Armour 住宅房產信托 (ARR) 2016 Q1 法說會逐字稿

Ladies and gentlemen, thank you for standing by. Welcome to the ARMOUR Residential REIT, Inc. First Quarter 2016 Conference call. During the presentation, all participants will be in a listen-only mode, afterwards we will conduct a question-and-answer session. (Operator Instructions) As a reminder, this conference is being recorded Tuesday, May 3, 2016. I'd now like to turn the conference over to James Mountain, Chief Financial Officer. Please go ahead.

Thank you, Christy. And thank you all for joining ARMOUR's first quarter 2016 earnings call. By now everyone has access to ARMOUR's earnings release, Form 10-Q and April monthly company update, which can be found on ARMOUR's website. This conference call may contain statements that are not recitations of historical fact and therefore constitute forward-looking statements within the meaning of the Private Securities Litigation Reform Act of 1995. All such forward-looking statements are intended to be subject to the Safe Harbor protection provided by the Reform Act.

Actual outcomes and results could differ materially from the outcomes and results expressed or implied by the forward-looking statements due to the impact of many factors beyond the control of ARMOUR. Certain factors that could cause actual results to differ materially from those contained in the forward-looking statements are included in the Risk Factors section of ARMOUR's periodic reports filed with the Securities and Exchange Commission. Copies are available on the SEC's website at www.sec.gov. All forward-looking statements included in this conference call are made only as of today's date and are subject to change without notice. We disclaim any obligation to update our forward-looking statements unless required to do so by law. Also our discussion today may include references to certain non-GAAP measures, a reconciliation of these measures to the most comparable GAAP measures is included in our earnings release, which can be found on ARMOUR's website. An online replay of the conference call will be available on our ARMOUR's website shortly and will continue for one year.

ARMOUR's Q1 core earnings were $30.3 million or $0.72 per common share. That represents an annualized return on equity of 9.9% based on book value at the beginning of the quarter. ARMOUR does not use hedge accounting for its GAAP reporting. As we've seen in prior quarters, fluctuations in fair value of our open interest rate swaps remain the dominant factor in GAAP P&L while the inversely related mark-to-market on our agency securities flows directly in the stockholders' equity. These factors are responsible for our GAAP net loss of $279.5 million or $7.33 per common share. And together they contribute about $120 million or $3.25 per common share to our book value decline for the quarter. At March 31, 2015, ARMOUR's book value was $24.48 per common share and we estimate that book value has improved by 4.25% to $25.52 per share at the end of April.

We paid monthly common dividends of $0.33 per share in the first quarter. Consistent with our discussion last quarter, we paid common dividends of $0.27 per share for April and declared common dividends of $0.22 per share for May and June. ARMOUR also completed its previously announced $85.2 million acquisition of JAVELIN Mortgage Investment Corp. on April 6, 2016. The JAVELIN acquisition allowed ARMOUR to acquire an attractive portfolio of non-agency and agency securities totaling $664 million at a discount to current market prices. In Q2, ARMOUR will recognize a one-time gain on the transaction of approximately $6.5 million after reserving for fees on JAVELIN's management agreement. We recognized approximately $1.5 million in transactions cost in Q1 and expect to book another $1 million in transaction cost in Q2.

Now I'll turn the call over to our Co-Chief Executive Officers, Scott Ulm, and Jeff Zimmer to discuss ARMOUR's portfolio position and current strategy.

Thanks, Jim. Good morning. In addition of the customary SEC filings, we also provide a company update, which is furnished to the SEC and available on our website, www.armourreit.com as well as EDGAR. The company update contains a considerable amount of information about our portfolio, our hedging and financing on a timely basis. As a result of the update, along with the comments we made during our last earnings call, the Q1 financial reports that we filed last night should contain no surprises for any of our equity analysts or shareholders.

During the first quarter, we had disappointing book value results. We also had the successful and accretive to book value tender offer for JAVELIN Mortgage Corp, and perhaps most significantly, the addition of significant non-agency assets to ARMOUR's portfolio both from the JAVELIN portfolio and through direct purchases by ARMOUR. We believe that our investment in non-agency assets will work to stabilize returns going forward, limit our interest rate exposure and swap risk and lower leverage.

Our book value decline is the direct result of increased mortgage-backed securities spreads and a reversion of swap spreads to their negative levels from a brief recovery at the end of last year. Volatility is still strong as shown by the 6.5% book value decline we suffered from spread, swap, and rate moves in the third week of February alone. Since we did not make major portfolio moves, we can recover. As of Friday's close, our estimated book value has recovered 4.25% since the end of the first quarter and is approximately $25.52 as of Friday. The portfolio moves we've made, which we'll detail now, have contributed to those recent gains and have lowered our risk from swap moves in the future.

For us, the most important events are the successful completion of our merger with JAVELIN and the expansion of ARMOUR's investment program into non-agency assets. The JAVELIN merger is a book value accretive transaction for ARMOUR shareholders and represented a substantial price premium prior to the merger announcement for JAVELIN shareholders. In the merger, we acquired agency assets that fit perfectly with ARMOUR's existing book as well as non-agency assets in the legacy NPL/RPL and credit risk transfer areas. Contemporaneously with the merger, ARMOUR added credit risk transfer (CRT) and nonperforming/re-performing assets at very attractive spreads. Today, ARMOUR owns $479 million of CRTs, and $105 million of NPL/RPL securities. In the CRT transactions, we take the credit risk of recent Fannie and Freddie underwriting, which we feel is quite good in return for very attractive spreads in an untapped floating coupon.

In addition to great carry, these assets have provided attractive book value gains so far this quarter. NPL/RPL transactions have relatively short maturity, fixed rate issues that are driven by the improving housing market. We continue to believe that housing trends and strong mortgage underwriting will give a robust underpinning to the credit quality of these sectors. We see relatively little new to do in the areas of legacy MBS from 2008 and prior, though our existing assets continue to perform well as they run-off. ARMOUR owns $109 million of legacy MBS.

Like many market participants, we continue to hope for a revival in the jumbo securitization market, but see plenty of opportunities elsewhere while we wait. While we have owned more significant amounts in the past, our new issued jumbo MBS exposure is only $11 million at this point. On the agency side, the portfolio is comprised of six major components. 30% of our portfolio is 15-year pass-throughs of which 88% of those have loan balances less than $175,000, great convex assets. 26% of our portfolio is comprised of 20-year fixed rate assets maturing between 181 months and 240 months with a weighted average seasoning of [68] months. The seasoning also provides great convexity to those asset classes. 15% of our portfolio is comprised of Fannie Mae multifamily bonds or DUS, which is delegated underwriting and servicing bonds, which are generally locked out from [prepay] for the first 9.5 years of their 10-year expected maturity. Now that portfolio right now has an 8.3 year weighted average maturity to the balloon date. The lack of amortization causes these assets to roll down the curve, particularly as they approach benchmark areas like the seven year and that provides great potential to trade tighter. 12% of our portfolio is comprised of 30-year maturity fixed rates of which 84% of those have $175,000 loan balance relapse.

While we exited the dollar roll transaction last year as its yield declined to levels equivalent to owning and financing bonds, we've recently reentered the dollar roll market with a current value of just over $1 billion. For example, we see the Fannie Mae 3% dollar roll as a 47 basis point pick up over owning Fannie Mae [threes] outright and funding in the repo market. We actively monitor the attractiveness of risk in returning dollar roll and may increase or decrease depending on market conditions. Our $96 million hybrid ARM position has a weighted average of 13 months to reset.

Last quarter, we noted in our earnings call that our dividend rate was considerably higher, more than 25% higher than our peer group and that you could expect we would be closer to average in the future. We've now adjusted our dividend to $0.22 a share for the months of May and June or return of 10.3% on today's estimated equity. This is very competitive with our peer group average, and we feel is sustainable in the current environment. Our notional swap position has been reduced from $10.8 billion at December 31 of last year to $6.7 billion today, driven by a small agency book requiring less rate protection. Forward start swaps have declined to 16% of our swap book. We did not repurchase any stock during the last quarter primarily because of two factors, we needed to conserve liquidity and constrain leverage due to the impending JAVELIN merger and our stock moved to a significantly higher valuation than where we purchased most of the stock in prior periods.

We evaluate stock repurchase on a continuous basis and maintain a current authorization from the Board to purchase up to 1.87 million shares or $39.4 million at yesterday's closing price, if repurchase is compelling it should be clear from our purchase of 17% of our outstanding shares last year that we will use that authority. Financing remains consistent for us and reasonably priced for our business plan. ARMOUR maintains MRAs with 38 counterparties and is currently active with 30 of those for total financing of $10.1 billion. We have $100 million advance from the Des Moine FHLB that does not mature until December of 2016. We frequently analyze opportunities for financing for periods of the year and longer and will add to this [store] book when it is attractive. New counterparties and new structures under review hold promise for additional compelling sources of portfolio funding as well.

Our overall outlook for our business is constructive. The addition of non-agency assets will give us an attractive opportunity that lowers rate sensitivity, swap exposure and leverage. The backdrop of the US economy making painfully slow but relatively steady progress bodes well for credit exposure in the residential sector. The extremely slow pace of domestic expansion combined with international headwinds we believe lowers the ultimate scale of rate risk. Nonetheless, we remain wary of volatility and will continue to carry substantial protection against interest rate risk. We currently see our equity allocation of non-agencies at approximately 23% of our capital base. We define that equity allocation as a percentage of our equity tied up in haircuts for repo and we expect over the relatively short term that this will increase to about 25%. While gross portfolio allocations will show a much larger quantum of agencies on our balance sheet, we think the purest way to think about capital allocation is equity committed to financing haircuts in each sector.

Non-agencies have substantially higher haircuts but equivalent or better equity yields. Equity is not tied up in financing haircuts as are liquidity, and that liquidity is available to any part of the portfolio. Hence, our focus on equity and financing haircuts is the real bright line showing capital allocation. I'll now open up for questions.

(Operator Instructions) Trevor Cranston, JMP Securities.

Thank you for the color on the capital allocation targets and other trading credit assets. A follow-up on that, can you talk about how you're targeting overall leverage for the Company as you move towards more credit positions because just looking at the last monthly update with the $700 or so million of credit assets, it looks like your overall leverage was roughly similar to where it was at 3/31, so can you just talk about how you think about that?

So when we complete our program, which will ultimately have us between $1.50 billion and $1.2 billion of our non-agency assets, Trevor. Then according to the plan that we've laid out would have our leverage between 7.5 and 7.8 but that may vary depending on how much dollar roll we have, but that would be inclusive of the implied leverage of dollar roll.

And with the credit assets, CRT spreads have been fairly volatile over the last few months. Can you talk about where you're seeing those today and any particular kind of specific sub sectors of the CRT market that you guys are focused on?

Sure, so we're on the M2s on the Fannie Mae side and on the stackers we're the 3s, a stacker deal is priced today, which we were allocated a lot at [4.65]. So when we started buying most of our purchases were between [500] and even north of [600], we allocated a capital at a really good time for that transaction. I'm looking at a report today that would show the increase in the value of my CRT position, our CRT position is actually up almost 4 points. There is a point in spread time, Trevor, much inside of this where it doesn't make as much sense for us. Today's stacker dealer's price made sense but we might sit back and watch the market for a little bit if they come in from there.

On the NPL side, as Scott said in his comments, spreads have remained very relatively stable and I would note that during the January 16 to February 15 kind of period when the markets got very volatile, the NPLs performed very well, and we drafted the price points that we can get from actual trades done and we're very impressed with how consistent the price was, despite the fact that CRTs backed out from 350 or 650. So another reason that we want to own our CRTs and wide spreads and not too many tight spreads.

Douglas Harter, Credit Suisse

The first question, can you just talk about which assets led to the sort of the underperformance you guys had in book value or how they are hedged relative to some of your peers that have reported smaller book value declines, just to get a sense as to the kind of what caused the magnitude of the decline?

Sure, generally it was almost 50% asset widening and 50% on the rate side, which is our swap position. For example, we started January 1 with DUS at 85% to 87% versus swaps, and by Feb 1, we were trading at 105%, these are huge moves, I think we owned $1.6 billion DUS at that point, you know what happened at collateral because you can follow that on a daily basis probably a little bit more than DUS, and then during that time the tenure went for [227% to 166%] and we are under [180%] again today. So a majority of portion of that move was the decline in the swap position, retrospectively you might say, despite having started the year at 0.22 duration, that perhaps you were a little heavily over swapped if you might say, but I would also notice that the swap basis during that period of time also got more negative, we had gotten back only negative 5 versus the 10-year note and now we've got the negative 18 during the period of time too, which increases the negative valuation of our position versus the swaps.

Noteworthy today is that's back to 11 and that's part of the reason along with our CRT position, the book value has recouped and probably more than four in a quarter result, as I said this morning.

And then just thinking about, how you're thinking about leverage obviously you've had a fair amount of book value volatility so far this quarter in the positive direction, but given that volatility how are you thinking about your comfort with leverage or your willingness to take down leverage in the face of that volatility?

During the less volatile markets of 2010, 2011, and 2012, our firm versus some of the others that you may study in the sector, it always had slightly higher leverage, slightly higher hedges and higher returns across the Board in terms of income producing capabilities, and the way we look at that now is in the increased volatility market, that's not as good as business model. And so although we did well in earnings in 2015, that model hurt us in 2014 and obviously in the first quarter this year as we get ready for the JAVELIN acquisition and are prepared to reduce leverage did affect us negatively there. So I believe to your point, we do like a lower leverage business model going forward. I provided Trevor Cranston with some targets a moment ago, 7.5 to 8.5 and that depends on how effective we can be in investing our money in the non-agency sector. The higher leverage definitely hurt our book value during the first quarter.

Just to be clear that 7.5 to 8, on just the agency portfolio or on a blended basis?

That's on a blended basis, but also includes the implied leverage, Doug, that we would have on dollar rolls, which by the way I have no haircut on those essentially, right, but you do have to be prepared as they go up and down in value right.

(Operator Instructions) David Walrod, Ladenburg.

Just couple refines, on the book value you increased quarter-to-date, you put out your monthly Company update and as of April 20, it looks like it was closer to $25 and then it's up $25.50 here in the basically week, week and half after that. Is there anything specific in that last week that caused that nice move or is it just general trends?

No, the CRTs have come in quite a bit and we've got CRTs a week and a half ago at [500] and they price a deal today at [465]. I'm imagining if the pricing people that we hire look at all the CRT markets tomorrow, we're going to have that pole position up another three quarters of a point. So that's been very, very helpful. I would also note the basis trade, David, that I talked about a little bit before, the 10-year swap rate versus 10 years continues to get more towards zero than not and that's at 11, it was at 13 a week ago.

On the capital allocation you said it trended up, you've taken up from 23 to 25 in the near term, when you think about a year or two years from now, do you anticipate that going higher, maybe a third or 50% of the capital.

A lot of it depends on financing and the liquidity that we need to provide for changes in those asset prices or the volatility. Right now, we're very comfortable with owning somewhere, as I said, between $1.050 billion and $1.2 billion of non-agency assets and the average haircut is around 25%, some are way lower and some are a smidge higher, with that structure we really don't want to allocate anymore. If the haircuts come down on that or we can work out a facility on these three or four banks that are looking and we'll discuss facilities here, we can get a better haircut rate. I think we'll increase that allocation. But that being said, you still have to maintain a large liquidity position for change in the markets and we talked I think with Doug a second ago on how CRTs went from [600 to 465] in about six weeks. So we have to be prepared that of some [exogenous] event makes them go back out again and we want to have the liquidity for that.

I would note that there was a period where that asset class seemed to trade in tandem with high yield bonds and we think that there is a decoupling of that, which is evidenced by the way - it's a polygraph and I'll show you. So we get great comfort on that. The home price appreciation numbers are very, very good again. And so I think as that market matures, if we look more of what's the mortgage market doing in the economy and home prices rather what high yield is doing.

And then my last question is on prepays, if you can just update according your update that you picked up some in the month of April, if you can just talk about your outlook for that?

So we modeled it for ourselves being up another 10%, okay, over the next couple months and every time we do it internally, it always comes in less than we say it's going to. So that's what we're internally saying. If you take a look at our agency assets and, as Scott commented on those quite a bit in his comments, we have a lot of convex assets, low loan balance stuff, 85 stuff, 175 right across the Board a majority of our agency assets are of convexity features. And a question that Doug had asked earlier, what kind of hurt book value? I should probably also exclamation the point that, that asset class well underperformed during the risk off period, when the 10-year note ran, it should be that people get scared about prepayments and the pay-ups on our [LBs] to go from up [12 to 24] very quickly and it did not happen. They are slowly now starting to gain some momentum there. So quick answer is we don't anticipate prepayments for our portfolio to be up as much as the market maintains.

Chris Testa, National Securities Corp.

Just with the capital allocation question again, how long do you think it's going to take you to get to the 25% or are you looking to get there quicker given the opportunities on the non-agency side, or should that be a more measured approach?

It is measured, but the opportunities have been there right now. So we've been buying as much as we can. I look at my credit risk transfers with the allocation today, we're going to be closer to $520 million of that asset class. I don't know how much higher we'll go there with the NPL sector, Chris, our underwritten deals and the best time to buy a block of those is when a deal is underwritten and they come out every four to six weeks. So we participated in the last couple, we've done some secondary trades there, so if the pricing is good we can get this done in the next quarter and a half, if it's not it will take a little bit longer.

Scott also mentioned the other sectors of the non-agency space don't provide what we see as great opportunities. However, we are holding on to what we already own.

And just to verify I think I missed it when you said it earlier on a dollar basis, what are you looking for the non-agency to potentially grow to?

Based on our capital today, for the $1.050 billion to $1.2 billon, that's our target range.

And I know obviously the professional fees were more elevated given your acquisition of JAVELIN, should that carry over a little bit into the current quarter.

I don't think the professional fees would be elevated in the future, what you would see, however, is for example, Intex, which we had to a pay for JAVELIN is now being paid for by ARMOUR, and perhaps Jim Mountain might have some other comments on them.

So we did say in the opening remarks that we'll probably have another $1 million worth of transactions cost to book in Q2 for things that just didn't get done by the end of the month since our transaction closed on the sixth and seventh, after that sort of out-of-pocket professional costs ought to be down. As Jeff just said, there are some data and analytics costs that are going to be largely new to ARMOUR as a result of getting into non-agency space, if we would have just done that on our own, those costs would have been there but the fact that we already had relationships with some of those providers through JAVELIN just makes that transition easy, it means that we can get up and running very quickly, but it's primarily going to be things like data and analytics that you will see running up ever so slightly going forward.

And with the acquisition of JAVELIN, did you take a significant amount of the credit people there to assess the credit risk within ARR now?

Well, just to be clear, Chris, ARMOUR capital management manages both those companies, so all the employees were already housed here, so everybody (multiple speakers). It actually might be an important point to not only did we not let anybody go, we've added three significant staff positions here over the last quarter, a senior partner from Deloitte was added to be Jim's number two person in the presentation of financial reporting. We hired a Wall Street veteran on the investment side who worked for asset management ARMOUR, one of the Wall Street firms and we've just added another seasoned veteran in the repo area to help us develop and be more acute in that sector. So we are actually expanding our employee base to make us better.

And just with the dividend reduction again down to $0.22 despite you essentially earning better yields on the non-agency assets, which you acquired a significant amount of, just how do I reconcile the two? On one hand you should be earning more yield even though the leverage is going to come down a bit and we would expect maybe that the earnings trajectory on a core basis would be up but the dividend [was cut] pretty significantly, so can you just try to reconcile that?

Sure, a couple of things. Number one, the dividend now at $0.22 is average or slightly average than the sector. And this was brought up by a number of analysts in the past. We had wanted to use the forward starting swaps effectively and we did in 2015 and either the swaps come online and they hurt earnings a little bit or we cancel them and roll that out. So even though in the implied leverage basis on the non-agency, we could equal a 4, 5 times amount in the agency side, so that would be like the difference. We have decided to reduce a little bit more. So there is a trade-off that you can have here and it's a financial trade-off, so our monthly Company update would have listed the average swap rate that we have. If you go ahead and look at today's average swap rate, and you were going to just add assets, it would be a lot lower in the marketplace, right. So we have some swaps that are out of the money, we have swaps that are out of money that if we reset them all to today's rate, we could be earning quite a bit more per month. What we've decided to do is keep much of the swaps that we had on even though they are out of the money, so in order to be able to regain and recoup some of the book value losses because of the hedge position. So don't look to see us take all those out and regroup because you can transfer that for earnings, which is temporary or you can let them ride, which is more permanent regain the book value as those roll down in the curve. And so for the most part, that's the decision that we've made and that is the primary reason that the earnings are going to be a little bit less than might have been anticipated although I was very clearly in the Q4 earnings call earlier this year that we were going to be a market payer and that [we'll be above] the market in the future.

Absolutely, and just last one from me, just a housekeeping item. The $30.3 million of core earnings it doesn't seem to foot the $0.72 given the diluted shares outstanding, am I missing something there?

Yeah, one of the things that some people have problems when they do that calculation is that they forget to take out the dividends paid to preferred shares, so just like when we do the EPS on a GAAP basis, that's about $3.9 million of preferred dividends, you need to take core and reduce it by that same amount before you divide by the shares outstanding. So if you look at the income statement in the 10-Q, put your core number right beside [$279 million and $475 million] net loss, take out the same $3.9 million and get core available to common and then do your division.

Brock Vandervliet, Nomura Securities.

You've talked in the past about the DBO1 metric, which I wanted to just touch on from page four of the supplement. So it looks like it's moved from negative $150,000 to positive $622,000, the negative $150,000 that would have implied your position for higher rates. So maybe that's explaining some of the weakness in the first quarter. But just how you're thinking about that and what the take away should be in terms of your positioning?

So we started the year on January 1, at 0.22, net balance sheet duration is slightly positive balance sheet duration, and you see in the graph of the 10-year note and what happened over the next two weeks and then we took a breather two weeks more. So we've maintained our position throughout that time. So you went from 0.22 to negative 0.6 in about eight to 10 trading days very quickly. What we didn't want to do when you have extremely volatile moves like that is go ahead and unwind things at losses, which we did not do and so able to regroup our ability to get those swap positions back on the books for permanent book value. The duration on the balance sheet today Brock is 0.58 on a morning report I'm looking at right now and the DBO1 on a [$652,000]. I would not be surprised if you saw the duration next time we speak at something closer to 0.7 or 0.8, slightly more positive and the DB1 for rates not spread move would be somewhat similar.

And that implies you're leaning pretty hard?

Well, one of things we do - so you are a research analyst and you listen to all the calls from all the other REITs, all the other firms in our space. So some firms may say that we have a zero duration but the net duration gaps are actually higher, so I would tell you that our duration where it is now is considerably lower more than most firms actually report, we internally have the empirical data to support that.

I guess it would make me feel better if you had more of a neutral posture but I understand what you are saying?

The neutral posture is 0.2, which would have been at the end of the year turned into substantial book value losses in the course of three weeks, and so now at a point in time where you have the opportunity, the earnings opportunities are good and the swap position is - we can already see has regained some of its balance opportunity to absorb some of that. Once again, the 0.58, what I said this morning is considerably lower than anybody else in our space when you've taken - we don't include repo in our net duration. We just don't include it, if we did that number will be way lower, probably closer to zero.

We think that that doesn't clearly reflect the duration of the included repo, that's a 45-day liability, right.

And just as a follow-up to one of the prior questions with respect to your cost of funds, was it some aspect of the forward starting swaps that drove the increase in cost of funds or something else going on this quarter?

Fed rates raise in December. So what happens is, for a period of time we were out 60 to 90 days average on all our repos and we actually were there before the fed rates raise. As those came off and we had to be at new rates, that's where we saw the increase in the cost of funds, but interestingly enough since it looks like a June roll, June federal funds rate hike at least for - until this morning was off the table to [lock our swap] this morning. We actually shortened up a little bit and have benefited from that. And I think you might see it has extended a little bit. So imagine this at the end of the year, we had a lot of 60 day stuff as those rolled off and we had to go to the new higher fed funds rate, that's where the cost of funds. Most of the forward starting swaps that will be either rolled or taken off.

(Operator Instructions) Mr. Mountain, there are no further questions at this time.

Well, thank you. Christy, and thank you all for joining us and we look forward to talking to you again in another quarter.

Ladies and gentlemen, that does conclude the conference call for today. We thank your participation and ask that you please disconnect your lines.