Earnings Call Transcript - AGNC Q3 2023

Operator, Operator

Good morning and welcome to the AGNC Investment Corp. Third Quarter 2023 Shareholder Call. All participants will be in listen-only mode. After today's presentation, there will be an opportunity to ask questions. Please note that this event is being recorded. I would now like to turn the call over to Katie Turlington, Investor Relations. Please go ahead.

Katie Turlington, Investor Relations

Thank you all for joining AGNC Investment Corp's Third Quarter 2023 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 facts, constitute forward-looking statements within the meaning of the Private Securities Litigation Reform Act of 1995. All such forward-looking statements are intended to be subject to the Safe Harbor protection provided by the Reform Act. Actual outcomes and results could differ materially from those forecast 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 this call include Peter Federico, Director, President and Chief Executive Officer; Bernie Bell, Executive Vice President and Chief Financial Officer; Chris Kuehl, Executive Vice President and Chief Investment Officer; Aaron Pas, Senior Vice President, Non-Agency Portfolio Management; and Sean Reid, Executive Vice President, Strategy and Corporate Development. With that, I'll turn the call over to Peter Federico.

Peter Federico, CEO

Thank you, Katie, and good morning. Despite signs of improvement early in the quarter, the selloff in the bond market intensified in the third quarter as treasury supply concerns and the threat of overly restrictive monetary policy weighed heavily on investor sentiment and drove benchmark interest rates and interest rate volatility materially higher. The treasury market continues to be in the midst of a historic multiyear repricing event. To put the move in context, the increase in the 10-year treasury yield over the last three years is the second largest ever recorded over such a short time period. In percentage terms, the treasury market has never experienced anything like this. The Bloomberg Long Treasury Bond Index, which tracks maturities of 10 years or more, has experienced a total return loss of close to 50% over the last 2.5 years, a loss equal to what the S&P 500 experienced following the dot-com bust in 2000. The move higher in Treasury rates began relatively early in the quarter as supply expectations were revised materially higher due to the growing fiscal deficit. The bearish sentiment accelerated late in the quarter following a hawkish message from the Fed that short-term rates would likely remain higher for longer. In environments like this when treasury prices fall abruptly and the market struggles to find its new equilibrium, Agency MBS typically underperformed. That was indeed the case in the third quarter, as Agency MBS performance relative to benchmark rates lagged meaningfully. In aggregate, Agency MBS spreads to comparable duration treasuries widened 20 to 25 basis points across most of the coupon stack. Since quarter-end, Agency MBS have remained under pressure with spreads widening by a similar amount in October. At this point, Agency MBS spreads are close to the widest levels reached during the height of the pandemic in March of 2020. The sharp steepening of the yield curve also caused Agency MBS performance to vary significantly across the yield curve. Agency MBS hedged with short and intermediate-term instruments performed materially worse than Agency MBS hedged with longer-term instruments. The volatile market conditions that we are now experiencing are a result of a complex set of domestic and global factors. In addition, the Fed is nearing a critical inflection point in monetary policy. During these types of transitions, economic data is highly scrutinized by the market and can have an outsized impact on the trajectory of monetary policy. Once the downward trend in labor and inflation data becomes certain, a more favorable monetary policy stance will undoubtedly emerge. As challenging as this period has been for all bond market participants, the current investment opportunity in agency MBS, on both an unlevered and levered basis, is without question. On an unlevered basis, new production par-priced agency MBS provide investors with the opportunity to earn a yield of close to 7% on a security that benefits from the explicit support of the US government. Importantly, this yield is now at least 150 basis points higher than every point on the treasury yield curve and materially above highly rated corporate debt instruments. For levered investors, in addition to the very compelling base yield of close to 7%, it is becoming increasingly apparent that a new trading range is developing for Agency MBS which materially improves returns. Over the last six months, the spread between current coupon Agency MBS and a blend of five and 10-year treasuries has ranged between 150 and 195 basis points. The average has been 170 basis points, and currently, the spread is near the upper end of the trading range. Like all bond market participants, our financial results have been negatively impacted by the unprecedented speed and magnitude of this Fed tightening cycle. The rapid rise in long-term interest rates, the increase in interest rate and financial market volatility, heightened geopolitical risk, and growing US government dysfunction. As the Fed recently stated, however, this tightening cycle may be nearing its conclusion, and the economic balance of risks is now two-sided. Although this process has taken longer and has been considerably more difficult than anticipated, we continue to believe that a durable and very attractive investment environment is still ahead of us once the uncertainties associated with the current environment subside. With that, I'll now turn the call over to Bernie Bell to discuss our financial results.

Bernie Bell, CFO

Thank you, Peter. For the third quarter, AGNC had a comprehensive loss of $1.02 per share, resulting from the significant rate volatility and spread widening that occurred during the quarter. Economic return on tangible common equity was negative 10.1% for the quarter, comprised of $0.36 of dividends declared per common share and a decline in our tangible net book value of $1.31 per share. As of late last week, our tangible net book value was down about 11% for October. Despite the decline in our tangible net book value, leverage at the end of the quarter remained well contained at 7.9 times tangible equity or only moderately higher than 7.2 times as of the end of the second quarter. With our average leverage for the quarter being 7.5 times compared to 7.2 times for the prior quarter. Our liquidity also remained strong throughout the quarter and in line with our typical operating parameters. As of quarter end, we had unencumbered cash and Agency MBS totaling $3.6 billion, or 52% of our tangible equity, and $80 million of unencumbered credit securities. During the quarter, we also issued $432 million of common equity through our at-the-market offering program, which at a significant price-to-book premium was accretive to our net book value. Net spread and dollar roll income, excluding catch-up amortization, was $0.65 per share for the quarter, a quarterly decline of $0.02 per share. The decline was largely due to a 23 basis point decrease in our net interest spread to 303 basis points for the quarter, driven by higher funding costs, which more than offset an increase in our average asset yield. Lastly, the average projected life CPR on our portfolio at the end of the quarter decreased to 8.3% from 9.8% as of the second quarter. Actual CPRs for the quarter averaged 7.1% compared to 6.6% for the prior quarter. I'll now turn the call over to Chris Kuehl to discuss the agency mortgage market.

Chris Kuehl, CIO

Thanks, Bernie. The start of the third quarter was relatively favorable for both US treasuries and Agency MBS as inflation data continued to show a downward trajectory and regional bank sector stress faded into the background. In fact, treasury yields rallied and Agency MBS tightened over the first few weeks of July. This favorable sentiment shifted, however, when the treasury released its borrowing estimate, which was larger than anticipated. Following the September FOMC meeting, the sell-off in treasuries accelerated with five and 10-year treasury yields ultimately increasing 45 and 73 basis points, respectively, as of 9/30, with two-year yields moving only modestly higher; the yield curve steepened significantly during the quarter. Against this challenging backdrop, Agency MBS underperformed both treasury and swap-based hedges, with performance varying considerably depending on hedge positioning on the curve. Coupon positioning was also a significant driver of performance as lower and middle coupons materially underperformed production coupons. Our portfolio increased modestly from $58 million to $59 billion as of September 30th. Within our agency holdings, we continue to move up in coupon at more attractive yields and wider spreads. During the quarter, we added approximately $10 billion in 5.5 through 6.5 versus lower coupons. We also converted a material portion of our TBA position to a mix of both high-quality and low pay-up specified pools. Our remaining TBA position was largely comprised of Ginnie Mae TBA given attractive valuations and better roll implied financing relative to UMBS. As of 9/30, our hedge portfolio totaled $63.2 billion; given the duration extension in our assets, we increased the duration of our hedge portfolio primarily by adding treasury-based hedges at the 10-year point of the curve. As a result, our duration gap remained low throughout the quarter and was 0.2 years at quarter end. As of 9/30, approximately 70% of our hedge portfolio duration dollars are at the seven-year or longer points on the yield curve with approximately 50% of our duration in Treasury-based hedges. Looking forward, our outlook for Agency MBS is very favorable. Spreads have decoupled from treasuries and corporates due to supply and demand technical factors that we expect will ease over time. As Aaron will describe in more detail, spreads and other fixed income sectors are close to post-GFC long-term averages, while spreads on Agency MBS are in the 95 plus percentile area. This is especially remarkable given that the fundamentals for Agency MBS have rarely looked as compelling as they do today. Organic agency supply is minimal, prepayment risk is deeply out of the money, and repo markets for Agency MBS are deep and liquid. With spreads as wide as they are today, we believe investors in Agency MBS are well compensated for elevated rate volatility over the near term. That being said, geopolitical tensions have increased the near-term risk quotient. While we believe we are in one of the most favorable earnings environments in our history, we will remain disciplined with respect to managing rate risk and leverage. I'll now turn the call over to Aaron to discuss the non-agency markets.

Aaron Pas, SVP, Non-Agency Portfolio Management

Thanks, Chris. The significant interest rate volatility and the sharp move higher in yields by the end of the quarter had a more material impact on interest rate sensitive products than credit spreads. Credit spreads were mixed in the third quarter and many sectors of the market generated slightly positive excess returns. As a proxy for credit spread moves in Q3, the synthetic IG index was eight basis points wider, while the Bloomberg IG Index representing spreads on cash bonds was actually two basis points tighter over the quarter, both outperformed mortgage spreads. The majority of fixed income credit spreads are currently at valuations far from extreme levels, unlike Agency MBS. Taking a step back, while MBS are near their widest spread since the GFC by almost any measure, CDX IG spreads are roughly at their average over the past 15 years. A brief overview of several consumer fundamentals suggests that weakening is developing, albeit slowly. Auto loans, credit cards, and marketplace lending delinquencies have ticked higher. Additionally, the resumption of student loan payments this month will stretch a relatively broad segment of households even further. While excess savings numbers have been revised higher recently, lower income and renter households have likely drawn down a material amount of savings as a result of increased exposure to inflationary pressures. Taken together, the outlook for the consumer has declined and will contribute to further economic slowdown. That said, we expect a much more muted impact on the mortgage side due to the significant amounts of homeowners' equity resulting from several years of strong HPA growth. Turning to our holdings, our non-agency portfolio is roughly unchanged in size, ending the quarter at just over $1 billion in market value. Within CRT, we turned over roughly 15% of the portfolio and continue to skew holdings to reposition into likely tender candidates or migrating further down the credit stack in seasoned and de-levered profiles. The favorable technicals in the CRT market that we discussed on last quarter's call remain in place; low issuance volumes, coupled with the GSE desire to extinguish older credit protection via tender offers, this drove further spread tightening on our CRT portfolio during the quarter. The meaningful valuation improvement throughout this year has correspondingly led to lower expected go-forward return expectations. Nevertheless, the risk side of the equation has declined as we expect lower spread and price volatility for many of our CRT securities as certain segments of the market have become anchored to some degree. With that, I'll turn the call back over to Peter.

Peter Federico, CEO

Thank you, Aaron. Before opening the call up to your questions, I want to briefly discuss our current common stock dividend. We do not provide forward guidance for our dividends, but I do want to share some thoughts on the dividend in the current environment. As I have mentioned many times, one of the primary factors that we evaluate in setting our dividend is the economic return that we expect to earn on our portfolio at current MBS valuation levels. You can think of this return as the mark-to-market return on our portfolio. Given the significant spread widening that has occurred over the last two years in Agency MBS and the subsequent decline in tangible net book value per share and stock price, the dividend yield on our common stock has increased notably. At the end of the third quarter, the dividend yield on our tangible net book value of our common equity was close to 18%. While this number is informative, it does not include the benefit of our lower-cost preferred equity, which is also permanent capital. Including this preferred component, our dividend yield on total equity, including both our contractual preferred stock dividends and our current common stock dividend was approximately 15% at the end of the third quarter. Including our operating expenses, the required yield on our total capital was just over 16%. Said another way, as of the end of the third quarter, we needed to earn a 16% return on our total tangible equity capital base of $6.9 billion in order to satisfy all of our operating costs and dividend obligations. At current valuation levels, the expected levered return on Agency MBS, depending on coupon, is in the mid-teen to low 20% range before convexity and rebalancing costs. The important takeaway from this analysis is that our common stock dividend remains well aligned with the return that we expect to earn on our portfolio at current valuation levels and operating parameters. That said, we continuously evaluate our dividend as market conditions, expected returns, and risk management considerations are always changing. With that, we will now open the call up to your questions.

Operator, Operator

We will now begin the question-and-answer session. Our first question will come from Crispin Love with Piper Sandler. You may now go ahead.

Crispin Love, Analyst

Thanks. Good morning, everyone. I appreciate you taking my questions. First off, can you just give us your updated thoughts on leverage? Just based on the preannouncement last week, you were at, I think, about 8.2 times as of October 20th. But what are the ranges that you're comfortable operating at? And what is the max level that you would operate at before needing to bring it lower?

Peter Federico, CEO

Thank you for the question and good morning, Crispin. You're correct that we reported our updated leverage at 8.2% about a week ago, an increase from 7.9% at the end of the quarter. Currently, our leverage is closer to where it was at the end of the quarter, around 7.8%. We are quite comfortable operating within this leverage range. Mortgages are incredibly affordable, and we would prefer to utilize higher leverage given the low mortgage rates, but we must also remain aware of the volatile market conditions impacting all fixed income markets, especially the treasury market. Additionally, as we disclosed and as Bernie pointed out, we maintain a substantial unencumbered cash and security position from a leverage perspective. It was 52% at the end of last quarter and remains in the mid-50s today. This gives us significant capacity as we look for the right opportunity, which we believe will emerge when the uncertainties the market faced in the third quarter begin to settle. I can't provide a specific answer regarding maximum leverage as it depends on the environmental conditions, including interest rate volatility, expectations about spreads, and current spread levels. However, given the favorable mortgage conditions, it is an advantageous time for investment. Ultimately, we need more stability in the financial markets, particularly from the Fed, and consistency in the treasury market. I'll stop there.

Crispin Love, Analyst

Thanks, Peter. All very helpful there. And then just kind of on your point on how cheap mortgages are. Can you just give an update on your outlook for spreads? They remain very cheap. But curious on your outlook and if it's changed at all over the last several weeks and months on spreads being range bound in your expectations there?

Peter Federico, CEO

Yes. The mortgage market has many positive attributes. In my prepared remarks, I highlighted that the current coupon is nearly 7%. What I find intriguing about the mortgage market right now is its strong appeal to a diverse range of investors from a return perspective. Agency MBS provide exceptional value, yielding almost 200 basis points above the treasury curve for securities backed by the US government. When comparing Agency MBS to high-grade corporate securities, the yield difference is substantial. I believe it's very attractive on an unlevered basis. As noted, the Agency MBS market appears to be establishing a new spread range, likely around 150 to 200 basis points when compared to the current coupons of the five and 10-year treasuries as a baseline. Notably, in the third quarter, we've reached the upper end of that range multiple times over the past year. Recently, mortgages hit that upper limit and have begun to retreat, which I see as a positive development. I anticipate that in the near term, mortgages will likely stay within the higher half of that range, possibly around 160 to 180 basis points, given the existing uncertainties in the broader fixed income market related to the Fed, treasury supply, and interest rate volatility. Over time, as these uncertainties lessen and interest rate fluctuations decrease, I believe mortgages can operate more comfortably within that range. While uncertainties remain in the short term, it is encouraging that mortgages have recently retreated from the upper limit of the range and are beginning to stabilize.

Crispin Love, Analyst

Thanks, Peter. That's it from me. Appreciate you taking my questions.

Peter Federico, CEO

Sure. Thank you, Crispin.

Operator, Operator

Our next question will come from Rick Shane with JPMorgan. You may now go ahead.

Richard Shane, Analyst

Thanks, everybody, and nice to talk to you this morning. I'd like to talk a little bit about the decisions, the tactical decisions to sell securities during the quarter and repositioning yourself within the stack. Obviously, the market saw substantial realized losses. And I guess, to some extent, just by rotating within the stack, you're realizing losses, but as long as you're reinvesting further up the stack, you will benefit from spreads tightening ultimately. Can you talk about that decision? And can you also talk a little bit about the tax implications of that trade?

Peter Federico, CEO

Yes, I appreciate that. We don't really consider it from the perspective of realized or unrealized losses. Ultimately, our securities are all marked to market, whether through our income statement directly or through comprehensive income, as we still hold some securities available for sale, though that represents a small part. Our focus isn't on whether to realize or not. Throughout the quarter, and on a daily basis, we aim to find the right mix of securities that we believe will provide the best risk-return trade-off. Moving into higher coupons has benefited us significantly, as the highest expected returns are associated with those higher coupons. We continued this approach during the quarter and will likely keep doing so from a sales perspective. We're always looking to position ourselves in a way that optimally manages risk for the current environment. Our decisions to buy or sell securities are not influenced by tax implications or the appearance on our financial statements.

Richard Shane, Analyst

Peter, it's great to have that context. We've been following your company for a long time, and when you were managed externally, there were different incentives tied to selling assets at losses. Even then, you were very thoughtful about your investments. My primary question concerns the current balance sheet. These assets were not intended for sale. Can you discuss what you experienced during the quarter? Were there challenging moments when you felt compelled to take actions you would have preferred to avoid?

Peter Federico, CEO

No, that's a great question. The answer is absolutely not. There wasn't a moment where we felt any forced action. If you consider our position during the quarter, we started taking steps due to a notable shift in market sentiment that happened late in July. As Chris mentioned in his prepared remarks, the bond market was actually quite positive for the first three weeks of July, especially in the treasury market where the lower 10-year treasury rate was around 380 at that time. Mortgage spreads tightened during those first three weeks. The market was in a good position until discussions from the Fed about the term premium started surfacing, and ultimately, what triggered the shift was the increase in treasury supply and expectations surrounding it. In August, market conditions were somewhat illiquid, and we took actions during that time to maintain a level of leverage we were comfortable with. Our leverage never exceeded the 8.2% we reported. Our cash position remained in the high 50s each month, consistent with previous levels. We were never in a situation where we had to act against our wishes. What you're hinting at is the challenges emerging in both the treasury and agency MBS markets. Specifically, the agency MBS market tends to underperform in downturns like we recently experienced because it often faces illiquidity. The flows, especially in the agency MBS market, have become largely one-directional with the Fed stepping back and reducing its balance sheet. Given the potential constraints on banks, the money manager community has become the primary buyer or seller of securities. During the environment we faced in August and September, fixed income sentiment turned negative, and bond fund flows became outflows. When bond funds experience outflows and redemptions, they typically sell the most liquid securities available, which are treasuries and agency MBS. As Aaron pointed out, corporate bonds didn't move much at all, and corporate spreads remained relatively stable. It seemed favorable for corporates, yet the selling pressure from money managers occurred in a one-way flow. That trend appears to have eased. The market can reverse direction quickly. For our portfolio, we aim to act proactively and make smaller adjustments over time, ensuring we are never forced to deleverage; that certainly was not the case in the third quarter. We were operating with a position we were comfortable with throughout the quarter.

Richard Shane, Analyst

I guess the challenge of having permanent capital is that there are decisions that you have to make that someone who just has experienced outflows doesn't have to.

Peter Federico, CEO

Well, that's exactly right. And that's the challenge. Sometimes you have to make decisions. For example, we made decisions to sell some securities, which unfortunately, you don't like to do because they are cheap, but there will also come a time when we're comfortable adding more securities and the outlook from a spread perspective will be a lot better than it was at times during the third quarter. Maybe some of that improvement is starting to reveal itself right now.

Richard Shane, Analyst

Thank you. I apologize, I've taken too much time. Thank you, guys.

Peter Federico, CEO

I appreciate the question, Richard.

Operator, Operator

Our next question will come from Trevor Cranston with JMP Securities. You may now go ahead.

Trevor Cranston, Analyst

You've talked about the sort of new trading range you're seeing for spreads. Can you maybe talk about sort of how much conviction you have in the upper range of spreads given the sort of weak demand picture, in particular, for MBS at the moment? If we were to see, for example, like another significant move higher in the 10-year, who do you think the buyer could be that steps in to contain additional widening? Thanks.

Peter Federico, CEO

I appreciate the question. I do have increasing confidence that the upper end of the spread range can hold, but that doesn't guarantee it won't be breached. As you've noted, the events in the third quarter were not related to mortgages but were instead influenced by challenges in the treasury market, such as supply issues, runoff, bank constraints, and the potential for a government shutdown. The treasury market faced numerous challenges that could return and potentially lead to more weakness in Agency MBS. There is certainly a chance that mortgage spreads could widen from here. However, I believe that even if they widen, it won't be sustainable. Given that the agency MBS security currently has the full backing of the US government post-financial crisis, it doesn't make sense for that high-quality security to trade 200 basis points over the US Treasury. Over time, investors will likely shift to that security, especially on an unlevered basis. This relates to your question about marginal demand for Agency MBS in the near term, particularly with the 10-year yield near 5% and Agency MBS yields close to 7%. The rotation in fixed income will arise from unlevered funds moving out of other products like investment-grade corporate debt that present lower yields and higher risks. This rotation will also come from corporations and equities as we move into a slowing economy. Ultimately, this process will drive demand for US Treasuries and specifically Agency MBS securities. However, it's a gradual process; investors need to physically transfer money from one security to another, and accessing the agency MBS market can be more challenging. That's why I believe the upper end of the spread range will hold. At 200 basis points, the additional return for Agency MBS is excessive.

Trevor Cranston, Analyst

Got it. Okay. That makes a lot of sense. Thank you.

Peter Federico, CEO

I appreciate the question, Trevor.

Operator, Operator

Our next question will come from Bose George with KBW. You may now go ahead.

Bose George, Analyst

Hi, everyone. Good morning.

Peter Federico, CEO

Good morning, Bose.

Bose George, Analyst

Peter, thanks for the comments on the dividend. In terms of the ROE that you noted there, I mean, is it another way to kind of think about it as looking for us to look at the leverage on the common? So if you do the math of the 180 basis points or whatever the spread is, we should be thinking really about not the 7.9% at risk leverage, but sort of adding the leverage as given by our preferred and that kind of gets us to more like a higher high-teens ROE on your invested capital.

Peter Federico, CEO

That would be a fair comparison. I'm approaching it from that viewpoint. My intention with my comments on the dividend was to emphasize the importance of considering potential earnings. It's essential to evaluate what we can earn from our portfolio compared to our overall cost of capital. This is significant because our preferred stock provides us a substantial advantage with its fixed dividend of about 7%. Over time, this dynamic will evolve. Currently, we're utilizing around 23% to 24% of our capital in preferred stock. If you performed that calculation, as you suggested, you would find ROEs in the low 20% range, which aligns closely with the 18% I mentioned at the end of the third quarter.

Bose George, Analyst

Yes, perfect. That makes a lot of sense. Thanks. And then just to confirm, the 11% decline in book value that's after accruing for the dividend, right? So it's like a 7.25 mark-to-market.

Peter Federico, CEO

Yeah, I'll stick with the 11% as opposed to giving you a point estimate. But, yes, that does. The estimate that obviously, what you can tell is that when we gave our numbers out a week ago. By the way, we released them because that's typically when we would have released them. This call happens to be about a week later than it normally is for scheduling reasons. The market was, in fact, weaker a week ago than it was just last week, as I mentioned. Mortgages have found some footing and have begun to improve, and 11% is a reasonable number for right now.

Bose George, Analyst

Okay. Great. Thanks.

Peter Federico, CEO

Yeah. Thank you for the question, Bose.

Operator, Operator

Our final question will come from Eric Hagen with BTIG. You may now go ahead.

Eric Hagen, Analyst

Thank you. Good morning. I would like to follow up on the structural leverage and the balance between preferred and common stock, specifically how you view that leverage and what your comfort level is over both short and long timeframes. Additionally, how do you perceive that leverage impacting the valuation of common stock, and what tolerance do you have for it at this moment and in the longer term?

Peter Federico, CEO

Yes, thank you. Over the longer run, we've operated with our preferred mix right now in low 20 range for the last several quarters. It's been 22%, 23% at the end of last quarter. It's ticked up as the book value has declined and has been absorbed by the common that percent has gone up to around 24%, and we're comfortable operating in that range. I don't expect it to change much. It gives us the benefit that you're talking about with respect to our overall cost of capital. Importantly and this sort of gets to part of your question, when you think about our sensitivities, from a risk management perspective, the sensitivities that we disclose, for example, our interest rate sensitivity and our spread sensitivity is based on the common component of that. From a risk management perspective, we are looking at that sensitivity, obviously, as a driver of how we're making decisions about our overall leverage position, our overall interest rate position, and our overall liquidity position. I think it's appropriate to think about tangible at-risk leverage as being based on your total capital base because our preferred is permanent capital that we can use. But from a risk perspective, you want to look at the sensitivity on your common only, and that's why we break it out that way.

Eric Hagen, Analyst

Yes, that makes sense. Okay. Just a question on the hedging here and how high you envision the hedge ratio getting just given the shape of the yield curve and maybe even anything you're looking for this week in the Fed meeting that can maybe change your posture towards taking a duration gap going forward. Thanks.

Peter Federico, CEO

Yes. Let me start with the hedging question. What we're trying to do with our hedge portfolio, if you think about it at a high level, is to achieve two purposes. The two purposes are, one, you want the right mix of hedges that you think gives you the best opportunity to offset the value changes, the market value of your asset portfolio. I think that's the way most people think about it. But there's also another objective of the hedge portfolio, which is to give you the most stable cost of funds. To have a very stable cost of funds, you have to have essentially a 100% hedge ratio, meaning all of your short-term debt is hedged with the same notional amount of hedges. So we're trying to find a mix that achieves both those purposes because both of those objectives are really important. With respect to the hedging portfolio in the third quarter, as both Chris and I mentioned in our prepared remarks, there was a lot of variation in mortgage performance across the yield curve because the 10-year moved up so much while the five-year didn't move as much and the two-year hardly moved. If you think about the market value exposure of a mortgage, you can think about that duration being broken down across the key rates: the two-year part of the curve, the five-year part of the curve, and the ten-year part of the curve. For our portfolio, for example, if you took our mortgage portfolio duration and broke it down across the curve, it would be something like 20% of the sensitivity of the mortgage would be to the two-year, about 30% around the five-year, and around 50% to the back end of the curve. If you look at our hedge portfolio, often I think people look at the notional and they say, 'Well, AGNC has a lot of short-term hedges.' We do from a notional perspective; 44% of our hedges, for example, are three years and in. But when you think about the market value sensitivity of our hedge portfolio, only about 18% of the duration of our portfolio is coming from the two-year part of our curve. We don't have a lot of interest rate sensitivity from our short-term hedges. In fact, we have from a model perspective, if you will, just the right amount of two-year hedges. So we have 44% notional hedges, but only 18% of our hedge sensitivity comes from our two-year part of the curve. I point that out because as you ask about the hedge ratio, we've operated with a really high hedge ratio and a mix of hedges across the curve to try to achieve both those purposes. We've talked about for several quarters now that we've moved more and more of our hedges to the longer end of the curve. In fact, Chris mentioned in his prepared remarks that 70% of the duration exposure of our hedge portfolio is seven years or more. We will likely shift more of our duration to the longer and intermediate part of the curve as we expect the yield curve to steepen and as we expect the Fed to pause. I think at this next meeting, the Fed will, in fact, stay constant. Again, I think they'll talk about the fact that the economy and the outlook is continuing to move in their direction. I don't expect the Fed to make any move. I expect the next move from the Fed to be an ease ultimately down the road. As the Fed gets closer to the pause point and reflects that in the market, then ultimately, I think you would expect us to operate with a lower hedge ratio over time as we want to benefit the portfolio from a decline in short-term rates. But that, I think, is further out. Obviously, that will be something in 2024, where the Fed actually starts to ease. For now, we like having more of our hedges in the longer part of the curve. We'll likely do more of that shift. Over time, I think you'll see our hedge ratio come down. I wanted to give you that sort of explanation on the key rates because I think it's important for people to understand what the notional of our portfolio means and what the duration dollars mean.

Eric Hagen, Analyst

That's really helpful. Thank you, guys. We appreciate you.

Peter Federico, CEO

Appreciate it.

Operator, Operator

We have now concluded the question-and-answer session. I'd like to turn the call back over to Peter Federico for closing remarks.

Peter Federico, CEO

Well, appreciate everybody's time on the call today, and we look forward to speaking to you again at the end of the fourth quarter. Thank you for participating.

Operator, Operator

The conference has now concluded. Thank you for joining the call. You may now disconnect.