Great Ajax Corp (AJX) 2018 Q3 法說會逐字稿

Good afternoon, and welcome to the Great Ajax Corporation Third Quarter 2018 Financial Results Conference Call and Webcast.

(Operator Instructions) Please note, this event is being recorded.

I would now like to turn the conference over to Larry Mendelsohn, CEO.

Please go ahead.

Thank you, operator.

Thank you, everybody, for joining us on Ajax's third quarter investor call.

I want to wish everybody a happy Election Day.

And with that, on the second page of the presentation, I want to have everybody take a quick look at this harbor disclosure and from there we can go on.

As an introduction, we had a very successful value-building quarter.

We bought a bunch of loans at good prices, we expanded our loan seller base, we added nearly $200 million of co-investments with institutional partners.

We closed 2 securitizations in the last week of September with very good execution, both in advance rates on the senior bond structure as well as the cost of funds.

And we continue to see rises in net interest income, even though average loans on our balance sheet in Q3 was lower than it was in Q2.

Other than expected REO impairments, which I'll discuss a bit later in this call, the quarter was all positive.

On Page 3, business overview, many of you have seen this page before but we've had a lot of things that have -- these things have advanced.

Our sourcing network continues to be an extremely important piece of our business.

It's very important to our ability to acquire the types of loans that we want, in the locations that we want and the prices we pay relative to others, you'll see that in a couple of things as we discuss today, one, some loans that we sold but also in what's going on in Q4 and the brand that our sourcing network has created.

We use our manager's proprietary analytics to price each pool, we analyze large amounts of data to determine target loan characteristics and to develop pattern and algorithms for both pricing loans and servicing loans.

And you'll see in the numbers, just what that means in terms of what it does for our interest income and our yield increasing and the cash flow velocity that we're seeing on the loans that we're acquiring.

Our servicer just as important, if you remember, our servicer, Gregory Funding, we own 8% of it, and we have warrants on another 12%.

We think the servicer has significant value optionality, especially given the growth of our institutional co-investments and additional third-party institutions that have reach out to us about buying loans with us and through us.

The other thing are analytics and processes of our manager and servicer enable us to do is to broaden our reach through joint ventures with institutional partners and it also enables us to see loans from even larger sellers more frequently and in accustom and loan-by-loan pricing way rather than poor competitive bids.

We use non -- moderate non-mark to market leverage, average leverage in Q3 was actually lower than that in Q2 and lower than that in Q1 as well.

So you'll see that ROE actually has increased in Q3 versus Q2, even though leverage was a little bit lower.

There's -- before we get into the actual numbers, I just want to kind of do a little walk-through of the different pieces.

There's several important pieces to the per share numbers, the per share map in Q3 2018, so like we did last quarter, I just want to walk you through the different pieces.

First and the most important piece, we increased our interest income and net interest margin as our cash flow velocity continues.

This increased loan yields and interest income despite a lower average loan balance during the quarter, on the flip side, you also see a small uptick in interest expense, resulting from accelerating the amortization of deferred issuance costs on our outstanding securitization debt as a result of the higher cash flow on the underlying loans.

Higher cash flow velocity leads to faster pay down on the related debt, which leads to faster amortization of the related deferred issuance costs, and as a result, you have a little bit uptick in costs.

Second, we called our 2016-A and 2016-B securitizations in September -- in the last week of September.

We took a charge of $836,000 or approximately $0.045 a share, this amount would have been amortized in full during the 12-month period ending Q3 2019, because there was 1 more year remaining in the amortization period.

By calling the bonds early, we accelerated the remaining amortizing amount to a onetime charge.

We believe that financing costs and advance rates from new securitizations more than make up for this in the securitizations we did in the last week of September, one was an 80% advance rate at 3.75% at par, and one was an 82% advance rate at 4.30% at par.

Third, our quarterly REO impairment was approximately $0.048 per share, as we've mentioned on previous quarterly calls, REO impairment happens first under GAAP and sales gains happen second.

Our REO portfolio continues to have expected future gains on a net basis.

Additionally, any foreclosure that results in a third-party property sale rather than becoming an REO is accounted for as a loan payoff in loan pool accounting for GAAP and does not offset any REO impairments.

In Q3, we had 46 foreclosure outcomes, of which 27 ended up as third-party sales and 19 became REO.

The profits of the 27 third-party sales became interest income and not REO gains and losses.

Approximately, half of the REO impairment comes from 2 properties, 1 in Western Passaic County, New Jersey, and 1 in Royal, Colorado.

The interiors of the property seem to have been purposely impacted during the resolution time line.

Fourth, we took an impairment of $0.02 per share on the remaining $23 million of UPB from our $100 million 2014 NPL purchases.

Based on the property conditions and expected time lines, we have reduced our yield expectation on the remainder of this $23 million by 0.4% or 40 basis points, which results in a onetime impairment of approximately $380,000 or approximately $0.02 per share.

As our portfolio yield and interest income increases indicate, we're seeing the opposite effects in our later acquisitions.

From a GAAP perspective, if we were to account on a mark-to-market basis similar to many other companies in the mortgage REIT space, the aforementioned impairments would be blended into our large built-in gain on our loan portfolio.

However, we choose not to mark-to-market loans because unlike securities loans don't really have readily available dependable marks.

As a result, we get to take charges for the negative impairments but we don't get to write up assets when they are worth more.

Altogether, if you take these 4 pieces, it's approximately $0.11 a share, which would normalize the EPS to approximately $0.46.

On the interest income side, one thing that's important to note in the second bullet point on the Highlights Page is that we only owned all of the $65 million we purchased in Q3 for on average 9 days.

So we really received no interest income during the quarter from what we bought in Q3, but we did have all the related loan diligence expense to those loans in Q3.

So we'll see the benefit of the $65 million in Q4 with the expense having been in Q3.

The other thing, I want to mention is we sold some loans in Q3 right at the very end of the quarter into a joint venture, and we retained a 63% interest in that joint venture.

For the 37% that we sold, we have a gain of $2.1 million on the sale, which implies if you were to sell 100%, about $5.7 million, however, since we still own 63%, we can -- we continue to consolidate the loans on our financials, and the gain as a result, will be amortized over the life of the loans, even though we did receive the sales proceeds and the cash representative of that gain.

If we jump to Page 5. You can see our portfolio overview, we have about 3% of our portfolio is nonperforming loans and about 97% RPLs.

RPLs continue to increase as a percentage, and you can see in Q3, the lion's share of what we bought was RPL.

REO is principally held for sale, it turns into cash over a relatively short period of time.

We've seen a decline in REO value at the higher end in certain states, materially affected by the limits on deductibility under the new tax law.

So if you were to split certain states or locations within states, Northern New Jersey, for example, parts of suburban New York City Metro and Connecticut, if you were to split those into deciles 1 through 10, with 1 being the lowest and 10 being the highest, we're clearly seeing the highest impact in deciles 8, 9 and 10 in those states, and in fact, the property in Western Passaic County, New Jersey would fit into that bucket that we took the impairment on.

On Page 6, our reperforming loan portfolio.

We continue to buy lower LTV loans with overall RPL purchase of 61.9% of property value and about 85% of unpaid principal balance.

The price of property value does not include any home price appreciation since acquisition.

Basically the strategy is continue to play offense and defense at the same time, and we're able to do that because of our private negotiated sales with sellers.

So we're able to pick the loans we want in many of the acquisition cases in each quarter.

And one of the things we concentrated specifically is our location concentration and the deciles in the underlying property values that we're looking for as a protection against any future declines that may happen in home prices.

On Page 7, our nonperforming loans.

Nonperforming loans have been declining in absolute dollars for us.

In Q3 2018, we created 19 REOs through foreclosure, but sold 32.

REO impairments continue to be primarily driven by borrower-induced property damage and to a much lesser extent home price declines in certain specific locations.

For our NPL portfolio, the purchase price to property value is approximately 53.7%, as you might imagine higher LTV NPLs become REO sooner, and create impairment sooner and lower LTV NPLs are more likely to become performing loans or to pay off in full.

On Page 8, you can see our map, we have not changed our targeted markets between Q2 and Q3, California continues to represent the largest segment of our portfolio, particularly, Los Angeles, Orange and Sand Diego counties, we're seeing consistent payment and performance patterns in these markets, particularly California, urban centers.

We expect the California percentage of our portfolio to increase from our Q4 2018 pending acquisitions.

On Page 9, this is very important in terms of its reflection on NAV and why we think we have a very significant kind of built-in value gain in our loan portfolio.

Approximately, $1.05 billion of our portfolio is 12 consecutive payments or better, up approximately $400 million from year-end 2017.

Approximately, $1.2 billion is seventh consecutive payments or better.

For the loans that we buy, and the way our servicer manages these loans, our analytic data suggests that once a loan becomes 7 for 7, there's a 91% to 93% probability that becomes 12 of 12.

Once loans are 12 of 12, an NBA current for that period, they're worth very close to par if not above par depending on their coupon and their LTV, and we see significant numbers of AAA-rated transactions getting done with these clean pay loans.

In addition to increasing cash flow and net asset value, the significant outperformance of our loans also lowers our asset-based cost of funds over time.

We've seen from our last 7 securitizations including 2 in September of 2018, that our loan cash flow velocity enables us to push advance rates up and lower cost of funds.

While our total average asset base debt cost increased in Q3, it was driven by prepayment and cash flow velocity causing more rapid amortization of deferred issuance costs rather than rate.

Page 10, what happened since September 30, and the answer is a lot.

It's been a very busy fourth quarter already.

We've already closed $103 million of purchases post September 30, $28.3 million of which was acquired through a nonconsolidated joint venture with third-party institutional investors.

And then we have approximately $670 million of loans under contract, much of which is in joint ventures with third-party institutional investors.

We also have another $3.5 million of small balance commercial loans under contract, and we have 3 properties, 3 small apartment buildings in urban Miami, Houston and Raleigh, North Carolina.

We have already 14 total transactions either closed or in process, we're also currently looking at 5 or 6 more that sellers have reached out to us and asked if we could get done by year-end for them.

So we're evaluating those loan portfolios currently.

One consistent pattern, however, if you look at the numbers of everything under contract is the low price to collateral value.

On the RPLs, 55% of collateral; on the NPLs, 57% of collateral; on the SBC, 65% of collateral.

And the overwhelming majority of these underlying properties are in what we call deciles 4 through 7 in each location.

The other thing I would say is approximately $250 million of the $560 million RPLs are in the state of California.

If we jump to Page 11, some financial metrics.

One of the things every joint ventures of institutional investors does is to the extent we own more than 20%, we have to consolidate the loans on the balance sheet, which skews some of the numbers proportionally.

So we -- here, we excluded the consolidation of our 2017-D and 2018-C, where we don't own 100%.

And we've effectively, the first column, deconsolidated it.

The -- one of the things to see is the continuing materially increasing cash flow velocity on loans is good for yield and NAV creation, the yield of 8.9% includes the reduced yield taken on the 2014 NPLs, so you can see that our yield increases have been pretty steady and cash flow velocity is the real driving force on that.

And if we could drop-down a little bit, if you take out the impairment -- the REO impairments, our ROE was actually 12.1%, return on average equity was actually 12.1% during the quarter, up from Q2 and up from Q1.

Even though average leverage in Q3 was actually a little bit lower than it was in Q2 and Q1.

And then, if we look at the ending leverage ratio at the bottom of the page, you'll see that average leverage during the quarter was less than ending leverage, that's because we closed 2 securitizations with significant advance rates right at the very end of the quarter.

And as a result, ending leverage is a little bit higher than it was throughout the quarter, and we would expect to see that in Q4 as well.

The -- after that, we have the income statement and balance sheet, and I'm happy to take any questions that anybody might have.

(Operator Instructions) The first question comes from Tim Hayes with B. Riley FBR.

My first one, looks like you're raising that dividend from $0.30 to $0.32, that's a little bit below where taxable income has been running this year on average, and just wondering what the thinking was behind setting the dividend there and if there's anything that would drive taxable income lower in the near term, that's reflected in that new dividend?

Nothing that we know about.

The one thing after discussions we've had with accountants on a special dividend front, we would be able to include it into quarterly dividends next year, so our board took the position at our more -- most recent board meeting a week ago, that what they would like to do is just have consistent increases in dividend rather than onetime jump.

And that's the only logic behind it, and that was a board decision.

Okay, so I just want to make sure, I'm thinking about this correctly.

At -- so assuming you had some leftover taxable income that you hadn't paid out this year, you can just factor that into your quarterly dividends next year, as you continue to ramp them?

That's right.

Yes, that's in equal installments basically.

But we would also -- and that would be separate from any taxable income you would have from next year also.

Right, very right.

Okay, got it.

So what -- our board has taken the position because they want to see consistent -- consistency in the dividend rather than any onetime jump and any onetime substantial jump in the dividend, so they want to keep on a kind of a keep leasing basis and get to a number that kind of then just becomes the new normal.

Makes sense, okay.

And then you went into some detail about the impairments and obviously, some of those seem pretty idiosyncratic and you continue to originate and acquire small balance commercial assets.

Just wondering can you just talk about dynamics of the small balance commercial market and maybe how much more similar that is to the resi market and just kind of be why the broader credit fears you hear about the broader commercial real estate market might not really affect the properties you're looking at as much.

Sure.

Yes, the kind of small balance commercial we look at is generally urban intercity locations, and the same, call it 8 cities or 10 cities that we have 80% of our residential portfolio in and the kinds of assets that we look at are specifically either small or multifamily or they are ground-floor retail with apartments above, also small -- relatively smaller buildings.

And we really look at it not that much different than the single-family to rent business, other than we think it has more -- the ability to have more concentration because instead of having 100 houses in 20 blocks, you could have 100 units in 2.5 blocks.

And it makes it more readily available to have your people on the ground, specifically because we already have lots of loans on the residential side in these exact same locations.

It's -- on the small balance commercial side, it's very driven by analytics of the specific places we want to be, and we're really looking at it as, which area is going to be different 5 or 10 years from now, not which area is going to be different tomorrow.

So it's a way of getting pretty good yield -- on the property side, pretty good yield and having kind of a neighborhood gentrification analytic strategy.

And then on the loan side, we basically make loans to smaller borrowers, typical loan size is $3 million or smaller and it's either a bridge for an acquisition with some repositioning money or it's a longer term, one of the programs that we have in our origination business is on the small balance commercial, is we're really trying to make it a portfolio business rather than a quick, we see all the time, people originating loans to resell them.

So we have a program where basically we make the loan at a relatively high interest rate for a couple of years, and then if the borrower performs, they can automatically extend it at a 1% to 1.5% lower interest rate for 8 more years with the 25-year amortization.

And so effectively, they're able to refine to a slightly cheaper loan without having any expense, and it gives us more duration in the portfolio and the ability to just keep clicking on that ROE as a long-term asset as opposed to where you see a lot of these bridge loans prepay rapidly in most places.

So we just look at it a little bit differently, and we look at it as almost like a cross between single-family to rent and commercial on the property side.

And we look at it from the loan side as a way of creating slightly longer-duration assets with higher coupons, all on properties that if something happened, we'd be perfectly willing to own for the amount that we've loaned.

And then just one more from me...

Now on the -- now one thing I would add, Tim, is on very large commercial, if you look at 300 unit residential buildings, one, you typically don't find them in the specific locations we have.

You -- I think you'll find them 5 to 10 years from now in those locations but you won't find them now.

But number two -- because of the nature of the location.

But number two, we've definitely seen the market a little bit softer on the very large kind of institutional 300 and 400 unit apartment buildings, we've definitely seen rents come down a little bit in that market.

Got it, okay.

And then just one more from me.

Just curious kind of how you see the pace of joint ventures going forward, are you seeing even more increasing demand or just conversations from other institutional investors looking to get involved?

Or you've kind of had a lot of activity there over the past couple of quarters?

Just wondering if you see that pace holding up or slowing down for any reason.

We have 4 that have reached out to us -- we have 4 that we already do joint ventures with, we've had 2 more on top of those 4 reach out to us.

And then we've had another large institution ask, if we'd be willing to put in place kind of a pre-securitized program, where if we created the joint ventures, they could be guaranteed a certain amount of the senior debt at pre-structured terms, which -- that was a surprising phone call to get.

And is that something that you're entertaining?

Yes, and that is something that we're entertaining, we started working on that last week to try to see where it goes.

But the amount of total portfolio that we've been asked to find, I'm not sure we could ever find.

The next question comes from Stephen Laws with Raymond James.

So a couple of follow-up questions to Tim.

I guess, first on the multifamily stuff.

Do you think you'll branch out in maybe new markets, you're going to stick to kind of the existing footprint that you guys highlight your portfolio on Page 8?

Or how do you think about how those opportunities might expand your business footprint?

Sure, I think there's 6 or 8 cities that we want to get to scale in relatively quickly.

And -- but I think there's about 12 cities overall, that we have interest in, just some of the cities, it would be harder to get to the scale than in other cities.

So we're focusing on certain cities, where we already have feet on the ground, and we know a lot about those locations block by block by block by block already.

And in fact, I would say that probably half of what we buy are properties that are because of connections or people we know as opposed to listed.

So off market I would say.

And then following up on the JV and the investments there, significant amount of activity.

Can you maybe talk about if you've got the capital and financing capacity to execute all of that, that you have under contract for 4Q?

I assume JVs takes 5% to 10%, but can you talk about that a little bit please?

Sure.

We -- depending on the JV, we take anywhere between 9% and 25%.

And it depends on which assets they are.

So the more NPL they are, the smaller the percentage we take.

The -- and different investors would like us to take different percentages, that being said, we get to decide in each case because one of the things that we require is we don't want to commit to buy a pool, if we're not willing to do 100% of it.

So we have to be comfortable we're doing 100% before we are willing to do a JV also.

So we don't -- we're not willing to kind of stretch and buy loans and we wouldn't ordinarily buy just because there's a JV.

What we do is have the JV effectively be a fee payer to entities that we own 20% of, the servicer and the manager.

So as a result it makes our values of ownership in those entities more valuable, it's one of the reasons why we structured it this way and why the board wouldn't let us do any joint ventures until we owned 20% interest in the servicer, because it didn't want the servicer to be distracted in any way.

But -- so from a capital perspective, we're comfortable.

Now if someone showed up in 8 different transactions should we agree to buy another $2 billion, I would say that we'd have to figure out how to do that.

Okay, well, a bigger problem to have, I guess, but for the stuff under contract subsequent events, you do have the capital and capacity to finance those...

Yes, yes, and we have a lot of leverage availability.

I mean, we're only -- I mean, at quarter-end leverage we're only 3x levered, and at quarter-end, we had $60 million of cash.

So we have a lot of leverage capacity as well.

The next question comes from Scott Valentin with Compass Point.

Just with regard to activity in the fourth quarter, I know, I think, typically seasonally fourth quarter is pretty active because of what the sellers are interested in clearing balance sheets but it seems that there is more activity than normal this quarter, is that fair to say, you are just seeing more activity?

More activity normal from 2 different places.

Place number 1 is banks looking to stop people from yelling at them, particularly what I'll call the 10 or 15 largest banks.

So they're still clearing things out of their balance sheets that are still there, especially reperforming loans.

And then number 2, we're seeing a lot of hedge fund activity looking to sell things and that they tend to do at least through us, they don't see us as a competitor because we don't raise money in the same places.

And as a result they'll come to us and allow us to kind of give them loan-by-loan pricing and for whatever purpose they need it.

And then we can negotiate loan by loan by loan and put together kind of optimized pools for us in that regard.

And the ability to close when we say we're going to close and do what we are going to do kind of has given us the brand that they know that they can trust us to close.

So we get a lot of -- in fourth quarter, we get a lot of hedge fund activity as well.

And from the -- from what's closed, the $103 million that's closed it's about 80-20 hedge fund versus bank but in what's closing the -- in the $600 million, it's about the opposite about 80% bank and about 20% hedge fund.

Okay, that's helpful.

And then you referred to or I guess talked about the implications of the tax law change on real estate tax?

Can you delve more into that.

Is that -- is there any change in strategy maybe avoiding some of the high real estate tax states or dial back, I mean, you're active in California and in...

Well, clearly -- well, one of the things we find is California is actually a relatively low real estate tax state, it's a high other tax state.

But -- and -- where we -- what we have found is in each of these states it's not really affecting the average house to the

(technical difficulty)

It's effecting, call it deciles 8, 9 and 10.

So for example, in the -- we've -- we started seeing impairments in REO in properties above about $800,000 or $900,000 in Western Passaic County probably about 4 or 5 months ago, this is the third 1 of 3 that we've had an impairment on.

Now most of that is borrowers wrecking interiors of houses, but the time line of New Jersey allows more wrecking, but also the change in the tax law has caused the higher end properties in that part of New Jersey to decline faster than the middle.

So it was kind of like a double whammy in that.

We've clearly seen it in Connecticut, where I wouldn't say that Connecticut is a free fall but we've clearly seen now for probably 12 consecutive months on average prices coming down in Connecticut.

Less so at the lower end, more so call it above about $700,000 or $800,000.

Okay.

And so as a result, are you shifting strategies avoiding some of those properties and those price ranges?

No question about it, clearly, of the things we're buying in this quarter, but that being said, the breakdown of our portfolio was much more California than any place else and a lot -- and typically Georgia, Florida, places that have been benefited from the new tax law, and I would say that maybe 20% or less of our portfolio was really in states that the tax laws hurt.

And then probably, a small subset of that is in the deciles that have been impacted.

But we would expect kind of higher end outside the city in New York properties in the Metro area to be impacted somewhat at the higher end where we've not seen it in the lower end.

So we've -- since about early 2016, our focus in the RPL market has been very low LTV loans and to buy loans at very low prices relative to property value.

We've seen that our competitors take kind of a different approach, and the principal reason is if you buy 105 LTV loan, if you put home price appreciation as a variable in your model, then home price appreciation makes those loans more valuable so you have more upside.

We don't want to make a bet on home price appreciation, we want to make bet on don't be wrong, not be right.

So it's sort of like play defense and play offense by playing defense so that if we're right, we do pretty well, and if we're wrong, we do a little bit better.

That's kind of our mentality.

All right.

Fair enough.

And then one final question.

On the margin, you mentioned -- and the margin was up linked-quarter.

Just wondering if you can keep expanding going forward.

You mentioned you're getting better execution on the ABS transactions.

I imagine loan yields are creeping up, so it sounds like a combination of you can still see margin expansion from here.

Yes, we think that's true, and our prepayment on our loans is much less dependent upon rate sensitivity than you'd see say on a typical newly originated agency loan or something like that.

And part of it is just demography.

For example, if you look at the typical new -- newly originated agency loan, the borrower is typically 40- or 41-years-old and with younger children and things like that.

Because most of what we buy is 2007 or earlier and a lot of it is 2004 -- 2002 to 2004.

On average, we get a lot of emptiness pay off.

So what matters to us and what we see is the biggest driver of payoffs is not rate, it's absolute dollars of equity.

So that when people in different markets hit a certain absolute dollar of equity number and they're in a certain age bracket and a certain life cycle location, they tend to be sellers.

So most of our payoffs are sales of houses not cash out refis.

The next question comes from Kevin Barker with Piper Jaffray.

I just wanted to see if we can get an update on some of the strategic options you've been looking at across the entities.

Could you just give us a layout on some of the -- what you're -- what the latest is on some of those discussions?

Sure.

The holders of -- well, let's kind of do it piece by piece, there's kind of the Great Ajax side and there's the servicing side.

On the Great Ajax side, we have looked at some corporate acquisitions, we've not been able to get to what I'll call price levels on those acquisitions that people seem to want or need.

On the kind of REIT versus de-REIT, I think, we'll know a little bit more tomorrow, once we have some more information of where the tax law maybe, although I don't think it will be really to guiding given what all the headlines seem to be.

I think our board has really kind of pushed that decision into later next year, based on other things that -- opportunity sets that we might have.

On the property side, they clearly want to grow that business and expand that kind of as a subsidiary structure to our operating partnership.

And then on the servicing side, we own 20% of the servicer, actually we own 8% plus warrants on 12%, on the servicing side, I think the owners of the servicer plan on actually growing the servicer pretty materially and might even look at doing a capital raise inside the servicer, given that the rapid growth it seems to be having through co-investment from institutional parties and the demand for its services that we're seeing in the last 6 to 9 months.

So that's something that Russel is spearheading as the person who runs the servicer on a day-to-day basis, but our Great Ajax board has one of its outside directors appointed to the board of the servicer and is part of that decision-making process.

Okay.

And regarding your comments in the beginning around pricing, it sounds to me like you won't look to pursue some consolidation of the different entities given...

Oh, you mean, of the manager and the servicer combined?

Yes.

All in one.

I wouldn't say, we would never do it, I would say it's not imminent.

There's -- from a servicer perspective, we've already bought a piece of it, and I think we as a board think it's not quite ready, if we wanted to consolidate the servicer, we would want it to accomplish a couple of more goals that the servicer board has -- is working on.

From the manager perspective, I think that we already own 20% of it, there's been -- in mortgage REIT land there's been lots of internalizations, many of which investors of the REITs that have internalized have complained about the price, we don't want to be in that position, that being said, since we already own 20% of it, we'd be paying money to ourselves for a chunk of it.

So that is something the board has looked at but we own 20% of it, we have a 0 basis in it.

Obviously, the joint ventures with these institutions make the manager more valuable, it doesn't show up in our balance sheet really that it becomes more valuable because we own it for 0, so it really doesn't show up in our balance sheet.

So the board is actually looking at ways to try to get it to be more magnified in our actual financial statements as well.

If you start to go down the route of not consolidating, is there -- and the REIT has a stake in these entities, is there any way to get visibility on the value of that stake or a little bit more clear on the potential value?

I would say, we -- well the -- as part of the purchase of the shares of -- and warrants of Gregory Funding, we've had a valuation by Houlihan Lokey as part of that.

The value of the servicer has increased since then because servicing will go up by first quarter of next year by 30% to 35%, since we made our acquisition in first and second quarter of '18.

And that would, I would say, probably increases the value of it by -- perhaps it's because of that the margin profitability probably increases the value of it.

And -- now this isn't scientific, there's no kind of third-party valuation.

But I -- my own feeling is that increases the value by approximately 30% to 40%.

The -- on the manager side, we've had some people give us kind of the way managers get valued and I would say that based on that it would imply that the managers' worth somewhere between about $50 million and $65 million, so our 20% stake would be about $0.70 plus a share as is assuming no growth or no changes in fee revenue.

So it's on our balance sheet at 0, and unless we consolidated and bought more, it would still be on our balance sheet at 0.

Okay.

Considering the potential growth in these entities and the stake that the REIT holds, is it possible to disclose that value going forward and being able to...

We do in our SEC -- quarterly SEC filings, we do put it in the footnote.

Okay.

But it's not science, it's kind of an estimate.

It's not we got 8 bids and here's the number, you know what I mean?

Yes, of course, of course.

Yes, I know.

We don't pay for that valuation every quarter.

That's exactly right.

That's exactly right.

All right.

And then you provided some color on the REO impairments that you incurred, when you look out over the next year or so.

Do -- is there -- are we still at risk of seeing several more REO impairments?

Or is it something that's winding down at this point, given the REO portfolio as of now?

Yes, yes.

When we have the beginning of the impairments from REO from NPLs, we talked about how impairments would peak starting about second and third quarter of this year.

So we would expect impairment to become a lower and lower number until there's no more NPLs or until there is a small number of NPLs and very little new REO getting created.

So the answer is, yes, the impairment started about 5 quarters ago, and we said, the peak would be about 18 months out.

So we're right about kind of the reporting 5 quarters ago, so call it 16, 17 months ago.

So we said the impairments would peak around kind of 18 months out, so we're right about right there.

Okay.

And then you always talk about...

I mean, think about it in our 2014 NPL pools, there's only $23 million of loans left.

Yes, I know it is kind of pretty small.

Right?

It's pretty small now.

So -- and that being said, loans that become REO are adversely selected properties, and loans that don't become REO are depositedly selected NPLs, right?

So the problem is that we don't get to write up the good ones, we only get to write down the bad ones.

Yes.

Well, it all comes back to earnings, eventually.

Yes, eventually.

That's right, eventually.

Okay.

And then you keep reverting the NAV estimate -- I'm sorry, I missed it earlier, what was the number that you laid out?

I didn't say a specific NAV number, but here's the kind of the simplest way to look at it.

So in Q3, we sold the 37% interest in some loans, about $100 million of loans.

The 37% was a $2.1 million gain which implies about a $5.7 million gain, so call it 5.5 or 6 points.

And that was a pool of loans, so in our 2016 A and B securitizations we separated out the clean pay loans from the nonclean pay loans.

And we did not sell into the JV the clean pay loans, we only sold the nonclean pay loans, as you might imagine, there's a lot higher value to the clean pay loans.

But we had a 5.75 point gain on the nonclean pay loans, that we sold to 37% interest in.

So when you think about it, if you took your clean pay loans and you sold them at the same price, clean pay loans are now 77% of all of our loans.

If you took the clean pay loans and sold them at the same price, and you say had a 5.75 point gain on $1.2 billion of loans, right, that's $70 million.

That's kind of a back-of-the-envelope calculation.

Now I personally think the clean pay loans are a little bit -- worth a little more but it is what it is.

The next question comes from Greg Mason with Ares.

I wanted to talk a little bit about the prices on the RPLs on the subsequent events.

If we look on Slide 6 and look at your RPL portfolio, looks like price to UPB is around 85%.

And then if we look on Slide 10, you've got 90% to 91% on the new RPLs.

So can you just talk about the change in that pricing?

Is it the competition?

Is it mix?

What is causing that change?

Sure.

So one thing I will add is in the RPL in that $561 million, there's 1 pool of $250 million that has an additional $11 million of owing balance that's not in that.

So as a result, the price to UPB would come down by a little bit.

But because it's not technically UPB, we can't add it into the UPB calculation.

But it's owing balance on performing loans of what I'll call, past due interest on performing loans, weighted average LTV at the past due interest is 60, so as a result, it's collectible, but it's not -- from an accounting perspective, it's not UPB.

So the UPB would actually be about $572 million and change versus the $561 million.

So that would bring down UPB.

The other the other thing I would say, is these are much lower loan-to-value loans.

So -- and we're looking at an environment that we think you're going to actually see some home price changes to the negative of somewhere between 3% and 5%.

So as a result, we're very focused on what I'll call relatively faster time line states in case of default.

And where your IRR goes up in case of default, rather than your IRR going down in case of default.

So we buy these RPL, and but it's sort of like if they default, I do better, and if they don't default, this is what I make.

So we've kind of -- since about early '16, we've taken that route.

And even more so throughout this year, so there's a little bit more safety factor.

The other thing though is there is more prepayment speed from these loans.

There's also a higher coupon on these loans relative to past purchases.

Got it.

And then the warrants that you have for the servicer, what is your thought about exercising those warrants, why would you do it?

Why would you wait?

If you could just kind of talk about that, that would be helpful.

Sure, sure.

The -- in servicing land to own above 9.9%, you then have to go through the whole control -- become a control person in 65 jurisdictions and the board of Great Ajax similar to a couple of the institutional shareholders are also on 9.9% of the servicer, clearly doesn't want Great Ajax as of now to be a control person.

I think their position is they want to be a control person if they own 100%, but not if they own 11% so as a result -- but they also wanted as much optionality as possible versus as much putting as much money in as possible.

So from the board's perspective, I think they see a lot of optionality in the servicer, a lot of kind of value-creation.

I don't think they would exercise absent a sale of the servicer or a purchase of the servicer, I don't think that they would exercise their warrants.

But one thing they built into the warrants is that to the extent the servicer pays the dividend, the warrant strike price decreases by the amount that the dividend would have been received had the warrants' been exercised.

Yes.

Got it.

And on that kind of dividend commentary, you talked about the potential balance sheet, book value impact from the servicer and manager.

Are you getting any cash flows off of those that positively impact the income statement dividends from either of those 2?

We don't currently from the manager, the servicer, we do expect the servicer to pay some dividends.

The other thing is that and this is kind of an interesting one, the Great Ajax owns 20% of manager, but 50% of the fee it pays is in the stock at Great Ajax as the higher market or book.

And there's a 3-year lockup on those shares.

So as the 3-year lockup shares roll off quarterly, the manager distributes those shares to the owner of the manager, so Great Ajax gets a dividend quarterly off its own shares back from the manager.

Will those shares then be held as an asset on the balance sheet or will you just retire those shares, how would that work, when that starts happening?

They become treasury stock.

Treasury stock, okay.

Great.

And then on the -- if the servicer pays a dividend, obviously, you record the dividend off of the proportion you own outright.

I assume there is no income on the warrants since the strike price is reduced?

Or is there a...

There is no income, but the warrant value would go up, because you'd have a lower strike price, right?

So the valuation of the warrant using the option models would cause the warrant to increase because of lower strike price.

But that wouldn't be income, that would be carrying -- that would be mark-to-market, so it wouldn't even be carrying, right.

This concludes our question-and-answer session.

I would like to turn the conference back over to Larry Mendelsohn for any closing remarks.

Thank you all for joining us on Election Day for our third quarter conference call.

Feel free to contact us if you have additional questions over time.

And we'll be back in about 3 months with our Q4.

Thanks very much.

And look forward to talking and catching up with everybody.

The conference has now concluded.

Thank you for attending today's presentation.

You may now disconnect.