AGNC Investment Corp. (AGNC) Earnings Call Transcript & Summary

Good morning, everyone. Thank you for joining us for this fireside chat with AGNC's CEO, Peter Federico. We have a number of prepared questions we'll be going through. If anyone in the audience would like to ask a question, click on the Ask a Question button in the upper right-hand corner of your screen and follow the prompts to submit your question, and we'll do our best to address it in the time we have today. We've also prepared a number of audience polling questions that we would encourage you to answer during the presentation, and we'll be publishing the results in our report, summarizing the takeaways from the conference. With that out of the way, let's get to the discussion.

Peter, thank you very much for joining us. I think you put out a disclosure just last night on book value. Could you just update on how that came in, maybe some qualitative comments on kind of what drove book value movement during the month?

Sure. I'd be happy to. And first off, let me say thank you to yourself and Barclays for setting up this day and having us participate and for all the good work that you and your company do for our industry. It's really important, and we really appreciate it. As you say, we did put out our book value last night. And really, the quarter is shaping up sort of as we expected. If you go back to our second quarter earnings call, we really outlined an expectation that we expected mortgages to drift a little bit wider in the third quarter as we approached the tapering. And in fact, that really was the case for the first 2 months of the quarter. In July, we put out our book value. In July, it was down about $0.11 or about 0.7%, not very much of a movement, but showing some drift wider in mortgages. Again, in August, our book value last night was $16.25, down only $0.03 from the previous quarter or 0.2%. So for the first 2 months of the quarter, our book value was down roughly about 1%, a little less than 1% as mortgages have drifted somewhat wider in the quarter. All that said, in September, the tone in the mortgage market has been a little bit better. Mortgages have performed fairly well. Part of it was some of the economic data and maybe some expectations that the Fed may end up delaying their tapering a little bit. We'll talk about that later, I'm sure. But generally speaking, the quarter is shaping up as we expected. Mortgages, particularly lower coupons a little bit wider. But our book value, not very much changed after first 2 months of the quarter.

Okay. That's helpful. Given where things stand today, could you just update us on your expectation for returns in the agency market and what, if any, impact that's had on kind of your asset allocation here?

Sure. I'll talk more broadly first, and I'm sure we'll get into more details about specific return opportunities with specific coupons. But just broadly speaking, our return expectation for Agency MBS over sort of multiple cycles is that Agency MBS should return something in the low-double-digit ROE range, particularly if you think about it against the backdrop of the way we operate our business. From a leverage perspective, for example, operating traditionally in the 7x to 9x leverage range, keeping our interest rate risk low. For us, that means less than a year duration gap typically and often less than half a year. Generally speaking, mortgage-ish Agency MBS should give us the opportunity to generate on a gross basis low-double-digit ROEs. There are times certainly when mortgages are a lot tighter, and that was the case for a good amount of last year. And there are certainly opportunities when mortgages are a lot cheaper. But generally speaking, over the long run, that's what we expect and that's supportive of our business model. From a portfolio composition perspective, in the current environment, and we can talk about why this is the case. But generally speaking, we've been operating with a very well-balanced asset portfolio, meaning a high concentration of lower coupon TBAs as well as a significant concentration of higher coupon specified pools. And we think that, that's giving us the sort of the best balance in the current environment. And we've seen why that's important over the last 3 quarters because of the performance that sort of uneven performance between those 2 segments of the market in the first quarter, for example, higher coupons outperformed a lot. They've underperformed in the second quarter, actually outperforming a little bit here in the third quarter. So we haven't changed really from an asset composition perspective. We think it's the right fit for the current environment, and it's actually paying off for us.

Okay. Just a follow-up question on that, Peter. If you think back over the last decade and on your -- more specifically your comment around kind of expecting double-digit returns on equity. It feels like over half of the last decade the Fed has been deliberately distorting the market. For QE to stimulate the economy, does that feel like the right return in -- I don't know if we ever -- what normal is, but in the environment the Fed is on the sidelines, right? And it's more of your traditional players who are investing in the market. Would you expect higher returns in that type of scenario if we ever get there?

You raised a really great point, and it's one that we've been thinking about. And I think that the specific reference that you made for over half the time is actually really light. When you go back and you look at what happened from QE1 until now, there were times when the Fed stopped the actual QE. But in between those times, the actual reinvestment activity of the Fed was really significant. So if you go back all the way to 2008 until now, you've had the Fed basically in some way or another, really having a strong influence on the mortgage market. And that's both good and bad, if you think about it, right? One is that it stabilized prices and lowered our returns. But the flip side is it also, I think, speaks to the importance of this asset class and why this asset class lends itself so well to a levered investment strategy, right? At the end of the day, the Fed is creating a lot of liquidity, a lot of transparency in this market, great funding opportunities. And so it's really supportive of the business model. But I think you're right in that if we think about where the mortgage market is today and going forward, all other things equal, I do think that returns will be a little lower than they would have otherwise been because I don't see a scenario where in some way or another, the Fed isn't basically participating or willing to participate. So it could lead to lower return opportunities somewhat going forward, but it also reduces the volatility in the mortgage market, which is also really good. But I still think that, again, with our leverage strategy and our risk management strategy, we can still generate really attractive returns for our shareholders. And all other things equal, I think it's a healthier market because of that with the Fed and its participation.

Okay. Great. Turning to more granular discussion of your investment portfolio, what coupons do you currently keep active for your Agency MBS? And can you discuss how returns on pools compared to TBAs and kind of what the near and intermediate-term outlook is for roles set.

Yes, it's really an interesting time. And as I mentioned, with our book value being down just slightly over the first 2 months, what that would imply is that mortgages have a little bit wider, and ultimately, that means returns are a little bit better. But there really isn't much change, generally speaking, from the return opportunities that we talked about at the end of the second quarter. And more specifically, if you look at lower coupon TBAs, 2 and 2.5 coupons, in particular 2.5s more specifically. The returns in those coupons is still in the low-double-digit range, call it, 10%, 11%, giving us a NIM, for example, of around 100 basis points with our leverage would translate to sort of a base gross ROE in that very low-double-digit range. But that doesn't include the potential return opportunities from specialness. And in those coupons and in particular the 2.5% coupon, the implied funding for those coupons continues to be really attractive, meaning that if you look at the funding for a TBA versus an on-balance sheet portfolio or pool, we're picking up somewhere between 25 and 50 basis points, say, roundly for specialness. And if you multiply that by the leverage, you could easily pick up 2%, 3%, 4% additional return for some period of time for those coupons. So the marginal return opportunity is best in the TBA market in those 2 coupons. On the specified pool side, what I would describe it is that those pools are sort of fairly well priced right now for the prepayment environment that we're in. And we saw a big backup in those prices, and spreads widened significantly in the second quarter to reflect the faster prepayment environment that we're in. The returns on those pools is really closer to high-single-digit range. And that's where it sort of has leveled out over the last few months. But what I would say, and we'll, I'm sure, talk about the prepayment environment later. As we sort of make our way through this prepayment environment, we should start to see slowing in the prepayment speeds of those higher coupon specified pools. We're starting to see it a little bit now. And ultimately, what that will mean is that as you reprice those pools to a more benign prepayment environment, what you're going to see is the returns on those things expand. And I expect those pools also over the next, say, 6 months or so to start to reflect a better prepayment environment and ultimately a higher yield. So I expect the returns on an ROE basis to move up into that low-double-digit range as well.

Okay. Great. That's very helpful context. Just switching briefly to credit. Are you considering any initiatives that would broaden your investment set in mortgage credit?

Well, nothing specific with respect to sort of the generic mortgage market. We still obviously have a few percent of our capital exposed to the non-agency market. And the challenge in that market right now is that the returns there just are not nearly as attractive as they are on the agency side of our business. So our marginal capital dollars would go toward the returns that I just talked about, the high-single-digit, low-double-digit return opportunities. Credit, as everyone knows, is just really, really tight. And so the returns are not nearly as attractive. So we continue to sort of be on the sideline with respect to the -- our marginal capital dollars. We certainly have the capacity and the flexibility to expand over time. But in the current market, I don't expect much change.

Okay. Makes sense. Funding costs remain close to 0 here. How do you expect funding costs as monetary policy tightens and the Fed begins to taper?

Well, it's a great question. And you're absolutely right. Let's start with sort of where we are today in the funding market. You really can't ask for a better funding environment for Agency MBS. If you think about the return -- or the borrowing cost from overnight out to 1 year, it's somewhere between 5 and 15 basis points. So -- and you've seen our cost of funds and our repo costs really ratchet down into the now low-teens. So you really can't expect funding to improve any from here. But that said, I don't expect it to deteriorate either. And I think some perspective here is important. If you go back and you look at the repo market for Agency MBS and you sort of go back to 2018 and '19, in particular, where we had significant volatility in the repo market for both treasury and Agency MBS, 2019, in particular, in the third quarter. That really was a pivotal point in time for the Fed. And what the Fed has done since then is incredibly supportive for our business model because they have sequentially taken a series of steps that have greatly enhanced the liquidity of the repo market for our asset class Agency MBS as well as U.S. Treasuries. And you see the impact of that now. And just recently, the Fed announced yet another action to sort of make permanent what they have been doing all along, which is adding liquidity to the market or certainly making it available when we need it. We now have a repo facility that the Fed has put in place that it essentially established just a lower bound in its reverse repo facility. And now as of a couple of months ago, they just now have the permanent standing repo facility, which will establish an upper bound. So to your point, what I think the Fed has done is put themselves in a position when they do start transition into a more normal monetary policy and higher rates. They have put tools in place that will keep the agency repo market well anchored to its Fed funds target rate. And that was really the lesson that I think that they learned in 2019 and said, look, we need to make sure that we keep all of the money market rates essentially tethered more closely to the one rate that they do control. And they've done so very effectively. And so I think what it means for us over the long run is that we should enjoy a much more stable funding market for our asset class, which is really important.

Okay. Excellent. Turning to prepayments, how they performed quarter-to-date relative to your expectation?

Well, the theme, I think, in prepayments is that they continue to be stubbornly fast. In this last month, in August, the speeds that we just got, due to day count and due to the driving mortgage rate, the market expected prepayments to increase rather materially about 15%. And in fact, they did increase, but not nearly that much, 30-year prepayment speeds increased about 11%. They went from around 22.2 CPR to 20 -- I think it was 24.6 CPR on average. So there was an increase in speeds this last month. But importantly, it was lower than the market expected. And that's sort of what we see now for multiple months is that we've seen a little bit of improvement, but we haven't seen the big slowdown. And that's in part because the Fed has done such a great job of keeping the mortgage rate relatively stable. And there are certainly lots of technology innovations and enhancements in the process to refinance the mortgage that we all know about working from home, all of the automation that has happened, all of the online processing. Those are real technology enhancements that have made mortgages more refinanceable, and we're seeing the impact of that. We're also starting to see speeds start to slow, but just at a really moderate pace. This last report was actually really interesting because while in aggregate the speeds were a little bit lower than the market expected, which was good. There was also very big variation in the speeds between coupons. And in fact, the lower coupons, the 2 and 2.5s that were recently created mortgages, those saw huge increases in speeds. 45%, 25% increases in 2 and 2.5s, generally speaking. So they drove the increase or the acceleration in prepayments, which means that originators are going back to recently created mortgages and being able to refinance those, churn those mortgages really fast. It's important that that's happening, by the way, in the coupons that the Fed is actually addressing because the Fed is continuing to absorb that new issuance in the supply. If you look at the higher coupons in this last report, even though the aggregate increased 11%, lower -- higher coupons like 4 and 4.5s only increased around 1% month-over-month. So it's starting to show that some of those higher coupons, and it goes back to the comment I made about the ultimate return opportunities in those coupons. We are starting to see some good news develop in the higher coupons. Burnout is starting to show itself. It's just happening a lot slower and a lot more methodically than I think the market expected. But generally speaking, good news, but slow coming.

Okay. Great. Can you update us on your thoughts on when you expect the Fed to announce and begin taper and what the pace of the taper would be? And do you still think that's largely priced into the market at this point?

Yes, it seems so. And I give the Fed tremendous credit for being transparent. And I think since the second quarter, we've learned more about what the Fed thinks and what they -- the way that they view the market. They essentially came out early in the quarter and said we've achieved what we wanted to achieve with respect to inflation. We're making good progress on the employment front, but we haven't yet achieved substantial further progress. We want to see more job growth, which we expected in July. We obviously got a really strong nonfarm payroll number north of $1 million that was setting up perfect for the Fed. And if we had gotten a series of those reports, I think the Fed would have been in a position at the September meeting to essentially make an announcement about tapering and then actually begin tapering probably in November. But then, of course, we got the August payroll report, which was much weaker than expected, I think, largely driven by the risks associated with the Delta variant. And it essentially has given the Fed, I think, a little bit of pause. But that said, I think the Fed is still in a good position to essentially announce a tapering at November meeting, certainly by December, and I do expect the employment outlook to continue to improve, as I think the Fed does. And that will allow them to start to taper either at December or in January. So we've gotten a lot of information, and I think it's down to essentially 1 month or another. It's not significant from the market's perspective. And so to your point, I think the market now has largely enough information to adequately pricing the tapering. So I think that means that mortgages, while they still could widen a little bit further, are probably closer to where they end up stabilizing. The real question that I think the Fed and the market now face is not when the tapering is going to occur, but it is how fast is the tapering going to occur. I was of the view that they would likely follow the same process they followed in the past, which is over 8 meetings, they would reduce the pace of both treasuries and MBS sort of on an equal pro rata basis. There was a report from The Wall Street Journal yesterday, maybe that would be over 8 months instead of 8 meetings, which would accelerate the tapering. That seems like maybe a logical compromise between the sort of other extreme, which is to do it over, say, the course of 4 meetings, which I think would be the absolute fastest that they were tapering. So my expectation is it's probably in the 8-month range of tapering, at the very longest 12 months. And again, I don't expect the market to really have a hard time digesting that. Certainly, with rates starting to drift up a little bit, that will ultimately obviously lead to a little bit slower mortgage origination. And then importantly, the big question for the market goes to the discussion we've already had is once they've concluded the tapering, then the question is how long are they going to reinvest for. And that's going to be a really important fact for us to sort of get our arms around as we go forward because it's going to have a significant impact on the mortgage market.

Got it. So how do you sum it -- obviously, the Fed has been a lot more transparent than they were in the past, and it feels like your view is the market is better prepared. But how do you expect this period of tapering to be different from the last?

Well, I mean, the key from the last is that we have such a dramatically different environment in terms of what we know and understand from the Fed, right? When the Fed first grew its balance sheet in QE1, 2 and 3, it was going from 0 to ultimately 4 -- $4.5 trillion. I can't remember what the actual number was. And the discussion and the concern was how long were they going to have that balance and how quickly could they get it back to 0 essentially was the question. In today's world, we have 2 important facts that we didn't have then. And one is that we know actually how they're going to taper because they've already done it. And two, we know that they don't have the same view of the stability of their balance sheet, meaning that they are obviously much more comfortable operating with a significant balance sheet indefinitely, right? And when we went through the tapering the first time, we didn't know that. And in fact, the Fed I don't think actually knew that because it ultimately realized the impact and the intersection between its balance sheet and liquidity and excess reserves in the system and ultimately had to make an adjustment for the funding markets. Today, we have a -- whatever it is $1 trillion balance. And so we know how they're going to taper, but we don't quite know how the size of their balance sheet might affect their tapering. And what I mean by that is that may ultimately alter the -- their willingness to reinvest. I don't know if it means that they're going to reinvest for longer and not care or if they might actually shorten the reinvestment period. After QE3, for example, they reinvested their pay downs for about 3.5 years and then ultimately tapered the pace of reinvestments over about a 12-month period. So they were really, really transparent and methodical in that. If they did that again, that would be incredibly supportive for the mortgage market. I don't know if they'll follow that same process or not. That's something that we're just going to have to learn over time. But that's where I think the difference is going to come in.

Okay. Got it. How, if at all, is your hedging strategy changed as we enter into this period of expected tapering? What are your expectations for rate volatility? And how is that impacting strategy?

Right. Well, I think what I've been sort of describing is that I don't expect tremendous volatility. I don't even really expect significant volatility on the mortgage side of it. I do think that the -- what we'll see in this goal around with respect to tapering is maybe potentially a little bit more interest rate volatility. And that's because, obviously, as the Fed goes through the tapering process, it's telling us something about the underlying economy, right? We will have moved through the Delta variant, the vaccinations will sort of be putting COVID behind us. There's tremendous pent-up job demand out there, 11 million job openings. So you can see a scenario where the economy really starts to heat up. We obviously all know that there's lots of inflationary pressures. How long those inflationary pressures persist, nobody really knows. But what it might mean is that we have a little bit more upward rate volatility. We probably will see some more steepening of the yield curve. So what that means for us is that as we go through this tapering process, you have to be really sort of disciplined in your hedging, which means that we're going to continue to operate with a relatively high hedge ratio as we have. We'll keep our interest rate risk really well managed. And ultimately, the amount of hedges, but also the nature of the hedges will matter. You've seen us add more option-based hedges to our mix over the last several quarters, we'll likely do that, giving us some out-of-the-money protection against sort of outsized rate moves that will be important as well as having longer-dated hedges to give us protection against increases in the back end of the curve and the steepening of the yield curve. So the quick answer is you really have to be active, and I do expect us to see a little bit more rate volatility as we go through the next 12 months.

Got it. And just given that backdrop, how are you thinking about leverage in the current environment? What conditions do you need to see or you look to increase the leverage?

Yes, great question. What I described on our last earnings call was that if you look back over the last probably 5, now 6 quarters, you've seen us sequentially take our leverage down from around 9.5x to 7.5x leverage at the last quarter. And what I described on our call was that I really thought that sort of 7.5 level was sort of a sweet spot, meaning that it gave us the flexibility to take our leverage lower if we really felt like there was going to be a lot of volatility and mortgage spread widening risk or the opportunity and a lot of flexibility in capital to put to work and take our leverage back up to 8, 8.5x if we felt like the return opportunity was there and we had enough transparency on the Fed. What I would say today is that I still think 7.5 gives us a lot of flexibility. It's sort of a nice spot to be. But I would also say that I think we're starting to realize and learn that the downside risk is starting to fade away because of everything that we've already talked about. And really, we're probably closer to the opportunity where mortgage, I think, people will look at the mortgage market and say the returns are really attractive. They've widened. They're reflective of the current environment and return opportunities there are probably more durable. So all other things equal, that would be an opportunity potentially as we get to the tapering and understand some of these last nuances of the Fed exactly what they're going to do. It will give us an opportunity to potentially add mortgages at attractive levels.

Okay. And then what are your expectations for -- I think we talked about this a little bit already, but is -- where returns level out once the Fed has done tapering? Is that kind of still the low-double-digit ROE?

Yes, I think it is. Because I think the market is pretty well priced for it. It could widen a little more like it did in the first couple of months. Lower coupons, I think, are going to be a little bit more susceptible because obviously that's where the Fed is concentrated. But I do think it's going to settle out not far from where we are, which when you think about it from a business model, sort of fits well with our business model, meaning if you can get gross ROEs in that low-double-digit range, you have to factor in your cost of running your business and the cost to rebalance your portfolio over time. So there's some downward pressure on that gross ROE. But ultimately, that translates to ability to generate a really attractive dividend level as well. So I think we're probably pretty close to the end of the mortgage widening cycle. But we got another, call it, month, quarter to go, something like that.

Okay. And do you expect that the end of the Fed taper will kind of close much of the arbitrage and returns between TBAs versus pools?

Actually, no, because it really is going to depend on the reinvestment horizon, right? Because even though the Fed is not growing its balance sheet, if they're just replacing runoff, and what will be critical is if they are replacing runoff, which they're going to reinvest for some substantial period of time, we just don't know how long, I think. Then you have to say, what is the issuance environment like. If mortgage rates are higher, which I think they will be at that time, mortgage issuance is going to be lower, and the Fed is still going to be replacing a significant amount. So on a net basis, they actually still could be having a very significant impact on the mortgage market, and that will continue to lead to some specialness. So while we do expect specialness to come down gradually, like it has -- at times, it was close to 100, now it's probably half that. I do expect there to be some specialness for a reasonably extended period of time.

Makes sense. You discussed your latest thoughts on the potential for GSE reform or new housing policy and how it might affect your business.

I don't expect it to affect our business. But what I would say is that I do think that just generally speaking, there's a lot less sort of risk of unexpected developments in the GSE market under the current regime. And the director there, Sandra Thompson, seems to be doing a really, really great job. And what I would say is that under the last regime, there was risks of announcements and fees that the market didn't expect and at times created a little bit of volatility. I don't expect that same sort of volatility under the current regime. And I also expect that the current regime will push the GSEs back toward their sort of main objective, which is continuing to find ways for the GSEs to expand not only homeownership, but affordable homeownership. That's really the mission of the GSE. So I expect them to be really focused on that part of their mission. And to some extent, I expect the GSE footprint to expand over time, whereas under the old regime, they were trying to really contract and really put themselves in a position to come out of conservatorship. I don't see that happening anytime soon under the current leadership at the FHFA and under the current leadership at the White House. So nothing substantial for our business though.

Okay. The stock has been trading around tangible book or book. So what are your thoughts on potential for equity issuance, and what type of price-to-book premium that you need to see before thinking about raising equity?

Yes. Well, certainly, we view capital market activities as a real opportunity to generate returns for our shareholders. So it's an important part of our business. If we do it in a very disciplined way like we have in the past, when our stock trades at a meaningful discount, we can buy our stock back that's an accretive transaction from a book value and from an earnings perspective. At the current price, it's around $16 this morning. I already said that our book value last night was $16.25. So as you point out, we're trading at just a slight discount, nothing substantial there. On the issuance side, the same sort of discipline holds when our stock is trading at a meaningful premium. Being an internally managed REIT and a relatively fixed cost, we can issue stock and do that accretively to our shareholder. For our shareholders, we'll certainly do that. I don't expect in the current market really much of that activity in either direction. But it's something that we continue to focus on and certainly use those activities as a way to generate value for our shareholders. But a lot also has to -- when we issue stock, it not only has to be accretive to our book value, but we also want to be able to adequately deploy those proceeds at attractive returns. And so things are coming together on that side. But with our stock trading where it is, I don't expect much activity on either side.

Okay. One last question for you before we close. Your core earnings have exceeded the dividend over the last couple of quarters, though the yield remains close to 9%. How do you view the dividend in the context of this dynamic?

Yes. Well, it certainly has, and we're very pleased with the earnings profile of the portfolio. As you mentioned, it's been in the $0.76 range for our net spread and dollar roll income for the last couple of quarters. It's been in excess of that really following the pandemic. So from an accounting earnings perspective, what you're seeing is the portfolio generate really attractive earnings. But when we step back and look at the dividend, I think you have to look at it sort of from a higher-level perspective. And I actually expressed this on our earnings call last time, but it's worth repeating is that for us we're not trying to achieve a particular dividend outcome, right? What we're trying to do as stewards of our shareholders' capital is to generate the best economic return we can for our shareholders, which is the dividend that we pay and the book value change put together, right? That's the way I think our shareholders will look at their investment return. And we want that to obviously be as good as possible. And the composition of that excess return is really important. We want to pay a very attractive dividend. And to the extent possible, also have some book value accretion. We think the combination of those 2 things would obviously be most attractive to most shareholders because you'd be getting great current income, and you would also have sort of upward pressure on your stock price. At today's dividend of $1.44 a year and $16 a share, you're looking at a 9% yield in an environment wherein the 10-year is a trade net less than 1.5%. So we think that the 9% yield is a really, really attractive dividend, but we're certainly always evaluating the appropriateness of that. And what's important there is that we don't necessarily look at the current earnings profile of the portfolio, but you really need to look at the economic return of your portfolio to really have a view of the durability of your dividend. And sustainability and durability is really important to us. So for that, what we're always looking at is, is our dividend consistent with the return environment that we're in. You can think about that as the mark-to-market return on your portfolio, which would be equal to, for example, the mark-to-market return or the return opportunities on marginal investment opportunities. And as we said, marginal investment opportunities are right now sort of in the low-double-digit range, not too dissimilar to where our dividend is. So those are important inputs into our overall equation, but we're certainly always looking at our portfolio, making sure that our dividend policy is consistent with the economic return of our assets. And as I said, in the current environment, things are improving. Things look attractive. Maybe not where they're ultimately going to settle out, but we're probably not far from that. So I think we're in a really strong position with our dividend at the current level.

Okay. Well, I think we're going to need to end on that note, but we'll thank you for all your time and insights. We really appreciate it.

Thank you for having us. I appreciate everything you guys do for us.

All right. Take care.

All right. Thank you very much.