Thank you all for joining AGNC Investment Corp's third quarter 2026 earnings call. Before we begin, I'd like to review the Safe Harbor statement. This conference call and corresponding slide presentation contains statements that, to the extent they are not recitations of historical fact, constitute forward-looking statements within the meaning of the Private Securities Litigation Reform Act of 1995. All 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 those forecasted due to the impact of many factors beyond the control of AGNC. All forward-looking statements included in this presentation are made only as of the date of this presentation and are subject to change without notice. Certain factors that could cause actual results to differ materially from those contained in the forward-looking statements are included in AGNC's periodic reports filed with the Securities and Exchange Commission.

Copies are available on the SEC's website at sec.gov. We disclaim any obligation to update our forward-looking statements unless required by law. Participants on the call include Peter Federico, President, Chief Executive Officer and Chief Investment Officer, Bernice Bell, Executive Vice President and Chief Financial Officer, and Sean Reid, Executive Vice President, Strategy and Corporate Development. With that, I'll turn the call over to Peter Federico.

Good morning, and thank you all for joining our third quarter earnings conference call. Fixed income markets experienced significant pressure in the third quarter, particularly in the final weeks of September, as a confluence of adverse conditions caused bond market sentiment to turn decidedly bearish, interest rates to move sharply higher, and interest rate volatility to spike. Periods of instability in the Treasury market often drive more pronounced weakness in the Agency MBS market.

This was indeed the case in the third quarter as Agency MBS was the worst-performing fixed income sector. As a result of the underperformance of our asset class, AGNC's economic return in the third quarter was negative 12.7%. The challenging fixed income environment in the third quarter was driven by a number of negative catalysts, including substantial monetary policy uncertainty, mixed inflation data, the crowding out effect of historically high corporate debt issuance, heightened geopolitical risk, elevated energy prices associated with the Middle East conflict, and stronger than expected domestic economic growth.

In a typical quarter, the impact of any one of these factors would have likely been considerable. Collectively, they led to historically poor quarterly fixed income and Agency MBS returns. Monetary policy was perhaps the most significant negative catalyst. At the Jackson Hole Symposium in August, Chairman Warsh signaled a more restrictive monetary policy stance. Consistent with that message, the Fed raised the federal funds rate by 25 basis points at the September meeting, the first increase since 2023.

Although the Chairman characterized the September rate hike as simply removing a dose of accommodation, the market's interpretation and subsequent repricing of monetary policy were much more dramatic. In late February, the Fed funds futures reflected the expectation of 3 rate cuts by year end, consistent with the Fed's own projections and its gradual progression toward a neutral policy stance. Following the September rate hike, however, the market priced in 4 rate hikes by the middle of 2027.

In total, the expected path of monetary policy increased by about 200 basis points over a very short time period. In effect, over a six-month period, monetary policy expectations pivoted from a fixed income-friendly view that the Fed would provide additional accommodation by year end to the belief that the Fed had embarked on a full tightening cycle. Against this backdrop, the yield on the 10-year Treasury increased 82 basis points over the course of the quarter, with 54 basis points of that move coming in September alone.

A quarterly increase of that magnitude has occurred just on 4 other occasions over the last 30 years. This Treasury market weakness spread to other fixed income asset classes, as evidenced by the negative 3.5% total return of the Bloomberg U.S. Aggregate Bond Index in the third quarter. When Treasury yields rise sharply, as they did in the third quarter, it is not unusual for Agency MBS to experience greater weakness.

This is evidenced in the performance of the Bloomberg Agency MBS Index, which posted a negative total return of 4.25% in the third quarter. Moreover, on a duration match basis, Agency MBS underperformed U.S. Treasuries by nearly 120 basis points. Since 2010, the Agency MBS index has underperformed Treasuries by a wider margin in only 2 other quarters. Those 2 periods were the third quarter of 2022, when the Fed raised short-term rates by 150 basis points and was rapidly reducing its balance sheet, and the third quarter of 2011, when the S&P 500 index dropped 14% and 10-year Treasury yields fell below 2%.

Despite this more challenging environment, there are a number of reasons to be optimistic about the outlook for Agency MBS. First, the return profile of Agency MBS is considerably more attractive at current valuation levels. At quarter end, the yield on a new production current coupon Agency MBS was close to 6.5%, which, for a security backed by the support of the U.S. government, should be appealing to a wide range of investors.

For hedged investors, the return opportunity is also compelling, with the spread between current coupon Agency MBS and benchmark rates ending the quarter near the upper end of a multi-year trading range at about 135 basis points to Treasuries and 165 basis points to swaps. Lastly, Agency MBS look increasingly attractive relative to investment-grade corporate debt. Second, mortgage cash flows have become more stable.

When interest rates increase, the duration of Agency MBS also increases. This is the extension risk inherent in mortgage-backed securities. At a certain rate level, however, mortgage durations plateau and become relatively insensitive to further rate increases. Given the magnitude of the rate move in the third quarter, that inflection point has largely been reached, meaning the vast majority of the duration extension in the mortgage universe has already occurred.

Prepayment risk for Agency MBS has also diminished. A simple measure of prepayment risk is the percentage of outstanding mortgage-backed securities that have a 50-basis-point refinance incentive. With the primary mortgage rate meaningfully above the average note rate of the mortgage universe, less than 2% of the outstanding mortgage universe has a 50-basis-point refinance incentive. Moreover, if mortgage rates were to drop by a full percentage point, the refinanceable population would increase to only 10% of the outstanding universe.

In summary, the aggregate cash flows of the mortgage universe are more stable today because extension risk is minimal and prepayment options are further out of the money. Finally, the technicals for Agency MBS continue to point to favorable performance once interest rate volatility subsides. With the primary mortgage rate near 7.5% and favorable year-end seasonality, the supply of new Agency MBS for the remainder of this year and into next year should be materially lower than previously anticipated.

At the same time, the demand for Agency MBS should remain steady due to heavy bond fund inflows, the finalization of bank capital rules, the reemergence of the GSEs, and an uptick from other investors as they seek to take advantage of attractive return opportunities in the Agency MBS market. Taken together, these conditions should benefit Agency MBS performance over time. That said, interest rate volatility and macroeconomic uncertainty are likely to remain elevated over the near term, and Agency MBS performance will likely remain sensitive to that backdrop. With that, I'll now turn the call over to Bernice Bell, our Chief Financial Officer, to review our financial results in greater detail.

Thank you, Peter. For the third quarter, AGNC reported a comprehensive loss of $1.28 per common share. Our economic return on tangible common equity was -12.7% for the quarter, consisting of $0.36 of dividends declared per common share and a $1.45 decline in tangible net book value per share due to the underperformance of our mortgage assets relative to our interest rate hedges. As of last Friday, our tangible net book value per common share was up 3% since quarter end, or a little over 2% net of our October dividend accrual.

During the quarter, AGNC was added to the S&P MidCap 400 index, an important milestone for the company and a testament to the scale that AGNC has attained over our 18-year history. The announcement and subsequent index rebalancing activity drove significant trading volume in our common stock. We took advantage of this outsized liquidity to issue $992 million of common stock through our at-the-market offering program at a significant premium to tangible net book value.

Our access to capital at attractive levels enabled us to maintain our leverage profile within a reasonable band while growing our asset portfolio by $2.5 billion on a cost basis at wider spread levels. Even with this growth, leverage increased only modestly to 7.9 times tangible equity at quarter end from 7.4 times as of Q2. In addition, our liquidity position remained strong throughout the quarter, ending with $6.3 billion of unencumbered cash and Agency MBS, representing 57% of tangible equity.

Turning to earnings, net spread and dollar roll income was $0.37 per common share for the quarter, down $0.03 from Q2. The decrease primarily reflects 2 factors. A 14-basis-point decline in our net interest spread as higher funding costs more than offset higher asset yields, and a slight earnings drag from capital raised but not yet deployed. Lastly, the average projected life CPR of our portfolio declined nearly 2 points to 6.7% at quarter end, consistent with higher rates. Actual CPRs for the quarter fell to 8.2% from 13% in the prior quarter. With that, I'll turn the call back over to Peter to discuss our portfolio.

Thank you, Bernie. The underperformance of Agency MBS was fairly uniform across the coupon stack in the third quarter. As measured by the Bloomberg Agency MBS Index, the negative excess return by coupon ranged from 110 to 140 basis points. Middle coupons, which were widely held, marginally underperformed as lower coupons benefited from a lack of supply and bond fund inflows, and higher coupons benefited from sharply reduced prepayment risk.

At quarter end, the market value of our asset portfolio totaled $94 billion. During the quarter, we purchased $2 billion of primarily intermediate coupon specified pools. We also shifted the coupon composition of our portfolio somewhat higher, given the more attractive yield profile of this segment of the market. As a result, the weighted average coupon on our portfolio increased to 5.13%. Through purchases and portfolio repositioning, the percentage of assets with favorable prepayment characteristics increased to 83%.

A common misconception regarding specified pools is that these securities only provide a benefit in faster prepayment environments. Certain pools, however, can provide both call protection in declining rate scenarios, and importantly, extension protection in rising rate scenarios through seasoning and faster borrower turnover. Our hedge portfolio, excluding receiver options, totaled $86 billion at quarter end, an increase of more than $12 billion from the prior quarter, driven primarily by the addition of longer-term Treasury-based hedges.

This hedge portfolio repositioning reflects our decision to offset nearly all of the asset duration extension that we experienced during the quarter as it occurred. Through these actions, our duration gap at quarter end was little changed at 0.8 years. Moreover, at current rate levels, the duration of our asset portfolio is close to fully extended, and as such, relatively insensitive to further rate increases. With that, we will now open the call up to your questions.

Good morning.

Yeah, there's a lot there, Crispin. First off, I think you're absolutely right to sort of identify a shift that may be appearing now post quarter end. Obviously, I talked about the fact that for a good part of the third quarter, the sell-off was sort of contained, and there really wasn't that much damage, if you will, to the bond market. Really, it sort of unraveled in the last 2 weeks of September. There was obviously some positioning that had to take place, repositioning of portfolios, maybe some forced transactions, not specific to MBS, but just in general in the rates market.

Given that 54 basis point move in the 10-year Treasury, that's always going to happen. But since quarter end, I would say that perhaps we're starting to see the market in what I would kind of define as a consolidation phase, in meaning that the market has to trade at this new range for some period of time. People need to be comfortable with what the upper end of the range is and what the lower end of the range is. Obviously, when we ended September with the 10-year Treasury at 5.30-ish percent, the question was, is the upper end of the range 5.75 or is it ultimately 6?

I think the important trading behavior that we've seen since quarter end is that we've bumped up against that 5.30, 5.35 range probably on 4, maybe even on 5 occasions. Often in that same day, it bounced off that level and ended the day lower. I think that was a really good sign for the market. As you point out today, we're back to maybe 5.15 or so this morning. So our expectation is perhaps we're starting to see the new range, and maybe that new range might be something like 4.90-5.40.

You're right, though, to point out that there are a lot of other macroeconomic forces, not the least of which might be what's happening overseas, instability in foreign bond markets, and particularly in the French bond market. That was one of the catalysts that I think that ultimately started the rally back in the U.S. With respect to corporate issuance, or not corporate issuance, but maybe AI. We did have some big transactions go through at the end of the third quarter.

Importantly, I think what we started to see after those transactions, that Paramount transaction in particular, that was like a $45 billion transaction. We did start to see some of those tranches widen materially after issuance. Maybe that is also a sign that ultimately corporate issuance in the fourth quarter may not be quite as high. The market may not have as much capacity to absorb all of that flow that we might see.

In fact, when you look at the amount of corporate issuance, it was sizable in the third quarter at a little over $200 billion. It was actually less than what was in the second quarter. It is going to be really interesting to see going forward. Again, there is a lot of moving pieces, but I am somewhat optimistic with what we are starting to see now with the Treasury market starting to perhaps find a new range. I will pause there and let you have a follow-up.

Yeah

Yeah. Sure. I ended my prepared remarks, the first part of them, with really the outlook for Agency MBS, and I really do believe that the outlook for Agency MBS is really strong. We just have to get through all of this other macro uncertainty, and ultimately, I think those underlying conditions will really flow through. It really starts with the return profile. When we have seen mortgages at these levels in the past, they have historically been good buying opportunities for both levered and importantly, unlevered buyers.

I really do expect rotation into fixed income from unlevered buyers. I pointed out that the current coupon at 6.5% is really attractive. I expect fixed income flows to be heavy. In fact, I think it was the first week of October, we had $25 billion in fixed income inflow. So unlevered returns are really attractive. The return on the index, the yield is at a little over 6% or thereabout 6% right now. So I think you're going to see unlevered money.

On a levered basis, when you look at where those spreads are, I quoted spreads at 135 and 165 basis points to swaps in Treasury. That's across the curve. That was at the end of the quarter. They're a little bit lower than that today. But when you take those levels, you have to sort of make 2 adjustments when I do the calculation for what our expected return is. One is we have to have the right mix of swaps in Treasuries.

Typically we use about an 80/20 mix as kind of our long-run average. So you would weight those 2 things 80/20. But then importantly, you also have to make a duration adjustment because looking at a current coupon MBS to the curve like I just described, would not be duration neutral. There's actually a benefit because the hedge portfolio is too long of a duration for a current coupon. That's probably worth around 10 basis points.

So you put all those together. If spreads were at 135 and 165, you would average those, make those adjustments. I think that would be a net spread of around 170 basis points after making the duration adjustment. Lever that return the way we typically would operate with leverage at around mid 7 and seven and a half range. Add to it where current coupon yields are, which are now well above six and a quarter today, six and a half at the end.

You're going to get an expected ROE right around 19, maybe a little bit above. So I would say expected returns are in the 18%-20% range based on where mortgages are trading today. I really do believe that mortgages look compelling relative to corporates as well.

Sure.

Good morning, Marissa.

Yeah, I think that what often happens is that when you have this big move in rates, it is very common. In fact, it happens probably every time you have a move of that size in Treasury rates. The mortgage market is going to feel the pressure more. It is the spread product that people can trade. So it is kind of a victim of its own liquidity. When things move like that, and I talked about the duration extension. There was unwinding of mortgage positions due to duration extension, which was significant.

Mortgage durations moved on average almost a full year in the quarter. So there is going to be some natural rebalancing due to that. We also heard of structured mortgage trades, some cap trades that there were some forced liquidations there where a cap came into the money, maybe was not expected for a mortgage cash flow that was converted into a floating rate security. Those things naturally happen, and so the mortgage market feels the pressure of that.

The other thing that happens is that liquidity just generally goes down in that environment. So small transactions have the effect of moving mortgage spreads more than they otherwise should. And once that stops, it just generally stops, and it stops pretty sharply, and you can have convexity buying just as quickly as you had convexity selling. The other thing that sort of takes time is when the mortgage rate moves as sharply as it did, and now up at 7.5%, you do not see the full benefit of the supply side reduction for some period of time.

We are starting to see that right now. A dealer put out yesterday that they expect daily originations now to be down into the $1.5 billion range. I think they were actually in that range just yesterday. But we have not seen that level of daily originations in a long time. Just a couple of months ago, that was in a $3 billion and $3.5 billion range. So it takes a little time for the market to recover from those forced transactions, illiquidity, and ultimately the fundamentals will start to show back up. So it will take time, but I think it will occur.

Well, certainly the level of the premium does factor into our consideration, and as Bernie mentioned, it was really fortunate for us. It was a confluence of positive factors that led to our sort of larger issuance, particularly the inclusion into the S&P 500. That created a one-time demand for our stock that was probably in the $1.5 billion range. So it was very substantial, and we were very proud to have joined the S&P 400.

It was a great accomplishment for our company. Unfortunately, it came in a difficult quarter. But it did allow us a lot of flexibility. And in certain markets, we're going to be able to use that ATM. As Bernie mentioned, it gave us great risk management flexibility, and then ultimately, it's going to give us great investment capacity going forward. But yes, as you point out, the level going forward will certainly dictate whether we use it more or less and also the market conditions.

But right now we're in a really strong position, ending a really difficult, volatile quarter with leverage below 8 and, as Bernie mentioned, really strong liquidity position. We got a lot of flexibility going forward. We just need market conditions to stabilize.

Sure. Thank you, Marissa.

Morning, Doug.

Yeah. I've talked about this before. The way we sort of approach our leverage, and this is important, this is one of the reasons why we ended the quarter where we did, is we started the quarter, if you recall, with leverage in the low sevens. And our expectation when we set our leverage level, we're always thinking about what will that leverage be when the market moves positively and when the market moves adversely like it did.

And when we do our projections, we say, what will happen if rates move against us and spreads move against us? What will that do to our leverage? And are we comfortable with leverage at that point? Because obviously, as a levered investor, what we're trying to do when we set our leverage level is set it at a range such that when the market does come under pressure, we are not forced to adjust our asset portfolio. That was in fact the case in the third quarter.

Importantly, we did not have to adjust our asset portfolio. In fact, we were able to add assets during the quarter. Going forward, I would just say there's no change to that framework. We have plenty of capacity to operate at higher leverage should the market push us there. We have plenty of liquidity, plenty of capacity from a risk management perspective. But I think it goes back to making sure that you enter these periods with the right leverage, and I think we did enter this period with the right leverage. There's going to be some natural movement up and down as spreads widen and tighten.

Sure.

Yeah. It goes back to really my answer to the first question. There is definitely capacity to take leverage up opportunistically. It really just depends on when we see enough signs of stability that we feel like the macroeconomic backdrop is more conducive of that position. We obviously had to prioritize for some period of time, risk management over opportunity. That is just a prudent thing to do in environments like these when there is that much uncertainty about what the forces are at play.

But once we gain confidence, and maybe we are starting to gain confidence, although there is still a lot of macroeconomic uncertainty, at least the rate market is giving us some confidence that there could be an opportunity, yes, to take advantage of these levels. Ultimately, I expect that is going to be the case.

Sure. Thank you, Doug.

Morning, Bose. Bose, we can't hear you. You might be on mute.

Good morning, Trevor.

Hey. Hey, Peter. Good morning. Follow-up question on your comments on marginal ROEs.

Yeah.

I guess if you look at the current dividend level on book value is around 20%.

Yeah.

I was curious if you could kind of give us an update on your thoughts around the appropriate dividend level and if the Fed hikes that are currently priced into the market were to materialize, does that kind of incrementally change anything with your views on the dividend? Thanks.

Yeah. I'll start with the latter part first, and the answer is no, it doesn't with the Fed moves. First off, our expectation is not that 4 rate hikes will be realized. But if they were to be realized, it would not have that meaningful of an impact. As Bernie mentioned, our hedge ratio was significant. We've added a lot of hedges. When I look at our hedge ratio, I look at it adjusted for our receiver swap option position.

On that basis, our hedge ratio was 97%. It essentially means when that's close to 100%, that we have very little exposure to short-term rate movements at all of our debt is synthetically hedged. From that perspective, with respect to ROE and dividend, I would say that our dividend continued to be well-aligned with the economics of our business. You've heard me say this before, but the first thing I look at is what is our total cost to capital.

Obviously, given the amount of movement we had in rates and spreads, our total cost to capital, as you would expect, didn't increase during the quarter, not because the dollar of the numerator increased, but predominantly because of the denominator, the book value going down. It ended the quarter probably somewhere right around 19%-ish as a total cost to capital. Then you ask how does the economics of our business, what is the economics of our existing portfolio?

Our existing portfolio generated $0.37 of net spread and dollar roll income on an ROE basis. That's around a 20% return. Obviously, given the dividend, it's consistent with the book value dividend yield of around 20%. The new book of business, that goes back to the ROE I talked about. A new book of business comes on at around 19%. That also is aligned with our total cost to capital. There is good alignment still. The other thing that I would say is in a quarter when you have this amount of volatility and this amount of movement, we don't hyper-focus on any one quarter in terms of the long run alignment.

I think the long run economics are still aligned. There is a lot of moving pieces that drive these numbers. They will change rapidly quarter to quarter. But I think the long run fundamental alignment is still there.

Yeah. Okay, that's helpful. Thank you.

Yeah

Well, it's a good question. I would say it's more tactical in that because of the way the rates moved and trended, the long-term rates during the quarter, it was a quarter where delta hedging was really important. As we talked about at the beginning in our prepared remarks, I mentioned that our duration gap was essentially unchanged at 0.8 years. We started the quarter at 0.7 years. Our mortgage duration is extended by a full year.

So we essentially offset all of that mortgage duration because we wanted to maintain our duration gap where it was, and that meant that we had to add hedges. So I would say that was tactical in that, of course, if rates were to go the other direction, we would likely then take those hedges back off. So it was in response to the movement in the mortgage market. The other thing I would say about hedging, and I think this is really important, is that we made a very conscious decision to end the quarter with a duration gap that was similar to the previous quarter and a positive duration gap of 0.8 years.

You look at, we show this in our sensitivity table. I think this is really informative about the duration extension left in the mortgage market. Our duration sensitivity table shows that if rates were to go up 100 basis points per year, our duration would only extend 0.4 years. If rates move 50 basis points, it's only 0.2 years. So it's essentially saying that our portfolio is nearly fully extended. That same applies for the mortgage universe, by the way.

The mortgage universe duration, someone showed me this this morning, has a positive convexity today for the first. There's only been 3 other occasions in the last 30 years where the index has had positive convexities. That means the mortgage universe is fully extended. On the flip side, though, when you look at our duration sensitivity table, we do have more exposure to rate rally. Obviously, mortgage durations are not going to move when rates go up, but they are going to shorten.

It shows that our duration would shorten by about a year. So we want to operate with a positive duration gap because from a risk management perspective, the prudent risk management action is to have a long duration gap in the event that mortgages rally. We also obviously have some options in our position that give us some down rate protection as well. So that's the way we're looking at it from a risk management perspective in terms of our duration gap and our interest rate risk management.

Yes

Yeah.

Yeah. First off, with respect to the TBA going forward, I would expect that there's not. When you look at dollar roll specialness and the benefit of TBA funding versus on-balance-sheet funding, that has compressed recently. On some coupons, I would describe it as being back to the historical norm of maybe 10 to 20 basis points of specialness. Not that much meaningful in terms of TBA specialness relative to on-balance-sheet funding. That would say all other things equal, there will probably be less TBA going forward. In the quarter.

In the quarter, we did bring it down opportunistically, and there were 2 factors that drove us bringing it down. One was we rotated out of TBA into specified pools because there were some attractive opportunities to acquire specified pools at attractive payups, at least what we believe are attractive payups. The other was that because of the funding differential, we actually ended up taking delivery of some of our TBA position to take advantage of better funding in the repo market versus the TBA market. Those 2 things together ended up bringing our TBA balance down.

Sure. Thank you. You too.

Good morning.

We are not fully deployed already, and I cannot speak of the timing of it. That will be market and opportunity-driven. I will have to tell you that after the fact, not before the fact. We wouldn't want to say that before the fact.

Oh, when was that raise?

You can sort of judge for yourself in that you could figure out the average dollar price of that capital. And the average dollar price, I do not know exactly, but it was a little over $10. What was it? $10.17, Sean just told me. $10.17. So that gives you an indication where it was raised. You can also look at, Sean, what was the inclusion date of the. You can also look at that two-week period around when the S&P 400 inclusion date was. But so the $10.17 give you an idea of when that capital was raised.

Well, it definitely can. There's no doubt about it. I mean, the fact that just the inclusion required an additional $1.5 billion of demand for our stock says everything you need to know. It is certainly positive. There'll certainly be more buyers of our stock because we are in the S&P 400 on a go-forward basis. There's only 1,500 companies in the S&P universe. So we're really happy with that. And obviously it will be a new source of demand for our stock. Anytime anybody buys the S&P 400, they're buying some share of AGNC stock, which is great.

Sure.

Morning, Jason.

Yeah.

Sure. First off, on the drag, Bernie mentioned it was slight. I would say it's maybe in the neighborhood of a penny, since we deployed essentially $2 billion, maybe 20%-25% of the capital, something like that. So it was about a penny. On a go-forward basis, I would say that obviously what will first be important is establishing confidence in the new Treasury range. So that will be one. Obviously we're going to get some more information on monetary policy.

I just think the amount of uncertainty around monetary policy has really been challenging for the market over the last 4 months or so. We'll have a Fed meeting coming up. We'll have another one in December. Economic data and inflation data is obviously going to have a big impact on that. I expect the Fed to be on hold at this next meeting and then possibly raise again in December. Any sort of change in monetary policy outlook will be important for us.

With respect to the GSEs, we don't know what their purchase activity was in September yet. That's not yet been reported. I do expect to see that there will be some portfolio growth that had occurred in the month of September. The GSEs really can play, and I think do play a really important role in the market now. If you go back to look at the GSE portfolios from April to August, they were essentially unchanged. In the first quarter, they grew at a pace of around $12 billion a month, or $35 billion-$36 billion in the first quarter.

It would not surprise me to see the GSEs acquire mortgages at a similar pace in the fourth quarter as they did back in the first quarter. Because when you look at where mortgage spreads are at the upper end of this sort of multi-year range, it is a great opportunity for the GSEs, it's great for taxpayers, and ultimately having them participate in the market when the market needs it from a liquidity perspective, really strengthens the mortgage market. I think that it's likely that they did buy in September, and it's likely that they're going to buy going forward. It's really beneficial to the U.S. housing finance system that they do that. I expect them to be active in the fourth quarter.

Sure.

Bose, good to get you back.

Yes. When we have a market like we did in the third quarter and we're very active in delta hedging, we do tend to delta hedge more in Treasuries. In fact, it was almost exclusively in Treasuries. As a result of that, our hedge composition shifted down. I think at the end of the quarter, we were in high fifties, maybe 58, maybe 60% of swaps versus Treasuries. That's lower than our long run average. When I talked about our ROE, for example, going forward, I use usually a mix of 80-20.

That would say that over time, I would expect that hedge composition to shift back up toward a greater share of swap-based hedge. Obviously, they have better carry profile to them. But often the first move that we make, we rotate into Treasuries, and then ultimately we rotate out of Treasuries into swaps as market conditions change and we kind of get a feel for what the new run rate is of our hedge portfolio. So I would expect more swaps going forward. Said another way, for example, when we acquire mortgages on a go-forward base, they will likely be hedged 100% swaps.

Great. Thank you.

Again, appreciate everybody participating on our call this morning, and we look forward to speaking to you again at the end of the year.