MGIC Investment Corporation (MTG) Q2 2020 Earnings Report, Transcript and Summary

Ladies and gentlemen, thank you for standing by, and welcome to the MGIC Investment Corporation Second Quarter Earnings Call. [Operator Instructions]. I'd now like to hand the conference over to your speaker today, Mr. Mike Zimmerman. Please go ahead.

Thank you. Good morning, and thank you for joining us this morning and for your interest in MGIC Investment Corporation. Joining me on the call today to discuss the results for the second quarter of year-end 2020 are Chief Executive Officer, Tim Mattke; and Chief Financial Officer, Nathan Colson. I want to remind all participants that our earnings release of this morning, which may be accessed on MGIC's website, which is located at mtg.mgic.com under Newsroom, includes additional information about the company's quarterly results that we will refer to during the call, it includes certain non-GAAP financial measures. We have posted on our website a presentation that contains information pertaining to our primary risk in force, in new insurance written and other information which we think you will find valuable. I also want to remind listeners that from time to time, we may post information about our underwriting guidelines and other presentations or corrections to past presentations on our website that investors and other interested parties may find valuable as well. During the course of this call, we may make comments about our expectations of the future. Actual results could differ materially from those contained in these forward-looking statements. Additional information about those factors, including COVID-19, that could cause actual results to differ materially from those discussed on the call are contained in the Form 8-K and 10-Q that was filed last night. If the company makes any forward-looking statements, we are not undertaking an obligation to update those statements in the future in light of subsequent developments. Further, no interested party should rely on the fact that such guidance or forward-looking statements are current at any time other than the time of this call or the issuance of the 8-K or 10-Q. At this time, I'd like to turn the call over to Tim.

Thanks, Mike, and good morning, everyone. I hope everyone who is listening is safe and well. I want to express my gratitude to and admiration for my fellow MGIC coworkers and their families. Their efforts day in and day out over the last several months to support our customers through local communities and fellow coworkers, while coping with their own unique circumstances brought about by the COVID-19 pandemic have been remarkable. So thank you. The safety and health of our coworkers and their families is a responsibility I do not take lightly. That is why we continue to operate in a remote work environment, while we provide critical support to the housing market, especially first-time homebuyers. As we navigate through this unusual period, we continue to execute on our business strategies with a goal to position our company to prosper over the long term. We strive to achieve that goal by, among other things, working with the GSEs and servicers on loss avoidance programs, offering competitive products and services to our customers and maintaining a sharp focus on the sources and uses of our capital. We think this is the best approach for all stakeholders and is particularly relevant as we manage through the current environment. I will kick off this call by spending a few minutes providing a high-level summary of our financial results for the second quarter and our current financial position. Then Nathan will cover some more details of the financial results. Finally, I will wrap up by discussing the state of housing finance reform and then open it up for questions. As for our financial results, GAAP net income for the quarter was $14 million. The decrease in net income from prior quarters primarily reflects the increase in loss reserves that we established in response to the material increase in new delinquent loans that were reported to us in the second quarter. Nathan will get into more details in a few minutes. During the second quarter, the volume of both purchase and refinance mortgage originations was very robust. The demand for single-family housing has been very resilient and seems to have actually increased despite the pandemic. Of course, the low interest rate environment makes refinancing very attractive for many borrowers. As a result, we wrote $28 billion of new insurance in the quarter and despite lower persistency on our existing books of business, our insurance in force increased by approximately 8% year-over-year. Refinanced transactions as a percent for monthly new business writing peaked at approximately 44% in April and May. That was back down to 33% in July as demand for purchase mortgages increased and refinanced transactions slowed a bit. The combination of our application data, lender reports and the MBA indices provided us with a reasonable visibility into NIW over the next several months. And while our current pipeline remains robust, there's considerably less certainty about mortgage origination levels or credit performance beyond the near term, given the uncertainty impact COVID-19 will have on economic conditions. As a result of that uncertainty, as further discussed in our risk factors and in the 10-Q, it is difficult to confidently forecast our future financial results and capital position. Therefore, we will not be providing any guidance about the potential path or outcomes for insurance in force growth or credit performance, but we'll continue to provide the market with monthly credit metrics. We entered this period of uncertainty with the book of business that had strong credit characteristics. In addition, we are supported by a balance sheet that has a low debt-to-capital ratio, $6.3 billion in cash and investments, contractual premium flow and a robust reinsurance program. Despite the increased number of loan delinquencies and the corresponding increase in minimum required assets required to be held under the private mortgage insurer eligibility requirements of the GSEs, or PMIERs, we estimate that at the end of June, our available assets exceeded the minimum required assets by $1.1 billion. In addition, our policy holder position was $2.9 billion in excess of the minimum state capital requirements. While delinquency notices received in the second quarter were materially higher than the first quarter, there were 38% fewer notices in June than in May. Approximately 67% of our June 30 delinquency inventory and 80% of June new delinquency notices were reported to us as a COVID-19-related forbearance plan. The delinquency rate ended the quarter at 6.35%. Although there remains much uncertainty about the potential impacts to credit performance in our business caused by this international emergency, notably the potential for higher incurred and ultimately higher pay losses, we are encouraged by the July new notice and cure activity. With that, let me turn it over to Nathan.

Thanks, Tim. I will spend a few minutes talking about the second quarter, and then we'll turn to some of the uncertainties that Tim mentioned. In the second quarter, we earned $14 million of net income or $0.04 per diluted share which compared to $168 million of net income or $0.46 per diluted share from the same period last year. The difference is almost entirely the result of higher losses incurred that are primarily COVID-19-related. Net premiums earned increased 1%, and the net premium yield declined to 42.7 basis points from 46.5 compared to the second quarter of 2019. Net premiums earned and the net premium yield have several components. The largest component is what we call the in-force portfolio yield, which reflects the premium rates in effect on our insurance in force. The biggest driver of the lower net premiums earned and premium yield in the quarter was the decrease in the profit commission on our quota share reinsurance transactions. With the increase in losses due to the increase in the new delinquencies in the second quarter, more losses were ceded to the reinsurers, which lowers our profit commission. While I'm on the topic of profit commission, I want to remind listeners that profit commission is calculated on an annual basis, and there is no clawback of profit commission earned in prior years. In the first quarter, we earned, what I would call, our normal level of profit commission as losses were very low as they had been for the previous several years. As a result of the material increase in ceded losses in the second quarter, our year-to-date profit commission as of June 30 was lower than the year-to-date profit commission as of March 31, which shows up in our financial results as a negative profit commission in the second quarter. Of course, we did receive the benefit of reduced losses as well. The amount of profit commission we earn in future periods will be primarily influenced by the amount of losses incurred that are ceded to the reinsurers. In addition to the profit commission, net premium earned and net premium yield were affected by the lower average premium rates on our insurance in force. Accelerated premiums from single premium policy cancellations had a favorable impact on net premium earned and the net premium yield, an increase from $11 million in the second quarter of 2019 to $33 million in the second quarter of 2020, reflecting the strong refinance market. Net losses incurred were $217 million compared to $22 million for the same period last year. In the second quarter of 2020, we received approximately 58,000 new delinquency notices compared to 13,000 in the same period last year. The estimated claim rate on new notices received in the second quarter of 2020 was approximately 7%. This estimate was influenced in part by the actual performance of delinquent loans, expectations for home price appreciation and the expected performance of borrowers that suffered a financial hardship as a result of COVID-19, and these loans have entered a forbearance plan. Of course, there remains a great deal of uncertainty about the ultimate loss performance of these delinquent loans. When we established loss reserves, we monitor the level of new notices received, the level of delinquencies carried, the uptake of forbearance plans and the current and expected economic activity. Then, using that data, we establish reserves that reflect our best estimate of the ultimate claim rate and claim amount or severity on both new and existing delinquencies. The ultimate claim rate represents the percentage of delinquent loans we expect to result in mortgage insurance claims and are net of expected cures, including cures due to successful loan workouts after a forbearance period is over. In the second quarter of 2020, our reestimation of reserves associated with previous delinquencies resulted in approximately $10 million of adverse loss reserve development, primarily attributable to an increase in expected severity. We also increased our incurred, but not reported reserve or IBNR by $31 million. IBNR reflects estimated losses from delinquencies occurring prior to the close of an accounting period on notices of delinquency not yet reported to us. To establish the IBNR reserve as of June 30, we estimated the number of loans whose borrowers had missed their June 1 payment, but that had not been reported to us as delinquent. Reflecting the low level of the delinquency inventory that existed in March and the foreclosure moratoriums enacted by the GSEs and others, the number of claims received in the quarter declined by nearly 60% from the same period last year. Primary paid claims declined 44% from $52 million to $29 million. Although most foreclosure moratoriums are set to expire August 31, we expect claim payments to remain modest over the next several quarters due to their effects and the effects of forbearance agreement that are in place. In addition to the effects of foreclosure moratoriums and forbearance agreements, we expect the impact of unemployment and economic uncertainty will cause the delinquency inventory to increase further, although we are encouraged by the July new notice and cure activity. We continue to diligently monitor net underwriting and other expenses. Before ceding commission, they totaled $59 million in the second quarter of 2020, which is flat to the same period last year, while writing substantially higher volumes of business. We previously reported that MGIC did not request or pay a dividend to the holding company in the second quarter. Future dividend payments from MGIC to the holding company will be determined on a quarterly basis in consultation with the Board and after considering any updated estimates about the length and severity of the economic impacts of COVID-19 on our business. We also asked the Wisconsin OCI not to object before MGIC pays dividends to the holding company. And until March 31 of 2021, we will also need to seek the GSE's approval before MGIC pays any dividends to our holding company. As previously disclosed, the holding company Board declared a cash dividend of $0.06 per share payable on August 28. Any future dividends will be determined in consultation with the Board. As of June 30, we had approximately $530 million of cash and investments at the holding company. Our next debt maturity is in approximately 3 years, and our interest expense is approximately $60 million per year, of which $12 million is paid to MGIC on the holding company debt that it owns. At quarter-end, our consolidated cash and investments totaled $6.3 billion, including the cash and investments at the holding company. Investment income was modestly lower year-over-year primarily as the larger investment portfolio was offset by lower yields. The consolidated investment portfolio had a mix of 81% taxable and 19% tax-exempt securities, a pretax yield of 2.8% and a duration of 4 years. Our investment portfolio had a net unrealized gain of $265 million at June 30, 2020; $83 million at March 31, 2020; and $147 million a year ago. At the end of the second quarter, our debt-to-total capital ratio was approximately 17%, and MGIC's available assets for PMIERs purposes totaled approximately $4.5 billion resulting in a $1.1 billion excess over the minimum required assets. As many of you know, the PMIERs generally required us to maintain significantly more minimum required assets for delinquent loans than for performing loans. The PMIERs required asset factors for delinquent loans are based on the number of mispayments and whether a claim has been received. PMIERs allows for these factors to be reduced by 70% under certain circumstances, including related to COVID-19. During the quarter, the GSEs clarified that for loans that become delinquent between March 1 and December 31, 2020, the 70% reduction is applicable for at least 3 months and longer if a forbearance plan is in place. As a result of our strong positive cash flow during the quarter, which increased our available assets, and the application of the 70% reduction of minimum required assets for COVID-19-related delinquencies, our PMIERs excess increased by nearly $100 million in the quarter. With that, let me turn it back to Tim.

Thanks, Nathan. Before moving to questions, let me address a few regulatory and political topics. During the quarter, the FHFA reproposed the GSE capital rule with comments due at the end of August and have continued the work of preparing the GSEs to exit the conservatorship at some point. What, if any, impact these potential changes have on the MI industry is not clear at this point. That said, we do believe that the FHFA appreciates the benefits of an insurance company structure to provide to the housing finance system relative to capital markets and other less regulated solutions. The FHFA is also continuing the review of all GSE activities. And in June, the CFPB proposed changes to the definition of qualified mortgage and the so-called GSE Patch. The revised definition would replace the borrower's debt-to-income ratio in the definition with a pricing threshold. Based on work that our trade association, USMI did, it is estimated that less than 10% of the private mortgage insurance market will be impacted by the proposed rule, if enacted as is. The 60-day comment period for the proposal ends September 8, and the proposed changes have a targeted effective date of April 2021. While other market options for credit enhancement are scarce or unavailable, our industry and our company continue to provide credit enhancement solutions to lenders, borrowers and the GSEs. While we are focused on prudent solutions in responses to the current environment, we continue to be actively engaged in discussions regarding housing finance policies. We continue to advocate for and remain optimistic that any changes will include the use of private capital, including private mortgage insurance. Long term, we remain encouraged about the future role that our company and the industry can play in housing finance. It continues to be difficult to gauge what actions may be taken and the timing of any such actions. Private mortgage insurance offers many solutions and a great value proposition for lenders and consumers to overcome the #1 barrier to homeownership, the down payment. We are navigating this period of uncertainty with a book of business that has strong underlying credit characteristics and to be supported by a balance sheet that has a low debt-to-capital ratio, an investment portfolio in excess of $6 billion, contractual premium flow and a robust reinsurance program. In closing, as I mentioned at the beginning of my remarks, in addition to the well-being of our employees, we are focused on: one, continuing to provide critical support to the current housing market; and two, positioning our company to prosper over the long term. I want to remind listeners that our company was founded in 1957. We have successfully navigated many economic cycles and have continually provided borrowers and lenders with affordable and prudent low down payment options. I'm confident that we have the right team in place to navigate through this period of uncertainty, and we'll continue to deliver the quality products and service our customers that have come to expect from MGIC. With that, operator, let's take questions.

[Operator Instructions]. Your first question comes from the line of Jack Micenko.

I wanted to ask about the July cures, very good trend there, I think, over 100% cure to the full. Do you seem to have a sense of what the total delinquency portfolio looks like from a forbearance standpoint? Can you -- do you have a sense of how much of those July cures or forbearance loans that either come out of forbearance or, I guess, as they come up [indiscernible] if you have any level of detail?

Yes. Jack, it's Nathan. I'll take that one. We did put out the new notices in cures, but are still kind of working through our review process on the forbearance-related information that we have as of the end of July. So don't have an update relative to the composition of new notices in cures relative to the forbearance at this time.

Okay. And then you said 7% claim rate assumption on the book drove the loss incurred number for this quarter. I think you were at 9% last quarter. So curious what if the model really changed? Obviously, forbearance, outcomes are probably going to be a lot different than true defaults. But just curious, was it the improvement that you saw sequentially through month-to-month of the quarter? Or was there something [indiscernible] to drive that change?

Yes. I think if you recall back to the end of the first quarter, still a lot of uncertainty about forbearance plans what the uptake would be and what the rules would be. I think a lot of that stuff has gotten more formalized through the second quarter here. So from -- I think from our perspective, economic conditions may have improved slightly. But really, it was the, I think, the uptake in forbearance and now the list of options that the borrowers in forbearance have, including the 12 payment deferral for COVID-19, at least, are related forbearance that we think that there's some benefit to that and ultimately, a lower claim rate on those. And given that 67% of the inventory and 80% of the June new notices were in forbearance that benefited the claim rate relative to last quarter.

Okay. If I could just sneak one more. You're down about 4 basis points on the in-force yield year-to-year. How are you thinking about sort of terminal velocity? Where -- when -- at what level do we sort of -- the new money yields and the old money will sort of begin to sort of net out. Is that -- at a low 40s? Or is that mid-4? How do we thinking about it?

Jack, it's Mike Zimmerman. So it's going to be of funds right off that profit commission and the ceded losses. I mean, that's the big driver in the quarter. When you look at the sequential decrease in yield, we get a nice -- put a reconciliation into the press release on that. But you can see that the -- even though we had accelerated single premiums because of the refis, that was more than offset by the increased ceded losses, which reduces our profit commission. So like a lot of other things, what we're looking out as to whats the impact the yield is going to be and does it bounce back, does it -- where does it go to, to your point. It's going to be a function of how these losses develop in this going forward. As far as new notices coming in, curers that might happen off the existing notices, all that's going to drive into it. So really that -- great deal of visibility, I would say long term, it's kind of where we thought this would be heading over time. Clearly, in the quarter, it was a big step down, but that's all a function of just of the accounting due to the losses.

Our next question comes from the line of Douglas Harter.

I guess, touching on Jack's last point. Can you talk about pricing during the second quarter and kind of how that trended compared to the increase you guys saw and talked about in the last earnings call?

Yes, it's Tim. I mean, I think from my perspective, we continue to be focused on making sure that our premium rates, obviously competitive, but prudently growing the insurance in force and long-term value. And we wrote a lot of volume this quarter. I think we felt really good about sort of the risk return dynamic on that, but I don't want to get into a lot of specifics related to competitive pricing.

Okay. And I guess, do you have a sense, I know it's not everyone's reported yet, but a sense as to kind of where -- how your NIW might have compared to the market? Do you think that, that you might have gained share or lost share?

I'd say, it's tough to know at this point. Again, I think we feel really good about the book of business that we wrote in the quarter. And the fact we grew our in-force portfolio. But until all the others report, I think, it's difficult to know exactly where we landed other than we're really happy with the business that we wrote.

Next question comes from the line of Geoffrey Dunn.

I wanted to break apart new notices a little bit with respect to reserving. I think you said 80% of notices were forbearance. And obviously, the remainder were not. 7% incidence, how does that break apart right now in your view of non-forbearance versus forbearance loans?

Jeff, it's Nathan. I mean, we didn't set separate factors for forbearance versus non forbearance loans. So I think we ended the last quarter at 9%, a little less forbearance there, obviously, lower, but with a very high forbearance rate. So don't -- just don't really think about it that way. It's more the forbearance becomes a qualitative factor in the decision-making process ultimately around setting claim rates. Although, I think as everyone is aware, it's subject to a lot of uncertainty right now, just given that most of what we're experiencing is unprecedented.

Yes, so what I'm trying to get a read on is when forbearance goes away, if we're facing kind of the current economic situation, is that a 10% incidence environment or something that's 12%, 13%? Any kind of qualitative comment around that?

Yes. I think forbearance plans are -- there's still uptake in plans, although it seems like from some of the broader market data that there's also a lot of people coming out of them or curing from forbearance. We should have better visibility on some of those things over the next couple of months. But and I think part of what the claim rate would be in the future on non-forbearance items will be largely dependent on the economic conditions at that point. And what the prospects are kind of going forward from that point. So I think it's somewhat difficult to say. I think directionally, absent forbearance plans, I think, our claim rate in the second quarter would have been higher. But I wouldn't want to guide to any specific number, especially not out into the future.

Next question comes from the line of Bose George.

You noted that 70% discount that's in place for PMIERs is in place until the end of the year. What happens after that? Or does that stay in place as long as COVID-related forbearance is being offered by the GSEs? Or is there sort of visibility on that?

Yes. Sure. This is Nathan. I think you think about it in 2 parts. The way the kind of clarified rule will work relative to capital under PMIERs is for all delinquencies that are in, really the balance of the year here, will be considered kind of COVID-related, at least for 3 months. So for the first 3 months of that delinquency, we will be able to use the 70% haircut. But for all items that are in a forbearance plan, we continue to get to the benefit of the 70% haircut, while that loan is in a forbearance plan, obviously, loans that are going into those kind of plans now, those plans might extend out into 2021. And they also clarified -- the GSEs data has also clarified that during, I'd say, a post forbearance workout period like a repayment period on a repayment plan or a trial period on a modification plan, we would continue to get the 70% haircut during the post-forbearance workout period as well.

Okay, great. That's helpful. And then just in terms of the ILNs, I was curious, do you guys see any value in some of the high attachment point ILNs that we've seen from a couple of competitors, so it looks sort of like statutory capital ILN structures? What are your thoughts on that?

Yes. I guess, regarding -- this is Nathan again. Regarding the ILN market, I think, it's really encouraging to see multiple MI companies be able to access that market. I think the terms are not as attractive in terms of the attachment points and pricing levels is what we saw pre-COVID, but I think that's to be expected. We continue to evaluate that market and would seek to do things if we thought it made sense. But the very high attaching structures, I think, you could think about those as being maybe for other purposes other than risk mitigation or PMIERs capital. So while, I think, there's some value there, that hasn't been our primary mode and just remind that we do have kind of a pretty large quota share program that is a little different than some of our competitors, who have used the ILN structures a little more in different ways than we have.

Okay. Let me just sneak in one more. The 2007 and '08 books where you've seen obviously a pretty decent pickup in the delinquencies. So what are the mark-to-market LTVs on those books?

Bose, this is Mike. First off, the delinquencies did pick up but significantly less than we saw in the '09 and forward book. I don't have that right at my fingertips here, but maybe during the course of the call, we'll come back and give you that mark-to-market LTV for those books.

Your next question comes from the line of Mark DeVries.

How should we think about the ILNs impacting incurred losses going forward. If you continue to have elevated new notices, is there a point at which kind of the incurred loss per new notice starts to go down as you start to breach some of the threshold delinquency levels that bring back your reinsurers on risk?

It's Nathan. No, that's exactly right. Right now, any losses incurred that we are experiencing on the loans that are covered by those ILN transactions are still within our retention players on those deals. So we are absorbing the losses on an accounting basis there. If it did get to the point where the cumulative losses incurred were above the attachment point, we would begin ceding those losses incurred to the ILNs.

Okay. And do you have a general sense of kind of what delinquency rate, we'd start to see that become a meaningful driver of the incurred number?

I don't have -- I think because those are closed pools and they've been running off very quickly, just given the relatively low persistency we've had, I'd have to kind of double check right now what the of effective attachment point. But I think what you're looking at is at some point out in the future, I would say it's -- they'd be fairly high. I don't know -- I don't have an exact number, but something above, say, 5% in terms of expected loss rates, I would expect, just given that those deals continue to delever as the loans prepay.

Okay, got it. And then just on the profit commission, I think, Nathan, you indicated that those are done on an annualized basis, and there's no clawback. But is that determined on incurred losses or paid claims?

It's on an incurred basis. So it's really on a kind of an accident year basis. So we have all of the notices associated with a given accident year. The ultimate losses associated with those becomes the loss number in that calculation.

Okay. And so to the extent to which there are favorable developments down the line, because maybe you assume too high of a claims rate on these new notices, there's no -- that you don't get that money back?

No, we would. Right now, we're assuming, obviously, very little, if any, of this has resulted in paid claims. So this is incurreds primarily just establishing reserves at this point. If those reserves turn out to be too low and losses are ultimately higher, then we would receive less profit commission if loss -- If reserves are too high and ultimate losses are lower, we would receive more profit commission. So it trues up, but really within the accident year construct. So 2020 is kind of a stand-alone year, but we'll continue to true it up based on any revised estimates of ultimate losses.

Okay. Got it. That makes sense.

Just want to circle back to, Bose, to answer your question on the mark-to-market LTV. So done at the CBSA level, right? So they give you the comp, right there. You're looking at below 70 on those 2 book years and probably for all the '08 and prior and quite frankly, for -- really as you look at that mark-to-market LTV at the CBSA level really from the '18 book back is probably at 80% or less than those real old books, significantly lower.

And your next question comes from the line of Adam Starr.

Housing prices have been quite firm, I guess, because of demographics and low interest rates and tight supply. What are you assuming for the loss per loan? And what happens if the holder of the mortgage gets all their money back? Do you get a recovery on the loan part of the originals reserve?

Sure. It's Mike here -- from here. So if there is no loss on the property, then the claim most likely would not be submitted to us or if it would be submitted to us, it would be what we would call $0 paid claim. But if the lender or the insured gets full recovery of what's owed them through the sale of the property by the borrower, then there's no claim for the mortgage insurance.

But what of its foreclosed and sold for more than the amount -- the uninsured balance of the loan?

So there's a few states, and I'd have to get the list of them where there's ability to go after and to get the technical term, or where you go to the borrower resources to see if there's any valid any assets that the borrower has. And there's only a few states that have that ability to kind of go back to the borrower. Quite frankly, what we've seen in our experience over the years of doing that, very little recovery comes from those borrowers because quite frankly, they don't have the asset levels.

What if the lender forecloses and the lender sells the property at a -- for more than the outstanding uninsured balance? Do you get back some of the claims that you had paid? Or do you not settle up the claim until the house is sold and hence only pay part of the insurance balance?

Yes. Title needs to transfer for us to pay the claim. So if the lender foreclosed and took title, then a claim would have been submitted, and we would evaluate whether or not there was a loss there or not. If the lender, though, at that time, if we assess whether -- because we have the right to acquire a property versus paying the claim. So we will be seeing the same things that, that lender would be seeing and say, oh, geez, the property values are up, they foreclosed. The borrower didn't sell the property to avoid the foreclosure. So we have the opportunity to acquire that and hold. And so that would then reduce our loss if we did pay one. So that's kind of the check and balance there. We have the right to acquire the property if it could be sold for more and reduce that loss.

Got you. So -- but when you set up your initial reserve on the default notice, you assume a total loss on the insured balance?

With the severity, so like this last quarter, it was right around 100%. But yes, we would assess what our severity would be. And if it's -- like in some of those loans in New York, New Jersey that were -- we paid claims that are delinquent 7 or 8 years. That's a very high severity. Now we're paying much more 120% or so. More recent claims coming in because of price appreciation, we probably see less than 100% severity. So we make those adjustments when we established the reserves.

Okay. But rising house prices do give you a lot of protection then on defaulted loans?

Absolutely.

Your next question comes from the line of Phil Stefano.

I wanted to go back to the 70% haircut in the capital requirements and just understand, I guess, 1 thing I've been struggling with. As the time in forbearance gets longer, is there an increase in the capital requirements that you have on that mortgage? And then the 70% haircut applies? Or are we locked into this 2- to 3-month kind of aging capital requirement over the duration of the forbearance program?

It's Nathan. You had it right in saying that the 70% haircut applies to the underlying PMIERs factor that is based on time and delinquency. So while the 70% haircut continues to apply, the factor by which it's haircutting continues to increase, as the loan ages in delinquencies. So the amount of capital required does go up as the item ages, just by -- not by the full amount of what would be indicated in PMIERs but by 70% less than that full amount.

And Phil, just as a quick reminder, right, that first step up moves to 55%. And when it goes to that 2 to 3 category. And then by the time it gets out to before claim, I think it goes to 85%. So the incremental cap -- yes, there is incremental capital, but the big step-up comes in that first move.

Yes. Part of my fear was in not understanding this, could there be a double whammy next summer, as you get the step-up in the aging and you also get the potential falling off of the haircut, but it doesn't seem like that's the case. Or at least I don't want to risk getting ahead of myself, right? But no, I think that makes sense. When you had talked about gross expenses in the quarter of $59 million were flat year-over-year. Is there anything you can point us to from a benefit of people just being stuck at home and your salespeople can't be out on the road, doing what they do as a salesperson? Was there any expense benefit to this shelter in place world that we live in?

This is Tim. I would say, for us, it's modest. I mean, there are some savings there, obviously, from a travel and marketing expense. But I would say, for us, our expectation is that, that's not significant.

Got it. Okay. And then going back to one question that was earlier on the profit commission. I think that it makes sense to me that it's done in a year-to-date basis. I think part of what the question I was trying to get at was, if there's a huge favorable development number that -- not asking for guidance, but if there is a big favorable development number that comes next year on the reserves for this year, that wouldn't impact the profit commission in either 2020 or 2021, right?

So again, the -- you think about the years as accident -- this is Nathan. Think about the years as accident years. So say the new notices for the second quarter of 2020, those are going to be associated with the 2020 accident year. And as we continue to update our estimates of ultimate losses, that will flow through to the updated estimates of ceded losses, which will impact the profit commission. So there's like a -- there is a true-up, but it's contained within the accident year. So losses next year don't impact the profit commission for 2020. But any development on the notices that we have received in 2020, where we will continue to true-up both ceded losses and the profit commission.

Okay. Maybe to ask it differently, we've seen adverse development this year, which is potentially on the 2019 reserve losses. Did the 2019 profit commission change at all based on the adverse development that we saw this year?

I'd have to look at -- I don't know exactly what that adverse development is attributable to in terms of book years, but there weren't a lot of losses in 2019. But if you did assume that, that adverse development was associated with 2019 and that the loans that were -- it was associated with were part of the reinsurance, the quota share deal, then yes, the profit commission would have gone down by the amount by which ceded losses went up. That's most of that adverse development was on kind of more legacy type loans that are less likely to be in our reinsurance deal or covered at a lower percent. So my '19 example there was just illustrative not that we think that the development was coming from 2019.

No, no, I'm just mechanically wanted to make sure I understand it, and that's perfect.

And your next question comes from the line of Alex Klipper.

Can you give us any guidance on what your end of July delinquency rate was versus the 6.35%? I know things went down, and it looks like it came down meaningfully. But one of your competitors' reports actual delinquency rates. And so I'm just trying to triangulate what that number is.

Mike Zimmerman here. It probably ticked down maybe a few basis points because the inventory just declined maybe about 1,000 or so units. The number of loans outstanding, it's just one more month. So that went up. So maybe we dropped down a tenth or so. But I'll -- maybe we'll try and get that inch up around $6.2 million, which would be my guesstimate of that number.

Our next question comes from the line of Mihir Bhatia.

Firstly, I just want to start with the COVID related forbearances and delinquencies. I appreciate that 67% of delinquent loans are related to COVID-19 forbearance. But is there a statistic you can share going the other way that is what percent of your forbearance loans are already reported delinquent? And so like what could potentially be coming if they miss another payment, et cetera?

Yes, it's Nathan. I think we've -- typically, our reporting is more robust for delinquent loans than for performing loans. And while we do get forbearance reporting from the GSEs and from certain servicers on performing loans, it's not as robust as the information we get for nonperforming loans. So I think we're a little cautious that I think we would have maybe a good number for one side of that statistic, but not jump -- not feel so strongly about the other number, just given that our reporting on performing loans typically isn't at the same level as it is on nonperforming loans.

Understood. No, I appreciate that. So I guess, I think, you mentioned you expect delinquencies to increase as we go into 2H, into the second half of the year. And I was just wondering, is there something you're seeing in your data that's telling you that? Or is it just more given the economic environment and the high unemployment rates, et cetera?

Yes, it's Nathan. I do think the July results were -- I think we're certainly encouraged by that, but still feel like there is a lot of economic uncertainty at this point. And there also seems to be some items that entered forbearance that maybe didn't either want to be in forbearance or at least anecdotally, or are curing to refinance. These are things that we've heard. So obviously, the ultimate -- the peak, let's say, of delinquent inventory is quite uncertain. I think we were certainly thinking about it in July as being still trending upward from where we were, but seeing the results for July, I just feel like that's certainly favorable to what we were thinking.

Understood. And just one final question. I just want to make sure we get this right, on the PMIERs and the 70% discount. If a loan is in COVID-related forbearance. The 70% discount will apply throughout the COVID-related forbearance period. Is that correct? I'm not talking of like just the first 3 months that you get [indiscernible] if it's actually a COVID-related forbearance plan. You will get the discount for the entire COVID-related forbearance. Is that right?

That's correct. For that loan, as long as it remained in a forbearance plan, we would continue to get the 70% haircut. And if it were in a post forbearance workout, such as the repayment plan, or a trial period on a modification, we would continue to get the 70% haircut during the post forbearance period as well.

[Operator Instructions]. No further questions from the phone line. Presenters, you may continue.

Okay. This is Tim, again, just want to thank everyone for their interest, and I hope everyone is able to stay healthy and safe out there. Thank you again. Have a great day.

Ladies and gentlemen, this concludes today's conference call. Thank you for participating. You may now disconnect.