American Capital Agency Corp. (NASDAQ:AGNC)
Morgan Stanley Financials Conference Call
June 12, 2013 3:35 pm ET
Gary Kain – President and Chief Investment Officer
Cheryl Pate – Morgan Stanley & Co. LLC
Cheryl Pate – Morgan Stanley & Co. LLC
Good afternoon. For our final presentation today, I’m pleased to welcome American Capital Agency, an agency mortgage REIT that has had one of the most sophisticated strategies in terms of asset selection and perseveration of book value. Here to present for American Capital Agency today is Gary Kain, President and Chief Investment Officer.
Prior to joining American Capital, Mr. Kain served in several investment management roles at Freddie Mac from 2001 through 2009, including Senior Vice President of Investments and Capital Market, Senior Vice President of Mortgage Investments and Structuring of Freddie Mac during which he was responsible for managing all of Freddie Mac’s mortgage investment activities for the company’s $700 billion retained portfolio.
And with that, I’d like to pass it over to Gary. We’re going to skip the polling question given that we have some great color in the slide deck today. Thank you.
Thanks. Thanks, Cheryl. And I want to thank Morgan Stanley for inviting us to participate in this conference and to all of you for your interest in AGNC.
Now, given the volatility that we’ve witnessed in the global fixed income markets and in the REIT equity space, I really did feel it was important to focus this presentation on current market conditions and to give you some insight into how AGNC has responded today to the current challenges. And with that said, let me turn to the presentation. And I’m actually going to skip page three and jump right into page four where we can review what has transpired so far during Q2.
Now, first of all, as many of you are probably aware by now, the second quarter and especially the month of May has not been kind to fixed income in general, and to the Agency MBS space particular. And as you can see on the top left-hand side of this page of the slide, five and ten-year treasury and swap rates have increased around 35 basis points quarter to date.
Now agency mortgage shown on the right side of the page have performed considerably worse on a relative base. 30-year 3% TBAs were down over three points through June 7, while 3.5% fell almost 2.5 points. Again, for comparison purposes, five-year treasury bonds dropped only around 1.5 points during the same period. Higher coupons, which are typically much less sensitive to interest rates, were still significantly lower in price for the quarter.
15-year mortgage is also underperformed with 15-year 2.5% TBAs down about 2 points, even higher coupon 15-year mortgages, which really are typically one of the least sensitive to interest rates kind of in the mortgage market were down over 1 point in price pretty much across the board in all relevant coupons.
So with that, let’s turn to page five and we can look at the performance of specified mortgages so far this quarter.
Now I want to caution investors that pay-ups have been very volatile over the past several weeks, especially, in the 4% coupon. And so – and we can also expect that this volatility is likely to continue. As such these numbers should be only viewed as ballpark indications and not exact marks for the product shown.
That said I would like to give investors some visibility into where pay-ups might have been quoted a few days ago. So as you can see with a 3.5 coupon lower loan balance pools, the pay-up dropped 41 basis points during the second quarter through June 7 versus 73 basis points in Q1.
But, as importantly, keep in mind that the smaller decline in pay-ups in 3.5% coupon specified pools so far this quarter also occurred against the backdrop of a much larger move in interest rates than we saw in Q1. So hedges could provide some offset to the move this time versus the move last time.
So for that reason, pay-ups accounted for, in a sense, and also accounted for less or much lower percentage of the aggregate kind of price declines of a specified pool versus kind of just the move in TBAs that we saw on the other page.
Now while the pay-up declines in the 4% coupons so far this quarter were generally larger in absolute terms than the declines that we saw in Q1. The decline relative to the move in rates and to mortgage prices was still considerably smaller. Now if we look ahead given that the majority of the pay-ups on specified 3.5s are now going to at or below a 0.5 point, the bulk of future pay-up risk is now concentrated in higher coupon MBS such as 4s and above.
Now to this point, the aggregate pay-up on our entire pass through portfolio inclusive of our TBA positions was only about 63 basis points or five base of a point as of May 31. As such, we expect the pay-up risk going forward on our pass through portfolio to be considerably lower than where it’s been over the past year or so. And I think that’s kind of a key point to keep in mind.
Now if we turn to slide six, you can clearly see the under performance of mortgages over the past several quarters. The graph depicts current coupon mortgage spread, after adjusting for the option cost inherent in the securities. And despite the massive purchases from the fed, which total over $600 billions in September 12, when they announced QE3, mortgages have widened significantly and are now back to levels much closer to historical norms.
And as a function of the under performance of generic MBS and lower specified pay-ups, we have seen continued pressure on our economic returns and on our book value again this quarter. As of June 7, 2013, our estimate for the decline in book value for Q2 was similar to what we experienced during the first quarter. This estimate is based on asset and liability valuation as of June 7 and does incorporate an estimate of our yet to be declared Q2 dividend.
Now I want to remind listeners that this high level estimate is not derived in the same manner as those that we disclosed in our quarterly financial statements. For quarter end results, we have a very thorough revenue process audited by external auditors, which is not possible to implement on a daily basis obviously.
Additionally, I want to remind you that the current environment is characterized by substantial volatility as we’ve seen in both interest rates and in mortgage valuations. And so the actual performance for the quarter could easily deviate significantly from the quarter to date estimate that I just discussed.
So with that said, let’s turn to the next page so we can discuss how we are responding to this market environment. And first and foremost, and this is important, as a levered investor, AGNC must be respectful of both market volatility and potential signals reflected in market valuations. Moreover, the driver of recent volatility has been widely attributed to uncertainty surrounding the current and future actions of central banks around the globe such as the Fed and the DOJ.
And in addition to and that’s in addition to questions regarding the strength of the U.S. and global economies. And given these kind of factors, it’s more difficult to discern whether the recent changes in rates and in MBS valuations are merely reflect short-term fluctuations that we see periodically or maybe therefore saddling larger moves in the future.
And so with this uncertainty in mind, AGNC took proactive and almost daily actions to rebalance both our assets and our hedges throughout the quarter. As a result, our leverage during the quarter through June 7 remained within reasonable bands and our current total leverage and when I say total leverage, that’s leverage inclusive of both on balance sheet assets and any TBA forward positions. As of June 7, it was not significantly different from our Q1 average of 8.2 times. Moreover, despite the decline in book value during the quarter, our leverage was not significantly higher than this at any point during the quarter.
So what I’m trying to kind of imply there is our current leverage was not achieved by one-off sales executed late in May or in early June just to bring leverage back down, but rather the consistent actions throughout May and June. And I also want to be clear that we do believe that we are well positioned to run at higher than 8 to 8.5 times leverage under the right circumstances. And at some point in the future, if we feel the risk return trade-offs are compelling, we might choose to do that.
But in addition to just managing aggregate leverage during the quarter, we also took some incremental steps to adjust the composition of our portfolio for the current environment. Two such examples of this include reducing some exposure to low coupon 30-year fixed MBS and also by moderating our exposure to higher pay-up 4% coupon specified mortgages.
Now on the hedging side, we also increased the duration of our hedges in order to mitigate extension risk. Again, like on the asset side, these actions were kind of taken on a relatively continuous basis throughout the quarter and they weren’t confined to at the end of May after things had moved a lot.
Now I do want to be clear that realistically there are pros and there are cons to the approach that we took on both the asset and on the hedging fronts. On the positive side, we strongly believe that these actions provide us with additional flexibility to proactively respond to future market volatility. However, they will tend to reduce our earnings potential in the current environment to some extent over where they would otherwise have been, because (inaudible) hedges do cost money and our asset portfolio is a little smaller.
Now there may also be somewhat less upside to book value if interest rates fall quickly or if MBS spreads just tighten materially back to where they were. And that’s a reality of the business.
So if we turn to slide eight, I want to kind of give you a little more information on our view of kind of what our interest rate risk is like and how it’s evolved during the rather sizable moves that we’ve seen in rates and mortgage prices this quarter. And as you can see, our current duration gap has extended to 0.7 years. But our duration gap, if interest rates rise another 100 basis point, will then only increase 0.5 year to 1.2 years.
So what’s really important about that is if you look at the April 30 duration gap on the top kind of middle of the page at 3.6 years went from 3.6 to 5.2 over the last five or so weeks. Even if we got another 100 basis point move, our models kind of predict that we’d only have extension of 1.1 years or less than the extension that we’ve seen before. And if that’s true of the market as a whole, the market should be better positioned to handle the next move up if it occurs than it was before.
Another thing to keep in mind on this slide is up 200 basis point scenario, which is somewhat analogous to what we saw in 1994, our duration would still be only 1.3 years, in other words, it’s very little extension from 1.2 to – from 100 basis points to 200 basis points in rates. And so we feel very comfortable with our – how we’re positioned against large moves.
this is a pretty meaningful reflection of how low management believes are current level of the interest rate risk is, because think about that 1.3 up to 100, and that’s below where a lot of levered financial institutions are willing to operate on a day-to-day basis, and that’s the position we would be in, if we had an immediate up 200 basis point move. And this number is considerably lower than where we have operated in the past.
Now one other thing to keep in mind is that if a move of either the 100 basis point or 200 basis point move occurred, even over a two-week period, we would likely take rebalancing actions and so our ending duration gap might be noticeably lower than what’s on here, which assumes an immediate shift with no rebalancing.
I also want to reiterate one thing that we’ve highlighted consistently in the past, which is this duration gap analysis is based on estimates calculated by models that are dependent on inputs and assumptions that are provided by a third-party provider as well as by the AGNC management team and as such, actual results could differ materially from these estimates as is the case with any model derived numbers.
Now if we turn to slide nine, I think this is an important slide and we believe that AGNC has significant equity available to absorb potential reductions to book value and we think that this concept may not be well understood by market participants with respect to the REIT industry as a whole. So the table on the left shows an example of approximate uses of equity and is based on our March 31 positions, which are, of course, not equivalent to today’s position. In this illustrative analysis, we assumed aggregate total leverage of 8.1 times and this was comprised of 5.7 times on balance sheet leverage and 2.4 times leverage via TBAs.
The blue bar shows that repo haircuts, which are assumed to be 5% in this analysis, account for 29% of our equity. the payment delay related to prepayments is depicted in orange and is estimated around 7%, which assumes a 12% CPR. Now the upfront margin requirements on TBAs totals 8% based on a 3% margin assumption. And then other uses of equity shown in grey totaled 8% and these include things like clearing deposits and upfront margins on swaps. Now the large purple bar, which is the residual in this example shows that 48% of the equity is unencumbered and available to meet margin calls or other unexpected cash needs.
In addition, repo haircuts are generally fixed during the term. So investors sometimes overestimate the impact that changes in haircuts could have on our liquidity in the near-term, especially given that the average term – remaining term of our repos is around 120 days. Should also point out that we have not seen, over the course of the quarter, any upward pressure to date on repo haircuts, and on another note, we’ve actually seen our repo borrowing rates decline over the course of the year.
Now the table on the right depicts the hypothetical decline in book value or equity in some scenarios where Agency MBS significantly underperform their hedges by either one, two or three percentage points. But we then show how these would change under different leverage scenarios. For example, an entity operating at 10 times leverage that experience a two point under performance of mortgages could expect a decline in book value of 22%.
At eight times leverage, a massive three point under performance of MBS versus hedges, again, this is not just the price drops by three points, it’s three point versus all of their interest rate hedges. That would produce an estimated 27% decline in equity. interestingly, that would only utilize just over half of the 48% of the available equity in our example to the left. therefore, the sample portfolio would still have substantial cash resources after meeting all expected margin calls even with a three point of kind of go forward widening of MBS versus its hedges. Again, this analysis assumes no change in haircuts, TBA requirements or other unexpected developments.
Now, before I conclude, I think it’s important to discuss our current views on the mortgage market, because this should be a key component of the value proposition for a more total return oriented investors in today’s environment.
In summary, we are bullish on our outlook for Agency MBS over the next three to 12 months. Now that said, investors should clearly expect significant volatility over at least the near term. So why are we bullish? MBS spreads have widened to levels relatively consistent with historical averages despite ongoing Fed purchases and the almost 25% of the Agency MBS market that is now held by the Fed. Furthermore, even if the Fed begins gradually tapering purchases in September and stops QE3 related purchases by the end of Q1, 2014 as some observers are certainly predicting, they’re still likely to purchase between $350 billion and $450 billion of additional MBS over that timeframe.
These purchase volumes are still very large against the backdrop of declining new origination volumes given the increase that we’ve seen in mortgage rates and the predicted declines in all kind of refinancing activity not including HARP, where you’ll still see decent refinance volumes.
So to this point, gross issuance of Agency MBS averaged just over $150 billion in Q1 and we project that to average around $100 billion, let’s say, by the end of Q3. So that’s a decline of $50 billion in gross issuance and that would likely provide a pretty significant offset to any tapering activity that from a reduction in purchases by the Fed.
Additionally, since QE3 was announced in September, money managers have likely reduced their allocations to Agency MBS. Foreign investors have been rumored to have net sold mortgages and demand from banks has been surprisingly light prior to, let’s say, the last month or so. However, given today’s prices, higher yields, wider spreads, domestic bank interest in MBS appears to be picking up and that clearly could provide some support to mortgages. Also levered investors such as hedge funds, REITs, dealers and servicers have also likely executed a meaningful amount of their typical convexity hedging given the recent volatility. And so positions of these firms are likely more balanced than they may have been earlier in the year.
So in conclusion, the combination of the meaningful slowdown in origination coupled with the ongoing Fed purchases and more balanced investor flows should facilitate a strengthening of MBS valuations once we see some of the current volatility subside. And furthermore, if the Fed does continue QE3 into 2014 without tapering up, which is still a very meaningful probability and I would actually add to that that the Fed may start to taper. But then feel that they have to go back to kind of full purchases or even increase purchases and under either of those scenarios is clearly a lot of upside to MBS prices and spreads.
So with that, let me stop while we have time and make sure we open up the floor to questions.
Cheryl Pate – Morgan Stanley & Co. LLC
May be I’ll start by kicking it off. Can you talk about obviously a lot of volatility in the market and pricing, but how you’re thinking about the relative attractiveness of some of the investment opportunities including the former TBA market?
Sure. In the current environment, we have talked on our conference calls, the last couple of conference calls, a fair amount about TBAs and favorable financing available in the dollar roll market and we still feel that TBAs make up fair amount of sense and the financing advantages are still there. We also feel that even though the prices of specifieds are lower again in Q2 relative to hedges and given the move in rates, we actually feel they were cheaper really at the end of April when rates were lower and the prepayment protection was more important. So, I mean, on a relative basis, we’re generally would lean more toward TBAs versus specified. And the coupon distribution really is dependent on kind of the aggregate portfolio mix. But the reality is the wider spreads clearly have led to more attractive kind of opportunities in terms of holding assets going forward.
Hi. How much of the book value decline in the first quarter would you say was related to the pay-ups versus just the under performance of the generic mortgages? And just going thinking a little further, if you look at the second quarter and there is less pay-up risk yet and hedging has actually been extended. So why would be book value in what you mentioned, why would it be rent same as it was in the first quarter of the book value decline?
Both very good questions. So, let me start with your first one, with respect to Q1, if you went back and said okay, at a high level how much of Q1’s under performance was related to specifieds versus TBA or how would you allocate it and it’s somewhere in the 50-50 kind of area. Actually in Q2, we would attribute a much lower amount of performance of specifieds. Specified pay-ups are down, but as we reviewed on the prior – on the slide, especially in the 3.5s, you could obviously see that the pay-up dropped about – dropped almost half the amount for an interest rate move that was four times as big. So a bigger interest rate move you would have expected a bigger move in the pay-up, you actually had the bigger move in the pay-up last quarter. And even though the move in 4s was larger than the absolute move in the pay-ups on 4s was larger in Q2 and Q1. Again, the interest rate move was three to four times larger. And so actually even 4s, the pay-ups performed better during the second quarter.
So what I would say is we’re early in the quarter, there is a lot of volatility around this and I’m not going to try to break it down. But it’s noticeably lower in our minds and what we would say is, it’s generally the performance of agency mortgages kind of as a whole and for that matter our portfolio is much more weighted toward just under performance of mortgage securities as a whole and much less focused on pay-ups or any individual specific composition of AGNC’s portfolio.
Thank you. I had a few questions. First, are you getting any pressure or suggestions from the banks or the prime brokers to reduce your leverage?
Absolutely nothing. No, we’ve had – we chose to manage our leverage and keep it in the same vicinity as where it was totally on our own from kind of immediately throughout the moves. And we’ve had no pressure whatsoever. We absolutely believe we could operate at higher leverage if we wanted to. And we have so much ample – we have very significant additional repo capacity and we have, as we talked about in the example, we would have significant excess equity to run higher leverage if we so choose. I mean, you just have to be cognizant that as a levered investor where spread volatility is relatively high that you want to do that when you really are sure you want to be there.
Last time I believe you were selling below NAV, you guys were buying back some stock. Is it something that might be in the cards?
We have to appreciate the question. We have a program in place for AGNC to where we can repurchase stock. And if the stock is at a material discount to book, then we are – we stand ready to buy it back. One thing I do want to be clear is either it needs to be a material discount to book and then the other thing is, you just have to understand that with any stock repurchase program, there are lots of constraints such as window periods and so forth where you can execute. So it’s not a seamless process as some people might think.
Cheryl Pate – Morgan Stanley & Co. LLC
Hey, one last one from me. You spoke to changing to put it on more hedges during the quarter, how are you thinking about sort of extending the term of the hedges and then also on the repo side as well as that something that you’ve been looking to extend out of in the current environment?
Sure, and I actually try to go back to one slide on the hedging, so slide eight, which if you look at this slide, I mean, when we say we increased our hedges, some of that is the duration of the hedges rather than just putting on another one. So you might pair off a shorter swap and put on longer swap. But if you look on this chart, what we show is that as of 4/30, this is from the earnings call, the duration of our hedges was 3.8 years and that’s on the left kind of up on the top. And at the end of – as of June 7, the duration of our hedges in aggregate was minus 4.5 years. So we had extended the duration of 0.7 years.
Now some of that comes naturally from the swaption. So not all of that is an extension of actual hedges, but a good chunk of that is. So you can see incremental hedging again. And then what I would say going forward is just keep in mind how little extension there is in the portfolio net from here, if we were to continue going up, which gives us a lot of comfort around managing the portfolio.
Now with respect to the second half to your question, about the repo, we’ve extended the term of our repo borrowings really over the last year in a pretty meaningful way. So the average term of our borrowing was not that different from a month or a couple of month and a half, a year or two ago and now it’s the remaining terms that are about 120 days and we have repo going out five years.
So we definitely are managing the term of the repo. And we are very comfortable doing more. But what I would say on that is there is a massive difference between one and two months and having a weighted average of four months. And even if whether you’re thinking about the circumstances, the haircut example that we talked about is really good one. Let’s just say that we woke up tomorrow, and again this is not going to happen, but if we woke up tomorrow and every haircut was going to go up 4%, if we take, let’s be realistic, every haircut was going about 1% or 2%, it would take four months at the average material – just using the average, it would take four months before half of that hit us, okay. From the form mobile liquidity, that’s really critical. So we feel really good about where are our borrowing is. We will continue to look to extend it at the right cost, but we again feel really good about where it is relative to where it was a year ago.
Cheryl Pate – Morgan Stanley & Co. LLC
Any last questions in the room?
You’d mentioned in the presentation that with the hedging that you put on in the past few months, it might result in lower earnings and lower distributions going forward. Could you give us some more guidance if things stabilize at the current level, what that might look like?
What I would say is I’ll talk about that statement a little more of a high level, which is, we implied there is clearly some decline to book value as we talked about. We maintained our leverage with so that it’s close to where it was 8.2 the average of Q1. So that implies our portfolio is a little smaller. So clearly a smaller portfolio is going to kick off less income. That’s really one part of that statement. The other part of the statement is as we just talked about on the hedging slide, we either added hedges or we added to the duration of the hedges. Hedges cost some money. You’ll notice it’s not a huge difference there. But there is some incremental cost; I’m not going to attempt to quantify those. They are on the margin, okay. They are not like – they are not massive. But there is incremental cost.
Now there are other factors that people want to think about, I did mention we’re on the slide with repo that repo costs and rates have come down. TBA dollar roll financing remains very favorable and we’re clearly in a lower prepayment environment, where prepayment estimates are likely to fall. And so there are things on the other side that are actually very favorable to returns as well and that should be bought up in the process.
Great, thanks very much.
Cheryl Pate – Morgan Stanley & Co. LLC
Okay. Thank you. Please join me in thanking Gary for presenting today.
Copyright policy: All transcripts on this site are the copyright of Seeking Alpha. However, we view them as an important resource for bloggers and journalists, and are excited to contribute to the democratization of financial information on the Internet. (Until now investors have had to pay thousands of dollars in subscription fees for transcripts.) So our reproduction policy is as follows: You may quote up to 400 words of any transcript on the condition that you attribute the transcript to Seeking Alpha and either link to the original transcript or to www.SeekingAlpha.com. All other use is prohibited.
THE INFORMATION CONTAINED HERE IS A TEXTUAL REPRESENTATION OF THE APPLICABLE COMPANY'S CONFERENCE CALL, CONFERENCE PRESENTATION OR OTHER AUDIO PRESENTATION, AND WHILE EFFORTS ARE MADE TO PROVIDE AN ACCURATE TRANSCRIPTION, THERE MAY BE MATERIAL ERRORS, OMISSIONS, OR INACCURACIES IN THE REPORTING OF THE SUBSTANCE OF THE AUDIO PRESENTATIONS. IN NO WAY DOES SEEKING ALPHA ASSUME ANY RESPONSIBILITY FOR ANY INVESTMENT OR OTHER DECISIONS MADE BASED UPON THE INFORMATION PROVIDED ON THIS WEB SITE OR IN ANY TRANSCRIPT. USERS ARE ADVISED TO REVIEW THE APPLICABLE COMPANY'S AUDIO PRESENTATION ITSELF AND THE APPLICABLE COMPANY'S SEC FILINGS BEFORE MAKING ANY INVESTMENT OR OTHER DECISIONS.
If you have any additional questions about our online transcripts, please contact us at: email@example.com. Thank you!