Two Harbors Investment Corp. (TWO) Earnings Call Transcript & Summary

And we're live. Good morning, everyone. Welcome to KBW's Virtual Real Estate and Real Estate Technology Conference this morning. I'm Tom Michaud, President and CEO of KBW, and I'm delighted to be kicking off our first-ever domestic real estate conference that's being done virtually. This is certainly a different view than what I traditionally have when I give the opening remarks to kick off this conference. But I'm thrilled, nonetheless, that you're joining us today in what we think is going to be a most insightful conference. We have a tremendous lineup of participants, and we have designed the format and the topics to be in tune with what's happening around the world today. I'm also very excited to have my colleagues who are leaders in the field of research in this sector, Bose George, Jade Rahmani and Eric Hagen, helped lead us through the conference today. Like I said, we'll be hearing from top executives in the real estate space. We'll be focusing on issues for mortgage REITs. We'll be hearing about their leadership during this crisis and the impact of COVID-19 on the mortgage market. We have over 320 investor meetings with 29 different companies set up over the course of the day. I do have a few housekeeping notes for you as well. On the virtual website, there is a companies tag that will highlight the management who are participating on site, and it will also include any handouts within each profile that a company may have included. The KBW team box has links to e-mails from our research analysts if you'd like to reach out to them directly. And without the coffee breaks or networking, we encourage you to do so. Lastly, during these presentations, there's an availability to type in a question in the Ask a Question box on each webcast link located just below the webcast screen. So send your questions our way, and our analysts will do their best to make sure that they are asked. Once again, thank you very much for your participation in today's conference. We value your friendship and relationship with the firm. We hope that you and your loved ones are safe during this time and that you will -- we will all have an opportunity to safely return to our offices, but we're certainly thinking about you. So we have a terrific lineup for our first panel. I'd like to turn it over to my colleague, Eric Hagen, who's going to kick it off and introduce our guests for us. Thank you. Eric?

Great. I hope I'm coming through. Thank you so much for the nice intro, Tom. It's good to be back for another year of the mortgage finance conference. Our first panel will feature leadership from 2 mortgage REITs as well as a repo lender to the mortgage REITs. In preparing for this panel, we wanted to help frame some of the key topics for investors as they navigate the residential mortgage REITs right now, especially those that are looking at the Agency RMBS market. And I think this is the perfect group of panelists to offer some perspective on recent market conditions, including a look into what the repo funding markets are doing and some of the things that are going on there, which have been especially important to the business recently. For our panel today, I'm pleased to first introduce Byron Boston. Byron is the President and CEO of Dynex Capital. Dynex is a $300 million market cap residential mortgage REIT with most of its investment portfolio concentrated in agency assets. Byron's 30-year career in mortgage finance and capital markets includes leadership roles at Freddie Mac, Crédit Suisse First Boston and Lehman Brothers. Representing our second mortgage REIT on the panel today is Matt Koeppen. Matt is the Co-Chief Investment Officer for Two Harbors, where he helps manage the company's $20 billion Agency MBS and MSR portfolio as well as its hedging and funding strategies. Matt brings more than 20 years of experience in mortgages and structured products to his role at Two Harbors. Finally, I'm very pleased to have Jim Tabacchi join us this morning. Jim is the President and CEO of South Street Securities. South Street is a nonbank broker-dealer, specializing in securities, finance and government repurchase agreements, and is a well-known provider of agency repo to many of the mortgage REITs in the sector. Prior to co-founding South Street roughly 20 years ago, Jim held various roles in securities, funding and bank whole divisions at Citi. Welcome, gentlemen, and thank you again for joining us today.

So I think what may be most helpful for the audience is to hear a quick snapshot from each of you about how you've navigated your business over the last few months and what you're most focused on as a leader of your organization, given the current environment. Why don't we go to Byron and then hear from Jim and Matt?

Sure. Thank you so much for having us this morning. At Dynex, we have always operated with a long-term view which led us to intensely focus on risk management such that we always remain in the game. This means we must run a diversified portfolio and remain nimble, where we're willing to adjust our risk posture appropriately for the changing market environment. So prior to this year, we always managed our business to survive the liquidity crisis of 1998 and the credit crisis of 2008. And now we all will add March 2020 to our list of market disruptions. So a few years ago, our disciplined top-down approach, macroeconomic approach to our portfolio decision led us to go up in credit and up in liquidity on our balance sheet. Hence, when we were prepared for a potential correction in market prices, we have planned for a scenario where we broke into all-time low in interest rates. And as 2020 unfolded, we executed on our plan. We maintained a substantial amount of liquidity throughout the crisis. We have made several portfolio decisions since late February, and all decisions have remained within the agency residential security space or the agency multifamily space. In the short term, we have a health crisis, an economic crisis and a fragile market environment. And government policy risk has increased. What will it ultimately mean? The unemployment rate remains high, more companies go bankrupt or, hence, our consumers or businesses fail to pay rent or mortgage payments. We're not sure. It's uncertain in the short term. Nonetheless, we are relaxed. We're patient. We're methodical. We're disciplined in our approach. And we've been able to reinvest some of our capital. So we deleveraged at some point coming out of March. Since that time, we have reinvested some of our capital at good returns. Our main focus today is preserving capital, taking the appropriate amount of risk and generating what we think is an above-average return as we navigate this complex environment. And it's also important to note that global interest rates, global yields, are at 0 or below 0 throughout most of the world. So the return environment has changed, the risk environment has changed and, as appropriately, we have adjusted our business focus here at Dynex.

Thank you. Jim?

Sure. Thanks, Eric, and a pleasure to be here. As I look back on the last 3 months, it's been a hell of a ride, as I'm sure everybody can appreciate. If I look at our business and how we navigate the business, it's very much customer focused. That customer focus intensifies, of course, when there's a little bit of a credit crisis. And I don't really believe that it is -- it was a credit crisis that was born basically on any of our customers' fault, so to speak. The -- I don't know how you could have anticipated what happened to the marketplace and to my clients over the last -- in that last period of March and early April. So we're a customer-focused business. We double down on the customer focus during the crisis. We talk to our clients on a daily basis, sometimes twice a day. We increase all of the reporting mechanisms. We have a phrase in the office called, hang in there with your clients. You need to be able to talk to people to do that. We try to be as creative and innovative as possible in doing that, structuring some terms and resolutions because the margin calls became, frankly, close to impossible. We understand that. We understand how important capital is and liquidity is in these businesses. And as long as our customers continue to talk to us, and they just about all did on a multi-times-a-day basis, we try to do our very best to hang in there with them. So for the most part, we're now having -- we had the financial crisis 10 years ago. We just had this pandemic crisis. A lot of people are always referring to these crises as 100-year events, which we seem to get every 8 to 10 years. I think there's structural reasons in the marketplace for that, which I believe, based on the agenda, we're going to get into further into this conversation. But I do think Byron mentioned an important point, being nimble, you have to be nimble. You have to be methodical. You have to be very well thought out in all of your decisions. I don't think the gunslinger mentality really has a place, at least not in my business. I don't think really in the REIT business either. Having said that, as I look forward coming out of the crisis, I think the prospects for the agency REITs probably are as good as they've been in quite some time, and it should be a very peak period. So I'm definitely looking forward. I know most of our customers really well, most of our active customers, certainly. And we hung in there with them during the crisis, and we're open to be able to help them rebuild their businesses and get some substantial profitability.

Thank you. Matt, why don't we hear from you?

Sure. Thanks, Eric, and thanks for having us on this panel, and good morning, everyone. So we're -- as they described in there, we're coming out of the most challenging market environment we've seen in the tender history of our company here. I think the real hallmark of this crisis, I think, wasn't necessarily the magnitude of it, but it was the speed of it. And really, that was overwhelming. And as we got into March, the violent volatility that we saw in agency market as we got into it, we're seeing, as you all know, 25, 50 basis point swings even in today in the asset class. So we, like others, we came intensely focused on excess cash. We delevered. We ended up selling about $18 billion in agency pools and TBAs during that time. At the same time, the impact on credit was unfolding. It wasn't just in -- so we had a legacy non-agency portfolio going into this. CRT markets, CMBS markets, credit markets, also a dramatic widening in spreads, so we became quite worried about ongoing portfolio liquidations in non-agencies. We became worried about the funding markets. There was a time that wasn't clear that you'd be able to roll positions in funding and manage the assets. We saw indications that haircuts were going to be increasing by 10%, 15%, 20%. Spreads were getting dramatically wider. So in order to put ourselves in a strong cash position, we ultimately liquidated substantially all of our non-agency position as well, which now needs us on to follow, basically positioned as mostly agency REIT. So we have a portfolio of servicing assets and Agency RMBS. So in the near term, within the last 30 to 60 days, we've been quite focused on managing our advanced responsibilities as servicer of record. We think we're in a position to be successful there. We've taken a lot of steps. We're managing that. We're working on advanced facilities as we speak. And so looking forward a little bit, we're going to get to a point here where we feel comfortable with our excess liquidity. We feel comfortable deploying cash again. There's interesting opportunities in the servicing asset. There's very nascent markets developing there. Global markets are starting to slowly come back online. There are some maybe indications on both transactions beginning to come online there. So we're cautiously looking in the near future to be able to deploy capital into those target assets.

Fantastic. Thank you for that, from all 3 of you. That was helpful. Maybe let's go back to the March, April time frame. And Byron, maybe you can walk us through the liquidity position of Dynex. I mean, as it was noted, agency spreads, among other things, reached very wide levels in March, which did generate a liquidity demand for REITs to meet mark-to-market margin calls. And thankfully, the worst of that, I think, has passed. But how does your liquidity position look today? And how should investors think generally about margin call and liquidity risk going forward?

So our liquidity thought process is the same as it was in December 31, 2019. At that point, we carried probably 3, 4x as much liquidity as we did going back, call it, 2012 or 2014. We had basically increased the outright liquidity on our balance sheet, which means cash and unencumbered securities. And the unencumbered securities themselves, we ensured carried the most amount of liquidity, except for treasuries, meaning obviously 30-year agency securities. So we're comfortable with our liquidity throughout March. We've got to a point where we were concerned with our liquidity. Just to give some numbers. At the end of 2019, we had about maybe $225 million in liquidity. Our lowest point in March may have been $80 million to $90 million in liquidity, maybe even potentially close to $100 million in liquidity. So we have -- and we were -- so we were managing and navigating that period carrying a lot of liquidity. What does it mean? What's the flip side of liquidity? It means that I could drag my leverage higher. That means I can run a 10, 11, 12x leverage fall. We have chosen not to do that, especially today. In the short term, we will consider the uncertainty high, meaning we have the health crisis, we have an economic crisis that is still unfolding. And so our leverage networks are lower, which means our liquidity is higher. That allows us to be more resilient. It allows us to generate an above-average return. And when we say above-average return, we're defining that in terms of context of a global return options that an investor may have to receive cash income. So we're looking to generate an above-average return, an appropriate amount of leverage, which then gives us an appropriate amount of liquidity. And we're always thinking at Dynex for the long term. So we're always saying, can we get through 1998? Can we get through 2008? And now will we be able to get to a 2020-type scenario? One of the focus that brought an [ elderly ] people like myself, one of the fascinating things in March was how illiquid the specified pools in 30 years happen to be. The pay-up on a specified pool is really technical, so much the other guy is willing to pay. And we saw those specified pool pay-ups collapse. TBAs maintained an enormous amount of liquidity as the Fed entered the marketplace. It clearly created some problems on the front end of the business as originated as CPA margin calls. But it's really important to understand, if you traded in the market 20 years ago, how this market has evolved with the separation in liquidity between specified pools and the TBA market. But at Dynex, we had a lot of liquidity at the end 2019. In May of 2020, we still have a lot of liquidity. We had concerns about the overall global risk environment at the end of 2019. In May of 2020, we still have some concerns about the overall global risk environment. And it's great that the Fed has stepped in and placed a nice comfortable mattress underneath the market. But at the end of the day, that's your cushion, and we're choosing still to carry a decent model for you.

Great. Matt, let's pull you back in, maybe follow up a little bit on mortgage spreads and returns in the business. Where do you see returns right now for agency specified pools and MSR or the pairing of the 2, anyway? And while agency spreads have narrowed off of the very wide levels that they've reached briefly in March, what do you think is still keeping them relatively wide or wider than the historical average right now?

Sure. I'll make a few comments, if it's all right. I think that interestingly, right, if you look at just Agency RMBS returns hedged with swaps, I think today, despite the round trip, I would tell you that they were pretty similar to -- they're pretty similar today with some levels that we saw back in February. We see that sort of low teens as kind of a growth indication. There's been a little bit of getting back in May as the Fed has sort of reduced its daily purchases now to less than $5 million. Nothing dramatic, but TBA spreads kind of in the middle of the stack have lined a bit. I think the -- I think with respect to the Fed, I think that their stated goal here was to stabilize the market, right? I don't think the Fed's intention was to drive mortgage spreads to 0 or negative, right? I think they wanted to bring liquidity back to the market and ensure sort of stable functioning. I think that they accomplished that. So I think for them to dial back their purchases is appropriate. And that's probably part of the reason that you might say that they're slightly wider than historical averages. And in terms of our portfolio, I think that the returns of the parent strategy like we run where we have servicing combined with Agency RMBS, I think those returns today are more likely in the mid-teens on a gross basis. So -- and we think that's attractive. And like I said a little bit earlier, we are working through our servicing advanced obligations and looking at still strongly focused on our liquidity, and those things are sort of changing and moving every day. But I think once we feel confident and are able to get back to business, I think we'll be able to sort of re-lever and take advantage of those interesting returns.

Excellent. And Jim, let's maybe pull you back in, talk a little bit about some color on the funding markets. The government repo markets did avoid or largely avoid a meaningful disruption in March, and I think that was thanks to the Fed's swift support. But how do conditions in Agency MBS funding markets look right now? How has the supply of cash been impacted by COVID?

I mean I would say that I would liken what the Fed actions to kind of like a rolling thunder. They increased their actions and included more and more assets into their process. Originally, they did the treasuries, then they did the agencies. They left DUS bonds out of the process for a while. That's created some havoc. But they kind of got on and got basically everything covered, and there's some nice support. I agree with Matt and what Matt said, that they provided a significant, I think also Byron used the term, cushion under the marketplace. They've ratcheted back their purchases. It's all on the TBA side, but that clearly has an effect going on to the pass-through. So the liquidity position right now is really good. They've done a number of other things. There's so many programs going on. I'll just tick off a few. First thing that I thought was most important is they raised the level of reserves in the marketplace so that there is a very wide and diversified source of liquidity out there for -- across the market. They increased repo operations. They bought securities and continue to buy agency securities so that the amount of liquidity chasing the securities in the marketplace outside of the Fed is very ample. So all of their steps have been prudent, well thought out. I got to give the Fed a lot of credit once again in a crisis. The one hesitation I have is how the Fed might take to pull back from the crisis. Now I don't think that's coming anytime soon, at least I don't -- and I think the Fed is going to be very meticulous in how it plans a pullback because the last thing you want to do is to have the Fed cause a disruption itself, maybe some significant prepayments. And I know we'll probably get into those things like the taper tantrum in 2013. We want to kind of avoid all of that stuff, if possible. So right now, I would say the spreads in the marketplace for our REIT clients is healthy. And they should be making certainly low double-digit returns as we move forward. And based on a risk-adjusted basis, I think that's a phenomenal return when you consider that there's no credit risk. And the Fed's proven, frankly, that they're going to support the business and as well as the GSEs. So I think the outlook for the agency REIT market is about as good as you can have coming out of a crisis like we had. And if there's any place that there's going to be a V-like recovery, it's going to be in the agency REIT market.

Excellent. I think we agree. Let's move back over to Byron. There is much more to the business than, of course, interest rates and the shape of the yield curve, but I thought this would be a good question to ask you. How should investors think about the returns in the business if long-term interest rates remain low by historical standards? What if rates head even lower? There's been a lot of chatter recently about the short end of the yield curve going below 0. What does that mean for the business? And I would also introduce the fact that by the same token, treasury issuance will likely be significant over the next number of years. And so what is potentially putting upward pressure on interest rates? What does that mean for your business? How do you balance all those things?

It's really fascinating, right, this time period. And there's -- as a trader, as an investor, I've been at it since 1981, there's always opportunities to make life. And at Dynex, most of the consumers are being very disciplined as we approach this market, especially in terms of where interest rates happen to be and where the yield curve happens to sit. We're at low levels. The Fed has pegged our financing costs at really low levels. We're not anticipating that to change for at least a couple of years. And then if I gave you my personal opinion, if I look out into the future, we've increased global debt enormously. So the concept that we're going to drive interest rates higher for any substantial amount in any time in the future is a farm-fresh idea, as far as I'm concerned. That's a personal opinion, but it is shared with those here at Dynex Capital. So we're looking at a yield curve where we see the 10-years in very, very low volatility. That volatility has been so low partially because the Fed has put a cushion under the marketplace. But there's an enormous amount of supply and maybe we see the curves steepen. All mortgage REITs should want the curve to steepen. Our potential ROEs -- or investors should understand our potential ROEs increase as the curve steepens and our funding cost remains a near 0. So it's an exciting time for mortgage REITs, starts with the fact that our funding costs are low. They're paying low. We have felt that we would ultimately get to a point where rates were -- our funding costs would be low. We didn't anticipate there'll be a health crisis. And so we believe that we can manage through. We'll have opportunities. We'll take [ heightened ] fund. We'll take the loan to fund. We can be an Agency CMBS. We can move on time with the agency sector if the opportunity presented itself, which it doesn't today. So there are opportunities. Yields are not -- yields rising some astronomical -- not on our radar screen. If I had to give you a range on the 10-year, you're 50 basis points to 1% at best during this period. And our short range of funding costs are pegged in near 0. And as an investor, the most important thing you need to understand is the funding cost is at 0. And that is a huge driver of returns for Dynex Capital and other mortgage REITs.

Absolutely. I want to spend some time talking about the prepayment outlook because that's been received a lot of attention recently in a COVID world. Matt, how are you guys thinking about your premium risk in the current environment? How are you thinking about the ability of borrowers to access the refi and purchase markets going forward in light of COVID? How do these various drivers and things impact your portfolio as it relates to prepayment speeds?

Yes, speed has been interesting. I would say the April speed report that came out, I think, for the most part, surprised most people in the market. I think there was a general expectation, right, given the shutdown that, that would slow down people's ability to close on purchases and to refinance, right? We would expect maybe it was tough to -- with the world shut down in March, basically, it would be difficult for people to get appraisals or maybe difficult to get to a live closing. So I think the market generally expected that to be a drag. It turned out that speeds in the April report, which was reported earlier this month, increased 25%, which I think was basically in line with what people would have expected, given the rate environment and day counts and MBA refi index indications. And it came in right in line. So there probably was some impact. Could it have been even faster had the pandemic not been upon us? Possibly. But it seems that the ability of people to get to closings is pretty high. There's 47 states that allow for remote closings, and getting remote notarization has been common. I think refi activity will probably continue on at a pretty good clip. I would guess purchase activity will be needed somewhat for the reasons that I just talked about. But I think we'll have to see -- we'll have to watch each month's report. But early indications are that things are still moving along there. In terms of our portfolio, I guess the thing that I would note is our entire specified pool of portfolio has some form of prepayment protection. It's mostly loan balance and geographic stories at this point. So the deviations on speed from expectations or from models will -- I would say, are not likely to be dramatic so that helps us to -- that largely -- that helps us to manage prepayments and worries about faster speeds, to a large extent.

Got it. Hey, Jim, this will be a good transition over to you as well. I mean how do you and your team calibrate your exposure to prepayment risk as a lender to the asset class? What kinds of metrics do you look at, maybe more generally, to assess your exposure to the residential mortgage REITs and mortgages, in general?

For us, it's pretty simple. I mean they announced the factors early in the month around the fourth or fifth business day, and the actual cash doesn't flow to the REITs until the 25th. That timing difference causes a significant margin call on all of our clients. Again, we are very conscious. Byron was talking about liquidity before. We also have a significant liquidity buffer. I think during the height of the crisis when most of our clients are going through margin health, I don't think our liquidity position ever went below $180 million or $175 million. So look, if the clients are talking to us, as I touched on before, and if they're having trouble with those margin calls, and we have an understanding about -- they got 15, 20 days that they have to get through, we're going to find creative ways. And several of our clients know that we've already done this, to hang in there with them, to use our liquidity position and our capital prudently, of course, and see if we can't get them to that 25th and when the cash starts to flow a little bit for our clients. Now that's why you worry a little bit about, again, how the Fed may pull out from here. And again, I'm trying not to jump too far ahead of ourselves. But the Fed tapering their purchases. And I think after the last crisis, I'm not sure the Fed ever actually started to sell mortgages into the marketplace, and they did reinvest the P&I back into the portfolio, which I thought was very prudent. So if they follow a similar forecast or procedure this time, I think we'll do just fine. But at the end of the day, you know your clients. You know the paper that they have and the type of paper that they have. You know how they manage. You know who is a little bit more aggressive with risk, who's a little bit less aggressive with risk. And you make your decisions based on all of those things. Now we are in a very fortunate place that I have 8 of my best clients are also partners in South Street, where they have not only given capital to South Street and made an investment in South Street, and we reserved a certain amount of our balance sheet for them. So when you have that type of familiarity with your clients, it allows you to do things that maybe other forms might not be able to do. So I think we do worry about the prepayments and more so based on the timing difference of the announcements versus the actual cash flow. But with abundance of communication and understanding of our clients' needs, we've been able to manage that, I think, quite nicely with all of them and support them as they've got through the period. And look, if you can do that in March and April, I think you can do it any time.

Excellent. Sticking on the funding, Byron and Matt, maybe you can just give the audience a quick snapshot for where you're rolling 1-month repo and what that spread is usually look like relative to 1-month LIBOR, Fed funds or whichever benchmark you pick, just to give the audience a little bit of a tangible data point.

Do you want me to go first?

Sure. Yes.

I think the relevant issue would be that the -- we're funding all -- let's compare back to the great financial crash. Funding costs were, call it, between 20 and 40 basis points, and that's basically the range that we're in today. Do I think that, that spread will change or go against us? No. I think one of the most positive things you can think about mortgage REITs today is that we're in this favorable funding environment. So we're seeing spreads versus 1-month lag. I mean just we're not assuming that, that spread is going to change and become stressful. And Jim, you can comment on this because dealers really play a big role in this in terms of where they want to have the spread versus where they can fund. But today, there's been an influx of cash into the short end of the yield curve. Liquidity seems to be very ample in terms of financing of government-backed agency securities. I think your question around funding is to start which assets are you trying to finance. I'd still be uncomfortable financing some certain non-agency assets and loans, and someone said further down the yield stack. But our main key is that we're seeing funding rates in the, call it, 20 to 40 basis points. I'm assuming that's where we'll be. If rates were to drop to negative, I'm not sure where they'll go, but I'm happy between 20 to 40 basis points. We'd love to see the spread between that funding rate in our assets, which may be between 1.50% to 2% move out wider. A steeper curve will make a difference with that. But at the moment, funding rates are very attractive.

Well, I agree with Byron's comments. And I would add, let's not forget that on the agency funding side of things, we had our own sort of crisis back in September, which seems like a million years ago and kind of a quaint crisis in retrospect. But back in September, October before the Fed came in with what then was a very large-sized term market repo operations, right, they came in overnight and termed in the hundreds of millions at the time. I think at the worst, the widest point there are all costings versus OIS, I think agency gross spreads might have gotten into the OIS plus 40, maybe even 50 at one point level. Now during March, the Fed, right, increased the size of those funding operations into the trillions, right, to ensure that part of the money market continued to remain functional, which it largely did. I mean I think there were some wide spreads in March, too, but we're seeing levels similar to what Byron was talking about where I might cut 1 month that OIS plus, call it, OIS is 5 or 6 or 7 basis points, call it, OIS plus 20 to 25, right, for 1-month repo. And I also agree with Byron that any reason to think that those spreads should be dramatically wider from those levels, right, it seems -- that market seems like it's redeveloping. We're seeing term markets redevelop, as time passes, the further we get away from March, right? It's moved out from available weekly to 1 month to now 3 months, and I'm sure longer-term markets than that will continue to develop as time goes by.

I think something you guys both said would be a great segue into shining some more light on what that repo market volatility was back in September last year. Jim, do you want to shed some light on what drove that imbalance in the market and what the Fed's response was really quickly? And then, I guess, the million-dollar question is, should investors be concerned that such imbalances resurface, particularly with the amount of government debt that's being issued now? Go ahead.

Sure. Basically, in a nutshell, what caused that repo volatility -- and frankly, that was building for the prior 3, 4 quarters prior to September. The cause of that is basically concentration risk. The top 4 or 5 bank holding companies have accumulated more and more and more of the concentration of funding. And so look, it's simple math. If you've got 5 players that are doing 80% of sources of funds for a marketplace, and for any given reason, that doesn't have to be a bad reason, it could be a major transaction that they're working on somewhere in the world, you have 1 of those 5 institutions or maybe 1.5 not be able to participate and have less liquidity than they would normally, which is what you saw back in September. You're going to have a spike in rates. If you were to ask the executives and the C-suites of those 5 large banking institutions, if everybody would benefit more from a much more widely diversified liquidity, why would anybody -- they would all agree, first of all, why would anybody not want private capital coming into the liquidity markets to add liquidity? So look, that was building for quite some time. And I would say that the biggest factor in what that changed after September was the fact that the Fed brought up the level of reserves. I look at the level of reserves at $1 billion -- $1.5 trillion, has basically the demarcation point because the level of reserves is the only way that you get liquidity out to everybody in the marketplace. It's very diversified the way they -- that liquidity flows. Now in talking to the future, if you look at the size of the reserves today, they're significantly -- I mean it's in the $4 billion to $5 billion -- $4 trillion to $5 trillion category. So as long as that stays there, I would say that there is going to be plenty of liquidity around. The other thing is they reduced supplemental leverage ratio or eliminated it for a year. I think, basically, what that causes all of the big banks to do is basically be governed more by what is the adequate GAAP leverage ratio for a well-capitalized bank. I think most would agree it's going to be a minimum of 6%. So that becomes the new demarcation point for the big banks. And I think they're all very well capitalized, and they'll be able to play at that level going forward. I do believe that we will see more disruptions caused by concentration of sources of funds in the marketplace, I think less so as long as reserves stay at such a high level. But I do believe, at some point, you'll see that come back only because we will again hit a day like September, where you have a significant amount with all the treasuries being issued of settlements. We're going to have a tax day again. We're going to have some type of international event that may cause companies to call in all of their liquidity lines on the big banks. And if you don't have more diversification in the liquidity markets, you could see a disruption again. I don't think you'll see it anywhere near like September when funding rates got to 10% against treasuries. Think about that for a moment. But you will -- you might see some disruption again, especially as we move forward 2 or 3 years and the Fed's level of reserves comes down.

Great. Great. That's fantastic color. We'll switch over to talk about capital structure. I think that's been a popular topic for a lot of investors. Byron, I know that you and I have talked about the capital structure of mortgage REITs pretty recently. What's -- how do you guys think -- Byron and Matt, how are you guys both thinking about the capital structure in your companies, including what I really mean by that is the mix of common and preferred stock? And I guess one kind of forward-looking question is, what do you think the optimal mix is between those 2 buckets of capital over time? Byron, let's start with you.

Sure. Earlier this year, we successfully refinanced our highest cost of preferred. We did $100 million preferred issuance with a [ 6.90% ] coupon. We believe the optimal mix or the right side of the balance sheet for -- especially depending on what we invest in, have to be available to mix the preferred and common. The theme should be an unreg rule around the Street. Oh, should it be 30%, 35% or 25%? And we don't hold ourselves to those particular rules. Here's the long-term thought process. To be resilient -- again, we have a long-term vision at Dynex. To be resilient over the long term, we believe Dynex will have to grow over time. Are we going to just try to grow under any cost, under all conditions and harm our existing shareholders in certain ways? No, that's not the philosophy. Just over time, that's what we'll look to do. We'll look to do it with both a combination of preferred and common stock. In addition, we are managing aggressively the right side of our balance sheet as we do the left side of our balance sheet. So -- and I think it was 2019 where our stock became grossly mispriced versus where we saw the value of our balance sheet, and we're willing to buy back our stock at that time. It was grossly mispriced. They're an enormous amount of where the short sellers came from. But the fact of the matter was, it was grossly mispriced. We saw more value in buying back our stock than investing in another agency, security or putting on more leverage on our balance sheet through the investing process. So is there some unwritten rule at Dynex Capital about the preferred versus common? Over the long term, we would like to grow. We like the idea of having, obviously, far more common than preferred. Both instruments provide us an opportunity to grow to take advantage of certain investments, and we're aggressively managing both sides of the balance sheet.

And asset. Matt, what's your thought there?

Yes, I agree with Byron. I don't think that there is necessarily a magic number or an optimal number. Maybe there's a reasonable range, I would say. Like that's something that we're considering on follow. Obviously, given the impacts in our common stock during the crisis, our current mix has moved up to about 2/3 common, 1/3 preferred. We do always think about the risk to common shareholders. And we report in our -- excuse, recording our risk metrics shocks to interest rates and mortgage spread shocks with reference to common only, so we think about that all the time. I think that, obviously, having some at the limit, right, if you had some very high number, like 90% of your capital structure in preferred, that would obviously cause untenable risks to common shareholders. But when you're talking about numbers in the 15%, 20%, 25%, 30%, I think we're pretty comfortable with that. I think it's a reasonable range. And like I said, we always think about the impact of our positioning and risk-taking a manage on common shareholders.

Great. We have 10 or 15 minutes left. I wanted to open it up to the questions, which I'm receiving on my computer on the dashboard here. One that came in, wanting to know what the -- any book value update from Dynex and Two Harbors since the end of the quarter? Matt?

We don't have much of an update. There is some -- there are some offsetting impacts for us post quarter end that we talked about during earnings. And we recently released a deck this week, too. We talked about the impacts of specified pool reflation. Basically, those levels have come back to pre-crisis levels. And that part of the portfolio had about low teens, 12% increase. The book, as offset by us recording of data, we're paying to get out of our external management arrangements. So we've indicated that the net of those things is that book value is up slightly, but that's what we had.

Byron, did you want to add anything?

Sure, absolutely. We are -- we delevered the balance sheet, and we've been slowly releveraging the balance sheet in a phased approach. And our book value literally has been flat to up a few percentage points, and it's bounced back and forth. But with this much low volatility, we've been pretty solid. You can easily tell what's happening from our book value. We can pull a lever of leverage and open up with another turn, another turn, another turn. You'll expose more book value to that. But at this time period in the short term, we've talked about being 5 to 7x leverage target, being able to generate a decent return, and our book value has not moved an enormous amount. But I'd say you're between flat to up a few percentage points in this bounce back.

So here's a question from the web, I think, for Jim. You mentioned the supplementary leverage ratio for banks was tweaked by the Fed recently. Should that limit the potential for bottlenecks in the funding markets to manifest around quarter or year-end? And do you think banks will continue to support the agency repo market as meaningfully as they have in the past? Or is there room for nonbank repo providers, such as South Street, to grow their market share?

That's a lot in that question. I don't think by in itself, the relaxing of the supplemental leverage ratio is going to create bottlenecks around quarter ends. What really creates the bottlenecks around quarter ends is what are all the other funding opportunities that the banks have that -- and look, they're going to put their balance sheet to work at the highest returning rate. I do believe that relaxation of the supplemental leverage ratio gives them more options and more liquidity. So again, what causes the rate spikes is concentration risk. And again, we've done a lot of analysis on this. And the Independent Dealers Association that I chair published a white paper, the end of last year, with a number of trends in the marketplace that talks to many of these questions. If -- and again, a little bit of a quick math. Of the 4 or 5 players, 1 or 2 can't play rate spike. I don't think the supplemental leverage ratio has a heck of a lot to do with that because the new level of measurement is now going to be basically a 6% GAAP leverage ratio. And all the banks are going to adjust to that. It gives them a little bit more room. It gives them a little bit more room at a time when having more liquidity in the marketplace is better than less. Hard for me to imagine when that isn't a good idea. But -- so in terms of the regulation going forward, is that going to -- the only regulations -- in terms of the supplement leverage ratio, if the Fed, for example, wants to go back to a supplemental leverage ratio at the end of the year, I'd look at a phased-in approach rather than just banging it back into place, or at least plenty of early warning about that. So 6 months before, they say, look, we're thinking that this is going to be the case, that we're going to put it back in place. Something along a phased-in or early notification so that everybody, the banks can adjust to it, the market can adjust to it, you want to try to affect the least amount of disruption for any of these regulatory changes as possible. And I think this Fed under the present Chairman has been pretty good at doing that.

Great. Here's another question from the web. I think it's for Matt. Can you discuss what types of stress scenarios that you and your team have run on the servicing advance obligations? Do you expect you might require under different levels of forbearance within your MSR portfolio? And I know that there's a panel on this later this morning, so I don't want to preempt that or anything, but I thought it was a good question. Go ahead.

Sure. So first, let me talk about what we're seeing in our servicing portfolio. So we're right around 7% of forbearance in our servicing book. And the rate of increase has slowed pretty dramatically in the last month. We're early in the -- the beginning of this is the end of March and early April, the uptake numbers were pretty big. So people were then talking about scenarios of, oh my gosh, look at the rate of increase here, could it get up to -- could this number get up to 25%, 30%? Could it get to 40%, right, at the rate you're seeing at the beginning? I think as time has gone by and we've seen the uptake slow pretty dramatically, we've reduced our assumptions, our scenarios that we're measuring to, to have some 20% uptake be the -- be our highest stress. So we call our base case 12% and a moderate stress 15% and a severe stress 20%. And at the moment, given our experience and the market's experience, without seeing a sort of dramatic -- I'm not saying it's not going to happen, but without seeing a dramatic resurgence sort of COVID and a dramatic increase in unemployment, which ultimately we think these things are going to occur, too, I think as time is going by there, we're getting more and more comfortable that it's going to be pretty tough to even see something get up to 20%. But we are still considering that as a scenario and making sure that we've got excess liquidity available to manage such a scenario.

Great. Another one here is -- and I know that we haven't really talked much about the credit markets. A lot of our conversation here has been focused on Agency MBS. But is there any update or color that you can give on mortgage credit? And how securities or loans are trading in the market today? And I know that neither of you are really in credit anymore, but is that an asset class or a segment of the market that you could see yourself returning to over time?

Let me be real clear here, we are in every asset class any given day. We choose the places where we invest today with our opinion. It's been for a while, you're better off being up in credit and up in liquidity. So we're constantly looking at the credit market, and we're evaluating them in a very disciplined, methodical in detail. We have a problem here and a goal. There's cash flow disruption. It's cash flow disrupting at the consumer level. It's at the business level. And in the short term, we don't know yet exactly how much that cash flow disruption will continue to increase. As we move on in time here in the short term, more companies go bankrupt and more companies go bankrupt. Or they go from furloughing employees to temporarily firing employees. What will the unemployment rate be at the end of the year? The remaining thing is the unemployment rate will still be in the double digits. That creates a problem for creditors. There's cash flow disruption. Someone has to take a loss. And if you're a creditor, I don't want to be the guy who takes a loss. Therefore, we're an agency-backed government securities. I can understand the resi side of the agency securities market, better than the multifamily side. Now you know I've been a huge preacher of using the multifamily because it has good prepayment -- structural prepayment protection. But right now, what does it mean when renters don't pay their rent? I'm not sure here in the short term. And in the short term, I'm defining over the next, call it, 4 to 6 weeks. Through the end of June is really won't want to see what happens here in terms of overall underlying cash flow, credit cash flow, whether you're talking corporations or whether you're talking consumers, and let's not forget whether you're talking states or other governmental entities. We've exploded the amount of leverage throughout the global system. Global debt has increased enormously, despite the fact that increased enormously after '08. So we've got more debt, on top of more debt, on top of more debt, and then we still have a question mark on who exactly will be able to pay. So let me just leave that with that macroeconomic opinion on credit. You can take further risk as much as you would like to take it. At Dynex Capital, we're disciplined. We can't answer the question, how much will cash flows be disrupted? The last thing we want to do as a creditor is be the guy who takes the loss.

I would just add one quick comment to what Byron said, sort of you just discussing the fundamental aspect of credit assets, which is, of course, [ occurring ] and true. I would point out that one of our main concerns through this has been funding of that asset class, too, right? That was one of the main big drivers in our decision to exit the space. And so I think that remains to be seen, right? Is it -- I mean from our perspective in the near term, I don't anticipate us getting back into credit. Could I see a scenario down the road where we can get our hands around the fundamentals that Byron was talking about? And there's some sort of long-term, non-mark-to-market funding availability that makes us feel comfortable with our ability to fund the asset class. Can I see a place for it? Maybe. But I think those 2 things are the big considerations these days for that asset class.

Hey Eric, I just want to point, that's a big -- that's a very important point. We're leveraged investors. So you're talking -- we're different than a mutual fund that's investing with cash. We are leveraged investors. So I think Matt's point is extremely important to understand as an investor.

[ It's really our lifeblood ].

Yes?

Yes. No, we are running up on time. I think one parting question for maybe each of you, just really quickly, would be just any kind of general thoughts on things that you would point investors to as they navigate residential mortgage REITs or just the markets in general from your perspective? Not to say that anyone is missing anything necessarily, but what do you guys find that investors might be overlooking or any diversity?

Diversity? Be careful how you invest your money. You must diversify your management teams. Don't think of these as companies. These are not companies. These are management teams, and you need diversity. We all take different levels of risk. We all have difference of opinions. It isn't size. That's not going to be the issue. It isn't a cost ratio. Do you want to put one major factor? Yes, how well will these management teams react to uncertain events? So I would really urge you to diversify your management team on multiple companies. And be careful what you see. Look at me. I'm sitting here, have a suit and tie on. Did you know, I am on shorts? Ran the old suit and socks that I haven't washed in 5 months. Just be careful. At Dynex Capital, we're trying to be as consistent and transparent as possible. We want you to understand what we're doing. We want you to understand that we are invested alongside of you. And for Jim and for the rest of the creditors, we want you to sleep well at night. But we don't want you worrying about Dynex Capital. We have our money in the company, and we're investing over the long term.

I'd like to add to what Byron said, too. I think don't fight the Fed. If the Fed is going to come in and support a marketplace, I think that's a marketplace where investors should be involved just on 5%, and I'll leave it at that.

Good comment.

I heard someone described yesterday when they were talking about it that they likened this to a hurricane. And they said, we're not sure where we're at, but we get hit in the front of it, and now we're in the eye, and there's a calm moment here, right? I thought that was an apt description. I mean I guess I would generally say, right, as an investor, be cautious, be careful here. I think the worst in the liquidity crisis is over, but it just happened 60 days ago. It's still pretty fresh. I would say [ stir over ], be careful, be cautious still, and let's see how it is going forward.

Great. Great. Thank you, guys, so much for joining us this morning.

Thank you, Eric.

Thank you, KBW.

Fantastic discussion. And I think I will turn it over to Kendall, our webcast operator.

Thanks, guys.

Thank you so much again, guys.

Thank you, Eric. Thanks, KBW.

Great. Good morning, and welcome, everyone, to the panel on COVID-19 and mortgage servicing. My name is Bose George, I cover the mortgage sector here at KBW. This panel is going to cover the latest events in the mortgage market with a focus on mortgage servicing. So with us today, we have a panel of mortgage market veterans who will help us discuss trends in the servicing market and what that means for both mortgage origination and servicing. So let me start with an introduction of the panelists. First, Lee Smith, on the far top right, joined Flagstar in May 2013 as the Chief Operating Officer. Prior to that, he was a partner at MP Global Advisers, where he worked with numerous management teams and Boards across various companies and industries where MatlinPatterson had an investment interest. Prior to joining MatlinPatterson, he was a senior director with Zolfo Cooper in their New York office. He is a member of the Institute of Chartered Accountants in England and Wales. Next is Bill Greenberg. He's a Co-Chief Investment Officer at Two Harbors. Bill has primary responsibility for the investment and hedging strategy of the company's investment securities portfolio and company and conventional MSR holdings. Mr. Greenberg was appointed Co-Chief Investment Officer at the beginning of 2020 after servicing as Managing Director of Two Harbors since 2012, and he has over 25 years of experience, managing portfolios of structured finance assets. And on the screen, he's the one right below me. Next, on the top left, Jack Navarro is the President -- sorry, Joel Katz is the President of HomeBridge Financial Services, a private mortgage company, which has originated $9 billion through May and expects to originate another 20 -- around $20 billion in full year 2020. Prior to that, he was a Managing Director at Fortress Investment Group and also worked at Citigroup. He's run several other mortgage companies in the past, most recent of which was Union Planters Mortgage. He started his career as an attorney at Brown & Wood. And then finally, on the bottom left, Jack Navarro, President and CEO of the Servicing Division of NewRez. Jack joined Shellpoint Mortgage Servicing in 2010 and has held numerous leadership roles there, including Managing Partner. He has over 35 years experience in mortgage servicing and REO asset management. Prior to Shellpoint, he served as Managing Director in the Mortgage Group at Goldman Sachs, subsequent to their purchase of Litton Loan Servicing. He also served as a Senior Managing Director and member of the Executive Committee of C-Bass for more than 10 years. So thanks, everyone. Let's start with a quick look at some of the broader themes in the mortgage market, and then we'll dig into mortgage servicing.

So Bill, can we start with you? First, let's touch on the CARES Act, what that allows borrowers to do in terms of forbearance?

Sure. And thanks for having us at this panel here. This is really great under the circumstances to be able to see everyone and talk to everyone. So the CARES Act allows borrowers to enter into a forbearance program if they are impacted by the virus. There's no documentation requirement necessary, although servicers should contact the borrowers to talk them through what being on forbearance means and to provide other sorts of education. What the plan has laid out is that borrowers should be able to have up to 6 months of forbearance initial length at which time after that time ends, they should check in with their servicer. And if they need another 6 months, they can ask for that. And the law says that, that will be granted. In terms of exiting a forbearance, this is not in the CARES Act itself, but there's been subsequent guidance from the FHFA about how to exit forbearance and to get current again besides options of, obviously, just reinstating the loan and paying a lump sum back or to enter into a payment plan. Most recently, the FHFA allowed for a payment deferral option, which means that the missed payments can be tacked on to the end of the mortgage and therefore, either paid back in 30 years or upon refinance or when the home is sold. So this is a great public policy option to allow as many borrowers as possible to stay in their homes and to resume their regular mortgage payments once their difficulty has passed at the end of the year, hopefully.

Okay. Great. And then just wanted to touch also on the volatility that we saw in the mortgage market, especially back in March, if some of that has abated. Joel, can you just talk about some of what you saw, how you handled those challenges and what that meant for you guys back then?

Sure. First, good morning to everyone. I'm so glad to be with you -- here with all of you today. Wow. Those days were completely insane. And it feels like in the middle of March, toward the third week in March, the Fed and the Treasury and everyone else kind of woke up to the reality that the pandemic was going to spill over into and become actually a global financial crisis. And as the Fed snapped into action to help with liquidity problems that were occurring in the marketplace, the unintended consequences of the sheer magnitude of their actions were astounding. And for many, including my company and certainly many independent mortgage bankers, I would say, life-threatening. And while the liquidity was much needed and rectified a disastrous situation or perhaps even prevented one from becoming very, very bad, we woke up and found ourselves in the middle of a monsoon. And the Fed was purchasing $40 billion of MBS a day. To this point, I think they've probably purchased $650 billion of mortgages, $1 trillion of treasuries and a host of other securities. But in the $40 billion a day, the run-up in pricing in the MBS market was simply astounding and unprecedented. In a market that is measured in 30 seconds of a point, we saw moves of 0.5 point to 1.5 in a day. And my recollection is that over a course of a very short time period, some coupons, Fannie Mae 2.5s, were up 5 to 6 points. And the effectiveness and the unintended consequence of all this was that if you were hedging your mortgage -- lock mortgage pipeline as all of us do and doing so essentially by selling securities forward and shorting the treasury market, shorting the MBS market, if you will, the clash of $40 billion a day of purchases and hedging your lock pipeline created margin calls of a magnitude that no one has ever seen and few, quite honestly, could withstand. This came to a crescendo, in my recollection, around Friday, March 27. Or at least for me, it feels like -- that day coincidentally is my wife's birthday, so I remember that day. And over the course of that weekend, many, including myself, had spoken to the MBA, who really was doing yeoman's work and continues to do great work on behalf of the industry. And over the course of that weekend, there were conversations with the SEC, with FINRA and with the Fed. And the requests were 2, at the very least. One, please stop with the margin calls. It's plenty. Oh, and by the way, a lot of these dealers actually are also warehouse lenders who had tons of mortgages in inventory as collateral, the price of which was rising every single day. And so what happened through all this is that, thankfully, the Fed was also asked to stop its purchases or slow down its purchases and create just enough liquidity to matter without destroying an industry. And -- so basically, FINRA and the SEC backed off -- backed down and didn't do anything about dealers and margin calls. That never materialized as far as I know. But the Fed did slow down purchases. Today, that number is something like $5 billion a day or less. And over the course of time, the margin calls were recovered through a sell-off, a gradual sell-off in the bond market or an easing, at the very least, of margin calls and settling trades. And one of the silver linings, and I find this so amazing because when you need oxygen, you just need oxygen and you can kind of figure out how to get it. One of the silver linings is that dwell times and warehouse lines, I think, shortened dramatically. Certainly, at my company, the conversion from mortgages to cash is taking place at a much, much higher speed, a much greater velocity. Just as an example, I would say, before the crisis, the dwell times in our warehouse line are probably, on average, 23 days or something like that, 24. Today, it's about 15. And on Fannie Maes, where we do cash -- settles at the cash window -- cash sales at the cash window, we can do those every single day. The dwell time there is probably 8 or 9 days, if I had to guess. And in the context of Ginnie Maes, you can't settle bonds in the same month as you closed the loan, so it's going to be a longer time period. But managing through that volatility was really treacherous, and thankfully, the world has calmed down dramatically.

Okay. Great. So that was helpful. Let me just do one more on the mortgage market, and then I'll switch over to servicing. Just in terms of current trends, Lee, let me just switch over to you. Can you just talk briefly just on what you're seeing in terms of current trends, volumes, margins? And also just on the credit box, there's obviously been a lot of commentary about credit tightening.

Yes. Sure. First of all, thanks for having me on the panel, Bose. The origination trends are very positive given the low interest rate environment, particularly around refinance volume, but we're also seeing some strong data right now around purchase volume and certainly in the states that have started to open up. A good example would be Georgia, whose purchase volume through May 15 that caught up to where it had been the previous year. And the reason this is significant is if you look a week earlier than that, it was lagging 9.5%. Two weeks before that, it was lagging 18.8%. So there's clearly some pent-up demand on the purchase side. Furthermore, the MBA announced yesterday that purchase volumes turned positive for the week of May 22 year-over-year. So as a result of the increased volume, gain on sale margins have increased as originators. We don't have all the capacity that's necessary to handle the demand. And so we use margin as a way to control the inflow. In terms of the credit box and building on some of the things that Joel said, the credit box tightened at the end of the first quarter, absolutely. There was generally no liquidity for non-QM and jumbo loans. And most originators, including ourselves, we tightened the credit box around other products, just given the volatility in the market. And we're starting to see some of that come back right now, but it certainly hasn't come all the way back. And again, just given the amount of refinance volume and purchase volume, I don't think there's any need to take unnecessary risk in this environment from an origination point of view.

Okay. Great. So let's switch over and start -- dig into the servicing sector. And I'll start with the main topic I think people have been focused on probably for the last month or more, and that's really servicing advances. So Jack, can you kick us off, just what are servicing advances? Can you just talk about how big they get, the different categories, P&I, T&I? And just, yes, color on that would be great.

Sure. Thanks, Bose. Great to be with you all today in these unprecedented times. I'm excited to talk about this great topic of servicing advances. I'm going to assume for the sake of -- I know there's all different levels of experience probably on the phone and certainly a very sophisticated group on the panel. But I'll sort of start at the beginning. So advances are fundamentally dollars paid by the servicer, funded by the servicer through the servicing process. There are basically 2 types of advances. There are what we generally might call delinquency advances, which is basically the funding of principal and interest payments. If a borrower does not make a principal or interest payment in a month and it's a scheduled deal, the servicer has responsibility to fund those amounts to the trust -- the deal. The second type is protective advances, and those are generally what we would call either foreclosure advances or taxes and insurance. And those are just what it sounds like. Those are monies that are paid in order to protect the property and ultimately protect the lien, again, the responsibility of the servicer. And depending on the deal that you're talking about, the advances can be collected in different sort of time frames, sometimes after a modification, sometimes after a liquidation. And so that's sort of the fundamentals. And where -- how can advances vary? As you might imagine, advances vary with loan type or UPB. If a property tax is higher in a state, you're going to have higher tax advances. One thing that is -- we're all very aware of on the panel but some of the folks listening in might not be is that advances are very affected by speeds or prepay speeds because you can often use advances to offset your -- at least your principal and interest numbers. They're affected by delinquencies. And then PLS rules can be quite eclectic depending on what the advance requirements are for the various PLS loans. And then the last thing I would say is advances are impacted by investor type. Fannie and Freddie recently clarified a rule that said that principal and interest advances only had to be made for the first 120 days, which obviously, we found to be very beneficial in this difficult time of forbearances. It was very helpful. Ginnie advances have to be made through liquidation. And again, PLS advances are sort of based on the PSA agreement. How big do advances get? Traditionally, those of us who are used to servicing the GSE loans, the advances very well might be 0. In these times, they're increasing to be some number of basis points on a typical portfolio, some amount, maybe 10 or 20 basis points. And then for higher delinquency PLS deals, the advances can be much higher, all the way up into the hundreds of basis points. So I think that's a sort of a quick little review on advances. Hopefully, nobody went to sleep.

Great. And Bill, can you just talk about the GSE market in particular? Especially, I want to go back on the earnings call, you guys had mentioned a certain sort of outlooks that were, I guess, somewhat cautious about where advances could go. And where do you see that now, just given things that have happened over the last few weeks?

Thank you, Bose. In terms of how big we think the advances could go and forbearances could go? Is that the question?

Yes. And your expectations a month ago versus now?

Sure. So I'm happy to talk about our current forbearance experience to date. And certainly, our expectations and projections have changed over time as more time has passed and more clarity has come into the market. Obviously, as Jack said, some of the announcements by the GSEs and the FHFA to limit the amount of advancing obligations [ a search ] would have has been important. The 4 months that Jack mentioned is important. Another one is that reimbursement for P&I and T&I will be affected at the execution of the loan deferral. That's also an important thing, which flattens the curve and shortens the time line and gives a lot more certainty to the total advancing obligation. At Two Harbors, we are experiencing right around 7% in total forbearance uptake through May 21. We only have a GSE portfolio. We have no government. [indiscernible], that's important. But importantly, of our 7% uptake that we're having thus far, 2 things. Number one, that the rate of increase has slowed dramatically. For the last 8 or 9 days, our incremental uptick is probably 1 or 2 basis points per day. So that's slowed significantly. And importantly, of this -- of the 7%, 4% of all of our loans and forbearance have made their May payment, right? So if we talk about loans that are delinquent and forbearance, that's only 4% for us. So when we think about how big this can be, yes, there are other factors out there in the world that could make this rate of increase pick up again, maybe expiration of the PPP program this summer could have an effect, if there is a second wave or a third wave in the fall and further rise in unemployment, that could make this go worse. But generally, I say we think of the forbearance take-up rate as being linked to unemployment, right? And we can talk about how far we are in that unemployment wave, right? I don't know if whether people think that we're halfway through or 25% of the way through or 75% of way through or whatever. But if we're at 7% gross and 4% sort of with that delinquent cushion, we think a reasonable base case for us is probably 12%, right? That's roughly double of where we are today on a gross basis and 3x if you include the loans that are currently paying. And we like to think of the world in terms of stress cases, and we're managing our liquidity to those cases. So we've -- we considered a moderate stress case that if we think it's running around 16% forbearance up [indiscernible] we think a severe stress for our portfolio in our experience is around 20%. So -- and I'd say it can't go higher, but given what the trends we're seeing and all the impacts that we have, we think those are reasonable cases.

Okay. Great. That's helpful. And actually, let me just switch over to Lee. Can you talk about maybe if this be -- servicing advance funding market from your perspective as a bank that finances a broad array of mortgage assets? Where do you see the funding cost advance rates, et cetera?

Yes. Sure. It's not quite as easy as sort of saying x marks the spot because you've got a number of different variables. You've got to look at the counterparty that you're lending to and the strength of that counterparty, the underlying collateral, what type of advances, are they agency, nonagency. And for us, it obviously makes a difference whether we subservice the underlying loans or not. If we subservice the underlying loans, we have much more visibility into the collateral and performance. And so we're more comfortable advancing more versus if we don't subservice the underlying loans and we don't have that visibility. Overall, I think it's reasonable to say banks are being more cautious in this environment. No one wants to take on additional and unnecessary credit risk. But if you have a strong relationship with a solid counterparty, deals are getting done. And coming back to what Bill said, which I think is the most important thing, there's a lot more certainty now around sizing these facilities. What the FHFA did around the 4 months of P&I was extremely helpful. People are looking at the number of forbearances. They're seeing the curve flatten. They're seeing a big percentage of borrowers continuing to pay. And so people can size what the advanced facilities or needs may be, whereas 6 weeks ago, it was a lot less clear. And so I think having that visibility is certainly helpful and again, another reason why you're starting to see deals get done.

Okay. Great. And then actually, just switching over to Jack, can we just bring you in just to discuss the Ginnie Mae PTAP program? How extensively has that been used? And do you think that's going to be meaningful?

Yes. Sure, Bose, an important topic for those of us who are servicing Ginnie loans. I might just say at the outset that with the exception of maybe 1 financial institution in the marketplace, it's historically been difficult to finance advances for Ginnie for a variety of different sort of technical reasons. That does look like that's opening up. But the program that Ginnie put in place early on, they were really kind of leaders and amongst the sort of government entities on this and they put in a program called the Pass-Through Assistance Program, which is what PTAP stands for. And basically, it allows servicers to finance P&I advances. And it's meant to be stated in their program. It's stated as sort of a financing of last resort. So they want you to sort of try to get other sources of financing, but if you don't have those, that program is available. Enrolling in it is fairly straightforward. There's a number of different sort of forms you have to fill out and complete. It's available onetime per month and it's got to be repaid in about 7 months. And so that's kind of the structure of the program. And it's -- interestingly enough, it's 100% LTV, which is terrific for those of us who are looking for financing, and it's at market rates. The information is not public on how extensively PTAP has been used. So I thought I -- what I would just talk about is our experience on the NewRez-Shellpoint side and say that we have not used it yet. We don't think we're going to use it in June. And really the primary reason, we thought we were originally in our original modeling, and I'm going to refer to what Bill talked about and Lee talked about, and that is that even though we've all been modeling sort of every day, the -- some things have changed. More people have been paying on forbearances than we thought there would be. We've gotten some wonderful help from the FHFA and Fannie and Freddie and so -- and in this case, in Ginnie. But things have performed a little bit different than we thought they might. But the main driver on the not needing the PTAP program, so in other words, not needing P&I advance financing has really been the fact that prepays have stayed high. And because prepays have stayed high, higher than at least we expected, it's allowed us to not -- to utilize the prepays to offset the required P&I advancing. But we're ready to use it. We really like the program. Again, I want to be sort of commend Ginnie on their sort of getting out early on that, and I think it's a terrific program. So...

Okay. Great. Actually -- and just touching on the issue of prepays, can you just walk through the mechanics briefly about how the prepays offset -- can be used to offset the advances? Jack, do you want to do the Ginnie side? And then I'll ask Bill on the other side?

Yes. It's really pretty simple, and that is if that -- if you prepays up until the remittance date, required remittance day for Ginnie 1s and Ginnie 2s, again, a little technical, but it's a little -- the days are a little different. But basically, before mid-month-ish for most of these, you can basically utilize the prepays to offset your P&I advances. It's as simple as that. It's a pretty simple process. And -- but you cannot use it for other types of advances just P&I.

And Bill, on the GSE side, is it essentially the same process and the float levels are similar, et cetera?

Yes. Well, yes, it's essentially the same. There are some nuances about how the loan is pooled and how long you can keep the unscheduled principle for if the loan is sold at the cash window, then you have to remit the principal interest directly to the agencies the next day. For Freddie Mac, you only have a small number of days, 3 or 4, 5 days, although you can use the P&I for it to offset the advances. On loans that are so-called scheduled schedules, those who typically have, say, a month or thereabouts depending on when the loan prepays during the month and so forth, they can use the principals to fund your advances. Now it's true. It's not forever, like you have to pay that back the next month if that money has to go to security holders. And so there is a little bit of using each next month, each subsequent month to pay back the money that we used to correct the prior month to fund your advances there. So as long as prepayment stay high, you can keep doing that. As long as your pool is staying reasonably constant, you can keep doing that. And really -- the prepaid effect is a very powerful one. And speeds had to slow very considerably below 5, 6 CPR before that becomes sort of a problem.

Okay. Great. Actually, let's switch over to MSR valuations. [Operator Instructions] Joel, go ahead.

Could I just add one thing to what Bill and Jack were saying? Just to frame it slightly differently, the -- and it's implied. But in the context of principal and interest advances, that is a pool-level concept. So the unscheduled payments in the pool are available. In the context of taxes and insurance, that is a loan-level event so that recovery can only come through recovery of a particular loan. And I think that distinction is important and might help frame the way people think about it.

Okay. Great. That's very helpful. And then okay, let's switch over to MSR valuations. So Lee, let's kick it off with you. MSRs took pretty big marks in the first quarter. And since then, yes, treasuries have bounced around but been roughly flat. Primary rates have trended down a little bit. Just wanted to get your thoughts on where MSR valuations are generally trending.

Yes. And I can sort of take this from our own experience. I mean we actually had a good Q1. We saw strong hedge performance that offset the elevated prepay speeds. But I agree with you that primary rates have decreased a little in Q2. But we, Flagstar, do everything we can to protect our MSR asset, and that includes we hedge it. We have recapture that we undertake. And we expect our hedge to perform appropriately and offset the declines in asset value, which it has done historically. What I would say, though, is furthermore, the MSR value is going to be affected by market demand, which has been impacted by the uncertainty and volatility in the market because of COVID-19. So there were a number of variables that are going to impact MSR values, some of which you can counteract, and that's what we try and do. We'll put as many controls in place as we can to counteract as many of those variables.

Okay. Great. Actually, Joel, can -- same question for you on MSR valuations. Also, can you just remind us how you -- what you do with the MSR in terms of holding and selling, et cetera?

Sure. Well, in this market, it's kind of an easy decision for us. We, generally speaking, originate and securitize mainly agency and government loans. We like to sell servicing on a flow basis, on a co-issue flow basis. Historically, we have kept a piece of it, sold a piece of it. And generally speaking, it's, for us, price for multiple dependent. When someone talks about a multiple, they're speaking of a multiple of the net servicing strip that effectively becomes the price. And so we've generally tended to draw a line in the sand at price. Ordinarily, that's been kind of a blended 3 multiple for us. If the market was paying more than a 3, we're a seller. If the market paying less than a 3 multiple, we keep it. In this environment, it's, for better or worse, pretty simple because we're seeing agency bids in the 1.25 to 1.5 multiple range. That's down from 4 to 4.5, quite honestly. And in the government universe, we're really seeing no bid worth entertaining. In fact, without getting into specifics, I can think of 1 company that essentially will buy your servicing, but they'll charge you for it. So it's the equivalent of a negative yield in Germany on their tenure. And so that's a pretty unique and extreme concept. But hey, this is just another day in mortgageland. I mean sell me your servicing and pay me and we're good. So...

Okay. Great. And then actually, just on sticking to the MSR valuation theme. Bill, can you just talk about the -- has the forbearance activity impacted MSR valuations? And is that continuing to have any impact?

Sure. So I would say, first of all, I think our experience is pretty different than Joel just described. We think the MSR prices are higher than that. I think that there was a time some weeks ago where mults might have been that low, but we're not seeing that today. We have turned on all of our flow agreements again, and we're seeing competition reemerge from other flow buyers out there. And so I think we're seeing mults certainly above 2, probably in the mid-2s, at least. So I think in terms of valuation, it's been difficult to see visibility into some of these valuations because there's not been a lot of bulk trades taking place, certainly since March. At Two Harbors, we mark our portfolio by using broker [ markets ], independent broker [ markets ] so you don't model price or internally price anything. And so over at the end of March and even today, the brokers have generally taken the viewpoint that they didn't want to guess at what the prices might be in the absence of observable information. And so risk premiums have generally stayed unchanged for now until there is more visibility from observable trades. Now one trend that I think is certainly starting to occur and again, the cadence of marketing servicing is typically monthly. So if you look at the world in monthly time steps, this crisis just happened and it just ended. And so we'll see how things evolve. But one thing that is certainly observable is delinquencies, forbearances increased servicing costs and diminution of servicing cash flow from the strip as those borrowers in forbearance are not paying. So I do expect, at a minimum, those amounts of effects to make their presence felt in MSR prices in the near term. And I think as time passes, as forbearance rates become more visible and this whole thing unfolds and there becomes more comfort in the market and more trade started occurring in the marketplace, then pricing visibility will increase. And we'll incorporate all of that information.

Okay. Great. And actually, that kind of leads into the next topic, which is just the operational side of mortgage servicing. Jack, let's start with you. How are you guys dealing with the higher forbearance activity that you're seeing? And then what percentage of your borrowers in forbearance need a live interaction, can do it -- done it electronically?

Yes. I will echo some of the comments that Joel made early on. It's been a wild ride, that's for sure. Just a short, sort of geography statement. So NRZ, today, is the owner of almost $550 billion of MSRs and almost 3.5 million service loans. We're one of the subservicers at NewRez. We service about 1.2 million loans for NRZ and about another 400,000 loans for others as part of the special servicing business. And so when I put these comments in perspective, I just want to mention that I'm going to talk about it from the NewRez servicing standpoint, not from the NRZ standpoint. So if you guys don't mind, I just -- thanks for listening on that. So our primary focus, really as of the second week of March or so, has really been about these homeowners, the situation the homeowners are in, how do we get them through the process. And if I could just put this in perspective for everybody, we'll maybe -- we have a lot of delinquent loans because we take on those loans for our corporate clients. And we'll maybe do, on our portfolios today, maybe 5,000 work outs of a month, something like that. And so that's 60,000 work outs a year. We've gotten request for forbearance in the last 2 months of 180,000. So if you can get some idea of how big a deal it is and how challenging it's been from an operating standpoint. And at the same time, we've moved all our staff to work-from-home. So I'm sure everybody on the panel can really empathize with these kind of circumstances. It's really been unprecedented times. And so basically, what we realized early on about the first week of March was that we really needed to find a way for them, for borrowers to self-identify. And so we've -- we already had some pretty good web technology on the default side so we sort of beefed that up, and that became a way that borrowers could self-identify as being COVID-affected. They could understand what it meant to be on forbearance and how it was going to affect them, and they could request the forbearance, all at the same time in a pretty expedient process. And I'll be honest, we've done 155,000 forbearances at this point. We've gotten 180,000 requests and -- on ours and within our servicing operation. And without the benefit of technology, about 70% of those borrowers, those 155,000 forbearances were really done mostly online. In some cases, we talked to those borrowers but done mostly online. So that's the answer to the how many question, Bose. And it's been invaluable. We pride ourselves on the human interaction with borrowers and how do we help them. But in this situation, without the help of technology, we really could not have gotten it done, to be perfectly honest with you. And Fannie and Freddie and Ginnie were great helps in what they felt like we could do online and what they would support. And it was -- it's really been quite a terrific story. The average time to get a forbearance in our organization is about 6 days, and we're pretty proud of that. I mean I'm pretty proud of the team, candidly. We've increased staff by about 35% to the -- answer to the sort of the staffing question. Some of that's been variable labor, where we've sort of outsourced. We don't outsource internationally, but we have outsourced some stuff in the U.S. and called on some of our partners to help us with that. And some -- most of that has been permanent. So I think I covered most of the sort of the topic from our perspective.

Okay. Great. Actually, Lee, can we switch over to you and -- the same question, operationally, how are you seeing things?

Yes. Sure. Well, we definitely moved to get ahead of things. I mean first of all, we followed the CARES Act, which allowed borrowers to opt into a 6-month forbearance with the option for a second 6 months. So if you just started with 90 days, you're probably swimming upstream trying to deal with all of those borrowers right now. We also gave borrowers, similar to what Jack said, the option of opting in via the website as well as via the telephone. And we saw about 50% of borrowers use the website option. We've hired collections and loss mitigation staff and utilized vendors who can help in these areas. So that variable cost model that Jack also mentioned, we're utilizing. We saw it pretty limited a little bit, but we didn't see catastrophic erosion in our call center SLAs. And I think the website option helped balance things out so that everybody wasn't coming through the call centers. We've also stood up a task force to get ahead of the loss mitigation effort, and we're proactively reaching out to borrowers who are in forbearance just -- rather than waiting for the time period to expire. We're getting ahead of things. And I think, as everybody in this call has mentioned, we're definitely seeing a flattening of the curve in the rate of new forbearance requests. And so we feel pretty good about where we are right now.

Okay. Great. And actually, Bill, one question for you on the same subject. Can you just talk about how your subservicing agreements deal with the potentially higher costs? So how does the operational stuff impact you as a -- as an MSR owner?

Sure. So I mean, I obviously think everyone is aware that when loans normally go delinquent, that obviously comes with a higher cost of service. It's the current -- let me back up for a moment. So we use only a subservicing model. We outsource all of our servicing to 3 different subservicers. And the cost for current loans is, say, generically around $6 to $7 per month [ per month ] and delinquent costs are typically significantly higher than that. Now forbearance is -- as deemed by the CARES Act, is different. That's not something that was contemplated. Normally, it's not part of the regular fee schedule that subservicers typically have. It obviously requires more work than servicing a current loan, but presumably not as much as servicing a -- an old-style [ private ] loan in terms of trying to get them to pay and try to find them and try to encourage them to get a plan [indiscernible]. This is something you put them in a plan, you check back later and so forth. And so as I mentioned before, in monthly time scale space, this just happened. We are in active conversations with our subservicers about what those costs should be. Again, this thing just happened really. And so we're working through some of that. But certainly, it's supposed to be more than servicing a current loan and not as much as a regular way to [indiscernible].

Okay. And let me just bring Joel in on this as well. Just can you just talk about your costs and how this is impacting you operationally?

Sure. We are -- we outsource our servicing to a single servicer. Our portfolio is about $15 billion. It's probably 55,000 loans or so. And there is an increase. To Bill's point, it's not as extreme as a classic 90, 120 foreclosure bankruptcy-type loan, but it is expensive. It's a lot more expensive than the current loan. And it feels like the avalanche of calls is slowing. So the servicer is entering into a period of somewhat of a hiatus of servicing a bunch of these loans. I only have, out the 55,000 loans, probably 5,000 loans in forbearance. So it's not extreme. But the subservicer has a very, very, very large collective portfolio of a number of owners and so I'm sure they're [ overloan ]. They've done a very good job, by the way, of communicating and working from home and doing all of the stuff you would expect.

So -- okay, great. Actually, I do want to get to some questions, but just one last. I wanted to just touch on the regulatory side. And you guys have already sort of addressed a number of these issues. But just broadly, do you feel like the regulators need to do more? Or has the problem kind of been addressed in terms of advances, et cetera? And just -- I'll just start with Jack and work my way across. Jack, can you -- want to start with that?

Yes. Sure. I mean I think it's really important to recognize how much the regulators have done, how much FHFA has done, how much Fannie and Freddie and have done, how much Ginnie has done. I think between the PTAP program, the reducing delinquency -- reducing the requirement and delinquency thresholds at Ginnie and a number of other provisions they've made on the partial modification program, just the various announcements and bulletins that Fannie and Freddie have made, I think they've done a lot. And I think, Bill said this really well, that I think all of us are in the stage where if you asked us to predict the future, it would be really hard. But what we're all trying to do is look at the numbers every single day and see where the trends are going. And again, some of the trends are really clear. We mentioned this earlier. The number of forbearance requests has moderated. Prepays have stayed pretty high. The number of people, the percentage of people paying on forbearances to current has been much higher than any of us would have projected. And I think if we see ourselves in a situation where we can see unemployment moderating and ourselves getting back to work, I think you can make a case that with continued support and conversation with the regulators, they've done a lot, and we're in pretty good shape. I will just say one thing and I'll turn it onto the other panelists because I'm sure a lot of other people have things to say. But to some extent, the real work in this process has not really started. It's the work out of these individual forbearances and how do we ultimately resolve the forbearances and hopefully, a large percentage of people can go through the deferment process that Fannie and Freddie announced. That's going to be a tremendous program, very, very important. And then other people we're going to have to modify. But there's going to be a lot of one-on-one interaction required and a lot of work that has to be done in getting people from where they are today, which is on a forbearance, to a permanent work out. And so that -- and there's got to be a lot of interaction between the regulators and ourselves. And I think there will be. And I think they've shown themselves to be very interactive and supportive.

Okay. Great. Any view on -- do the regulators need to do more?

I largely agree with what Jack said. They've really done an awful lot. I mean I guess one thing that I would have preferred is if they would have given us some clarity as it was happening more in real time rather than announcing some borrower programs and then waiting, letting people like us sort of sit in our own sweat for a little while until we figured out how the reimbursement mechanism is going to be. But as things have played out and unfolded, I think things seem pretty good. Obviously, should there be a change in the forbearance take-up rates here, that might change my opinion about whether they need to do more or some of that. But as things are and if the trends that we're seeing stay in place, I think it's just fine, and they've done an awful lot already.

Okay. Great. Lee, how about you?

Yes. I would agree with that. I mean things moved incredibly quickly as we all know. So the CARES Act was introduced. And I think there could have been more educational material provided to the borrowers, in particular, about what it means to opt in to forbearance. And so what are some of the other effects or impacts of doing that. But again, I mean, I appreciate that things were moving incredibly quickly. And so you sort of solve 1 problem and then you moved on to the next. And by and large, I think the agencies and the regulators have done a decent job. I think the CFPB have particularly done a good job in terms of educating borrowers in terms of what it means to be in forbearance. So they put a nice video out on their website after we -- the COVID sort of situation really did take hold. And I've been very clear as well about doing what's in the best interest of the borrower. And if you do that, that's all anybody can ask. So there's a couple of pockets where we're waiting for a little bit more information to come out from the agencies. And I think we've all alluded to that in some way across this panel. But overall, I think the agencies have done a good job and the regulators, given the speed with which things moved here. And when you sort of think back to the last big downturn of 2008, 2009, I think we're all much better prepared, and we're all much more clear on what the expectations are.

Okay. Great. And then, Joel, a quick 30 seconds for you or...

Yes. The most underaddressed topic is Ginnie Mae, T&I, advance liquidity. That is an underaddressed, unaddressed topic, and it's a sleeper. It can be very, very large. And there's absolutely no facility. Main Street, which is a treasury facility, the directions around which have not yet been announced, is possible for a middle-market and other companies. But what mortgage banks are afraid of and don't know enough about Main Street yet is the rule that for lenders, that the debt that they undertake maybe have to be pari passu with warehouse debt, and that could be a problem. So if Main Street comes through, I think there is a chance that T&I, advance liquidity is taken can care of. Otherwise, I think that's the most underaddressed issue by regulators.

So -- okay. All right. And then actually, let's -- we've got a few minutes left just for questions. So let me just go through a few questions. First one, could you -- how much of the May books and forbearance are continuing to pay? I think, Bill, you gave a number for your portfolio. Can you just repeat that? And then I'll ask the others as well.

Yes. 44% of all of our borrowers in forbearance have made their full May payment.

And Jack, what was that for you?

We got Bill beat by 1%. I'm really proud of that. We're at 45% for -- basically, the same number at 45%.

Lee -- and Lee, is that something you have as well or disclose?

Yes. I have to be somewhat careful here because we're public. But at the end of -- when I was on our earnings call and we were looking at April, 50% of the 110,000 borrowers that had elected to opt in to forbearance, 50%, 5-0, had made their April payments. And I think I can just say I'm encouraged by what I've seen in May as well without disclosing any numbers.

So -- okay. And then there's a question -- sorry, go ahead.

I'd say one thing out first. This is Jack. But I'll say my data was as of May 21. I don't know what you're doing with your [ facts ]. So let's just [indiscernible].

The -- and then the -- actually, there was a question on MSR valuations this quarter. I mean I think you've alluded to that in some of the earlier comments. But just excluding hedges, is -- are MSRs -- are they up or down this quarter? I'll start with Lee. Any comments there?

I mean look, I think that they're under pressure. We -- again, as a bank, we hedge it. And so we do everything we can to protect that asset class. But for some of the reasons we talked about earlier, given your -- the question that you asked, there's no doubt that the valuations are under pressure. And it's going to be around just prepay speeds is number one. And the fact -- and Bill alluded to this as well, the market for MSRs has dried up since COVID-19, and it's -- we're beginning to see come back, but there have been no bulk deals, and there've been a lack of flow deals for the last 2 months. And so when you combine the elevated prepay speeds and the lack of market activity, there's no doubt about MSR valuations being under pressure in Q2.

Bill, anything to add on that?

Yes. I would just distinguish between prices being lower potentially because of interest rates being lower, right, which is an expected part of the price movements of the thing versus what I would call risk premiums being higher, spreads being wider because of lower demand or something like that. The things I mentioned before, maybe increased delinquencies, less cash flow, that's also what I would call more mechanical. That's just what -- at the same risk premium, the price will go lower because of that without any statement about the valuation of the [indiscernible]. And so I think we need to see where all that shakes out. And again, [indiscernible] you see more visibility from bulk trades and flow trades in the market, which is starting to come back slowly here.

Okay. Jack, anything to add there?

No. I'm lucky enough not to be involved in MSR valuation in our world, so I'll leave it to the experts on the panel.

Okay. Great. And then actually, there's a question just on MSR sales activity. And you kind of alluded to that as well. Can you -- actually, Bill, maybe I'll just start with you since you'd mentioned that. Can you just talk about what you're seeing in terms of activity, both in terms of bulk and then the co-issue, any signs of resurgence there?

Yes. So I'll reiterate what Lee said, there's been no trades that I'm aware of from the brokers on the bulk. On the conventional side, we actually don't own any government servicing, so I can't actually speak to that. There was one [ trading ] in the marketplace recently that did not trade. So there's not a real lot of information to be had from that. Again, I think our interest rate volatility has been low. Mortgage TBA volatility has been low recently. So I expect that as the forbearance experience becomes more stable -- again, these announcements from the FHFA and GSEs, again, really just happened days ago, maybe a week ago or 10 days ago. So that's provided a lot of certainty and a lot of comfort, I think, to market participants. And so I think I do expect there to be more activity in the near term as people have had become more comfortable with the environment. As I mentioned on the flow side -- and sorry, what Joel said earlier, certainly, in March and April, I think virtually every flow participant stepped back from the market and wasn't taking new loans or had set prices to 0. That's been changing. We have opened up our flow program fully again. We are taking lots from every one of our flow sellers, from every one of our subservicers. As I said, [indiscernible] rumors that other participants are [indiscernible] So there will be more price discovery as the factor that I mentioned come into play.

But to Bill's point, even at a 2 multiple on agency servicing, I mean, that ought to tell you a lot. I mean that's a significant discount to historical norm.

Well, except -- I would add, except 10-year treasury rates at 70 basis points.

I agree.

So the price should be lower by some amount.

But before March, it was still now pretty low rates, and prices were very different.

The -- actually -- and our time is almost up, but just let me just throw out one last question we got. Again, this is on the regulatory side. Is -- do you expect the regulators to do more in terms of a borrower sort of modification programs and things like that? Or would that depend on how the scenario develop in terms of the need for that. Let me start with you, Lee.

Yes. So I think some agencies have already done a lot. I mean we've talked about the FHFA and the deferral option. And as I mentioned, there are still some other pockets where we're looking for a little bit more information. So I can talk about this. The USDA, for example, we would like a little bit more information to come out of the USDA regarding what some options or other options may be for borrowers. But I think generally, the agencies have done a good job of trying to provide optionality and streamline options for the borrowers with an eye on what it means for the services as well because as Jack said, the real [indiscernible] now, the real work is going to be 90 days from now and beyond that. And so the more sort of streamlined options we have, the better experience it's going to be for the borrower, and it's also going to help the servicers and the MSR owners as well. So generally, I think it's been good. But as I say, there are pockets where we're looking for a little bit more information and optionality to come out.

Okay. Great. Actually, our time's up, but if anyone wants a last word on this? Anyone want to last comment on this question? If not, well, thanks very much to the panelists, and thanks a lot to the audience for joining us today as well. So -- and everyone, have a good day.

Thank you.

Thank you very much.

Thanks.