Great Ajax Corp (AJX) 2016 Q4 法說會逐字稿

Good day everyone and welcome to the Great Ajax Corporation fourth quarter and year-end 2016 financial results conference call.

(Operator Instructions)

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

Thank you very much, I appreciate it. Thank you everybody for joining the fourth quarter and year-end conference call for Great Ajax Corp. Before we get started I want to make sure everybody looks at page 2, the Safe Harbor disclosure and then we'll get right on into the material.

I want to make a general comment first. Number one, it was actually a good quarter in terms of the underlying economics and what I will call NAV creation. But we also, as you can tell and as we talked about on our third-quarter conference call, there is some GAAP noise from calling our 2014 securitizations and also some REO noise from accounting requirements. On the flipside there's also been, especially in the last three weeks, some really positive developments in the loan market and I will talk about those in more detail as we go through.

In general about Great Ajax, I can't ever say enough about our sourcing network. It is really important to our ability to acquire the types of loans we acquire, in a markets we want them, and at the prices we pay relative to others. And as you will see the prices we pay versus where clean-paid loans are now is even more dramatic, and we will talk about that later.

Over 90% of our acquisitions continue to be in privately negotiated transactions; in fact, it's probably becoming even more so and it's even more loan-by-loan than ever before. We have done 181 transactions since July of 2014 and 12 transactions just in the fourth quarter of 2016.

Proprietary analytics, another thing I can't say enough about. We spend a lot of time looking at data, data, data to determine the target loan characteristics, forecasting performance on those loans and looking at all kinds of different patterns and other statistical patterns regarding the loans that we look at as well as buy.

It also helps us in terms of neighbor determination in our small balance commercial portfolio as well as sub-MSA targets for our residential portfolio. Affiliated servicer, Gregory Funding, captive to us, very important that we were integrated in terms of the way we would look at portfolio management and the way we get in front of the curve with regard to working with borrowers.

Especially in the REIT world we are at a much lower leverage then most other REITs you'll find. Year end leverage was 2.37. We, obviously for the quarter on an average, we are lower than that. As we discussed in prior calls, as our portfolio has become much more re-performing loans versus non-performing loans, we have been comfortable with increasing our leverage gradually. But even at our quarter, or year end, 2.37 times leverage, relative to our peers, we're still very lowly leveraged.

Eight securitizations, we like nonrecourse fixed-rate funding. We have a great group of bond investors, many whom also co-invest directly in loans with us and many others who were working on setting up JV structures to co-invest in loans with us as well.

If we flip to the next page, we will get into the quarter itself and the year end. One thing I want to preface it by is, since early 2015, we have purposely focused on buying much lower LTV loans with underlying properties in relative market housing deciles, what I will call 4.25 to 7.5.

And when you think about that -- if you think of housing deciles 1 through 10, location by location by location, the price range of a 5 in one location is a different price range than a 5 in another location. So we care about prices relative to MSAs as opposed to absolute prices.

We also like minimum absolute dollar amounts and equity thresholds in the loans that we have really been targeting. Our analytics suggest that performance and prepayment are not just tied to LTV but to absolute dollars of equity. A $70,000 property with a $50,000 loan is much less likely to prepay than a $700,000 property with a $500,000 loan.

A lot of that has to do with the new origination rules but we have definitely seen that pattern and it seems to be accelerating rather than decelerating. Also larger absolute dollars of equity are natural hedge against potential housing price declines, against slower economic conditions, and against re-default or default probability. Although based on the stock market, none of that is ever going to happen again.

Additionally, nearly 50% of our loans are either ARMs or step rate loans, which are really just ARMs with a preset forward curve and these provide a bit of the built-in hedge against interest rate rises. It also -- we see that ARMs tend to prepay faster than fixed rate loans given all other characteristics being the same.

We are able to build our portfolio to target specifications because of our sourcing, in the many cases we negotiate on a loan-by-loan basis with sellers. And the classic example of that was in the last week of the year, we bought about $50 million in loans in several transactions and the sellers showed us well over $100 million and assets to go loan, by loan, by loan, versus their carrying value, so that we could get enough loans that they could, combined, get $50 million in proceeds.

So that -- every seller has there own reasons for selling and we go loan by loan to try to be a liquidity provider when they need it. If you look at what we bought in Q4, we bought $129 million of re-performers, with a purchase price, re-performers and non-performers. Re-performers we paid 86.9% of UPB, but only 63% of collateral value.

And we bought a small amount of non-performers, purchase price only about $2 million and change, but 59% of UPB and 53% of collateral value. So still, very low prices especially relative to property value, and $149 million of UPB that we bought in the fourth quarter, keeping in mind about $50 million of the purchase price happened in the last week of the year, which means we didn't gain the income from that $50 million in the quarter, although we did have expense related to due diligence.

Interest income continues to be just fine. Earnings $0.33 a share, and we'll talk about some of the noise in the taxable income, also $0.33 a share, so a significant increase in taxable income. That had as much to do with the increase in cash flow velocity from more loans, especially extremely more loans paying than we would have anticipated.

I want to talk a little bit about the two noisy pieces of the quarter, one the REO impairment of $1.3 million. We've talked about the concept of REO impairment in our November call. I'll go through more detail, we have a separate page on this in this presentation on page 9. But it's really not something that we don't know when we buy the pools, but GAAP requires us to report in a certain way that changes timing.

Also we have $600,000 of expense related to the early call of our 2014 securitizations. We called those a year early, so there were $600,000 of deferred issuance costs that would have been amortized over 12 more months, but because we called them and refinanced those with higher leverage at lower rates in October of 2016, we take the $600,000 all in Q4 rather than over the 12 following months that would have been, had we not call them early. We also talked about this as it was a subsequent event prior to our last conference call, and we talked about that on the call, that it was coming, so I'm hoping everybody was expecting to see that $600,000 charge.

We continue to see very strong payment performance from our loan portfolio with significantly less re-defaults that expected. It's great for cash collections and it's great for NAV. I'll go through why it's great for NAV in more detail on page 10. However if we just look at our Q4 acquisitions of RPLs, we paid 86.9% of UPB and 63.5% of property value for those loans. On those loans as we see -- we'll see on migration charts later, when those loans get to be 12 consecutive payments to us, they're probably worth 10 to 15 points more than what we paid for them. And we'll take a closer look at that when we get to page 10 in the presentation.

From [portfolio] perspective, it's re-performing loans, re-performing loans, re-performing loans. We've now reached a point where re-performing loans are almost 93% of our total portfolio, and non-performing loans are 7% of our portfolio. We expect the trend in re-performing loans to continue in the fourth quarter. Re-performing loans were probably 99% of what we bought, and we would expect the re-performing loan pathway. We think it's a better economic pathway for loans that we see, and where we see loan markets than non-performing as well. And as you can also see we have plenty of property value, plenty of equity in these loans, even on a UPB basis, and our purchase price is even lower than that.

We jump to page 6 on the re-performing loans. You can see from this chart we continue to buy low LTV loans in our overall (inaudible) purchase price is only 64.8% of the initial collateral value. So think of what that means. That's no HPA, that's 64.8% of the collateral value at the time we bought the loan, and 79% of UPB. Basically in re-performing loan land, we're playing offense and defense at the same time.

On the non-performing loans side, you can really see they continue to decline as a percentage of the portfolio. Remaining NPLs are 57% of the initial underlying property value. Again, 57% of collateral value at the time we bought the loans, keeping in mind that most of the NPLs we bought on our balance sheet we bought in late 2014, early 2015. So most of the RPLs have had at least two years of HPA from that 57% number.

We jump to page 8, no real changes from Q3 in our targeted markets. California continues to represent almost 30% of our overall portfolio. Southern California is 75% to 80% of that, so about 25% of our entire portfolio is Santa Barbara and south in California, but in our target markets did not really change from Q3 to Q4. We have increased our small balance commercial focus and an increased small balance commercial holdings in urban metro New York during the time period, and we have some more small balance commercial in the pipeline also that we've closed so far this year.

On page 9 we get to talk about real estate owned. I'm sure everybody's interested in hearing about it. We talked about it a little bit on our Q3 call but decided to actually just show the numbers in a table so that everybody can understand. If you think about how GAAP works, if we have 160 REO and 40 or 50 of them have built-in loss and 110 of them have built-in gain, we don't get to take the gain to offset the loss, unless we're willing to then go to an entire mark to market of our balance sheet for income purposes, which we think would not be a good accounting concept for stable re-performing loan portfolio.

So as a result we get to take GAAP losses before we get GAAP gains, even if none of them has actually closed in a sale. So we've gone through all of our REO, including the ones that we are considering renting, and we've said we're going to take an impairment on the ones we have reconciled will likely have a loss, and the ones that likely have a gain, we'll just show you on the list here that we think we're going to have a gain of about $4 million on the remaining REO post impairment.

Our experience over the years with non-performing loans, and almost all of these REOs come from late 2014, early 2015 non-performing loans, and our experience with non-performing loans over the years that we've been doing this is that unlike the [tail] happening at the end, in our experience of non-performing loans is the tail happens first rather than last.

This means that the better REO comes later and the worse REO comes earlier. The worse REO is usually related higher LTV loans that have low or even negative dollar amounts of equity. They tend to be in worse condition when you get them back, and they tend to be relatively lower market value properties within an MSA. So when we talk about, we really target deciles 4.25 through 7.5, these tend to be deciles 2 to 3.4 or 4. They tend not to be deciles 6 through 7.5 or 8 in terms of relative value [in] their MSA, not pure absolute price.

If you look at our REO portfolio overall, based on updated valuations and what we believe are reasonable expense expectations for repair and selling costs, brokers, title insurance, we believe we have material overall built-in gains, but GAAP doesn't permit us to offset this, again, unless we're willing to go to 100% mark to market for income statement purposes on our balance sheet. We don't think that is a good change to make, to go to 100% mark to market, even though it would probably make our book value closer to NAV, it would probably increase equity. But it would also create a bunch of choppiness and unpredictability to the balance sheet and income statement.

You go to the next chart, and this is something that we've been talking about for a while, which is our portfolio just keeps paying, even non-performing loans are paying, even loans that had only made two payments when we bought them, and people had personal credit histories that were going in the opposite direction, even they continue to pay. We're seeing re-default rates in some cases 50% to 70% less than what we've expected, based on certain loan characteristics.

And when we think about portfolio migration, it's very important to understand how that also has implications for implied NAV or implied future book to our portfolio. So if we look at this chart, this chart represents payment history for loans that we've acquired, for only payments made to us, so without regarding loans that had been made to the previous owner. So when you think about that, if we've owned a loan for less than 12 months, it couldn't have made 12 payments to us, and as a result it would not show up as 12 for 12 in this table. And in fact, if we bought a loan at December 31, it couldn't even show up for anything really showing material payment to us in this table.

So keeping that in mind, just from payments to us, we have $422 million that have made at least 12 consecutive payments to us. In the last three weeks, we've seen three significant pools of loans sell that were 90% 12 of 12 payments. One had a 3% coupon that sold for a $93 price. One that had a 5.7% coupon that sold for a $104 price. And one that had a 7% coupon and was 100 LTV that sold for a $104 price.

So when we start thinking about that, and then we think back of we're buying our loans all in for over the life, for under $0.80 on the dollar, and we're seeing 12 for 12s trading at what is now somewhere between a low 4% and 5% yield all in. And in fact there was one portfolio that traded basically at 240 or 250 to swaps, which makes it around 4%.

So when we start thinking about that, and then when we start thinking about, okay, if we were to take into account prior servicer payments, so if we look at loans that we owned and we add in payments made to prior servicers, so we take those payments plus payments to us, that $422 million number rises to be 60% to 65% of our entire portfolio. So we start thinking about 60% to 65% of our portfolio we paid in the high [$]70s for. Okay?

Even in Q4 we bought extremely low LTVs with big coupons at $86 price. As 12 for 12s, these are all worth into mid to high $90s and in some cases below over par. So from an NAV perspective, these developments in the clean pay loan market given our business strategy of we buy sub-12 for 12s and we make them perform, it's really created a good environment for us.

The other thing is that we've looked at also rated securities as well. All of our securitizations to date have been unrated. We've just met with all the rating agencies. We've seen a number of re-performing 12 for 12 and 90% 12 for 12, Securitization has been done by a number of large funds, and we're now working with the rating agencies for a rated structure. And we think that just in a rated structure, we can probably bring down the funding cost, the debt funding cost of our securitizations for those loans in a program by 1% to 1.5% per year, not 1% to 1.5% total.

So when we think about creating NAV, while all this payment has actually overall reduced yield on loans to us in terms of net interest income, because of its duration extension. On the flip side it's made them so much more valuable, and recent transactions in loan markets have really come to show that to us.

We've also thought about selling some of our clean pay loans. Subject to REIT tax rules we spend a lot of time with our tax folks over at Deloitte who've been very helpful about nailing down the best way to go about it, and the ways to do it within the REIT structure with regard to all these clean pay loans, given market developments for the value of clean pay loans.

On page 11, and this is something we've shown every quarter, but this is another way to look at NAV, other than just mark to market of loans, but this way looking through effectively bond equivalents of loans. We've updated this page through our 12/ 31/2016 portfolio. When we think about NAV another way, as forecasted what I call by the structured finance market, and also what this implies for our return on equity and return on average equity, if we were to sell our subordinate bonds at a little less than current market prices, spreads have actually have tightened in the subordinate bond market, and very much tightened in the senior bond market, return on average equity on loans would be, I say extremely, but probably a little bit more than extremely high.

This chart, if we sold our subordinate bonds, we'd have a remaining cash basis of about $36 million, and our residual after our senior and subordinates would be $267 million. So [effectually] you'd own 25% of all the UPB for 3 points, and your underlying collateral would be 93% re-performing loans that are outperforming expectations, 65% of which have made 12 consecutive payments. And even the non-performing portion you own at 57% of initial collateral value. And I think a lot of people would look at that and say, hmm, that seems really cheap.

Now we can finance this mortgage bonds and to some extent replicate this with a little less leverage than if we were to sell the subordinate bonds if we really want to. Today we financed a very small percentage of the subordinate bonds, but we always do have that availability if we choose.

In subsequent events on page 12, through much of January and early February, the loan markets were in what I would describe as a post-election paralysis, as many of the participants, both buyers and sellers, were trying to figure out what the election actually meant to them. I think it's only in the last three weeks or so that the conditions have become or reverted to a more normal environment. We've seen a lot of loans come out of some of the big banks. Fannie Mae and Freddie Mac are out with very large non-performing pools, and that's all happened in the last three weeks or so.

We're currently in loan by loan negotiations on several significant RPL pools, with a couple of different repeat sellers. The goal would be to close those in late March or early April, and if those do happen, it would probably be somewhere between $90 million and $115 million of underlying UPB. January and February acquisitions are so far pretty light, primarily because for about the first five weeks of the year the loan market didn't really function. It certainly functioned not at all pre-inauguration, and then it functioned completely dysfunctionally for the two weeks following inauguration.

One thing I do want to talk about is we've increased our focus on small balance commercial loans, particularly smaller multi-family and mixed use in urban markets in our target markets. We've created a small balance commercial origination platform here, and in Q4 we funded our first newly originated small balance commercial bridge loans and repositioning loans on mixed use properties on New York City metro locations.

We expect our small balance commercial origination, as well as our loan purchase strategies for small balance commercial, to grow materially in 2017 and 2018, and it's really become from our expectation a significant portion of what we expect to invest in as well over the next few years.

Dividend $0.25 a share. Record date March 15, payable March 31. You'll see that taxable income is $0.33 for the quarter. This dividend of keeping it the same at $0.25 is the final dividend related to 2016 income, with the $0.33 only a small portion of our dividend for 2016 will be return to capital. But it also, given that you see taxable income increased because of cash flow velocity, that has implications for future dividends as well.

And with that I'm happy to open up to any questions anybody might have.

(Operator Instructions)

Steve DeLaney, JMP Securities.

Thanks, Larry, we appreciate as usual the thorough update of what's been going on at Great Ajax. This decision you made to sell all your REO obviously is significant, if your valuations hold up and the other side of the trade agrees with you, it's a nice almost $0.25 per share, and whether we want to call it earnings or book value, it's a nice number. I'm just curious about the process here.

We obviously have a number of large single-family rental enterprises including a big IPO. Is it possible this is bulky in terms of how it's disposed of? Or is it going to be a more tedious one-by-one over the next year or two? Just a thought about how you expect to liquidate it?

Sure. I would say it would be more tedious than I would like, but there may be a small amount of bulk. By late 2014 and early 2015 NPL purchases had a fair amount of concentration in what we bought. They were heavily concentrated in Florida, New York, New Jersey, and Maryland and so as a result of that, the REO has some concentrations.

The early REO is less likely to be in bulk than the later REO, just because the later REO is probably at the middle to higher end of typical REO to rent, and the early is probably at the very low end of REO to rent. So for some of the later REO, including some that's on our balance sheet already, as opposed to what's going to come on our balance sheet from loans in the next 6 to 12 months, some of the early REO would likely just be sold one-off, because it's not as concentrated and it's lower value.

But some of the later REO I would think that we will lump together for bulk buyers, even if they're bulk mom-and-pop buyers. Because one of the things people don't, I don't think really realize is that as big as the REO to rent market is, and all the public companies that have large portfolios of REO to rent, it's a still tiny fraction relative to the mom-and-pop market of REO to rent. One of the things that we've learned over the years, particularly in some of our markets, is the number of people who have 5 and 10 rentals is much larger than anybody believes it is.

Yes, very much a cottage industry for sure. That's helpful. So it sounds like we should plan on timing that out over maybe the next 4 to 6 quarters rather than thinking that it's the next 2 or 3.

It's certainly not the next 1 or 2 quarters. I think a significant piece of it will be this year.

Okay. And just one quick follow-up if I may, just looking at your eighth securitization, and I'm not personally familiar with these, I guess I am a little bit, but it looks like I assume they're being driven as far as advance rates on the senior notes off a percentage of UPB, and this one looks like it was 65% of UPB, that makes sense. If I'm thinking about it, if your average acquisition cost is 80 of UPB, if I'm looking at a bond investor giving you cash at 65, you would put up, that's a 15% delta. So if I look at that, I get 4.3 times leverage to the senior. Am I thinking about it correctly?

That is the correct way to look at it, yes.

Okay. And then the subs you're just holding, so essentially you're securitizing part of your equity and you're keeping those subs as you showed us on the table near the back of the deck?

That's exactly right. And we put arbitrarily large coupons on the subs. The subs all carry a 5.25% coupon, so that as much linkage can come out of the securitization as possible on a cash flow basis.

Got it. Well thanks for comments, Larry.

And by the way, the securitization we did in October, 65, when we first started doing this securitizations in 2014, the advanced rate was 45 or 50. Now we've developed a group of bond buyers who don't just buy bonds from us, they coinvest directly in loans with us too. And they like the way we look at loans, they like our servicing and the way we service. They've come here and spent a day or two each looking at our servicer and getting a feel for everything and as a result our advance rate has increased significantly, and our all in cost of funds has actually come down as the advance rate's gone up, rather than gone up as the advance rate's gone up.

So we're really excited about it. We're also excited about -- I don't want to say we're excited to work with the rating agencies, because it's certainly a detailed process which will take a fair amount of time, but we're excited to get a program going with the rating agencies, because we think we'll actually even further the advance rate. If you look at the advance rates on some of the -- for example the tower point deals or [NRG] is out with their securitization. Those advance rates on rated deals are very high on the clean pay loans, and we think that our advance rate will actually go up, and we actually think the cost of the financing will go down by 1% to maybe as much as 1.5% on those loans per annum. So we spend a lot of time on the structured finance market.

Great. Good to hear, thank you.

Jessica Levi-Ribner, FBR.

Hello, guys, Ted Beachley here for Jessica. So how do you guys see the effect of lower regulation on loan sales, and how are you guys thinking of that?

It's going to make banks big sellers. We've already seen the banks start. It's partially a first order effect directly because it's less regulation, but I think the bigger effect really comes from a rise in rates and the steepening of the curve, or more yield in the curve.

But banks will take assets that they currently hold more capital than what they'd like to against those assets, and they're going to sell them because they can now replace it with something they can earn some money on, which a year ago they couldn't do. So from all the banks we've talked to, we expect a lot more is going to come out over the next 12 to 15 months than they had been sellers of previously.

Okay, thank you. And then to what extent do you guys think overall yield will be removed from improving credit?

For our portfolio, the more the borrowers pay, because we buy at such a discount to property value, the more the borrower consistently pays, it actually lowers our yield rather than raises our yield, because it extends duration, increases cash flow dramatically, increases money [multiples] dramatically and increases NAV of our balance sheet dramatically. Not what I'll call stated GAAP book but actual NAV of the underlying assets, it increases that dramatically also.

So we root for performance because we're deep value investors, we're not momentum investors, so we root for great performance from the borrowers, and the downside is a loan default, and it increases yield but decreases money multiple and decreases duration.

Okay, makes sense. Thank you, guys.

Kevin Barker, Piper Jaffray.

Hello, Larry. Regarding the unrealized gain on the $4.4 million that is expected to be realized, could you talk about the transaction cost in that sale, or your expected transaction cost?

Sure. So that $4.4 million number is net of an estimate of selling expenses of 8% and also net of an estimate of repair costs.

Okay.

And net of any payments we'll have to make for previous service or advances in the foreclosure or property tax or something like that on those properties.

So it's pretty fair to say that it's pretty close to the number that you expect to realize on the income statement, when everything's said and done?

Yes. It depends on the holding period. The real estate market shuts down and we wake up two years from now and we still own it, there'll be more expenses. But as of now, from what we know, can tell from the actual properties, from seeing the insides of them, from getting repair estimates and things like that, it's kind of our best guesstimate of the net.

So do you see this as a bulk portfolio where you can get rid of it pretty quickly, or is this something where you feel like it's going to play out over several months? I mean you just said it might be a couple years here.

So a portion of it could be a bulk portfolio, but not 100% of it and that's only because a portion of it is concentrated, and a portion of it is less concentrated. I would say more than 50% is concentrated, but there's a portion that is less concentrated and would be harder for a bulk sale other than to a mom-and-pop.

Okay. And then we're hearing some softness in the Miami area. Are you seeing any softness in the property values in those markets?

Yes, we're definitely seeing softness in the Miami area, but in a different spot or demographic than our typical loan. Where we're definitely seeing the softness is the higher end condo market, what I'll call more the flight capital market. And so condos, say about $600,000 and higher, we're definitely seeing softness in that market.

Part of it is due to the strength of the dollar versus foreign currency. Part of it is there's probably a 5-year supply of condos either for sale or under construction.

So we're not seeing softness if you go to the west side of Miami in the gated communities where all the houses are $450,000, we are not seeing softness in that. We're not seeing softness in the eight unit apartment building in northwest Miami. We're not seeing softness in the $400,000 condo on Brickell Avenue, but we are seeing softness on the $800,000 condo on Brickell Avenue.

Okay. Going back to my earlier question, when you think about the sale of the REO and a realizable gain, how should we think about the potential for that to flow through, through 2017? Is it something that's going to be maybe one big lump sum that could happen? Or is it something where you expect it to bleed through over time?

I would expect that for the next quarter or two it would be more of a bleed versus a lump sum. For the second half of the year, it's more likely to pick up quicker than it does the first half of this year.

Okay, and then one last question. I believe you mentioned about $90 million to $150 million in UPB that is potential purchases in the near term that you believe are in the pipeline?

$90 million to $115 million, not $150 million.

Oh $115 million, okay.

And those are in negotiations. In fact for the hour and a half prior to this call, I was going loan by loan with our acquisition team on a pool that we've gone back and forth with.

Okay. And then in relation to that portfolio, what do you expect -- it would appear that your leverage would increase right off the bat, if you were to purchase that right now. Do you think you can fund some of that with sales, or some of the cash flow that's coming off the existing portfolio, in order to keep leverage relatively stable with where it's at right now?

I think that we will likely do it and leverage would increase from about 2.37% to about 2.5%.

Thanks for taking my questions.

Robert Dodd, Raymond James.

Hello, Larry. Obviously the GAAP isn't particularly favorable to your economic value, right? I mean as things age out, yields decline, NAV -- (multiple speakers)

I'm still waiting for the FASB to ask my opinion about their -- (Laughter)

That would be nice, right? (Laughter) When I might look at it -- for example, you mentioned potentially selling some these assets. You've got $168 million UPB that's 24 for 24 (inaudible) 12 for 12, right? I mean, that's really valuable stuff.

You mentioned talking to talking to Deloitte. What's the planned time scale if you can give us any color on where you might start to monetize that? Just because obviously it's really -- and you shouldn't run your business on a GAAP basis, because that's a different thing on economic value, right? But it would potentially shift the needle on how those things show up in terms of GAAP earnings, in terms of recognition yields and things like that, and for modeling purposes those things can move around as you start to exit those. So do you have any idea what time frame what you might start to be looking to sell some of those early performing loans?

We've actually had some preliminary discussions with a couple of loan buyers. The tax rules have all kinds of limitations as to how much you can sell in a year, what holding period you might need, and what safe harbors there are, and whether or not you need to do some of it from a TRS and others not from a TRS, in order to be within the safe harbors.

The other thing we don't want to do is we don't want to put ourselves in a situation where we get too close to the safe harbor limits, and then something happens that we are just drooling over and we can't sell more at that time. So always want to have a buffer of availability to sell more. So we are both working with the rating agencies and evaluating some portion of those assets in sale as well. And my guess is we will during this year follow a path of some of each.

Okay. Got it. Thank you. I've got figure out how to model that now, but thank you. (Laughter)

That I can't help you with, unfortunately. (Laughter)

(Operator Instructions)

Brock Vandervliet, Nomura Securities.

Thanks for taking my question. I guess I got the counterpoint to the prior question. I know you don't want to be a slave to running the Company on a GAAP basis. But I feel like we all need some sort of level set on earnings power here and I add back the two special items that come back to around low 40%s --

The two special items were about $2 million, which is about $0.12 a share. Right?

Okay. So I get to the mid 40%s --

Mid 40%s, and then you have $50 million of acquisitions done in the last three days of the year, for which all the expenses in that quarter, so that hides some of the earning power from that, but let's call that another $0.01 to $0.02.

Okay. All right, adding the $0.02 on mid 40%s, consensus is 58% right now for the first quarter. Okay, that's helpful. It's just helpful getting some sort of a level set on earnings power here. Is there anything else --

Hang on a sec. You know me, that I'm much more -- since my whole background has been in, from years and years ago, was on the distress side. Is it cheap, is it cheap, is it cheap mentality, as opposed to what's my current earning, my current earning, my current earning. So I'm much more focused on the NAV side.

I totally get the earnings side and the consistency side, but from a running the Company perspective, I personally, I know I don't speak for everybody but I personally am extremely value focused, and are we creating built in value that eventually gets captured through realization, through pay off of loans, through sale of loans, through rated securitizations, things like that. But I totally get what you're saying on the consistency of earnings basis.

Yes, I think all the investors respect how you're running it and what you're generating. It's just I don't think there's much benefit in having EPS numbers out there that just aren't realistic, so I appreciate those comments. Is there any other callouts you'd make in terms of the expenses in loan servicing, or any other items that you would call special or whatnot this quarter?

Yes, and the loan servicing expense are little higher because we've had to pay some interim loan servicing expenses during the quarter which are a little bit abnormal.

Brock, we paid those - we get an invoice for those. We don't necessarily accrue for them, so if they come in higher than expected, it's a little distorted. And we actually had some pools that took a bit longer to board than we expected so the servicing expenses were a bit higher. And some of that was expense incurred in the third quarter, but we didn't see it until the fourth, so there's a little bit of timing difference as well.

That's right. If we don't know what's going to happen in the third quarter and the bill shows up in the fourth, it becomes a fourth. So there's a little bit of that.

Like I said, we closed a significant number of loans in the very end of the year, and all that expense shows up in Q4 without any offset. That has happened to us in some other quarters too. That happened to us in the second quarter, if you remember, not in the third quarter but in the second quarter. And that's something that will happen to the extent people need specific execution on a very specific day that's at quarter end, or in this case at year end.

Now, there's benefits from that in that we get to buy loans really cheap because of it. As you might imagine, the person who needs December 31 no matter what sells at a different price than the person who doesn't care between December 31 and January 1. But that's one way.

Another thing to think about also is on our funding costs. In one of our repo facilities, in the extension of that facility in November, the cost and the fees have come down, so we would anticipate that the funding expense related to that line would be lower going forward that it was in 2015 or 2016.

Got it, okay. Very helpful, Larry, thank you.

This will conclude 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 being on -- for joining in on our 2016 year end and fourth quarter conference call. Please feel free to contact us if you have additional questions over time, and we look forward to talking to you again in a few months. And with that, everybody have a great evening.

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