution of today’s underwriting, which is largely being shaped by what investors will literally buy as well as what will fit into the market’s developing regulatory framework, has been key in the wake of the downturn to the market’s future. What exactly this will look like remains uncertain, but based on a recent roundtable discussion Origination News organized last month with industry experts in New York for the Mortgage Bankers Association’s National Secondary Market Conference, it appears it will be a hybrid approach. This will likely involve a combination of manual and automated processes as players strive to produce enough data to show they have learned from the mistakes of the downturn while still operating with the efficiencies needed to survive in what continues to be a challenging market.
Some in the small group of mortgage market players and data providers brought together to provide a well-rounded view of the range of credit quality in the current market expressed frustration with the constraints of today’s market. While change has been clearly necessary, there is persistent concern about the continuing lack of a stable framework for it.
In particular, there was frustration with the developing definition of a “qualified residential mortgage,” which will determine which types of loans rulemakers will consider low risk enough to get a break from developing risk retention requirements for securitization. There also was the aforementioned debate over the degree to which underwriting can and should be manual, given the need to operate efficiently and competitively in the market.
Among other things, the manual portion of today’s underwriting seeks to incorporate an analysis of potential borrower’s “character” into decisions about whether loans should be made today. While this can be challenging to define and quantify, the group agreed that there appears to be some success in doing so when it comes to veterans, for example, as loans made through the government’s loan program in this area do have relatively strong performance statistics.
There also continues to be a great need in underwriting today for care in sizing up home values, which remain uncertain, but can differ locally by micromarket and type of home being sold.
Adding to the complexity involved in projecting home prices for underwriting decisions today is the effect of government involvement in the mortgage market and the extent to which it is likely to change in the future, when and how. The answer at press time seemed to be to incorporate and possibly model a view of the likelihood that government officials will proceed with their stated plans to withdraw their stimulus programs but do so at a pace that will unduly hurt the sensitive housing market,
Stephen M. Calk, chairman and chief executive officer of Chicago Bancorp and Generations Bank; Jon Daurio, former chairman and chief executive officer of Kondaur Capital Corp.; Martin Goodman, president of LoanMLS; and Avi Naider, chief executive officer, ACES Risk Management Corp., participated in the discussion organized by Origination News editor Bonnie Sinnock and moderated by Mark Fogarty, the SourceMedia Mortgage Publications Group’s editorial director and associate publisher.
DAURIO: ...The incredible influx of stated predatory lending laws are preventing lenders from being able to charge what is a reasonable rate or points commensurate with the risk. If you look historically, we are back where we were in the ’70s, where there was a vibrant “A” market and a vibrant hard-money market, “A” market meaning that if you had a lower downpayment with perfect credit scores, today you could still get 95%, back then it was 80% but back then hard money was 60% and today hard money’s really 50%. They say they do 60% but they are so hard on the appraisal of the property, it’s really 50%. So there is, just like at the birth of the B/C market, which is what, I mean “subprime” is a name that was attached to this flurry of hybrid and what I’ll call crazy products that came. Originally the market was called the B/C market because credit gradation was based on the alphabet. You had A, B, C, D credit and the credit score and the credit score was really an attempt at automated underwriting for that kind of risk gradation but then it got bastardized...But today we’ve got people that can—how many, I’m not sure—but there are people that have either greater than 95% equity in their houses or, even better, go back to the ’70s analogy, greater than 20% equity in their house or the ability to put greater than 20% down but they’ve got a 30-day late, or they’ve got a 60% just on consumer credit let alone on a mortgage credit and originators were they able to A, find a place to be able to park those loans in the secondary market, or B, be able to charge a commensurate points or rate or the combination of them both to assume the risk that you have with originating those types of loans even though you’re mitigating that risk with a higher LTV. For example, somebody with a 70% LTV or potentially a much lower debt ratio, contrary to the B/C lending of yesteryear where the worst credits were allowed higher debt-to-income ratios. Actually we might see in this round it’ll be the reverse that actually when you have worse credit you’re going to have to have a lower debt-to-income ratio. It’s hard to predict how long that’s really going to take for the market to be able to accept that. The great motivating factor is there’s a tremendous amount of private capital sitting on the side, desperate for yield. Therefore, if you can originate these kinds of products, there are so many states now and there’s more movement to restricting the ability to, regulating loan pricing irrespective to the risk, and pricing should really be reflective of the risk that the lender is assuming to mitigate the risk, not just debt ratios and equity in the home, but also by the compensation that you receive. Because the default ratios and loss severities of those products, especially in this economy, are going to be much greater where strategic default or you no longer have as much of a moral barrier to somebody defaulting based on the way the economy is today.
And secondly the way the economy is today, I’m a firm believer and I think there are many of us, that housing prices are continuing to deteriorate, that we don’t have stabilization in housing prices...it’s going to be, for lack of a better term, a return to historically “handwritten” underwriting as opposed to automated underwriting because...I think that what happened is that...one of the things that led us into some of the trouble we got was automated underwriting without a micro-analysis...If your LTV below 50%, even if you’re in an environment where housing prices are depreciating as they are today, you’re still safe. You have to deal with the normative issues though...Are you putting somebody into a house that they can afford?
CALK: The reality is we can’t return to completely manual underwritten industry for the simple reason that it’s just cost-prohibitive and until we get clarity on what QRM is and then build technology around that you’ll have no significant volume. They can’t scale without that clarity.
DAURIO: ...People talk about the three Cs...but people forget the fourth “C” which is character.
CALK: That level of decisioning that we had for years and years and manual underwriting...with FHA you saw that there’s a specific instance that was worth overlooking to do that loan, that character piece, that’s very hard to pick up with [automation].
NAIDER: It’s probably going to be a hybrid to some degree. To Steve’s point, you can’t return to a completely manual underwriting environment, it’s cost-prohibitive. On the other hand, when you think about what happened a few years ago, you’d have these automated underwriting systems, you pretty much put in the system whatever you wanted to put in...The mere fact that you were using technology wasn’t really the problem. What we’re seeing today for example on the prefunding side is you’re still using the automated underwriting systems but two things are very different. Everything is being verified, certainly with best practice lenders. For example, even the credit score is not necessarily being relied upon because we’re using undisclosed debt monitoring services. Similarly if you’ve got stated income that’s not going to be relied upon. We’re using third party verification tools to verified income, to verify Social Security, to verify assets, to verify all of those pieces of information on the 1003 but they are still technology tools. So what I think what we see emerging is you’ve got automated underwriting systems but in contrast to the past where red lights were ignore you are going to be using different pieces of technology to address those red flags.
DAURIO: Let me make a clarification. I’m not saying sole human underwriting. Absolutely automated underwriting on the front end but it’s a triage. There are situations where you get a human in there...or technology for the types of things like you said for verification purposes and on the back end in order to...be able to identify the higher risk loans that require a deeper human touch. So absolutely still technology...
FOGARTY: You’re talking about character and one of the places I see character coming in is in VA loans...
CALK: ...without question, [on VA loans] the numbers speak for themselves on their performance...
FOGARTY: That’s what I’m willing to attribute it to. It’s character. Anyway, somebody at the conference this morning was talking about that we need higher conventional interest rates to make nonconventional attractive to investors. Does that ring true for anybody? Why would that be?
GOODMAN: We need market interest rates and not artificial interest rates. That’s more important. The market would never keep them this low...
DAURIO: I agree. It’s the government that’s buying the securities that are out there and you want the free market to be free and that would allow interest rates to level out at whatever the market would bear...
GOODMAN: And they don’t want to because the housing markets going to come down a little bit, likely, if they let interest rates...adjust.
DAURIO: Do you allow [housing prices] to drop now and hopefully make this trough that we’re digging less shallow and not as long or do you just continue digging this trough that’s getting longer and lower?
GOODMAN: ...I’ve got to tell you we deal a lot in the micromarkets and we’re not seeing that because properties have hit the level our investors can pay cash...that is the bottom of the market. Now, upper-level homes that don’t track to rental values, I’m not really sure what’s going to happen value-wise with those but we’re hearing all over the country...there [are] battles for those properties.
DAURIO: From my perspective though I think that it’s a very tenuous floor that we’ve hit because people are relying on the 8% capitalization rate because they have assumptions about what default ratios and losses are going to be and that’s why I believe that the default ratios especially strategic defaults, they’re going to skyrocket for these particular properties...that’s going to lead to lower rents, which is going to lead to people heading out. So while you might see communities where there appears to be some stabilization, at least the way we’re approaching it is that while we think that the default ratios and loss severities have been realized, that’s what you are going to see. You have the second dip...whatever you want to call it.
NAIDER: I’m not sure where property values are going to go. I could see both sides of the debate because certainly in the South Florida market you can see what Marty’s describing. Nationally there is a double-dip but I think the key takeaway is with respect to your question mark. We’re operating in an artificial environment right now. With the government purchasing ultimately [around 90%]...that’s just not a sustainable environment, particularly when we are going to dismantle these GSEs over time.
