Great Ajax Corp (AJX) 2015 Q2 法說會逐字稿

Good day and welcome to the Great Ajax Corporation second-quarter and fiscal 2015 financial results conference call and webcast. (Operator Instructions). Please note this event is being recorded. I would now like to turn the conference call over to Mr. Lawrence A. Mendelsohn, Chief Executive Officer. Mr. Mendelsohn, the floor is yours, sir.

Thanks very much. I appreciate it and thanks, everybody, for joining us for our second-quarter 2015 conference call. Before I go through the presentation that we posted on the website, I wanted to go through a few highlights and themes of our second quarter. I don't usually show much emotion when I talk about our financials and quarterly achievements, but I really have to say we had a terrific quarter and we had a terrific start to the third quarter as well in the month of July.

We acquired about $200 million in purchase price of loans in the second quarter in July and over $275 million so far in 2015. We are at approximately 77% of what we've been, over the last year, been saying is about an annual guidance of about $360 million and we are at 77% of that in seven months.

Overall, our portfolio has been assembled through 96 different loan pool acquisitions and the purchase pricing has remained steady and in fact, on a price of property basis, has actually gotten cheaper. At June 30, we owned re-performing loans at 69% of property value and 77% of the principal balance and at the end of July, we bought another $46.8 million of RPLs at 59% of property value and at the same 77% of principal balance.

We own our nonperforming loans at 54% of property value, a very low number. Re-performers now represent over 80% of our portfolio as we said it would back at the end of Q4 as well as Q1 and nonperforming loans are approximately 20% and still we've been able to maintain more than 80% of the loans we own are in our target markets, our specifically targeted locations.

In Q2 and in July, we've targeted and achieved certain specific characteristics that we want for our loan portfolio. We specifically looked for lower loan to value loans and for adjustable rate and step rate re-performing loans. We like the lower LTV loans because they provide a good cushion for any frothiness in housing prices. In Q2 in July, we purchased significantly lower LTV loans and at lower prices relative to property value as well. We also increased our ownership of adjustable rate loans and step rate loans. These loans with built-in interest rate hedge protection now make up more than 40% of our portfolio and if we include July acquisitions, our re-performing portfolio is closer to 45% adjustable rate and step rate loans.

We are able to do this because of our ability to source loans from so many different sellers in privately negotiated transactions and those sellers have loans in target locations, not from auctions or large competitive bid sales. This network and the hard work we do to maintain and expand it gives us the ability to assemble our portfolio to certain specifications.

As I mentioned before, our June 30 portfolio represents 96 different transactions. If you look at some of the average numbers, our Q2 average loan assets were $355 million, $283 million of performing and $72 million of nonperforming. At the end of the quarter, however, we had $444 million, a $90 million difference between average loans on the balance sheet versus end-of-quarter loans on the balance sheet. If you just look at it in the most simple terms at say a 12% unleveraged return, that implies about a $0.17 a share difference from where the end of the quarter is versus the average of the quarter.

Similarly, on the equity and leverage side, if you look at 3/30, our March 31 numbers, equity to assets was about 0.69 and our debt equity ratio was about 0.43. So we had leverage of 0.43 where, at June 30, we increased our leverage throughout the quarter and we ended at a leverage ratio of 1.18. So more loans, more earnings per share unlevered and then we are able to put on non-mark-to-market leverage and securitized leverage during the quarter. Additionally, we did a securitization that closed July 2 and I'll talk more about that as we go on.

On the downside, buying in the private negotiated transactions that we do from many sellers creates some timing discrepancies as we see from the difference between average versus end of quarter. If we only bought at auctions, we know that auctions close when they are scheduled to close. 96 private transactions close when the seller can close them. And the seller can't close them until after they satisfy us from a due diligence perspective. As a result, we have less intra-quarter or intra-month control over the time of the closings than if we were to buy at auction, but we get to buy it cheaper and we get to buy what we want where we want it, not what a seller is selling.

While the timing of closings has a material effect on interest income in any given quarter as we talked about with average assets versus end-of-quarter assets, in any given quarter, the seller is the determinant of the timing, not us. That's one of the reasons why we focus very highly on creating net asset value rather than just earnings per share in a given quarter. It's more important to us that a transaction closes period than whether it closes in the first or last month of any quarter. The NAV increase that we get on a per-share basis gets created when we close the transaction without regard to which month we closed it in. We think we've created a lot of extra NAV, maybe as much as $3.50 to $4.50 or more per share and I'll talk about this more as I go through the presentation.

A quick summary of the timing of our second-quarter and July acquisitions. April acquisitions were spread throughout the month of April pretty evenly. For May, and May was our biggest acquisition month of the quarter, almost everything we bought in May closed in the last two business days of May. June was similar to May for acquisition timing in that it occurred primarily at the end and that is pretty typical for a quarter-end, that it occurs close to the quarter-end.

Almost all of July acquisitions also closed in the last three business days of July. We would prefer that these acquisitions close earlier in the period. However, when buying in private negotiated sales, the seller determines the timing because they need to provide us what we need before we are willing to sign off. Unfortunately, the sellers do that at their speed, not at our speed. We believe the portfolio specs that we put in place and the NAV benefits of buying in these private sales far outweighs any timing discrepancy created by a seller timeline.

Related to this is one material expense we outlined in the press release in Q2. We expense due diligence costs when we incur them rather than capitalizing them into the basis of the pools and amortizing them over time. This is a far more conservative accounting approach than what is typical. Most people use those as acquisition costs and expense them over the life of the assets. This expense shows up on our income statement under the category loan transaction expense.

In the second quarter, we expensed approximately $0.02 a share related to 550 loans that we kicked out of loan pools because of our due diligence results. These loans principally came from new sellers. One of the benefits of buying in private negotiated transactions is we get ample time to perform very thorough due diligence. While this is an extremely unusual kickout number for us, kicking out 550 loans, we think it was the proper decision for the long-run benefit of the Company. While there would be obviously some short-term income benefit by having the loans on the balance sheet in Q2, we didn't like the risk of these loans creating a very expensive tail.

If we do some quick income calculations, assuming an average loan purchase price of about $150,000, 550 loans implies about $82.5 million purchase price. The opportunity cost impact just from these 550 loans being kicked out on a net income available to shareholders basis from not buying the loans was approximately $0.03 per share per month, not including the cost of due diligence.

With that, I'm going to turn over to the earnings conference call presentation and go through it. I want to call your attention to page 2, the Safe Harbor disclosure. Our lawyers always insist that I mention it, so please read rather than me reading.

On the business overview, just a couple things to point out. Most of you have seen this before. We source loans in privately negotiated transactions from people all over our target markets. We've built our portfolio in 96 different transactions and it gives us the ability to tailor our portfolio to certain specifications like lower LTV, adjustable rate or step rate versus fixed, things like that. We use our own analytics that we've built over the years to come up with individual loan prices and individual target markets. We service loans our way as opposed to for the lowest marginal cost. We use moderate leverage. We are 1.18 at the end of the quarter and we were 0.43 at the beginning of the quarter. We've done four securitizations, the fourth one closing July 2 and on average, we've received about 2.3 times leverage from the seniors, although significantly more than that in the securitization that closed in July.

On page 4, a breakdown of our portfolio. Again, in the second quarter, we acquired assets with over $200 million UPB for $160 million; RPLs, $190 million of the $205 million; NPLs only $15 million; and again, re-performers represent over 80% of our portfolio right now. Our pricing has been pretty consistent. We have 2,620 mortgage loans with $598 million or $600 million of principal balance at the end of the quarter. We have 56 properties. Only two are rental. The other 54 are targeted for sales. We are not an REO to rent company.

We also declared a $0.22 dividend for the quarter, an increase over our dividend and part of our predictable increase in dividends over time to be the same as taxable income over time. Our GAAP net income was $0.36; our taxable income, $0.23. You may remember that nonperforming loans tend to be a little less tax-efficient. Our nonperforming loans, a big portfolio that we bought in November, started to kick in with its taxable income in the second quarter of this year.

On page 5 gives you a breakdown of our portfolio and where our property value is and it's pretty evenly dispersed. We have about $600 million of principal balance and we have about $660 million of property value. We tend to like lower LTV, we tend to like relatively larger average properties.

If you look on page 6, our unpaid principal balance. Again, 2,620 loans, an average UPB of $229,000. If we were to split housing into deciles 1 through 10, we prefer owning loans where the underlying properties are in deciles 4.5 to 7.5. We don't like the lower end. They tend to be lower value versus the fixed cost of both repair, as well as any servicing requirements versus -- and the higher end tends to have a little less liquidity. So we like -- the sweet spot for us is deciles 4.5 to 7.5 and that's where the overwhelming percentage of our portfolio resides.

If we look at page 7, I want to talk a little bit in depth on page 7. We've purposely grown our re-performing loan portfolio materially. We like the predictability. We like the way it sits relative to where housing prices have gone. Re-performing loans are a little less dependent on property values than nonperforming loans are. Note that our purchase price to property value is 69% and our July acquisitions, our purchase price to property value is even lower than that. The July acquisitions don't show on this page. This is just June 30.

As a result of our lower purchase price to property value percent, any increase in defaults has a very similar economic effect to an increase in prepayments. When you have a very low LTV loan and you buy it at a material discount to property value, default effectively becomes a prepayment over a few years as opposed to a long duration asset. So the low LTV loan in re-performing creates significant optionality. At June 30, over 40% of a re-performing loans are ARMs and step rate loans. We like the interest rate, the built-in interest rate protection they provide in our current interest rate environment and our July acquisitions, over 80% of what we acquired in July were adjustable rate mortgages and step rate mortgages.

On the nonperforming loans on page 8, the purchase price of unpaid principal balance is 56% and our purchase price to property value is 54%. We like lower loan-to-value nonperforming loans. It provides you a significant cushion against any change in how any negative change in housing markets and it provides you significantly more optionality in doing workouts with borrowers. As a result, we've maintained our nonperforming loans at about 20% of our portfolio and we've maintained and even lowered our purchase price relative to property values.

On the map on page 9, our target markets have not changed, but what has happened is the percentage of any particular markets has changed relative to our portfolio and probably the biggest increase, or the biggest change rather is in Dallas and Houston where Dallas and Houston declined to being a combined 2.3% of our portfolio. So while we are still relatively bullish, particularly in Houston, we, from a safety perspective, given energy markets and the decline in oil prices and what we think could be longer-term ramifications in those markets in housing, our portfolio now is only 2.3% in Dallas and Houston.

Page 10 is a complicated slide and I want to spend a significant amount of time on it because it's one of the things we pay very much attention to. We talk about internally that every time we send a wire for $1 million to a seller, we want to know that when the transaction closes for that $1 million that it's already worth $1.150 million. So that's kind of the mentality that we've put in our mind when we are buying assets.

If we were to look at where these what I'll call the fixed income market or the securitization market prices our portfolio, it's very different than where the equity market looks at it. We see this because we've done four securitizations and what we've done is we've separated into the average -- our execution on the average of the first three securitizations, which is the farthest to the bottom and then the middle, which is the execution of our securitization that closed July 2 and if we were to take our entire portfolio and subject it to securitization at these two different execution levels, these would be the approximate implications of it.

If you take our portfolio of $600 million at our cost of about $440 million and we apply, going to the bottom first, 50% seniors, 10% B1, 10% B2 at where the market would absorb those based on experience, we would end up getting proceeds where our entire portfolio we would have $32 million of cash remaining invested. Now we know that the -- and we would have equity of approximately -- equity face of approximately $180 million. We know from some unsolicited indications we've received that that equity face could be sold for somewhere between 30% and 50% of fixed. That would imply, if you just took the middle, $0.40, 40% of equity, $2.50 implied difference in NAV versus our book value, or our cost of those assets. If you were to do it at the high end at 50% on the $175 million of face, it would imply $3.59 of NAV, or $17.95 implied net asset value of the portfolio, or of the Company.

Now if we were to take our securitization that just closed and apply those outcomes, it has a material effect. You'll see a 5% change in securitization execution. It would take us from $32 million of cash basis on our entire portfolio to $3 million of cash basis. At 40% of the remaining bond face or the equity face of $150 million, that would be $3.50 a share. At $0.50, 50% on that bond face, it would be $4.50 a share. So that would imply the NAV could be somewhere between $18 and $18.80 based on our most recent securitization execution. And I'll talk more about that execution on a later slide.

Another way to look at this is what I'll call the aggregation difference between small and large portfolios. We actually saw today two portfolios sell and Citibank was the common party in the two of them. And they sold loans that had made 12 of 12 payments and based on their execution, their executions were between 15 and 20 points higher than where we buy re-performing loans, depending on which pool and their executions were on loans with lower coupons, similar LTVs. So it's what I'll call the small aggregation versus large sale where we see 12 of 12 payments on large sales for -- and the Citi pools were very large. One was $265 million, one was [$115] million.

And lastly, none of this analysis uses -- gives us any benefit from owning a 20% interest on our manager. That's on our books for zero and we think it has a value of somewhere between $0.50 and $.75 a share. So from a NAV perspective, based on market data, based on where securitization markets price our loans for execution and based on where large pools of loans are selling, we believe that our NAV is significantly higher than our book value. And that's -- NAV is probably the thing that we focus on the most versus any one particular month or quarter's earnings per share.

On page 11, you'll see our subsequent events. The fourth securitization closed July 2. We got approximately 3.3 times leverage just from the sale of the senior bonds. The senior bonds were significantly oversubscribed. If we were to sell the B1 bonds, we would get about 11 times leverage and if we were to sell the B2 bonds, we would in fact get more than 100% leverage and we would still have $47 million of UPB equity tranche remaining with a negative cash basis.

Our July acquisitions, $46.8 million of UPB, 77% of UPB purchase price, 59% of property value, very low percentage of property value; 100% of the loans we bought in July were re-performers, not nonperformers. In August, we have under contract $38.5 million of UPB. We closed about $5 million of those this morning. Our purchase price to UPB on our July acquisitions is about 67% of UPB.

One thing I should add is, in our financing, as you can see, we've increased our leverage to 1.18 versus 0.43 at the end of Q1. We have not yet sold any of our B1s or B2s bonds for any of our securitizations thus far. So we still own all the B1s and all the B2s from all four of our securitizations and as a result, they are not in those leverage calculations. The leverage calculations only included borrowings and money that we have borrowed as part of selling securitized bonds.

If we talk about the financials on page 12 and 13, the income statement we've talked quite a bit about. Obviously a lot is driven by average equity -- average loans during the quarter, as well as average leverage during the quarter and you can see that that ramped -- both ramped pretty materially throughout the quarter, from April 1 through June 30.

On the asset side of our 1.18 times leverage, $116 million at June 30 is non-mark-to-market borrowings and $153 million is securitized bonds. What you'll see is that in July, because of securitization, our non-mark-to-market borrowings will have gone down, but our total leverage will actually have gone up because our securitization, just the seniors, was a positive cash refinance of our existing non-mark-to-market borrowings.

That being said, at 1.18 times leverage, we are still very, very, very much less leveraged than the mortgage REIT space, significantly less leveraged compared to other mortgage REITs. And I want to just mention one more time our ability to source assets and conduct due diligence and service the assets our way is what really makes our NAV increases happen and we've seen from the performance of our securitizations and the bondholders' additional interest in our securitization bonds that it's starting to pay off in terms of our execution and the amount of available securitized leverage to us. With that, I'm happy to take any questions that you might have.

(Operator Instructions). Jason Weaver, Sterne Agee.

Good afternoon, everybody. Just want to say thank you for that color on the NAV and your securitization financing. It was actually very helpful. The first question I wanted to pose to you is kind of broad, but just based upon where you are talking about what you are seeing in the market today, I wonder if you could give some broad strokes as to what you are seeing on available prices in both NPLs and RPLs that you are targeting, as well as how that impacts your forward ROE expectations on those new assets you've put on?

Sure. What we call the bifurcation between large and small or the differential between large and small has gotten bigger rather than smaller. If you look at -- for example, in our August acquisitions that I had on the subsequent event page, if you look at that, our purchase price is 67% of UPB and 69% of collateral value. It's $38 million. That is eight different transactions.

If you look at very similar spec loans sold by Citi, one pool they purchased, one pool they sold. They with lower coupons sold in the mid-90s on UPB and collateral. So that's probably the most extreme we've seen on the re-performing front, which is a 25 point difference between small and large, keeping in mind eight transactions totaling $38 million versus one transaction or two transactions totaling $400 million in the Citi example I just gave. So we are seeing that differential getting bigger. We have not seen it cause our prices to change materially, a little bit on the price of UPB, but we have in fact lowered our purchase price relative to property value and part of that is by design because of a little nervousness maybe that housing prices, particularly on the highest end, have gotten a little too high, not so much in the middle.

On the NPL side, we've bought less NPLs for a couple of reasons. Number one is NPLs tend to be very dependent upon house prices and if you're at all even a couple percent bearish on house prices, we would rather buy more performing and less nonperforming unless the NPLs are in on our specific places. Number two is we've seen the prices of NPLs continue to increase in very large pools and call it pools of more than $60 million. That being said, with Fannie Mae now also selling loans, the supply of NPLs strangely is actually increasing rather than decreasing. The supply of performing loans, we've never seen as many performing loans as we are seeing now, or in as many pools as we are seeing now.

For us, our prices relative to property value for NPLs have gone down. Our NPL purchases have tended to focus more on smaller banks rather than in very large pools and our RPL purchases are from all kinds of sellers, big and small; many for regulatory ratio, many for getting a mark and many for non-economic reasons related to carry or something like that. But we are actually pretty excited.

That being said, the market is bifurcated even more. Part of that bifurcation increase is also caused by regulatory costs. It's more expensive than ever to be a servicer and also to be an owner of loans because of licensing and licensing upkeep and as a result, smaller buyers are rapidly disappearing. So the number of people we run into looking for $2 million and $3 million and $4 million and $5 million and $10 million pools is less than it used to be as well.

Okay, thank you. Very helpful. The only other question, and this may be detailed in your slide presentation, but I have to apologize, I haven't seen it yet, what is your current -- after the most recent securitization transaction capital position as far as liquidity goes? What's your dry powder?

We still have adequate dry powder and we also still own 100% of our B1s and B2s from all four of our securitizations. So the liquidity in the non-agency market is not as good today -- it's August -- as it was in June or even early July, but that's mostly because it's August. But that being said, we could still sell B1s and B2s should we determine to and we also have cash on the balance sheet. There is no immediate need for liquidity.

Very helpful. Okay. That's it for me, but I'd just like to say congratulations on another good quarter and looking forward to seeing more from you guys. Thanks again.

Paul Miller, FBR.

It's actually Jessica Ribner for Paul. One question we have on the model is what is the coupon that we can think about going forward? Can we leave it here at around in the 460s? Do you feel like it's going to trend lower given your focus on buying more adjustable rate and step-up product? How can we think about that?

Sure. The step-up coupons, we tend to buy them right at their step-up, either two months past their first step or a couple months before their first step, so we would expect the step-up to actually pick up a point right away. If our coupons declined at all, it's 460 to 450 something, it's not 460 to 300-something.

Okay.

We've actually been -- because of our ability to just target specific locations and specific kinds of loans, we tend to buy steps that have already stepped or are going to step quickly. So as a result, we get a little bit more built-in interest rate hedge from it rather than waiting two years or three years for the first step. We generally don't buy many of those.

Okay, perfect. Thanks so much.

Jim Delisle, Wasatch Advisors.

You kind of answered my question earlier about why the spread between the large pools and the small pools, but I would think with the return on equity differential between the large pools of re-performing and the smaller pools, it might become a place where you might see private equity starting to play. Have you seen any signs of that?

There's actually a lot of private equity already in the NPL space. Like the Oak Hills of the world, we see TPG through Roosevelt. So we see private equity in the NPL space and they tend to be sellers of performing rather than buyers of performing and part of it is for two reasons. Number one, duration is bad, even though, for us, predictability in duration is a good thing; and number two is don't underestimate how hard it is to acquire loans in small pools over and over and over and over and over again and to maintain the relationships with sellers and not care to the seller whether they close on July 1 or July 31. And if the seller doesn't have any of their loan files imaged to dispatch people to their office to scan, to go rent a scanner and scan everything for them so that we can then upload and do the due diligence so that they don't have to do it. So it's really -- some of it is customer service-focused too, which is only scalable if you've built a machine to do it. So we see private equity in the large NPLs. We don't see them materially in the re-performing space. We tend to see them being sellers of re-performers from their NPL pools as opposed to buyers of re-performers.

So I guess what I am hearing is that your ability to source things more efficiently through a better service and more flexibility does not seem to have any risks right now with new players coming to the market?

No, especially now in the small side. Clearly, on the big side, when there's $150 million or $200 million of re-performers for sale, there's a lot of interested parties, even insurance companies. For example, I'll give you an example, we've had probably at least five or six shareholders ask us if they could invest in JVs with us in -- directly in assets. And a lot of insurance companies are looking for longer duration fixed income assets that have higher yields than where I will say investment grade bonds are. So we do see some people looking for duration in assets who have longer life capital, but it's not private equity. The private equity tends to be much more focused on the nonperforming shorter duration stuff.

And actually you led me to a question that I had wanted to ask earlier in the quarter. When you're looking at the potential for JVs -- and I'm sure that number would be several times the seven or eight if people out there knew it was an opportunity -- is some of the exclusivity that Great Ajax has with the servicer, does that supersede the ability for you to do a JV with the servicer outside of Great Ajax? Could [Gregory] be used with something that did not accrue to the benefits of the Great Ajax investors?

If there was no benefit whatsoever to Great Ajax, that could only occur if Great Ajax literally didn't have capital to invest. If -- but the way we designed it is Gregory's allegiance is to Great Ajax, so the most likely scenario for Gregory to be able to service for a third party would be Great Ajax being a joint venture partner with somebody. For example, we own some small balance commercial mortgages where Great Ajax is the operating entity and we have a public REIT partner who owns a minority interest and Gregory services those lines.

Okay. All right, well, thanks again, Larry.

(Operator Instructions). [Rock Vanderlite], Nomura Securities.

Thanks for taking the question. I am sorry I'm jumping on late here, so you may have explained this, but I believe there was a comment that you made, Larry, in terms of a $0.17 a share delta between the end-of-period earnings power versus power over the quarter. Could you just clarify that?

Sure. So if you look at our average assets over the quarter, it was about $350 million. But if you look at end-of-period, it was about $450 million. If you just say it's a 12% unlevered yield, which is kind of the simplest way to look at it, that would be about -- it would be a difference of about $2.750 million per quarter in terms of value, maybe even a little more.

Okay, or $0.17 a share. All right.

Right. And that's just in unlevered -- if you just assumed an unlevered yield, right, just an unlevered yield. So -- now, the one thing you have to take into account is we ramped leverage materially during the quarter, not materially compared to a mortgage REIT, but materially for us, from 0.43 to 1.18. So on a leverage basis, there would be less equity in the denominator.

Okay. And just separately, maybe this has already been asked, but could you talk about your expectations for acquisitions the remainder of the year, if you can?

Sure. We have about $38 million under contract in August. August is generally the slowest month of the year. We closed about 15% of that this morning. A little surprised how much acquisition power we had in August so far. Keeping in mind that $38 million is eight different transactions, so it's a lot of little transactions. I would expect that acquisition volume would pick up. I would expect the first two weeks of September to be pretty slow. I would expect it to start picking up then and gradually increase over the fourth quarter with a flurry just before Thanksgiving and then a flurry around middle to late December. There's always some group of sellers that needs December 31 who calls us on December 23, but I would -- I think we all expect that our -- the target that we had set for the year back at the end of Q4 of 2014, I think we all think that we will be materially in excess of that.

Okay. Thank you.

If no other questions, we will go ahead conclude the question-and-answer session. I would now like to turn the conference back over to Mr. Mendelsohn and the management team for any closing remarks. Gentlemen.

Thank you very much, everybody, for joining us. We appreciate it quite a bit. We are around, myself, Glenn and Russell as well, for questions and we look forward to talking to you and we appreciate your support and the time you spent listening to us today.

And we thank you, sir and to the rest of the management team for your time also today. The conference call is now concluded. At this time, you may disconnect your lines. Thank you and have a great day, everyone.