PennyMac Mortgage Investment Trust (PMT) Earnings Call Transcript & Summary

Good afternoon and thank you for joining us today. I'm Barclays' consumer finance analyst, Mark DeVries, and I'm pleased to be joined by PennyMac Mortgage Investment Trust CEO, David Spector; and CFO, Andy Chang. They have some prepared comments to deliver, which will be followed, time permitting, by some polling of the audience and Q&A for management. [Operator Instructions] Before we start off with the prepared remarks, I'd like to lead off with a question for the audience. To participate, please click through to the polls on the left side of your screen. After you respond, you should be able to toggle back to the video of the discussion. Turning to that first question, what do you view as the biggest catalyst for PMT over the next year? Growth in MSR, excess spread portfolio, increasing investments in CRTs and other credit assets, better-than-expected mortgage originations, capital returns or other? With that out of the way, I'm going to hand it off to David for his comments. David, take it away.

Thank you, Mark. Good afternoon, everyone, and thank you for your interest in PennyMac Mortgage Investment Trust. I want to begin my comments with a brief overview of PMT's business model as well as its historical performance. PMT is a real estate investment trust that went public in 2009 with a sole focus on the U.S. residential mortgage market. Leveraging the operational capability of PFSI's more than 5,300 employees, PMT remains focused on the growth of its industry-leading conventional conforming loan production and the resulting investments. For 11 years now, PMT has followed a disciplined approach to risk management, which includes interest rate risk, credit risk and operational risk. PMT's strong balance sheet, supported by innovative and diverse financing structures, notably term CRT financing without mark-to-market exposure, enabled us to maintain substantial liquidity throughout the crisis with no asset sales to raise liquidity. We also have a strong track record of successfully managing capital. These activities include repurchases of approximately 16 million shares of PMT's common stock when it traded well below book value, and we also have issued new common stock when PMT traded at a premium to book value, issuing over 39 million shares through underwritten offerings and PMT's At The Market equity offering program over the last 2 years. Now let's turn to Slide 4 to review PMT's strong capital base and returns compared to other REITs. Book value per share declined from $21.37 at December 31, 2019, to $15.16 at March 31, 2020, as a result of fair value losses on PMT's CRT investments. However, PMT's book value per share has largely recovered to $19.39 at June 30, 2020. The recovery in book value was enabled by PMT's innovative financing structures for its MSR and CRT investments, along with its disciplined approach to interest rate risk management. We have grown PMT shareholders' equity some $315 million from the time of the IPO to over $2.2 billion today. And we expect PMT's book value to grow further as the earnings potential of PMT's investment strategies exceeds the current dividend level. Over our history, returns to shareholders have been extremely strong and have exceeded those of most other residential mortgage REITs and comparable indices. Now let's turn to Slide 5 and review PMT's 3 business segments. PMT's primary investments in MSRs and CRT are sourced organically from the conventional conforming loans we aggregate through our correspondent channel. We are a leader in correspondent production. PMT entered this channel in 2011 after large banks, which had historically dominated this channel, reduced their participation. As will be discussed in the following slide, there is significant opportunity in the current market environment. PMT retains the mortgage servicing rights, or MSRs, on all of its conventional conforming correspondent production. These MSRs represent the right to service the underlying mortgage loans for which PMT earns servicing and other ancillary fees. We own MSRs as part of a comprehensive interest rate risk management strategy that also includes investments in excess servicing spread, mortgage-backed securities and other interest rate hedges. Our CRT investments represent the upfront credit risk on the high-quality loan production acquired through our correspondent production channel and were delivered by PMT to Fannie Mae. The investment receives an interest-only strip, which results in spread income on the investment. As previously disclosed, PMT is winding down new investments in GSE CRT, having begun organically investing in front-end CRT back in 2015. At June 30, the unpaid principal balance, or UPB, of the loans underlying PMT's CRT investments was approximately $81 billion. Now let's turn to Slide 6 and discuss PMT's opportunity in the mortgage origination market. Economic forecasts for total originations in 2020 have increased to $3.1 trillion, the highest level since 2003. And forecasts for total originations in 2021 have recently increased to nearly $2.4 trillion, higher than the strong market we saw in 2019. These forecasts are supported by mortgage rates that continue to reach all-time lows, driving robust refinance and purchase mortgage demand. Forecasts for purchase mortgage originations have also increased recently as a result of higher demand, including in suburban areas. While margins remain elevated in direct origination channels, conventional correspondent margins have returned to margins seen pre-COVID as other market participants have returned amid less uncertainty. PennyMac is able to capitalize on the market opportunity ahead as a result of its capital structure, risk management disciplines and significant technology and infrastructure investments made by PFSI in recent years. Now let's turn to Slide 7 and discuss PMT's leadership position in the conventional conforming market. PMT has established itself as a leading producer in the conventional conforming production market. For the second quarter of 2020, PMT was the largest overall correspondent lender and fourth-largest producer of conventional conforming mortgages in the United States. PMT has been the largest correspondent producer of residential mortgage loans over the last year. PMT has generated record conventional conforming lock volumes through August 31, 2020, and production is more than 90% of the total for 2019. These record production volumes are driving PMT's ability to grow its MSR portfolio in a low interest rate environment. So now let's turn to Slide 8 and look at capital deployed to MSR and trends within the investment strategy. Investments in MSRs have accelerated along with PMT's increase in conventional conforming production volumes, in turn growing the UPB underlying PMT's MSR portfolio. At June 30, the UPB of MSRs underlying PMT's investments was $145 billion. As you can see on the top right, we created almost $850 million in new conventional MSRs in 2019. And with record production levels in 2020, we expect to maintain this strong level of capital deployment. The growing MSR portfolio drives increased servicing fees and income, excluding market-driven value changes in PMT's Interest Rate Sensitive Strategies segment. The decline in MSR fair value from 6 months ago is driven by elevated prepayment activity as a result of the significant decline in interest rates during that period and has been partially offset by new investment growth. Now I'd like to turn to Page 9 and take a few minutes to discuss COVID forbearances. 59% of forbearance plans that were in effect for PMT's servicing portfolio as of April 30 have been extended. As you can see on this slide, the percentage of loans in forbearance within PMT's MSR portfolio decreased to 5.2% at July 31 from 7.3% at April 30. A 13% increase in new forbearance plans added since April 30 was more than offset by 39% of borrowers in forbearance plans as of April 30 who have exited. The 39% of borrowers who exited forbearance were comprised of 31% that remain current or went delinquent and subsequently reperformed, 4% that are transitioning to active loss mitigation primarily for payment deferral and streamline modifications and 4% that repaid their mortgage. PMT's earliest CRT investments with scheduled severities will incur losses if the underlying loans migrate to 180 days past due. Forbearances related to loans underlying these investments have declined with rates consistent with PMT's MSR portfolio overall. Such cures out of forbearance related to these CRT transactions reduce the loss realization potential. Now let's turn to Slide 10 and discuss PMT's investments in GSE credit risk transfer. As I previously mentioned, PMT is currently winding down new investments in front-end CRT. And in the second quarter, total deliveries were $1.8 billion in UPB. PennyMac's strategy of owning the servicing relationship, along with the credit risk on the loans it produces, creates a strong alignment of interest and is instrumental in mitigating losses. We will discuss this in more detail on the next slide. Lifetime cumulative losses on these investments through June 30 were $13 million. However, we expect realized losses to increase as defaults related to COVID-19 hardships materialize. Now let's turn to Slide 11 and discuss the outlook for returns for PMT's CRT investments. PMT's CRT investments recorded $453 million in fair value gains in the second quarter, representing a significant recovery of the fair value losses in the first quarter. PMT's role as an investor in credit risk on the loans it acquires and PFSI's role as the servicer creates a strong alignment of interest in assisting distressed borrowers as we can work with these customers directly to find solutions to address their hardships and ultimately prevent losses. Since CRT investments are currently held at a discount to par, PMT avoids any actual losses and expects to recognize fair value gains on its CRT investments when the underlying loans pay off since principal is received at par. Or for CRT-6, the commitment amount is reduced. The majority of realized losses will ultimately be determined by credit fundamentals and long-term performance of the underlying loans. The credit profiles of borrowers with loans in PMT's CRT investments are especially strong, with weighted average FICO scores of over 750 and weighted average LTVs of 83% at originations. PMT's earliest CRT transactions, which represent 7% of the underlying unpaid principal balance, will realize losses if the loan underlying these investments becomes 180 days or more delinquent regardless of COVID-related forbearance plans. At June 30, these investments had a face amount of $292 million and a fair value of $271 million. If all presently delinquent loans proceeded to 180 days delinquent, unmitigated, scheduled losses on these earliest transactions would total approximately $90 million. Programs introduced by the agencies in response to COVID-19, combined with PMT's own loss mitigation efforts, are expected to substantially mitigate credit losses. As a result, our outlook for expected returns on the CRT investments remains largely unchanged since June 30. Now let's turn to Slide 12 and discuss the run rate potential for PMT's investment strategies. PMT's run rate potential represents the average annualized return and quarterly earnings potential PMT expects to earn on average each quarter from its strategies over the next 4 quarters. Our expectation for PMT's investment strategies is an average diluted EPS per quarter of $0.65 per share, which will result in an annualized return on common equity of approximately 13%. PMT's equity allocation related to Credit Sensitive Strategies is expected to average 37% with a run rate annualized potential return on equity of 16.1%. Our Credit Sensitive Strategies primarily consist of our investments -- of investments in CRT with a return potential that has decreased from previous estimates due to the increase in fair value in the second quarter, which decreased our go-forward expected rates of return. The CRT markets continue to reflect uncertainty related to COVID-19 and its associated losses. Equity allocated to Interest Rate Sensitive Strategies is expected to average 33% with an annualized return on equity of 11.1%. We consider the results in this segment in aggregate as mortgage-backed securities and hedge positions are primarily used to moderate the impact of interest rate volatility on MSR and ESS returns. Our expectations reflect higher return potential on MSRs driven by a reduction in the expected impact on servicing from COVID-19. Equity allocated to correspondent production segment is expected to average 14% with an annualized return on equity of approximately 38%. This is lower than prior projections as conventional margins have returned to normalized levels faster than originally anticipated. That concludes my prepared remarks. Thank you for your time, and we'll now open it up for questions. Mark?

Yes. But before we get to the questions, I have another question for the audience. If you would register your response. Next question is, what is the biggest risk to shares that you see here? Credit risk from distressed loan portfolio, interest rate volatility, weak market for mortgage originations, incremental regulatory scrutiny or other?

Turning to a list of questions for management. Can you discuss how prevailing macro factors affect PMT's business? And share any outlook you may have around those.

So, I mean, obviously, with where interest rates are at the moment, the heightened origination market that's right now running at an expectation of $3.3 trillion for this year clearly as an effect from the point of view that we're seeing more and more opportunity to buy conventional conforming production through the correspondent network. We've seen very nice growth in the correspondent channel, and having that increased volume gives PMT the opportunity to redeploy its capital and redeploy that's coming off of the CRT investments. So I would say that it's allowing us to redeploy capital in a very efficient manner and to do so at very good returns, as I pointed out. I think the other macro issue is coming out of COVID, we saw a period, albeit it was only a few months, where we saw margins increase rather dramatically in the conventional correspondent business. And PMT, being a leading aggregator, was able to participate in those increased margins in the second quarter. I would say that since that time, we've seen margins return to their pre-COVID levels. But still, I would point out, as we -- as I just mentioned, we think on a run rate basis, the annualized return on equity is still at 38% for the conventional correspondent business. And I think this macro story of low rates for at least through '21 benefits PMT. And on the converse side, the CRT investment gets reduced as prepayments increase, and -- but the cash flows that come off of those investments can be rather rapidly redeployed into investments in service side. So I think those are the -- I think it's a nice -- it's been a transition, as we're not creating new CRT investments, that I think has been helped by the fact that we can just invest more and more in servicing, and that's something that we're looking to do.

Yes. The other thing, Mark, with the macro environment that I think candidly we're still continuing to watch how it evolves is the ultimate change in realized losses on those CRT investments, right? We look at the credit profile of the borrowers in the CRT investments, as David pointed out in the presentation, that incredibly strong. We started delivering in 2015. And at origination, we're talking about average FICO scores over 750, LTVs at origination of 83%. But you can imagine, especially for those CRT investments that have seasoned, there's been a good deal of home price appreciation since then that's brought the kind of actual mark-to-market LTV down from the 83% at origination. And so we see some level of delinquency right now, a lot of it with forbearance. And as David went over the stats, it's still sort of early days in seeing people come out of forbearance because borrowers are entitled to up to 6 months of forbearance at the start through the CARES Act. So our expectation and I think the market's expectation is that ultimately, most of those that are delinquent and in forbearance should ultimately reperform through primarily the use of the payment deferral program that the GSEs have as well as loan modifications. But clearly, it's still somewhat early days in seeing how those loans come out. And to your question about the macro environment, I think we have to see how the -- what happens with the broader macro economy and how that plays into these very good, higher-end, creditworthy borrowers that are in our CRT investments.

Okay. That's helpful. Do you have any sense, I mean, you alluded to this, Andy, of if you looked at those loans on a mark-to-market LTV, where that would be? And also, just talk to how much lower the risk of ultimate default becomes when borrowers actually have real equity embedded in their loan.

Yes. I mean that's something that we track. We don't put that out in terms of what the mark-to-market LTV is. But, I mean, I think it's fair to say that again, on average, they were 83% LTV. And people have been getting, call it, 4% home price appreciation per year, and it's been -- they've been in -- it's been several years since the loan was originated. If you're talking about a borrower who's got 25, 30 points of equity, that's a -- and you think about having some disruption or -- in their payment ability because of COVID for however -- whatever reason, that 25, 30 points of equity is a meaningful differentiator in terms of their commitment to staying in the home and kind of how they address that hardship. Interestingly enough, one of the things that I think we see is for those who do truly have hardship and fit this kind of borrower profile is with the limited supply in the housing market, they have the ability to sell their homes, right, and to be able to -- and that -- so that's a tool that's available to them, again, because of the significant equity that they have in their homes. So as I said before, I think it's still somewhat early days to see how this plays out. We're just a few months in still from when the COVID crisis started and with the forbearance that the CARES Act allows the borrowers to have. But ultimately, it's part of what gives us a lot of conviction that the vast majority of these borrowers should perform.

Okay. This may be a tough question, but do you have a feel for what the returns could ultimately look like on your CRT business if losses ultimately come in closer to what you may believe as realistic here given some of all these mitigating factors like access to forbearance programs, embedded equity that really reduces risk of default? Could it be materially higher than where you're kind of assuming here based on where your portfolio is marked?

Yes. So we've given some of those sensitivities and scenarios previously in terms of there's a base case loss expectation that is embedded in the marks and that we've disclosed. And we did a presentation back in May where we showed, okay, what if the losses are 2x or 3x that? What does that do to returns? And our base case is what David went over in that last slide. We expect -- the CRT investments are marked at a discount. The discount rate is wider, which I think reflects some of the uncertainty around what ultimately defaults and losses could be. But in our base case, based on that, we expect the CRT investments to deliver an annualized ROE of something -- some 17% over the next year. And again, I think that reflects the fact that you got these investments marked at a discount and the base case loss expectation is higher than it was before COVID started. And -- but I think we expect that -- these investments to perform well and generate an acceptable return across all those different scenarios.

Okay. Got it. Can you just talk about your sense for capital needs going forward just given some of the potential draws or the improving outlook on the impact on higher delinquencies and just the investment environment you see here on the MSR side?

Yes. So I think that right now, I think we walked into the year with a really good liquidity position. I think it allowed us to perform really exceptionally well during the crisis. I think that we are -- first and foremost, we have capital allocated and then set aside to close our last CRT deal, which should be closing in December. I think when I -- when we look at the liquidity, we see fast prepayment speeds affecting the existing CRT investments, which means a return of capital that we can use to reinvest into mortgage servicing rights and to do so even in a scenario where we grow the number of purchase or the amount that we're purchasing in our conventional correspondent business. So I think that right now, we feel we're in good position from a capital point of view. We have a lot of different opportunities to raise capital if the right -- if we should find ourselves in a position where we wanted additional capital, whether that be additional convertible debt or potentially other forms of debt if -- we have a growing MSR position, and we have MSR financing. And so I think as we sit here today, we feel good about our capital position, and we feel good about the ability to continue to run our correspondent business and investing in MSRs going out -- looking out for a year and beyond.

Okay, great. Now one last question for the audience here. Final question is, would you expect over the next year your position in PMT to increase, decrease or remain the same? So turning back to the credit strategies. Obviously, you've lost kind of the access to that front-end CRT. How do you think about CRT investing in the future or alternative credit strategies?

So it's a great question. And I'll tell you, Mark, I think that while we are -- we have currently stopped aggregating for new CRT investments, I don't think that our ability to invest in CRT is nonexistent. I think that we have a few things going on. First and foremost, the new credit standards are about to be put up by FHFA probably by the end of the year. And I think the initial draft was a little bit punitive to CRT. I'm hopeful that through the feedback that FHFA has received from just about everyone in the mortgage ecosystem, that they're a little bit more accommodating to CRT. And I think on one end, you've seen Freddie Mac come back and issue their own STACR bonds. Fannie, I think, has put theirs on hold -- new issuance on hold for now, but I'm hopeful that we'll see CRT kind of resume its presence in the marketplace. I think in terms of front-end credit risk transfer, I think that while that program was discontinued, my hope is that we see out of this crisis the real benefit and advantage for the alignment of interest between the entity that's buying the loans, pooling the loans, servicing the loans and investing in the loans. I think that we've come up with some great proprietary programs ourselves to try to mitigate losses. And in the process, what it does is it keeps the outstanding insurance available for the GSEs. I think that we will -- the GSEs have been in front-end credit risk transfer long before the creation of STACR and CAS. They just called it a recourse guarantee fee. And I'm hopeful that we'll see the GSEs come back to something similar when the dust settles here. I think in terms of a macro -- kind of a macro situation as you think about PMT also is that in the event of a change of administration, you can have a new head of FHFA. And I think that, that's something that we're kind of keeping a close watch on. And I think that if you do have a change in leadership at FHFA, that could also accelerate the return to front-end credit risk transfer. So I'm not completely convinced that front-end CRT is permanently gone. I just -- I think that there's just a lot of different outcomes here that we're looking at. And I think in the next 6 to 12 months, you'll see that range kind of decrease to 1 or 2.