Tom Showalter, Chief Analytics Officer of Digital Risk, joins the broadcast to discuss QM and origination defect requirements and potential impact.
Showalter, hints that the bifurcation of the mortgage origination business is eminent. Showalter reflected that a mortgage company’s business is understanding consumers needs and wants then developing credit, marketing and secondary policy that meets market expectations. Because of the compliance risk and costs, manufacturing the loan or the process of originating will most likely fall to a few large companies that have the ability and balance sheet to invest millions into technology, compliance and process. Showalter goes onto to discuss the GSEs, short and long term outlook, impact of origination defects to price and performance.
Welcome to Mortgage Markets Today, my name is Louis Amaya and today is February 5, 2014. According to some the new QM rules bring stability to the loan performance by discouraging risky loan products and assuring borrowers qualified for mortgages, poorly documented loans, loose underwriting criteria and irresponsible loan products seemed to be a thing of the past. However, lenders will be faced with creating processes that adhere to the new rules while dealing with the fact that the cost originate alone is increasing rapidly. Plain and safe comes at a price as compliance and loan requirements have doubled the cost to originate a loan. Joining me to discuss the new QM origination or what's more accurately defined as the manufacturing process is Tom Showalter, Chief Analytic Officer at Digital Risk. Digital Risk is the largest independent provider of risk, compliance and transaction management solutions for the financial services market, manufacturing defects up next on Mortgage Markets Today.
Thomas. Welcome back to the broadcast.
Thank you. Appreciate the time to be here.
Before we get into the questions, I like to have you define the manufacturing process as it relates to the origination. Assuming that origination process to manufacturing loans, so what's the better definition or what is the definition?
Sure. Lets talk about that--the manufacturing process and the two basic kinds of risk that come out of it. There is a process in which obviously a borrower makes an application. The originator, the lender seeks to collect a variety of different kinds of information about the borrower. He also seeks to collect an appraisal and other detail information about the property and then all of that to put through to some sort of underwriting process and the underwriter then reviews the loan and then decides if in fact the loan should be funded or not funded then that set of to committee for final funding. So, the process of going from the borrower application all the way through all of these processes of data gathering and analysis to underwriting and then funding, it's called the loan manufacturing process. Coming out of that process are two kinds of risk. One, were very familiar with and that's the credit risk, that's the likelihood that the loan will default and there is a long history of tools and analytical methods to determine credit risk. Then there is a new kind of risk. It's actually been around for a while but it was a material or recognized as a big deal until the meltdown and that's called manufacturing risk. A manufacturing risk is actually one key element--key data element defining the loan are either corrupted or misrepresented, so for example, the borrower's income isn't really the income. The borrower's debt income ratio is not really accurate. The loan to value ratio comparing the appraised value to the loan is not accurate and as you know, some such things as DTI, debt-to-income, LTV, loan-to-value are key elements in assessing credit risk, so you can see if the manufacturing risk is high and these key information elements are corrupted then the credit risk analysis is also subject to some problems if not a lot. Does that make sense?
Yes. So the risk is always been there but it has been heightened with the implementation of the new QM rules. Let's start with the GSEs. What are the GSEs doing about the manufacturing risks and defects and possibly what the ramifications are to them?
Okay. I'm gonna step back for just a second and talk about QM because it has one particular information element that is very crucial. It's called the ability to repay and it's determined by the debt to income ratio which is the key info element on a loan file, so they have a demarcation line of the DTI greater than 43% is considered not QM and below 43% is considered QM but due to manufacturing risk, many lenders are concerned that they might not truly know the true debt position of the borrower nor might--they might not truly be able to accurately represent a borrower's income, so that means that the DTI statistic while people think it's quite relevant and meaningful, is also in the hands of the lender somewhat questionable because they are not sure they got it all right, so as you go from that you then see what the GSEs are trying to do and that they are trying to make sure that the manufacturing defects in this example that is that the DTI is actually the DTI. These defects are minimized so that when GSEs gets the loan and they analyze the loan. They are better able to project the future cash flows, price the loans accurately and not underestimate the risk. So, what the GSEs are doing is they're starting to publish and they just done a list of defects and defects standards, so there is going to be, for example, a target defects rate for a lender's population, in other words you have to be below that defect rate for your population or Fannie will not buy your loans and in turn Freddie's issuing similar kinds of guidelines. So what's happening there is that not only are the GSE is issuing defects or any guidelines for their sellers, they are also issuing what they called governance guidelines. Now this is a new twist. A governing guidelines means, not only is the GSEs interested in the defect rate of a population of loans...
They are also interested in how well the lender is organized to manage to a specific defect rate and control defect. So that means they are not only holding the head of the QC team that the lender accountable for the defect rate calculation. They are actually going all the way up stream to the CEO and holding him or her accountable as well. So you see here, it's gone from just loan quality as good to here is a specific target or standard for long quality, to here is the way we're going to govern and manage and loan quality and then finally there is the applications of penalties, fines and other kinds of retribution should. The GSEs feel like a particular lender is not really following the program.
So, what tools then. I mean, if someone's originating loan and they determined a DTI they've documented it through pay stubs or whatever the process maybe, what kind of tools are the GSEs using to determine that that information is not accurate.
Excellent question. There is a bit of past present in here. Here is what's been going on here and what's about to go on. What's been going on in the past is the GSEs would review maybe 1% or 2% of all of the loans they received. When I say review, I mean they submit these loans usually in batches of 2000-2500 to a team of underwriters called the quality control experts and these underwriters review every detail in the loan files of those 2500 loans to determine if something is misrepresented, if the appraisal is somehow or not done correctly, if the income has been miscalculated and so forth. So they spend, well approximately three to four hours per loan reviewing these 2500 loans and they do it each month and the major lenders do this each month. So, you have this process where 2% of the loans are subject to a manual review by experts and 98% of the loans are not reviewed at all and so that is the current process. However, the GSEs are aware that the bad loans can be put into the other 98% and therefore slip through and get in to their secure ties portfolio and therefore and they are either financially liable or in some cases the investor might be, and so therefore that's not acceptable, so what they're doing is developing models of manufacturing defect so they use a model to efficiently screen a 100,000 loans and identified the likelihood of a defect in those loans and identify those with severe defects most likely for some aggressive evaluation. Now this process, I am just describing or the use of tool they create an adverse sample and then go review the adverse sample. That is not in effect yet, that is being discusses. What is in effect right not at the GSEs is the random sample of the 2000 or 2500 and then the other 98% or so of the loans just flow through.
What they are trying to do is get to the point where they have enough confidence in their models that they can then identify a defect rate for a population using a model and then do a partial random sample and a partial adverse samples, so that they can get a more robust understanding of the long quality of manufacturing defects in the loans and also identify a larger portion of the loans with defects, so they can do something to call a secret remedy which they go back to the lender and say can you fix this. If it's an income calculation, can you get us the right income, if it's an appraisal evaluation, it's adjusted properly and so fort? I hope I am making some sense.
Yeah, absolutely. So, how is a typical lender or seller positions for this new environment with the liability running personally up to the CEO, I would imagine the lenders are putting a lot of effort in getting prepared to be able to manage under these new rules and strict adherence and review process?
You are correct up to point. The very largest lenders are hiring mountains of lawyers and quality control and business analyst and so forth to make sure that they followed all the QM, all the guidelines, all the defect rules so that they can be sure that as they produce and manufacture 100,000 of loans that they are meeting the defect quality that they can govern according to the defect standards and that is entirely within their scope and capability, but as you go down stream to the smaller lenders, the mid size and the small ones, they are having difficulty affording the breath and death of staff necessary to go do this loan quality review and manage to a defect rate and meet all the--check all the boxes with respect to what turns out to be a very complicated process. So, back to what are the lenders doing, the big ones are investing in staff and possibly some technology. The small ones are struggling figuring out how they're going to take their two person compliance team and have it be smart enough to handle this wealth of breath and death of new regulations. I might add to that, the lender communities, especially the larger ones, are starting to explore manufacturing defects detection technology similar what I described the GSEs are doing that will enable them to take a more efficient look at a broader population of loans so that they can stay on top of the defect rate of their entire population and not be limited to the potential misinformation caused by ignoring 98% of the loans and focusing on 2%.
So, what can the smaller to mid size lender do to better protect themselves. I mean, is there anything or anything that tool or processes that they can look from?
There is two trends that are going on. One is some of those folks are seeking out tools that will help augment their process, so that they can go survey loans electronically, automatedly and therefore not have as much labor absorbed in the process of analyzing and meeting a defect rate goal and that's happening. Now, the uptake on that is somewhat slow but it is happening. Another trend I've been reading about but I haven't really thought anybody was doing it and that is some are thinking about going a non QM route. Issuing non-QM qualified mortgages, and selling them on the private market and therefore not going to the GSEs for those and so that's another trend that may happen as people are looking at technology they're not sure of, staffing requirement they can't afford but still wanting to be in a mortgage business but can't really meet QM standards, so that's another avenue. I haven't seen that robustly presented but I have heard some people talk about it.
Are there investors that take out for the non-QM products at this point or they kind of waiting to see how the risk falls and the playing field stand out?
Your last comments, right on. They are waiting and seeing. In a private label market, every year I hear, how it's going to break a 100 billion and it's going to be like the good old days and I think this year, 2013, I think we got to 35 billion which is better than what it was two years ago which was almost zero but it's nowhere near what it was not that long ago at 500 billion a year. So I think the private label market is waiting and I think they are waiting for a few things. First of all, they are waiting for some index that will enable them to understand just how much manufacturing defect or how much loan quality they're buying. They're waiting to make sure that the credit risk that they're buying is reasonable and well-estimated and I think they're also waiting to see how the US housing market will rebound and how it will save a life, because as you know in the end, it's the collateral and the house that's backing up these loans and if that starts to erode again then the private label people won't have too many options. There is fourth thing they are looking for but I'm not sure were it's gonna come from and that is the equivalent of the government back insurance only offered by private insurers. Will somebody give them a credit to fall swab type play so that when they buy a bunch of loans from the secondary market that they can buy insurance against default and manufacturing defect. That sort of like everybody has dreamed it's just that there is no vendors of that right now.
So, what are the ramifications of a defect, so resonate a loan that a lender thought it was meeting all the requirements, they sell it to Fannie and Freddie, they comeback and say, it's not, what happened at that point?
Sure. Typically, assuming the defect has to do with some of the significant material information items in a loan file such as incomes, such as valuation, such as LTV and DTI. What our analysis have shown is that infected loans have a higher probability of faulting somewhere around 90 basis points increase. They have a much higher loss severity, sometimes as much as 12% point higher loss severity and therefore the overall lost rate on infected loans are somewhere north of 70 basis point above what's projected. So what happens is the loan defaults sooner than expected, it cost most if it defaults and the overall loss is generated to higher than expected and so if there is a high proportion of defects in your population the riskiness of your loan population has gone up and you didn't know it.
What is the likely outcome in the near term and long term, specific to this process?
Well, I think what's going to happen partly because volumes are low and I think the GSEs and everybody in the industry is worried about, if you move too fast, change to many processes too quickly a mortgage volume will fall even lower, the housing industry will struggle even more and more and more unemployment will result as banks lay off more and more underwriters and other people. So, I have a feeling there is going to be a slow procession from the old rules to the new ones. So the GSEs will introduce their defect rate. I am expecting them to introduce defect rates that they think the industry can meet, given current infrastructure and as they proceed and as they get comfortable that the industries are meeting their current defect rates then they will start to ratchet down or tighten their defect rate requirements so for the sake of this metaphor if they are willing to tolerate 5% defects to start, it will then go to four and then go to three then go to two and so forth. But each time they lower the defect rate that means the industry has to up its quality infrastructure in order to keep pace and that's the issue that is probably most concerning to the GSEs and that how can the industry respond, especially the industry at large I'm sure they feel the few biggest lenders can handle this but I'm not so sure they feel a whole industry can. So I would expect the slow roll out, I would expect a lot of feedback from the GSEs to the lenders about where their quality problems are and why and that will require a little bit of an additional information infrastructure on the part of the GSEs getting back to, okay, here is the kinds of problems your loans have, not only here is your defect rate but here are the kinds of problems you have and here's what we think are coming from, so you miss to a lender, please go back and fix this stuff and we'll take a look at it next week, next month and so forth. So, I expect that kind of roll out but at some point in time, in order to open up the private label market.
In order to enable the government to reduce its insurance backing., they're gonna have to get to a very, very low defect rate and that's going to eventually prompt the industry to go from its current manual methods to something that are much higher tech that enable it to be much more efficient and thorough without necessarily having to crack a loan file every time they want to figure out the defect rate.
You know that one of the thought that I've had over the past few months, that the cost originate loan, obviously is, well probably doubled from the last two to three years just because of the compliant and the QA requirement that are now require and really sometimes, I think that a mortgage company really needs to subside what is their business. Is their business compliance or is their business origination loans and secondary. Do you think that eventually as the market matures and evolves that the underwriting and the QA and compliance piece eventually become outsource to two, three very large players that has a balance sheet to back up their work and have become complete experts and have the technology that dollars to manages at a lower cost and the mortgage company is really just focused on self-marketing and secondary, is that a scenario that you would see happening over _20:24_.
Absolutely, because I think you are touching on the point. We've kinda handed that at a minute ago. First of all, this becoming on top of the compliance issues and getting the defect technology down and managing to a specific defect rate, ensuring that you are doing that in avoidance of federal penalty. That's really, really hard. We're in that business and we know how much money and time it takes to get to the point where you can get that stuff done reasonably well and it's not by tomorrow and it's not just for 10 cents, it's an expensive time consuming proposition and we got the _21:01_ to prove it. So, I think there's going to be more and more large vendors--large banks and small ones that are saying, our business is really understanding what's consumers want to borrow money from us and how to set up our credit policy that makes sense for those folks but all of these process of manufacturing the loan, gathering the data, quality assuring the data, lending the borrower, ensuring that all laws have been followed and that the manufacturing defect rate as well as the credit risk is in touch with the lender's credit policy and manufacturing defect policy. I think that might turn to a few vendors that have invested in millions in the creation of the infrastructure necessary to manage to those goals. Because I honestly think that what's gonna be placed on the industry is just far more than it's really going to be able to respond too quickly, especially when you get out of the mega lender category, I think that is gonna almost impossible for some of the smaller mid size lenders to really about. Does that make sense?
Yeah, it does, I mean. Did you see a scenario where the GSEs actually have an approved vendor list to manage the manufacturing process that they would recommend using and maybe price a little bit better for?
They haven't said anything to me specifically, meet personally but stepping back and knowing their processes what they normally go through is when the industry starts to outsource, they start to have criteria for vendors who can do outsourcers, outsourcing like for example appraisal management companies, you just can't open up an AMC and start getting business from Wells Fargo, for example, you just cant do that. Why? Because Wells Fargo has a very involved AMC compliance and lending process which also have to do with capital ratios and other things that need to be approved by the GSEs. So you can see how all of that kind of plays a role where the GSEs get involved because in the end they're kind of bearing the ultimate risk of all of this stuff. So in some way, shape or form they will set a criteria, a standard or maybe some sort of bureaucratic set of regulations that will cause them to have to cause the lenders to have to get those folks that they're outsourcing their underwriting to and it maybe either directly, here are the 15 qualities you need in a verified outsource as sworn to by the GSEs or the GSEs will regulate the lenders aggressively who have outsourcers and they didn't have to perform to at least the same or better standards, either one of those who thinks is highly likely. What's not likely is the industry just gonna turn the other way, let this stuff get outsource and then hope they can catch the quality problems. I don't think that's going to be likely.
I agree. So how the manufacturing defects affect pricing and loan performance over the long and short term?
The pricing and this is not yet fully agreed upon by the industry but we have data and we've done analyses to show that manufacturing defects increased loss frequency, loss severity and overall loss. So if that's the case then the estimate of manufacturing defect has to go into the pricing of the loans, especially loans for the next 12 to 36 months after origination because that's when some of the problems are most likely to occur. That also happens to fall within the window when the GSEs had put back options, so if those things tend to materialize you will start to see the market become aware that vendors or lenders with high defect rates end up with more early term losses and therefore have to be priced differently and so hopefully, for the sake of everybody they'll be able to get down to the loan level to identify the loans with the problems as opposed to just raising everybody's price. But I think that's highly likely.
And so as far as the private labels securities what is that look like, is there anything that, I mean, will it be any advantage once the private investors step in from going -- stay in pretty platform with the private?
There will be an advantage in that, a number of advantages assuming the private label world can get our arm around its quality issues and the integrity of its information that has supplies with each loan and its manufacturing defect issues and its credit risk issues assuming those things are well-managed and I think the investors and the private label market aren't going to warm up until they are and at that point I think investors will participate in that market. I think they'll be seeking higher yield because the Fannie Freddie yields aren't terribly high, so if you go by the thesis that money goes where it's treated best and they can get treated well on a private label market with certain risk assurances, you will see money starting to flow there but the private label market has to have some infrastructure changes in order for it to be really safe for the average investors because right now as you can see from the meltdown, huge loss has occurred within that market and people are afraid and they need to be convinced and as you know, when people lose a lot of money the time it takes and the effort it takes to get them comfortable again is quite substantial.
What do you see as far as the mid level mortgage market--mid level firms, I should say. Do you see as a result of the changing environment a consolidation of smaller mortgage companies by the larger ones, smaller ones which is going away? What is that landscape look like?
Sure, sure well let me talk first about the wholesale play which is full of what we call mortgage brokers. They're starting to dwindle at a rapid rate as obviously they can't meet the compliance requirements and those that buy their loans can't be sure that all the compliance and QM and manufacturing defect and credit risk and all those issues have been fully bedded and managed and the buyer of those wholesale loans doesn't necessarily have the technology or the resources to completely bet every loan it gets. So the wholesale channel starting to shrink dramatically and so that means that the typically mortgage broker is getting few and far between. In regarding small lenders like community banks and so forth, I have interviewed a couple of community bank or friends buying it and one was the president and he said he was also the chief compliance officer and he spent about an hour to a day at it. Well, you can imagine given the new rules and regs he can't cover even 1/10 of the whole issue in and out or a day and he can't really afford a staff of five to 10 people to go to a forum, so they're going to be facing some economic problems, like I mentioned one out for them is to go another route and go the non-QM route with fix amount of the out of some of the legal issues, not all. The other thing is I think you're gonna see a quest of market share, so that you're going to see bigger lenders buying up smaller ones so that they can increase their market share and the cost per loan for the compliance and manufacturing defect rate detection stuff. The overhead for that is spread across more loans and therefore less per loan. So, I fully see an acquisition, consolidation trend and I also see it for another reason.
That the mortgage loan volume is not particularly high and so as long as it's in the doldrums, I think there's gonna be a lot of cause pressure for folks to consolidate so they can just maintain their margins if not boost them.
We're just about out of time, Thomas. One last question, I think you just touch on it right then but your forecast for origination volume for 2014, is it going to equal or lower or higher than 2013?
You know, I was very surprised that 2013 ended up in excess of 1.5 trillion and the purchase money sector rebounded a little faster than people thought and the _28:29_ did not deteriorated fast. I would be very happy, if we could stay at 1.5, 1.7 trillion for 2014 and not regress to 12-13 trillion I do not see an emergence of the $2 trillion level or anything close to that, so I am expecting to stay as we are, maybe a little worse but I'm not expecting to see a big, big increase right now. I realize housing prices have gone up a little and I realized that interest rates are still stable but what concerns me is we continue to see a median wage and a median income going down and we continue to see the labor participation rate going down which means there are fewer and fewer are very well-employed people and those are the people you need to buy a house and most of the buying now is going to be on the purchase money basis as a opposed to a refine basis, so mortgage lenders are looking for a purchase money candidate and let's face it, the number of fish in that pond is just not what it used to be.
Right, right. Well, Thomas great information, I appreciate your time on the broadcast and we'll definitely have you back on for the third time some time in the future. Thanks again.
Louis, congratulations, best to you. Thank you.
Alright. That's it for this broadcast. This year's attendance at the 5 Stars Government Form Is More Critical Than Ever on March 25, 2014. Leaders of the Housing and Mortgage Servicing industry will engage government officials and regulators and meaningful discussion on housing and mortgage policy. This show was brought to you by Iserve Companies. Please connect to us via Facebook, Twitter, LinkedIn and at the 5starradio.com as well as dsnews.com and we will talk to you next time.
Sorry we couldn't complete your registration. Please try again.
Please enter your email to finish creating your account.
Receive a personalized list of podcasts based on your preferences.