Great Ajax Corp (AJX) 2016 Q2 法說會逐字稿

Good day and welcome to the Great Ajax Corporation's Second Quarter 2016 Financial Results Conference Call.

All participants will be in listen-only mode.

(Operator Instructions)

After today's presentation, there will be an opportunity to ask questions.

(Operator Instructions)

Please note, this event is being recorded. I would now like to turn the conference over to CEO, Larry Mendelsohn. Please go ahead.

Thank you very much. Thank you everybody for joining Great Ajax's second quarter 2016 conference call. Here with me are Russell Schaub, our President; and Mary Doyle, our CFO.

Before I get started, I want to just bring up on page 2 the Safe Harbor disclosure for everybody to read. And with that, I'll get into more discussion.

Before I get into the presentation itself, I'd like to make some general comments about the quarter. You may remember from our call on May 3, 2016, our first quarter call, we outlined that number one, we raised additional debt through our six securitization in early April. And as a result of that at the beginning of the second quarter, we had nearly $70 million of cash and cash equivalents. We raised all this so that we continue executing our plan and simultaneously play defense, you probably remember it was quite a bit of market chaos late in the first quarter and early in the second quarter. So we want to have liquidity to play defense but also to take advantage of the chaos and playoff hence later in Q2 2016.

We also talked about that we're seeing more loans than ever. And that's for a number of different reasons, everything from bank-regulation to risk retention rules to ratio management to more agency kick-outs than ever before, and we can be even choosier in our negotiated purchasing of loans. Following the same strategy, we raised a little bit of equity in June of 2016. And as a result of all of this, we went out more on offense in mid-Q2 and made a number of negotiated acquisitions in late Q2 and closed a bunch more in July 2016 which would be early Q3. As you'll see later on in this conference call, this has continued through August and early September with prospective closings as well. We think we've been able to play offense pretty well while at the same time being very defensive as we purchased lower LTV loans at significant discounts to property value and in specific concentrated locations that we'll talk more about also on this call.

A little brief business interview on page 3. Again over 90% of what we buy is in privately negotiated transactions and in what we bought in Q2 and what's already closed in Q3 and what is pending for Q3 that continues to be the case and even more so. But it's really important to understand our sourcing network, it's very, very, very important to what we do. It's important to our ability to acquire the types of loans we want, in the markets we want and at the prices we pay relative to what other buyers pay for loans. So also important understanding how much time we spend looking at data trying to figure out what we actually want to buy and where we want to buy it. We look at and conduct a lot of data analysis to figure out what qualities we want in loans.

We do a lot of transactions. We did 12 in Q2, we have a number already closed in July and we have a number more already ready to close in August. Small transactions are $300,000, big are $100 million probably typical is between $5 million and $20 million. Captive servicer can only service for us, very important that the servicer is integrated with our loan analysis and integrated with the way we think about outcomes for individual loans, gives them the ability to be strategic and have best outcomes.

On a leverage basis, we did a small equity raise in June of this year, in late Q2. Our leverage before we did the equity raise was about 1.65 times; post equity raise at the end of Q2, we're about 1.62 times and our leverage ratio will be higher than that by the end of July and higher still in August. We talked about when we did the equity raise that our net leverage would actually increase post close of the acquisitions and that has been the case.

We've done six securitizations. We actually went out this morning with our seventh about $130 million pool loans. In that deal structured slightly differently, we also have a prefunding account so that's some of our August acquisitions will be bought directly into our securitization. So we effectively pre-fixed rate non-recourse funded it into securitization much of the August acquisitions.

Turning to page 4. In Q2, we invested $52 million to buy $70.3 million in UPB. So about 74% of UPB was our purchase price, purchase price's collateral value was in the 60%s. Again we continue to focus on RPLs; RPLs are now 88% of our portfolio. Since early 2015, we purposely focused on buying lower LTV loans especially RPLs with underlying properties in housing deciles what we call 4.5 through 7.5. And if you were to rank housing prices 1 through 10, location by location by location, we tend to be in the middle, 4.5 to 7.5. We think those are the most liquid, the most stable, the most predictable. Those would be different price ranges in different cities. So a decile 5 in Brooklyn is different than a decile 5 in Atlanta.

We also like in these low LTV loans, we like minimum absolute dollar of equity threshold, our data working analytic suggests the performance in prepayment are not just tied to the LTV percentage, but it's also to absolute dollars of equity. So $80,000, 80 LTV loan and $280,000 LTV loans don't have the same re-performance rate and they don't have the same refinancing rate.

Larger absolute dollars are also a natural hedge against housing price declines. They are natural hedge against slower economic conditions and against recidivism, fancy word for re-default. Also 50% of our loans are either ARMs or step rates, which provide some built-in interest rate risk protection should the rate cycle that we're currently in reverse itself and we see ourselves with higher short-term rates. Again, it goes back to sourcing, we're able to build this portfolio to these specifications, because of the way we get loans from many different sources in small increments and negotiate loan by loan by loan from what a seller might have on balance sheet.

One of the things that's really important to understand is performance of our loan portfolio. Re-default has been significantly less than expected, that's a little bit of an earnings drag, although it increases cash flow over the life of these loans dramatically. We had a lot of cash during the second quarter. We significantly increased our secured borrowings in early April through our six securitization. The proceeds were only partially invested throughout the second quarter, which we talked about on the May 3rd call and that would be an income drag. If the proceeds have been fully invested throughout the quarter without any additional leverage levering up those proceeds, net interest income probably would have been about $0.05 to $0.06 a share higher.

Aside from having cash of nearly 20% of equity for the quarter, think about how much cash that is, 20% of equity, re-defaults on our RPL portfolio have been running significantly lower than expected. When you think about it, less re-defaults extends the duration of the loans, reducing current period interest income. However, over time less re-defaults imply much higher overall total payments and cash flow from those loans. In fact, in many of the scenarios of our current low re-default, cash flow could be as much as 5% to 10% of UPB over the life of those loans more than we would have anticipated. So on $600 million or $550 million of RPLs, that could be a $40-plus-million number. It also increases having less re-defaults increases the market value of the loans themselves.

As you might imagine, clean pay loans have gone up in value, especially in this rate cycle. There's a lot of big insurance companies and also funds that are pretty aggressive buyers at very clean pay loans. So if we're to -- given the lower re-default (inaudible) loans to market, which we don't do on our balance sheet or on income statement, they would obviously mark much higher than where we purchase them and where we carry them.

Some of the trends in the lower re-default can be explained perhaps by the fact that we've been buying much lower LTV loans with greater absolute dollars of equity, so maybe you would expect less re-default because of the selection bias, but we're seeing it throughout our entire portfolio, even in what we consider the less credit-worthy portion of the portfolio. So nearly all of our RPL portfolio is experiencing significantly less re-default than expected. In fact on average re-defaults are running at approximately 50% of what we expected. The only loans really in our portfolio that are re-defaulting close to the expected rate are higher LTV loans, in fact loans that are probably about 135 LTV or higher and those only in New York and Florida for the most part.

So we'll just talk about net income. There's a couple of small pieces that you should be cognizant of, one is about $0.03 a share of loan transaction expense that was run through our income statement in the second quarter is actually related to acquisitions we closed in July, which is in the third quarter. GAAP says you book the expense when you incur it not when you close the transaction related to it. So that's just a pure timing difference. That's $400,000 of expense we won't have in Q3 because we already expensed it in Q2. And if we think about what our earnings per share is, it's about 11.25% return on average equity with approximate leverage of 1.6 times and cash of approximately 20% of equity on the balance sheet earning down close to zero. So we actually think loan performance in our portfolio is doing very well.

Taxable net income is actually affected as well by having more debt on the balance sheet without deploying it because you had the interest expense which offsets taxable income. The other thing is that as more loans pay that actually differs some taxable income because tax is based on contractual maturity workout is based on expected life of the loans. However, in the current post-Brexit interest rate cycle, we do expect pre-payment to pick up. We would expect that to begin probably about 75 days after this rate cycle really took place in early July. So we do expect probably come September that we'll see material increases in pre-payment.

[Total] point down from there is the amounts we raised, $98.7 million net in secured borrowings during the second quarter and also $32 million net in public equity offering we did in June. We finished the quarter with just under $70 million of cash, which is about the same amount of cash we actually had at the end of the first week of April also post-securitization.

Talking about our portfolio on page 5, the first thing that you will see is that NPLs at year-end 2015 through today have gone down from 14% of our portfolio to 12% of our portfolio. We did not buy any NPLs in the second quarter. We just don't think that NPLs represent nearly as good a value as RPLs for the loans we see and in the places we want them. You'll see that REO increased materially again. Some of the REO increase, most of it is really just representing a portion of the NPL decrease as loans become REO.

On page 6, you can see that NPLs had an absolute decline of $18 million since the end of last year. So about 16% of our NPL portfolio was resolved during the first six months of this year. We would expect our NPLs to continue to decline and we'll talk about that during -- when we get into subsequent events.

On page 7, you can see that in the last 12 months, re-performing loans, RPLs increased by about $168 million on a net basis. But property value underlying our portfolio increased by $276 million. Keep in mind that that $276 million increase does not include any home price appreciation. These are all property values as of the acquisition date. So the property value probably increased a little bit more than that. What you've seen is the net increase in our re-performing portfolio has been done by buying at about 60% of property value. So again, low LTV loans, low dollar prices on those low LTV loans, low purchase price to property value, it's playing offense, while at the same time playing defense.

If we jump to page 8, non-performing loans continue to decline as a percentage of NPLs remaining on the balance sheet approximately 59% of underlying profit value, again no home price appreciation in these. Some of these NPLs required is long ago as late 2014. So the fact that we own NPLs on the balance sheet is 59% of underlying property value, the expectation would be, we actually own them a little cheaper than that if you were to factor in home price appreciation.

We made a few changes to our target markets. Probably the number one change we made is that Chicago is no longer a target market. We saw property management on the ground there and asset management there, but we no longer target Chicago. Our data analysis suggests that some of the education trends and crime trends have narrowed the scope of where we want to own underlying properties or loans backed by underlying properties and we just don't think that it merits enough attention to make it one of our concentrated markets.

And another change we've made is our re-performing loan portfolio in Brooklyn has increased. However, it's increasing at a decelerating rate and in fact, when we look at what we're acquiring in August and early September, you'll see that it's almost devoid of New York re-performing loans where we want to wait and see a little bit in parts of the New York metro area to see how world economics and some of the changes in foreign capital rules in LLC purchasing of real estate New York and how that affects on the ground prices.

California still represents nearly 30% of our overall portfolio. Southern California is about 75% of our entire California portfolio, so call it Santa Barbara in south, down the coastline not so much, Inland Empire. So LA or San Diego County has probably about 75% of that 30%.

If we jump to page 10, this is as we discussed before, we are less quarter to quarter earnings focus than we are about are we acquiring things really cheap, now they are performing well, now rebuilding value. This table we put in every presentation, we've updated this to our June 30 portfolio. So it does not include all the acquisitions we made in July, but if we think about NAV as forecasted by the fixed income market, so what does the securitized bond market think our balance sheet is worth and also we look at this from a return on average equity that if we were to sell subordinate bonds what do we own and what do we own it for.

So I look at it in two ways, one from a pure NAV, if we just took our entire portfolio, we securitized it at our April securitization. You can see that it would imply NAV by valuing the equity at about $0.40 on the dollar of about $18.61 a share, so about $3.68 more than book at the end of the quarter.

Just as importantly, as we think about return on average equity, if we were to sell our subordinate bonds keeping in mind that so far we've never sold any subordinate bonds and we've only financed about a third of them. But if you think of our -- if we were to sell our subordinate bonds and just hold the equity, we would have almost no remaining basis in the loans and about 26 points of UPB. So if you securitize the entire portfolio based on our April execution, April 2016, you sold this subordinated bonds at market prices, you kept all the equity, you would have about a $15 million basis in $209 million of UPB. On a return on average equity, that would be pretty remarkable especially since the underlying loans are 88% re-performing and they all re-performing at much higher rates than we would have expected versus expected re-default. So we're pretty excited about the intrinsic value of our portfolio and the performance of the underlying loans actually makes it better than we would have envisioned.

We jump to page 11, I guess the summary of page 11 is there's a lot going on. In July acquisitions, all of this is closed, we closed UPB of $148-plus-million underlying property value of $210 million. So price to collateral value of 59%, price to UPB of 83%, 880 loans in four transactions in July, all RPLs. We also bought a small amount about $700,000 UPB, about 50% of collateral value of small balance commercial loans. There are RPLs, there are low LTV loans and our low LTV and low price to property value strategy continues. It's evidenced further that our sourcing advantage also continues to be of refined loans in private negotiated transactions not at auction just like this.

You can see that we played some more offense in the second half of the Q2 and also in July. However we played offense in a way that has built-in defense when you're buying loans at 59% of property value and the average underlying property value was $240,000. So again deciles 4.5 to 7.5 and paid 59% of property value you can play offense and defense at the same time.

If you look at acquisitions that we've agreed to buy subject to due diligence, we call them pending acquisitions here, again 100% RPL, $124 million of UPB, $171 million of collateral value. You see price to UPB 82.6%, a little bit cheaper than what we bought in July and price to collateral 59.9% pretty much the same. 626 loans in eight transactions. Again, the ability to source loans all in privately negotiated transactions, none of these were bought at auction.

In Q3, we have continued what we started and these should close in August and September. Again low LTV loans all negotiated in certain specific markets. For our pending acquisitions only 1.5% of $124 million of UPB is in New York state, all of that is in the New York City Metro area. So as a result it decreases, our overall portfolio exposure to New York and any potential changes in real estate markets generated by worldwide economics.

Our Board declared a dividend of $0.25 payable August 31, record date of August 16 based on the performance of our loans, based on expectations of pre-payment based on the fact that more debt -- interest expense for more debt offset taxable income and loan transaction expense, due diligence cost for July pools that we booked in June, the Board felt that changing dividend to a lower number based on taxable income was -- would have been a little misleading. So they kept the dividend the same because they continue to expect overall increases in taxable income.

With that, I'm happy to take any questions you might have to go into as much detail in the answers as I can.

(Operator Instructions)

Jessica Ribner, FBR & Company.

The $400,000 in due diligence expenses, I just want to clarify was just for July, but not for the pending acquisitions?

That's correct, that's related to transactions that closed in July, but because we spent the money in June for due diligence, it's a June expense, even though it relates to July acquisitions.

And I know that you said that was about $0.03 to earnings, would that have been --?

Based on weighted average shares of Q2.

And then so my next question would be, what's the right yield for the RPL portfolio now that the duration is extending out on that portfolio or the expected duration?

We don't love saying exactly what our yield is, because we don't always want our sellers to know what we make on loans we buy. But I think if you assume it's somewhere between 8 and 11 depending on the underlying collateral, that's a reasonable range.

Or maybe a better way --.

When we would have said that if it defaulted as we expect it, it would have been a couple of percent higher.

And I don't know if you can disclose this, but what would you peg the expected duration to be?

I would peg the expected duration to be approximately 4.5 years or 5 years. Depending on what you think pre-payment is, I could argue that we think there's some pre-payment coming that's going to shorten that duration by maybe as much as a year, year and a half. So it's one of those things where you have either 10 different scenarios and you sort of take the average.

Although I would not have built a scenario that would have said the worst credit quality loans in our book are re-defaulting 50% less than expected, I wouldn't have built that, that's not one of the scenarios that I would have built.

Now I hear that, it's certainly an interesting market. My last question is just another clarification, in the press release you guys talk about net interest income for the quarter declining as a result of recurring payment performance. Can you --?

That's what happens is -- that's right. So that's yield on your underlying portfolio, extending duration reduces the top line on re-performing loans. The other thing you have is you have a lot more interest expense that reduces net interest income because in 2016 a securitization in the first week of April. So it's a combination of more interest expense that didn't have more loans attached to it, and lesser yield because of too good a performance on loans. If someone did ask me two years ago, if I was going to have a decrease in income would I be willing to have it from loans paying too well, I might have laughed about it.

Right, it's an interesting (multiple speakers). Can you quantify what impact that had on earnings?

I don't have the number off the top of my head, but it could be -- let me divide it by 4, so if you take it, let's say, we had $500 million of RPLs and if you just said the yield went from 11 to 10, each point would be on declining balance $5 million a year.

Kevin Barker, Piper Jaffray.

I just wanted to follow up on the transaction expense you mentioned in the second quarter. What was the transaction expense for 2Q -- for the second quarter acquisitions that was incurred in the first quarter?

It would have been a lot less because most of the acquisitions closed in the second half of the quarter. So, most of that would have already been in Q2. There would have been a small amount of that in Q1, so it was probably $50,000 or $40,000 or something like that.

So the overall, clearly, was a uptick just purely due to timing in this particular quarter?

We closed a large transaction on July 28, I think it was and there is just no way to get all the due diligence done in time, we agreed to the transaction in the second week of June and hence we did all the due diligence, all the compliance checks, all the property evaluations, all the travel to go look at properties, all those expenses were incurred in June. So as a result, if we had closed it in June while we wouldn't have gotten any interest income off of it by closing at June 30, it would have been at least been in the same quarter and on the balance sheet, but there is still wouldn't have been the interest income offset. The transaction, a lot of it relates to most of the July closings happened in the last three or four days in July, but most of the due diligence expense was in June.

And then you mentioned a material increase in prepayment speeds, in September, you expect (multiple speakers)?

We were starting to see it a little bit. Usually, we see prepayment increase about 60 days to 75 days after you see rates really go down. Some of that it's with under the new rules and the new closing requirements, newly originated loans take a little longer than they used to number of years ago. But given that you saw rates really collapse at the end of June, early July, we'll see that pick up 60 days to 75 days from the end of June, early July.

The other thing is that our portfolio has ramped over the two-year period that since we started Great Ajax and you need -- we see refinancing really pick up at about 18 consecutive payments so that as our portfolio becomes more mature on our own balance sheet, you will see a natural progression towards prepayments of those, coupled that with the decline in rates, you'll see a lot more loans eligible to prepay. The other thing is that since February 2015, so about 16 months ago, we purposely started acquiring lower LTV loans with more absolute dollars of equity, which are more likely to be interested in having bigger reason to benefit from a refinancing and as a result, we would expect as those have 18 consecutive payments you'll start to see the prepayment speeds pick up on those as well.

And so, when I think about the quarter-by-quarter changes in income from pickup in prepayment speed, I would assume that you probably see an acceleration in accretion income or potentially (multiple speakers)?

We have not modeled in any material increase in pre-payment in our portfolio and that's because our loans -- we haven't owned them long enough to know what those prepayment rates will be other than we expect it to increase, and we have seen some of that in the last few months. That being said, we can't pinpoint at loan by loan by loan other than you can say, yes, the ones with the most equity and the biggest coupon are the ones most likely to prepay. But one thing we have seen is though is our California ARMs tend to prepay especially when they have 18 consecutive payments. We see a lot of one-year treasury plus [275 ARMs refinance into new 7-1] ARMs and do a small cash out refi, we do see a lot of that.

So given that and the acceleration in prepaid fees, when that in theory cause an acceleration of accretion within the underlying portfolio in there.

Yes, we expect that pre-payment cause an acceleration of captured discount, it would also increase taxable income. Keep in mind, on a GAAP basis, we pool account. So unless pre-payment dramatically changes the total expected and timing of the pool wide cash flows, the pre-payment of any one loan is immaterial to the performance of the portfolio. But any pre-payment does accelerate GAAP -- I'm sorry, any pre-payment does accelerate tax income and to the extent that the pre-payment is a trend, and we can verify it statistically that also increases pool accretion for GAAP as well on a prospective basis.

And then there is a tick down in other income, was there anything you want to site in particular in regards to the decline in other income?

That was just less profit from the sale of REO.

Part of your portfolio that you originally started purchasing is getting to a point where it may be more attractive to other buyers in the market. Are you looking at potentially offloading certain portfolios in order to monetize potential gains or the market demand for that type of asset.

We absolutely are. Obviously there's some REIT restrictions as to how much you can sell and what you're holding period had to be to make sure you're not kind of that wasn't you're operating business plan when you bought the loan. That being said, there is no question that there is better owners than we are of [24 to 24s and close to 24-24s] and we've seen 660 FICO, 90 LTV, 24-24 sells to basically 2.5 points above Ginnie Mae securities. So to about a 4.5% yield. We are a seller, not a buyer of those.

And then you've also brought down, you mentioned the recent acquisitions have had a higher price to UPB, but obviously lower price to collateral value. So in theory, you would have lower risks, but also probably lower accretion income from this asset.

Low accretion income, but because of the absolute dollars of equity, combined with the lower LTV, you would expect more pre-payment as well.

So what would be the difference between the effective duration that you would expect from that type of asset versus some of the purchases you may have had in the first quarter?

We would actually expect the effective duration to be a little bit lower.

Would you expect it to be a couple of years or is there something more, six to 12 months?

I think it could be as much as 1.5 years or 2 years less, yes. We would expect a lot more pre-payment from the low LTV absolute dollar. So when you think about it portfolio that's weighted average LTV say of 90% and where you have a third of the portfolio is under 75 and you have average equity of $90,000, a lot of those are going to go away once they are 18 of 18.

But your total effective yield, do you expect it to be similar to your past purchases or do you expect it to be slightly lower given the risk adjusted?

We have modeled it lower, but probably expected ultimately to be the same.

Jim Delisle, Wasatch Advisors.

Looking at the book value and preferred NAV, we are kind of sitting right now at the end of second quarter, kind of exactly where we were at the end of fourth quarter, and for pretty much as a result, and building on Kevin's question really, as a result of the diluted equity raise. As we -- you can understand the optics of that, the optics of that are not nearly as flattering as the operating statistics the people spend the time, know how you're doing, what you're doing under the hood here. So I go to what Kevin -- I think building over what Kevin was just saying, how long do we have to wait to start seeing the monetization of the 18 to 18 and whatever re-performing periods that we get to the part of the curve where we can start financing these portfolio acquisitions through the sale of existing appreciated asset instead of through diluted share offerings?

One, you probably know that when we get our equity offering, we try to do the smallest possible offering as opposed to what the largest possible offering. We actually had order book materially bigger but we didn't want to issue more we just wanted to have enough equities so that we could make the acquisitions of things that we thought that were pretty cheap along with debt financing.

Number two, there are restrictions to some extent on what REITs can sell and how much of their portfolio and over what time periods. And we have analyzed those and our expectation is that we don't want to keep 24 and 36-month clean pay, 690 FICO, 90 LTVs on our balance sheet, we're not the best owner of those. There are other owners who have very cheap cost to capital, the MetLifes of the world for example, who have been our regular buyers of clean pay, there is also some REITs that buy some clean pay that we know of.

We've also seen a few banks buy some 24-24 high FICO clean pay. And they've -- in today's world deposits have negative rates not positive rates and so they are active buyers. So I would expect that we would be more involved in that. Some of it is waiting for our loans to be owned by us long enough for them to have that long of tenure. We can collect 24 payments on loans that we only own for six months. So we have to build the pay history and then we can create portfolios that we can sell or we can do other things with. But that's definitely part of the plan.

Reading into what you're saying -- would I be fair to read into that you had enough demand that you could have maybe offering last quarter larger if you did not believe that you would be somewhat self-financing going forward or is this the kind of situation where because you don't want the cash drag on the balance sheet we can look forward to a bunch of small offerings in the future until these things have seasoned long enough?

I don't think I agree with the second one, and I think the -- or the first one actually but our Board was pretty clear that if we could issue equity at $18 a share they would do a bigger offering and at $16 a share they would issue a little bit more than they did. But at the price they did, they had an absolute cap that they weren't willing to go over and they do believe that there are some benefits to scale in terms of liquidity of the shares. They do believe that we have and are going to have a lot more acquisition opportunities, they also think that given the execution of our securitizations and our financing facility that we just closed that we have in the press release cheaper financing facility that the leverage rates of return are actually going up rather than down. For what we're buying, so they want to do a little bit of equity, but not a lot of equity. I would not read into it, that's a regular pattern.

I would say that the regular pattern is we buy stuff cheap, we look at the optimal way to finance it and we have plenty of credit availability, plenty of bond availability in the securitization front and also plenty of JV availability to acquire things and earn fee income, rather than just interest income. And I think we'll be using all three of those approaches.

And if there's anything you can say on the JV side that hasn't been said already I'd love to hear it?

We're working on another one. We are going through and actually have to give numbers on some loans in our JV over the next few days. And I don't know whether the numbers will be adequate enough to buy anything, but we are negotiating with a bank on some loans in a JV structure, 100 and some odd million of loans, which we'll have a little more information on over the next few weeks, if something happens in that particular JV. That being said, we have a number of people who have reached out to us to buy loans together. We had a couple of funds, who actually want to go raise funds and use us and pay us fees for doing the analysis and being effectively a sub-advisor. We have and obviously, we continue with double line looking at pools of loans and trying to buy things together.

(Operator Instructions)

Brock Vandervliet, Nomura Securities.

I guess I'll take another crack at net interest income topic. And you noted in the release that the loans you purchased are only on the books for 42 days. You've already talked about the extension issue, if we kind of just try to parse those apart in terms of the effect on top line loan interest income, what was the main driver or was it both equally.

Probably -- Mary is here too, so, she'll correct me if I'm wrong. But the number of days probably cost us about $0.05 or $0.06 and the yield or what I'll call the too much performance or far more-than-expected performance probably cost us maybe as much as $0.06 -- $0.05 or $0.06 also. Some of it's prospective, so to the extent that you had some change in the second quarter, so maybe a little bit less. So call it combined $0.10 maybe.

I think the biggest -- one of the bigger driver is really just we bought a big portfolio in July, but that was funded partially in June.

Funded with debt in June, partially with debt in June, all on top of the securitization we did in April, as well as not getting all the April money invested for the entire quarter.

A large chunk of the portfolio came in on June 30 via 2Q acquisitions. So we didn't record any interest income on those, but the debt was outstanding for a good chunk of the quarter.

So we drew down our repo in advance of actual closing.

You mentioned leverage is stepping higher in July and then in August how -- where do you see it kind of topping out?

Our Board has told us they're -- given our RPL portfolio that they're comfortable with about 2 times. So we would expect it to over time migrate towards that area, not by July though.

And lastly, the other favorite topic of the call seems to be pre-payment speeds. Can you quantify, I would have thought in a portfolio like yours, the CPR if you expressed it that way, would be mid-single digit. Is that sort of the rate zip code and are we talking about --?

That's the right zip code based on pay history kind of in the first 12 months of ownership of a loan. But once the loan gets past 12, the 12 increases, once it gets to 18, 18 increases more and then once it gets to 24, 24 accelerates pretty significantly.

Okay. So are you saying --?

As you might imagine, if you think of FICO in the first and second derivative the more payments someone makes, the higher the first and second derivative are. The FICO increases in an accelerating rate the more payments they make. So as a result, it doesn't move linearly prepayment increases.

And so if you were to carry that forward, where do you based on those factors is 10% CPR, [12%]?

I can tell you in -- for low LTV loans we're seeing FICOs -- we are seeing for low LTV loans that have really good pay histories we're starting to see CPR get into the mid to high double-digit -- mid-to-high teens.

All right. Hadn't expected --.

That's before 24-24, that's really kicking in around somewhere between 12 of 12 and 18 of 18. For low LTV loans that have at least about $50,000 or $60,000 at absolute dollars of equity. Now the one thing we're not seeing Brock is 7 coupon $70,000 loans payoff. They never -- they are just not prepaying even though they are low LTV, but I think a lot of that has to do with the new rules of caps on the points and fees that originators can earn that they just don't want to do $70,000 loans.

And do you think this is mainly driven by your increasing preference for higher quality loans with more equity, and therefore an embedded desire to take advantage of prepayment opportunities than deeply distressing?

I think the higher the average underlying -- the better quality of the underlying property the higher the average UPB, the more absolute dollars of equity there is more economic incentive there is to refinance at current rates, no question about it. The other thing we're seeing though is a lot of people going who have adjustable rates going to fixed rates. In California, the [7-1 and 10-1] it seems like they were indented for the state of California long time ago and a lot of those [7-1s and 10-1s] are now linked to their period of being one-year ARMs, they change every year now instead of being fixed for [7 and 10] they are longer hybrids, we're seeing those being refinanced back to new hybrids.

(Operator Instructions)

There are no further questions, so this concludes our question-and-answer session. I would now like to turn the conference back over to Larry Mendelsohn for any closing remarks.

Thank you very much everybody for joining our second quarter 2016 conference call. We appreciate your interest in Great Ajax and we look forward to more calls and even more questions. Take care and have a good night.

The conference has now concluded. Thank you for attending today's presentation. You may now disconnect.