Orchid Island Capital Inc (ORC) 2020 Q1 法說會逐字稿

Good morning, and welcome to the First Quarter 2020 Earnings Conference Call for Orchid Island Capital. This call is being recorded today, May 1, 2020.

At this time, the company would like to remind the listeners that statements made during today's conference call relating to matters that are not historical facts are forward-looking statements subject to the safe harbor provisions of the Private Securities Litigation Reform Act of 1995. Listeners are cautioned that such forward-looking statements are based on information currently available on the management's good faith, belief with respect to future events and are subject to risks and uncertainties that could cause actual performance or results to differ materially from those expressed in such forward-looking statements. Important factors that could cause such differences are described in the company's files with the Securities and Exchange Commission, including the company's most recent annual report on Form 10-K. The company assumes no obligation to update such forward-looking statements to reflect actual results, changes in assumptions or changes in factors affecting forward-looking statements.

Now I'd like to turn the conference over to the company's Chairman and Chief Executive Officer, Mr. Robert Cauley. Please go ahead, sir.

Thank you, operator, and good morning, everyone. I hope everyone is safe and has been -- not been adversely affected by the pandemic and COVID-19. Hopefully, by next quarter's call, this will be largely behind us. However, it was all about COVID-19 this quarter.

I'm not going to follow the prior practice that goes through the slide sequentially. The events of the quarter don't really lend themselves to doing so, so please bear with me as I flip around slightly. Instead, what I'm going to do is focus on 6 main points and use the slides as needed in an effort. In any event, I will not cover all the slides, and again, just focus on the ones needed to support my points.

The 6 main points I'm going to focus on are: One, I'll briefly go over what happened in Q1, don't dwell on that too much. Second, I'm going to pause and just kind of revisit our investment strategy and say a few words about our strategy versus that of other mortgage REITs, try to draw some contrasts. Third, I'll come back and review our results -- I'm sorry, not our results, but what we did as these events unfolded during the quarter. Then I'll go over our financial results. And then the focus of the call hopefully will be our positioning, where we stand today, how we look at the world going forward and our expectations for rates, speeds and spreads. And then of course, we'll open the call up to questions.

So to it, I guess we could start on Slide 7, that's as good as any. It just shows you what happened to the treasury curve and the dollar swap curve. And the first point I want to make is that the financial -- as the crisis unfolded, the financial markets were the place where it was first -- took place. The stock market and then quickly the fixed income markets reacted very quickly, as the scope and the demand or the magnitude of the steps needed to contain the virus became clear and the effect of the economy, it was very obvious that it was going to be quite dire.

And so the first thing that we saw was a demand for cash and liquidity. This was driven by people anticipating losing their jobs, businesses losing revenue, investors seeking to protect against losses. And so there was a rapid rush to build cash balances. This, in turn, caused a lot of deleveraging on the part of levered investors or even in other asset managers that aren't leveraged, just redemptions. And so we saw a huge rush of sales that took place. This was mainly in mid-March.

And as you might expect, the markets that were impacted first were the most liquid or the markets that were impossible to gain positions or at least the smallest losses. As a result, this impact was the first immediately felt in treasuries and agency mortgage-backed securities. This selling then, of course, triggered more selling and more deleveraging and forestalling to meet margin calls.

So if you go to Slide 10, I'll give you a second to get there. And I think on the top left of this slide, it kind of paints a nice picture. And before I do that, just one final word on Slide 7. I mean it's interesting to note that the -- both curves, in the case of the cash treasury curve or the swap curve, essentially all rates are below 1%, and most rates are at all-time lows.

And then if you look at Slide 10, you can see it as the agency mortgage market. And these are just TBA prices over the course of the quarter, and they were gradually trending up. But even with rates falling in March, TBA prices actually went down. The performance of TBA mortgages versus hedges couldn't have been worse. We had days where the 10-year treasury was up 0.75 point and certain mortgages were down 0.75 point. So obviously, these liquid markets were the source of most selling. And on the slide on the right, we show pay-ups for spec pools. And these, of course, collapsed. The most high-quality spec poles are traded at modest premiums from basically record highs, as you can see on the chart.

And so given the magnitude and the significance of these 2 markets to the financial system, the Fed intervened very quickly. It was basically in 3 steps after initially lowering rates on the 3rd of March. On the 15th, they take more drastic steps. They lower the effective fund rates. The range to basically the 0 bond, the effect of 0 bond 0 to 25 basis points. And they announced asset purchases, $700 billion, $500 billion of which would be in treasuries and $200 billion would be in agency mortgages. So again, this was on Sunday afternoon. They announced this March 15.

By the end of that week, it was very obvious that this was not enough. As the Fed would come into the market and offer to buy securities, they would be met with offers to sell many times greater than what they were looking to buy. And it became clear that more was needed. In fact, really, by Friday, the markets were damn near ceasing. It became very difficult to sell assets.

On Monday morning, on the 23rd of March, basically the Fed announced they were going to buy whatever it took to stabilize these markets. And if you look on Page 10, you can see that in the case of TBAs, the market quickly reacted. We -- they started out with very large asset purchases, well over $100 billion per week. They've since tapered those. They're running at about $8 billion or $9 billion per week now. But importantly, they announced this week, during their meeting, that they are going to pledge to support these markets and do basically all that's needed going forward. And so you can see, at the end of the day, for the mortgage market, we ended up with very decent returns.

At this point, I'd like to pause and just kind of talk about Orchid's investment strategy and really in contrast to our peers. The agency market is obviously a very large market, second or third largest market in the world. And Orchid, since inception, has only been an agency REIT. But more specifically, even within the agency space, we don't buy Agency CMBS or multifamily. Very strong focus on pass-throughs, very few CMOs typically. We've used derivatives, IOs and inverse IOs in the past, less so today, but a very narrow slice of the agency market. And really, it's mostly spec pools, Fannie, Freddie and even very little Ginnie Mae's for that matter. So this has been our focus all along. It's a very narrow slice of a very deep market. And it mean -- it was meaningful given the events of this quarter.

So now I'm going to bounce again around, I apologize. I'm going to go to Slide 13. Slide 13 just shows the components of the aggregate index. And on the bottom of the page, we show '19, 2019. On the top was Q1. A very sharp contrast here. 2019 was all about what we call risk on. Risk appetites were large. Stock markets were making new highs throughout the year. The trade war ended. And as you look at returns across the different sectors, Agency MBS had a decent return that dragged meaningfully all of the higher forms of credit, investment-grade credits, emerging market investment grade, high yield, and of course, equities. In short contrast to the first quarter of 2020, just the opposite is true. And agency mortgages did very well with a positive return.

If you'll turn to Slide 30, it's in the back, it's one of the appendices. This is data that we get from BAML. And on the left-hand side, it shows securitized product returns, and they're sorted by their total returns for 2020. I don't want to dwell on this slide for long. But what I do want to point out, if you look at the top left, the total return in the first column, US Treasuries were 8.8%. Mortgage -- agency mortgages were third. There are only 3 categories that had a positive return. And if you look at the other sectors of the fixed income market, one, they're negative but also very meaningfully so, in many cases, double digits.

And so the niche the Orchid has carved out for itself, and we've been investing in ours since the beginning, has been very beneficial for us. This allowed us to -- this tight focus really is what allowed us to suffer much less book value erosion than most of our peers and also to recover quite quickly once the Fed intervened. Not only that, but even though we were able to -- we did -- we sold assets and reduced the size of the portfolio. The NIMs in the market, net interest margins available today are very, very attractive. Returns on equity are very compelling, mid-teens, if not higher. And even with a slightly smaller portfolio, once we've realized the full benefits of our lower funding cost, our earnings could rebound back to where they were. But I'll speak to that more a little further in the call.

Now my third point, which is basically just to go over what we did as these events unfolded. As they unfolded, our focus became very much on 2 things: basically maintaining our leverage ratio at a reasonable level in our range of 8.5% to 9.5%; and then very importantly, maintaining adequate cash and liquidity.

The first steps we took were to reduce our hedges. We reduced our hedges meaningfully. As rates rallied very quickly, it became very apparent to us that as more businesses and states shut down, the economy basically shut down as well. The outlook for the economy, both domestic and globally, was very obviously very dire. So we expected rates to stay low. So the first thing we did was to sell -- or reduce our hedge positions to reduce the margin calls associated with those.

Over the course of days, not weeks, as asset sales grew in magnitude, frequency became more distressed. An instance of distress would be early settle, T plus 1, T plus 2, even assets trading on weekends. Our asset prices collapsed as did spec pools pay-ups. As I mentioned, the performance of the asset class versus hedges was probably the worst we've ever seen. And of course, you would expect margin calls increased, and repo counterparties became very -- the angst was high. Many REITs of our peers who had exploited a credit were not making margin calls and in case -- many cases, had entered into forbearance agreements with their counterparties.

And so this level -- raised the level of angst amongst our repo counterparties. And then this as an aside is a byproduct of financial crisis of 2008. Most REITs today operate with MRAs, which has really advantageous language for the seller, and that is that they are the final arbiter of price discrepancy. So when the margin calls came in, it was much more challenging to fight these margin calls.

So in order for Orchid -- what we did, that was the backdrop. So what did we do? So to stay ahead of every -- with all the developments to maintain our leverage ratio in the range that we wanted it to be in and to have plenty of cash to meet potential margin calls, we began to sell assets on the week of March 16. And we completed all this by the end of the week. We sold that week about $1.1 billion of mortgages. We realized losses of about $30 million.

We -- our sales were focused on lower-pay-up specified pools or faster-paying bonds. And the reason we did this is 2 reasons: one, with spec pay-ups collapsing towards TBA, in some cases through TBA, we were selling the bonds where we would realize the smallest loss on sale but also realizing that next month, we would have pay-downs and margin calls associated with those. So we wanted to rid ourselves of any obviously fast-paying bonds.

But to summarize what we did throughout this is I want to make 3 main points: One, we satisfied every margin call. Orchid did not miss a single margin call. And in fact, once we got pay-downs on the 25th of March, we were able to maintain our unrestricted cash and unencumbered assets in proportion to our total assets or equity at basically historical ratios. So we were very comfortable with our cash position.

And then finally, I just want to make a point that all sales were initiated by management. In no case was any sale of any asset triggered by a counterparty. So we -- everything was done at our discretion.

Now I'd like to go over our financial results. If you will turn to Slide 4, please. For the quarter ended March 31, 2020, Orchid had a net loss of $1.41 per share. Earnings per share of $0.27 positive, excluding realized and unrealized gains and losses on RMBS and derivatives, which includes net interest income on our interest rate swaps. We do provide a reconciliation of this on Page 31, which I'll get to in a moment.

With respect to unrealized and realized gains, we had a loss of $1.68 on these instruments, both assets and derivatives. Our book value per share at March 31 was $4.65, a decrease of $1.62 or 25.8% from $6.27 at the end of the year, 12/31/19.

In Q1 2020, the company declared and subsequently paid $0.24 per share in dividends. Since our initial public offering, we have declared $11.16 in dividends, which includes $0.055 declared in April. Our economic return for the quarter was negative $1.38 or 22%.

As I mentioned, when the Fed intervened, our market has recovered quite quickly. That is carried on into April and through the end of April. And through the 29th of April, our book value is up to about $5.14, plus or minus, which represents a return or an increase in book value of 9.4% to 11.6%, so the market has recovered quite quickly.

Now I'd like to turn to our positioning going forward. So we've described what's happened in March, and everybody knows it was quite dire. So the question is what do we do and how do we look at the future from here.

First, the details. Our capital allocation is now 88.2% pass-throughs versus well under 80% at the quarter -- or end of the previous quarter and actually between 60% and 70% over the prior few years before that. Our pass-through allocation is almost all specified pools with a large bias to the best forms of call protection. We've reduced our exposure to CMOs, and our IO and inverse IO positions have been reduced in the case of the inverse IOs eliminated.

Now I'd like to go to Slide 21. And this is just the details of our portfolio as it stood at the end of the quarter. It does not look meaningfully different than the end of the year. A few points I want to make. We basically shifted down in coupon. And remember, this is a smaller portfolio. So it's $2.9 billion pass-throughs versus $4.1 billion. We did shift down in coupon slightly some by just selling more higher-coupon assets than lower. The weighted average coupon is down a few basis points to a little under 3.9. Our greatest concentration is now in 3.5 securities that represent 44% of the portfolio. That's up meaningfully from 31% at the end of last year. The allocation to 4s and 4.5s was reduced. In the case of 4s, almost reduced by half, and 4.5s slightly. The allocation of 5s has not changed meaningfully. And we did, as I mentioned, sell CMOs. We had a little under $300 million at the end of the year. That was reduced to $173 million. And the assets sold there were a lot of fast-paying CMOs.

Slide 22 is the detail -- is little more granular detail on our specified pools. And this was a slightly different presentation than we've shown before. What we show on the left column are the types of pools, $85,000 maximum balance, $110,000, 100% New York loans and $150,000. We call these higher payout or higher quality specified pools. And then other specified pools with less call protection, slightly larger loan balances, high LTV, FICO and so forth. And the point I want to make is we have 86.4% of our pass-throughs are in very high-quality protection -- high-quality specified pools with high call protection and almost nothing in the generic bucket, which would be kind of TBA or TBA like. That was as of quarter end.

Since quarter end, we've actually added to the portfolio approximately 2/3 of return of leverage. And what we did there was add very low pay-up specified pools. And the reason here, this is somewhat of a defensive step. If you recall, when we talked about our -- what we did as the crisis unfolded, we sold lower-quality pay-up pools just because they traded closer to TBA and we were trying to minimize the losses on sales. Basically, we're just replenishing that bucket. Some of these assets are very de minimis pay-ups, some form of call protection that may last a few months, maybe a quarter or 2. But in the event of a return of the turmoil in the market, these are assets that can be sold close to TBA levels without us incurring much of a loss.

With respect to our hedges, how are we positioned? If you go to Slide 26, we have a summary of our hedge positions. The top left, it shows our swap agreements. We do have one swaption in place. We've meaningfully reduced our euro-dollar exposure. And then of course, we have one final position in treasury futures. The notional amount of all these equals approximately 54% of our 3/31 borrowing balance or funding balance. We will probably look to build this up slightly. Going forward as volatility in the rates market has come off, actually, it's pretty much down to levels we saw before the crisis unfolded, probably look to use swaptions.

But just to draw a few points here. Another reason that when the hedge positions have changed is we really expect especially funding rates to stay near to 0 bond for quite a while, at least through the end of the year. You never know how the economy is going to unfold. Nobody really knows what's going to be on the other side of the pandemic and how much -- how quickly recovery -- the economy will recover. But we expect funding levels to be very low, at least through the end of the year.

Intermediate longer rates may drift slightly higher. There's kind of a balance that has to be drawn out between the treasury, which is going to be selling very large quantities of treasury notes and bonds versus the Fed, which has been purchasing them. I suspect that the treasury is going to issue more than the Feds are going to buy. So we may see some slight increase in longer-term rates, but we don't expect it to be truly dramatic. And as a result, we think using swaptions to protect movements on the long end makes a lot of sense. The absolute levels of rates are low, and the volatility is also low. So we think it's an attractive way. We also, of course, still have some IOs to protect us. With respect to our expectations going forward, as I said, we expect low-end rates to be quite low for the time being.

If you go to Slide 27. Slide 27 shows our interest expense by month. So it's more granular. This is not quarterly data. It's monthly going back at the beginning of last year. And we provide this granularity just kind of to show you the trend. And you can see that their interest expense in pennies per share per month has been trending down, and it peaked in the spring of 2019 to $0.15 per month, not per quarter. Looking at the most recent data for January, February, March, rates have come well off of their highs, in the mid- to high 2% range. And our interest expense was between $0.25 and $0.26 per quarter -- for the quarter of 2020 -- so this quarter.

As our repos roll off and we realize the benefits of the Fed's rate cuts, this number could drop to $0.03 to $0.04 per quarter. So a meaningful reduction. However, the portfolio has been shrunk as well. And the revenue side of -- on the assets will probably be down a fair amount as well. The key here, the wildcard in terms of our dividend performance going forward is going to be driven by speeds. Very, very high speeds result in high premium amortization that, of course, reduces our earnings. And then of course, the opposite is true if speeds are slow. So that's why we spent the time going through our positioning because what we're trying to do here is to try to protect ourselves as much as we can from speed. So to what do we think about speeds going forward? We've already discussed how we position for that.

If you go to Slide 14, I'll give you a second to get there. There are 3 charts on this slide. The top left just shows the refi index versus mortgage rates. In this case, the mortgage rate is the red line. And as you can see, it made a -- essentially an all-time low earlier this year. And it's since -- and since quarter end, it's come back down to that level. So rates are very low. And then of course, the blue line is the refi index. And it's, of course, spiked well above actually 6,000. But as you can see, it's come off some. So the question is why is this and what does this mean going forward.

So with respect to the crisis, the pandemic, it's had an impact on 2 things: One is the loan origination process. So whether it's social distancing, shelter in place, businesses being closed, it just impedes the ability for people to go through the process of looking for a home or refinancing and actually closing a loan, doing an appraisal and so forth. And so that probably explains that. If you look on the right-hand side, you see this is all the spread between the primary and secondary mortgage rate. You can see it's very elevated, much higher than it's been for quite some time. And this reflects the fact that mortgage originators are capacity constrained. They just have a difficult time dealing with the applications they're receiving.

And the second effect of the crisis is the loan approval process. And of course, as the economy is weak and we know that unemployment claims -- initial unemployment claims over the last 6 weeks are in excess of $30 million. Next Friday, we get nonfund payrolls. That number could be north of down $20 million. So that impedes borrower's ability to get approved. And of course, underwriting standards are becoming tighter. Recall, as a result of the financial crisis, many times when borrowers face difficulty and/or foreclosed upon, the new regulations are such that a lot of the burden is placed on the issuer of the loan and their assessment of the borrower's ability to make the loan. So as a result, it becomes somewhat gun-shy, and underwriting standards are actually tighter. So these 2 forces are going to offset what would otherwise be a very, very high refinancing environment where -- just driven by the absolute level of rates.

One final point. If you look at this bottom chart, it just shows the refi index versus the percent of the mortgage market that's in the money. And as you can see, the percent of the market that's in the money is in a multiyear high, yet the refi index is well off its high. So bottom line, the way we look at it, refi activity is expected to be high. In the next few months, it may be negatively affected, in other words, be lower than otherwise just because of these COVID-19 effects. But eventually, as the economy recovers, speeds should be quite high.

The next slide I'd like to say a few words about is 23. This is in addition to the way we look at speeds, which would just be straight CPRs. What we look at here are 2 things: The orange line is just the level of the 10-year treasuries. And you can see it's 67 basis points even lower today. The green line kind of is a representation of our pay-downs in dollars expressed as a percentage of the unpaid principal balance of the portfolio. And since it's a percentage, it kind of adjusts for the size of the portfolio, which varies over time.

And as you can see, when rates rallied starting late -- and really in late 2018, speeds became quite high. That's when we shifted the focus of the portfolio to much more call protection, higher-quality spec pools. And as you can see, our pay-downs have come down meaningfully. In fact, our portfolio, the pass-through portfolio prepaid at less than 10 CPR in the first quarter. And in April, which was released earlier this month, it did not increase meaningfully. So we are positioning ourselves to protect against speeds staying high for the balance of the year. And in combination with that, to the extent we're successful with lower funding costs positions us well, we think, going forward.

Let's see what else. So the final point I would like to talk about is spreads. I don't have a slide, per se, here. But if you look at mortgages, a common measure would be the current coupon spread to the 10-year or the 5, 10 years blend, still at very attractive levels, not at the wise we saw in March. In the case of the spread of the tenure, it reached 157 basis points. It's currently around 90. It's been as tight as the mid-70s. And you can look at it versus the 5, 10 blend, you get the same kind of story. In any event, mortgage spreads are still reasonably attractive. But given funding costs, agency mortgage-backed pass-throughs offer very attractive returns going forward. As I mentioned, the ROEs on these investments are between 15 -- even the very high-teens.

So we've already seen a fair amount of book value recovery as spec pool pay-ups have come back. But going forward, even though we did have to reduce the dividend last month just in reaction to the developments in March, there's the potential for our NIM to re-expand. And we look forward to trying to take advantage of our ability to protect against speed and maybe see some upward movement in the dividend going forward, but it's too early to tell.

And then finally, one other point I'd just make just with respect to the asset class. Generally how mortgages are going to perform in terms of how much tightening we see, it's really going to be a function of how we do versus comparable asset classes like investment grade and high-yield and CMBS and so forth. So for now, it's been doing quite well this month, and we hope that continues.

Operator, that is the end of my prepared remarks. We can now open up the call for questions.

(Operator Instructions) Our first question comes from Christopher Nolan with Ladenburg Thalmann.

Bob, leverage, the leverage ratio, kind of my calculation is about 9.3 or so, which is relatively high mostly due to lower equity. Are you guys looking to raise equity? Or are you looking to lower -- what are you thinking in terms of -- on leverage front?

Well, you're right, it was 9.3, but remember, we got a lot of book value back in the month of April. So -- and we did add some assets, but the effect of that is actually to bring it down. Our book value, as we say, is up and over 10% through the month of April. So that in and of itself brings our leverage down. The assets that we acquired on net probably -- it's even probably still down slightly.

As far as capital raising, obviously, that's just a question given where the stocks are trading. But the investment opportunities are very attractive. And even though that's -- this isn't addressing your question, per se. When we look at, say, for instance, repurchasing shares, do you want to sell assets to buy back shares? 2 points on that. One, asset prices are still recovering, so that makes me somewhat hesitant to sell a lot of assets just because I think they have upside price-wise. And two, the investment opportunities are so attractive right now.

Who knows what the future holds? But everything that we're seeing and hearing makes us think that our funding cost is going to be in the range of 25 to 30 basis points on the pass-through portfolio. And even with yields in the high 1s, low 2s, that's a very attractive NIM. And if you can contain speed, those are very attractive returns. So while capital raising isn't on the immediate horizon, if the opportunity presented itself, it would be an opportune time to raise capital.

Great. Given that you are experiencing higher investment spreads, assuming no change in leverage, what are you thinking about in terms of equity returns for the balance of the year?

Like I said, mid-teens, very attractive. We -- who knows? I mean I think of this as one wild card. It's -- we're seeing a lot of states reduce or eliminate restrictions. And so if we were to have a relapse and the markets will get thrown back in the turmoil, that could obviously change things. But absent that, it should be a very attractive year from this point forward. Obviously, the first quarter was quite challenging.

Sure. Final question. Turning to prepays. A lot of this turns on the ability of people to get mortgages in the first place. And given that the government is willing to do anything and everything to get this economy moving, do you anticipate where you can see regulatory changes, which almost like Community Reinvestment Act by the banks in the '90s, where they lowered standards relative to in the past where people to get mortgages? Do you think we can return to that? And if so, would that have a negative effect on prepays for you guys?

Well, 2 things. One, they've already done a lot. The forbearance program out of the CARES Act allows borrowers to not make payments for up to 6 months then into 12. A big concern that caused was obviously services having to advance against those. The FHFA announced that they would limit the services obligation to advance the 4 months, which is a big point for them in terms of liquidity. From that point on, the GSEs would advance. Importantly, the GSE has also said that they would keep the loans in the pool through the end of the forbearance period. Now at the end of that, to the extent these loans and these borrowers are unable to recover, you could see them enter or stay in default effectively. And in which case, they would have to be bought out. So you could see a spike in prepays.

But at the end of the financial crisis and the aftermath of that, there were a lot of steps put in place in terms of remediation steps that are mandated, that have to be taken to avoid borrowers losing their homes. And so the steps have been laid out. In this case, I suspect that somebody can't make a payment for 12 months. They're probably not going to be able to recover. They're going to be thrown into the modification pool, in which case they will see -- they'll go through the modification process. Their loan will be taken out of the pool. And they'll either have some principal reduction, rate reduction, whatever it takes to get the loan payment down to something they can afford or they ultimately do default.

But a lot of that's already in place. And the immediate effect for us is, to the extent they cannot make their payment, we'll still receive principal and interest. So it'll mute prepays with the caveat that you'll probably have a spike 13 months or 14 months from now.

(Operator Instructions) Our next questions comes from Jason Stewart with JonesTrading.

I wanted to add one thing before question. Great job going through what was a really difficult time. So this has been tough to watch, but nice performance.

The question -- Chris asked most of mine, but one on pay-ups and spec pools. If you think through your view on prepays, do you think that we see a relatively quick recovery in pay-ups? Or is this something that perhaps sticks around these levels and we wait 2 or 3 quarters until there's some clarity on access to mortgage credit?

This is Hunter. Yes, we -- the guys I talked to and transact with in the spec pool market, there's sort of a joke going around that we've seen a V-shaped recovery in spec pool pay-ups. And so yes, they recovered quickly. I expect them to continue to recover. My outlook is very positive on pools. We've seen even some of the REITs that were doing crazy things like selling on a Sunday afternoon have been in and had a very large appetite for adding specified pools. Dollar roll markets make it very easy to hedge these things. You can be long on one cash flow and short on something that's very, very similar and not have a lot of mortgage basis risk on the books. And in some cases, you can get paid for your hedge.

So short answer is yes, they have recovered quickly. And I expect them to continue to recover quickly.

Yes. The March cycle, they were very, very high. Of course, those collapsed. And our little picture in the slide -- I apologize, it's not that great. But they have come back 85%, 90% of what they were. They would be -- they should be higher because rates are lower, right? So they could potentially go through those levels. Next week, we'll get the auction cycle. And I would expect them to do very, very well. And don't forget refis in the most recent month were up surprisingly higher than expected, a very meaningful jump last month.

And so when people grapple with how speeds will evolve over the next few months, you have 2 forces working against each other. One is just absolute low level of rates. And the second one is the impacts of the pandemic, which tend to mute refis. It looks like, for now, that the former is seeming to win out. Most of the Street expects speeds to increase again next month. We'll see what happens in June and July, but everything that's happened so far has tended to support pay-up valuations. And then as Hunter mentioned, the roll market and the higher -- away from the production coupons, those are mostly negative. So it makes it very easy to hedge. So I would not expect spec pool pay-ups to do anything but be very strongly supported moving forward.

One final point on that is you have to remember that the Fed, since they stepped in and got involved in this crisis, bought $585 billion worth of the worst-quality mortgages. So that puts demand on the goods. That just shoves everyone else into that camp of looking for higher-quality assets that the Fed is not consuming.

Right. That's a good point. And then on Chris' question about -- or your answer, Bob, on repurchases of the stock relative to putting new capital to work. Your mid-teens number to me sort of is just levered cash on cash. It doesn't include any appreciation from basis tightening, et cetera. So one, is that correct? And then two, how does that change going forward relative to purchasing stock, assuming that does change?

Well, you're right, it doesn't include that. So the -- I think the assets have room to appreciate. And as we said, I think that we could see pay-ups go through the levels we saw in March. So selling assets to buy back stock even though we're trading at -- I don't know where we're trading at this very moment, but somewhere in the 80s percentile of book value, which is still a big discount. But still, those assets are going up in value plus the income-generating potential of them. I think those returns all in are actually higher. I try not to say those things on calls because then if it doesn't happen, people hold me to it, but they look very attractive.

We -- typically, when we trade less than 90% of what we want to buy back, this might be the exception where the returns on the assets, both the existing and from an income enhancement potential, are just too attractive.

Yes. I mean as evidenced by the fact that we put in our press release last night that we thought our return was somewhere in the 10% to 12% range for the 29 days that we -- of the quarter that we have been through so far. So that makes a pretty sharp justification for keeping the cash in the mortgage portfolio as opposed to buying the stock back at 10- or 15-point discount to book.

And I'm not showing any further questions at this time.

All right, operator, thank you. Thank you, everybody, for taking the time to join us. As always, if you have other questions or you happen to catch the replay, you can reach us in the office. The number is (772) 231-1400. We'll be happy to take your calls. Otherwise, we will talk to you at the end of the second quarter. Everybody, be safe. Thank you.

Ladies and gentlemen, this does conclude today's presentation. You may now disconnect, and have a wonderful day.