Dynex Capital Inc (DX) 2018 Q3 法說會逐字稿

Good morning. My name is Lisa, and I'll be your conference operator today. At this time, I would like to welcome everyone to the Dynex Capital, Inc. Third Quarter 2018 Earnings Results and Conference Call. (Operator Instructions) Thank you.

Alison Griffin, VP, Investor Relations, you may begin your conference.

Thank you, Lisa. Good morning, and thank you for joining us. With me on the call today is Byron Boston, President and CEO; Smriti Popenoe, Chief Investment Officer; and Steve Benedetti, CFO and COO. The press release associated with today's call was issued and filed with the SEC this morning, October 31, 2018. You may view the press release from the homepage of the Dynex website at dynexcapital.com as well as on the SEC's website at sec.gov.

Before we begin, we wish to remind you that this conference call may contain forward-looking statements within the meaning of the Private Securities Litigation Reform Act of 1995. The words believe, expect, forecast, anticipate, estimate, project, plan and similar expressions identify forward-looking statements that are inherently subject to risks and uncertainties, some of which cannot be predicted or quantified. The company's actual results and timing of certain events could differ considerably from those projected and/or contemplated by those forward-looking statements as a result of unforeseen external factors or risks. For additional information on these factors or risks, please refer to the annual report on Form 10-K for the period ending December 31, 2017, as filed with the SEC. The document may be found on the Dynex website under Investor Center as well as on the SEC's website.

This call is being broadcast live over the Internet with a streaming slide presentation, which can be found through a webcast link on the homepage of the Dynex Capital website. The slide presentation may also be referenced under quarterly reports on the Investor Center page.

I now have the pleasure of turning the call over to CEO, Byron Boston.

Thank you, Alison, and thank you all for joining us [on] our call this morning. I want to start the call up with giving a couple of themes that we want you to take away from this conference call. And if you got the slide deck, you can focus on Slides 3 through 6.

Now number one, first, despite 8 rate hikes by the Federal Reserve and over a 100 basis point increase in long-term interest rates over the past 11 quarters, Dynex Capital has continued to generate solid cash flow to be distribute to our shareholders.

And you can look at Slide 6, and you'll see a pictorial, you'll see the Fed funds rate moving up nicely as the Fed is tightening. Then you can see what we've done with our dividend, with our earnings. We've been focused on $0.18. And we have fluctuated earnings, fluctuating around $0.18, but you can see a pictorial, Slide 6.

Now number two takeaway. We believe that we are within striking distance of the end of the Federal Reserve's long tightening cycle. Rates have moved up significantly, and potential negative side effects are already developing in the global economy and global capital markets.

Number three. While we continue to generate solid cash flow in today's environment, we anticipate improving economic return opportunities as the Fed approaches the end of its tightening cycle.

And then finally, number four. Long-term factors continue to favor this business model. Now let me explain. Number one, there is and will continue to be a global demand for yield, which supports the long-term valuations of mortgage REITs. Securitized mortgage assets supported by the United States government are among the safest investments to generate yield in a globe burdened with uncertainty.

Next, there is a need for private capital in the U.S. housing finance system as the Federal Reserve Bank attempts to reduce its investment in RMBS -- Agency RMBS. Furthermore, Dynex brings significant experience and expertise in managing these securitized assets.

And then finally, the final long-term factor that we believe favors our business model is that demographics supports the need for more housing in the United States. Demographics also will support a long-term sustained demand for cash yield as the world's population ages.

So with those openings, I'm going to turn it over to Steve Benedetti to give you the facts about the quarter. Then Smriti will talk to you more about the -- our overall portfolio and macro themes in more detail, and I'll come back at the end of the call just to wrap up.

Thank you, Byron, and good morning to everyone on the call. Just a couple of quick highlights. We earned $0.19 in core net operating income for the quarter. That amount excludes the cumulative economic benefit of approximately $800,000 on $650 million in eurodollar futures, which matured in the third quarter of this year.

As you may know, we use eurodollar futures as -- similar to interest rate swaps. But for core accounting purposes, the benefit of those were excluded.

For the eighth quarter in a row, we declared and paid a dividend of $0.18. Our book value declined modestly around 2% from $0.18 per share -- or $0.18 per share from $6.75 -- $6.93 to $6.75. For the quarter inclusive of the dividend, total economic return was flat at 0.

Over the last 4 quarters, our portfolio has averaged approximately $4 billion, and leverages averaged approximately 6.4x. We reported this quarter an average portfolio of $4 billion and leverage slightly higher at approximately 6.7x. Most of the portfolio growth this quarter was in Agency RMBS.

Net interest spread for the quarter was 1.08% and -- versus 1.07% last quarter. And on an adjusted basis, which includes the cost of our swaps and hedges, adjusted net interest spread was 1.41% versus 1.51%. This decline reflects the increases in short rates on our repo, offset by the benefits from our hedge positions. On average over the last 4 quarters, we averaged 146 basis points of adjusted net interest spread.

With those quick comments, I'd like to turn it over to Smriti Popenoe.

Thanks, Steve. Let's go to our macro view now, and I'll review the slide starting on Slide 4 (sic) [Slide 5]. We've held this macro view now for some time. And basically, what we believe is that the ability for interest rates to rise rapidly and remain elevated over the long term is limited. Why?

First, the foundations of the global economy are built on extraordinary growth in central bank balance sheets and global debt. Second, we see that increasing global debt in and of itself is a significant headwind against growth, higher interest rates and inflation. Third, the ability of central banks to reduce monetary stimulus continues to be challenged by low inflation and tepid growth rates. In addition, the impact of global trade policies and varying fiscal policies inject considerable uncertainty into their ability to continue pursuing tightening policies. These factors, in our opinion, make global growth fragile and vulnerable to rapid adjustments. They also make markets vulnerable to surprise events. That's why we believe the ability for interest rates to rise rapidly and remain elevated over the long term is limited.

Now since the first tightening from the Federal Reserve in December of 2015, Fed funds have increased 225 basis points. 3-year treasuries have increased 200 basis points. And since the election in November 2016, the 10-year treasury rate has increased about 129 basis points. In contrast to prior tightening cycles though, interest rates have moved up in a low volatility environment. We now sit in a range on the 10-year U.S. Treasury between 2.75% and 3.25%. We believe we are approaching levels of rates that are having an adverse effect on the global economy and global capital market.

Regarding U.S. monetary policy, the market is pricing in 3 more increases in the Fed funds target rate through 2019. We believe the Fed will be data dependent, and the future path of short-term rates is largely tied to global economic outcomes. Due to the macroeconomic challenges I previously stated, we believe that rates cannot move above the current range on the 10-year U.S. Treasury for any meaningful period of time without negative consequences to the economy.

If we are indeed near the end of the Fed tightening cycle, as we believe, our view is that the outlook for our business and the mortgage REIT sector in general is bullish. As these tightening cycles end, the yield curve typically steepens, increasing marginal investment return opportunities. Even if the yield curve remains relatively flat, MBS spreads currently provide ROE opportunities in the low teens at reasonable leverage. Part of the current ROE return Agency MBS stems from market concerns over the Fed possibly reducing its investment in RMBS beyond its announced reinvestment policy. We believe this risk is very low.

Given this macro view, let me now explain our strategy since 2015. Over the last 11 quarters, despite increases in interest rates, we have paid dividends of $2.10 per share and posted a total economic return of $1.14 per share or nearly 15% on a cumulative basis.

This is what we've done since 2015. First, we've been willing to allocate our capital and earned income from the highest risk-adjusted return assets. In fact, Dynex has a consistent track record of capital allocation, being able to move in and out of sectors opportunistically. As one of the best recent examples, in 2016 and '17, we transitioned from less liquid hybrid ARMs with high prepayment risk to more liquid 30-year fixed-rate securities with better returns for this environment.

Second, we view investing in lower-rated credit investments as having an asymmetric return profile today, limited upside but severe downside potential, particularly when you consider the incremental liquidity risk. Historically, we have found that holding these strategies on a leverage basis late in the business cycle has proven fatal to many mortgage REITs no longer in business today.

Third, we have avoided the operational risk and overhead associated with inflexible illiquid strategies with a high degree of exposure to regulatory risk.

Fourth, we have been and continue to be unwilling to take a neutral or short-duration position, even if interest rates have risen. The historically high number of short positions in the fixed-income markets today creates the potential for a rapid short covering flight to quality move down in interest rates. The liquidity drain from such a rapid move down, very much ignored by many market participants, would severely limit our financial flexibility, particularly if we were short treasuries or derivatives, believing that our duration position was neutral.

Finally, we have been willing to remain long duration and earn income because when the Fed pauses or eases, we believe there will be multiple opportunities to grow book value per share in the future.

In summary, for this environment, we believe that up in credit and up in liquidity is the appropriate investment strategy.

I'll now turn it back over to Byron.

So I'm going to repeat for you kind of what I told you at the beginning because we want -- it's very important that you understand how we think. We haven't adjusted our thought process. We're very disciplined and methodical as we go through the multiple cycles in arriving at our opinions.

So let me give it to you again. I want to reiterate these key points. We continue to generate solid cash flow with a high quality diversified portfolio of liquid investments. We believe we are in striking distance of the end of the Fed's tightening cycle that we expect will provide opportunities to improve returns on our capital. We are excited about the long-term prospects of the business model. We are happy to be generating our cash flow and earning a solid return for our shareholders from our investments in the highest credit quality assets with the most liquidity. And then finally, we have a long track record of asset allocation and pride ourselves on running a transparent, understandable strategy with appropriate financial disclosure.

It is important to remember that we take a long-term approach to the business, which necessitates looking beyond periods when things appear calm. We are confident in how we are positioned, and we are very comfortable owning and holding Dynex stock in our personal portfolios.

So operator, I -- we can turn it -- open the call for the Q&A.

(Operator Instructions) And our first question comes from the line of Eric Hagen from KBW.

My first question is really just on, I guess, expenses and combined with risk. As your expense ratio has picked up a bit, just given the weaker book value, I mean, how do you think about the ability to offer like an efficient return to shareholders, including the value or differentiation, I guess, that Dynex offers for its higher expense ratio relative to peers and other investment vehicles? I mean -- and then how do you just think about taking -- along the same lines, I mean, how do you think about taking up your leverage and arguably just injecting more risk into the model at this time, which has the risk of pushing that expense ratio up even more?

So Eric, I'm going to point out something in terms of how we manage. Probably, it had been 1.5 years, maybe 2 years almost ago, I told you guys we had the ability to take the leverage up in our business model. And you noticed we really haven't taken it up at all. We've still been able to generate. So we made decisions, very, very disciplined risk decisions not to rush to just pile on a ton of leverage. We believe that leveraging this sector, the highest credit quality sector, is better in 2018 than it has been in my career. We believe the financing's better. We believe the support throughout the repo and the financial -- global financial system is better. And again, I'm going to emphasize, in a globe with this much risk, I am willing to take higher leverage on the most highest credit quality, most liquid assets. So that's the first point. Now the next point. We are never -- Chartpoint is never to be the largest guy in this industry. So I want to make sure you understand we think differently. We pride our ability to be nimble and generate returns from a diversified portfolio. And we've shown it over the last 10 or 11 years. We're moving at a slower pace as opposed to just blasting our shareholders away and trying to rush and become a $10 billion REIT. We have no desire to be there. It will probably limit our ability to take advantage of -- let me tell you some of the things we took advantage of: CMBS IO, the CMBS sector 8 years ago, nonperforming loans. We're one of the first REITs owning NPLs. We're one of the first ones owning a Freddie Mac K securities. We're one of the first ones owning DUS bonds. We're able to do that because we want to maintain a certain level of our business such that we can become nimble. And that's why we talk about the long-term because in the short term, you may get beguiled, meaning you, the analyst community or investors, by what will ultimately turn out in the long term. And in comparison, that will make us look back to 2005 or 2006 at the top REITs that were considered the outperform REITs, and they're not here anymore. So -- and the returns over the long term is what we think about through multiple cycles. So we'll continue to grow the business. We think it's important to grow it in a very disciplined, methodical manner over time. We're never trying to get to be the largest REIT in the business. And I think any differential in cost, look at the long-term returns. Look at the long-term returns. Size hasn't mattered. [We moved it along]. In fact, let's look at the largest REITs in 2005, well, their long-term return is 0, probably negative for most investors along them. So there's a lot of conversation out in the marketplace these days about size and cost ratio. It's overdone. Long term -- look at the long-term returns and then show investors where the cost ratio made a difference and go back and get the largest guys in '05, '06, go back and take the largest guys now and compare them to other people and then say, where exactly has this made a difference.

Our next question comes from the line of Doug Harter from Crédit Suisse.

And given the spread widening we've seen in Agency MBS so far in the fourth quarter, can you talk about your sensitivity to that, given the increase of portfolio size there and kind of your appetite willingness to take leverage up further to take advantage of those wider spreads?

Sure, Doug...

Hey, Smriti, let me point him out to Slide 18 and 19 first. So Doug, again, one of the main keys we did coming to this call was we really are trying to make sure everyone understands. We're trying to be as transparent as possible. So on Slide 18 and 19 in our presentation, we show you what our sensitivities are. So before anyone says, "Hey, what's your book value today?" You know what, this is a volatile environment. Book value is going to move around in a volatile fashion for any financial institution. But we try to give you the road map on Slide 18 and 19, such that you yourself will be able to track what's happening with our book value. There's not a lot of funky adjustments we have to make to really understand our business. So I'm going to let Smriti really get more into the details, but I just wanted to point out that if anyone wants to know book value, where it's been, please go to Slide 18 and 19. That's your road map. You'll be able to use that throughout this quarter and into next year to understand what's happening to the book of business.

And then you can also see on just Page 19 that, quarter-over-quarter, that sensitivity hasn't changed substantially. So the average earning balance, I think, has kind of sat right at $4 billion, Doug. And then the -- on Page 19, you can specifically see the sensitivity to spreads as a percentage of equity. So in our estimation, maybe mortgage spreads are out something like 10 basis points, DUS spreads are out maybe something like 8 since quarter end. So that's about 40% of the 25 basis point move here. And you guys can do the math. I mean, that's -- but it's not substantially different than what it was at the end of last quarter.

And then just...

And then you asked about spread widening, Doug. Spread widening will -- it's been our opinion for some time when we talk about going up in credit and up in liquidity is to be in the best place to try to weather a spread widening as central banks attempt to reduce their balance sheets. So underlying everything we say from a macro view, we are taking a skeptical view toward the central banks globally, their ability to all move in the same direction and reduce their balance sheets. We just don't believe they can do it ultimately over the long term without too much turmoil. However, we do believe that we're ready and waiting for some potential spread adjustments. And so we're carrying how much liquidity, Smriti, we're carrying more liquidity than we carried in the last 11 years to try to deal with this and have the ability to put money back in the market. So spreads widened, returns increased. The real advantage will come from us putting more money into the market at the wider spreads. It will offset any type of book value volatility over the long term.

And then, I guess, just looking at Slide 19, just to get -- is there any difference if you kind of look at the components of your portfolio, kind of Agency versus credit, where the sensitivity of one would be different than the other, just to kind of get into more detail on Slide 19?

Yes. I mean, I think one of the things I want to make really clear is that whatever credit position we have is in the CMBS IO book, and those are off of the AAA part of the cash flow. So it's really highly liquid relative to BBB or unrated parts of the cash flow stack. So the credit piece of our book is going to come in that CMBS IO sector. That sector incredibly has remained very, very stable. So that sits at probably close to $600 million or so in market value in our total balance sheet, and that has been a rock, essentially. So this is where the diversification of the book really comes in. So that sector has not moved in spread at all so far this quarter. The other 2 sectors, as I mentioned, the DUS book, which is a little over $1 billion or under $1 billion and then the pass-through book, which is about $2.5 billion, those are out about 10 basis points -- 8 to 10 basis points quarter-to-date.

Hey, Smriti, can you just talk about the roll-down impact over time of both the CMB -- the IO book and the DUS book in the illustration perspective?

Yes, that's the illustration, is that spread widening in the DUS portfolio does get offset by seasoning. As time goes on, those cash flows get shorter, and the spreads get tighter. So you do have that benefit with the marching of time on both that -- those asset classes. So you'll be covering some of that book value from spread widening just with the passage of time.

Just about -- just on that point, how -- what is kind of the average seasoning of those 2 books? And I guess, where do we stand on that roll down?

So the DUS book is somewhat barbell. So we have about half the book, I would say, is newer paper, and half the book is paper that's 7 years or less. So also, we have the opportunity to monetize some of that, Doug, as we get roll down and you're able to go out the spread curve and the yield curve. So that's an advantage. The CMBS book, most of it sits within, I would say, '15 -- '14, '15 and '16 vintages. We were not active in the '17, '18 vintages in IOs because the returns just weren't there for us. So most of it is probably 2- to 3-year-seasoned at this point.

Our next question comes from the line of Christopher Nolan with Ladenburg Thalmann.

Smriti, given your comments on IO being credit sensitive and also given Byron's comments on that we might be seeing the peak rates, should we expect the IO portfolio to roll off in coming quarters?

It is. It has been, Chris. We've been not reinvesting that book now for almost 1.5 years. So we've just been taking the pay downs and reinvesting those pay downs either in the DUS market or in the pass-through market based on where we think the relative value is.

Great. And then in your interest rate outlooks, Byron, I mean -- yes?

Chris, can I answer -- let me just add one other thing to what Smriti said. We love that CMBS IO product. It's just that it's not priced at the right place for us. But in terms of running the REIT, I mean, that's a core product for us, and it will continue to be, but we just need it to reprice itself.

I would also think it's sort of sensitive to where we are in the interest rate cycle. Also, Byron, in your comments on your outlook for where we are in the interest rate cycle, are you sort of positioning the possibility for an inverted yield curve?

We can deal with it. That's not as big of an issue from an inversion. What I want to be positioned for is when it steepens because that's where whatever happens, doing what I call the winter months of the business, the mortgage REIT business model, when the curve steepens, it will offset. Whatever it takes, it has the potential to offset whatever takes place during these winter months, which, I mean, I'm describing along the Game of Thrones theme. I'm describing the winter months of the business model is when the Fed continues to push up financing cost for financial companies. So does that make sense to you?

Yes, no, I hear it. I guess my follow-up question is on leverage. You're at 6.7. I think your peak, what you mentioned, was 8.0. Is that still a peak? Or how do you guys look at leverage and where it's going in the fourth quarter?

Yes, that is a policy issue that we put out there at 8x leverage, and you can see we haven't come anywhere near it. But let me just tell you, again, one of the biggest points here I really could pound the table on is we make risk trade-offs to generate income. We believe the best opportunity is to take higher leverage in the most liquid, high-quality assets for this environment. And it may not be clear to you all the time. It may not become clear to you until we're further down the road, but we're adamant on this point. Furthermore, we believe the system for leveraging these assets is better today than it's been. There is a stock effect of the Federal Reserve balance sheet. So everyone talks about the Federal Reserve selling Agency mortgage-backed security. They're doing it at a very predictable, slow pace. I heard Janet Yellen a couple of weeks ago say, "It will take a decade." They're not selling. They're allowing the portfolio to run off so there is a stock effect. So that's one. Number two, we believe the repo market is in better shape. And I will quote our financial regulators and central bankers. We believe that the financial system has far more capital within the system. So we believe that this is a better environment to take this type of risk than ever in any point in time in my career. And I would rather be leveraging these bonds at 8x leverage than taking some of the credit bonds at 2x, 3x or 4x leverage. And that's not always understood. But please understand, we are making very deliberate, disciplined, methodical risk trade-off decisions. And we've made one, and that is, we're going to take more leverage on the most highest credit quality, most highly liquid securities. And we believe that our shareholders should feel comfortable over the long term that they can sleep okay at night.

(Operator Instructions) Our next question comes from the line of David Walrod from JonesTrading.

A couple of questions. How are you viewing the TBA market versus specified pools?

David, it's Smriti. So right now, the -- I would say, we're kind of indifferent between the 2. The TBA market, the world had come off a fair amount going into the end of last quarter. Some of that was justified, in our opinion, some of it was not. So we preferred to own pools going into the end of the quarter. And now that's sort of changing as the market adjusts to sort of a new -- the new pricing and the coupon stack. Specified pools for us, right now, on average, our pay-ups at risk relative to other companies that invest in this type of security, our pay-ups at risk are much smaller. So we stayed away from like the high pay-up asset in general because we want to preserve the flexibility in the event that the TBA market does become attractive from a dollar roll standpoint. We want to preserve the ability to either deliver into the roll or to sell the specified pools at some minimal pay-up at the time. So for us, that -- while that trade-off sometimes exists in other people's minds to buy convexity in the [spectrum] market, it's really more of a financial flexibility issue for us. But now, right now, it's almost -- I'm indifferent between the 2. And we're really -- it could go either way for us at this point.

Okay. And then a question on the hedging. Given everything that you guys have talked about, will you be -- how will we be viewing your hedge portfolio relative to your repo and TBA exposure?

I think we disclosed that in the Q or in the press release.

It's in the press release.

It's in the press release, David. But basically, right now, we think our book is over 90% hedged versus repo. So just thinking about the benefit from the 1-month, 3-month basis, which was really a big factor in the first and second quarter of this year, that ended up being a drag in the third quarter. And we've actually seen that 1-month, 3-month basis come back into the low 20s here in the fourth quarter because of the move up in 3-month LIBOR. So that just start to come back in our favor again starting this quarter and going into next quarter if the Fed hikes again. Our book, we're relatively -- our book percentage has been about 85% or higher with respect to hedging our repo position.

Yes. In your press release, it says your average swaps to borrowings in TBA outstanding went from 0.8 to 0.7, June 30 to September 30. So -- and you also note you have no more exposure to eurodollar futures. So will you continue to just use swaps? And that 0.8 to 0.7, should we expect that to go back up towards 0.8 or to maintain at these levels?

So the 0.8 to 0.7, Dave, there's the -- obviously, the TBAs are included in that. When you exclude the TBAs versus repo, it's virtually locked up, which I think is where Smriti was alluding to a little earlier. In terms of the eurodollars and the hedge position going forward, as you correctly point out, we -- all the eurodollars we have currently have matured. We've gone to some lengths to try to explain how we account for them and present them because we could use those instruments again in the future. In terms of where the swap book may go from here and the hedge book may go from here, I'll let Smriti comment on that.

Yes. I mean, David, we're targeting -- we would want to keep that ratio, including TBAs, in the 80% or higher area.

And David, let me add a higher-level comment. And I'm sitting right here -- as I'm talking right now, I'm sitting on the edge of my seat because that's the way I am as I approach these markets today. I truly believe that, one, surprises are highly probable. And so -- and what I -- I'm leading my team to be prepared to adjust this position rapidly. And I am looking for the opportunity to see that yield curve do something that a lot of people won't necessarily expect it to do. As someone asked it earlier, "What happens if the curve inverts?" Everyone thinks the curve is going to invert. Everyone thinks it's going to stay flat. It's highly probable that the curve will surprise you because there's so many global events that could surprise the global capital market. So we're on edge to adjust our hedges as necessary. We're not wedded to any one idea, given our longer-term macro thought process. So we sit on the edge of our seats, looking to dynamically manage our book.

Dave, the last piece on that is that there's forward-starting swaps that will change that percentage as time goes on. Okay? So the current pay swaps will have one level of percentage coverage. And then once you include swaps that start in the future, it's different. So you can just look through that swap maturity table, and we can guide you through that as well.

Our next question comes from the line of Trevor Cranston from JMP Securities.

Most of my questions have been covered already. But I guess one more on the Agency portfolio. Can you talk about, in terms of your portfolio, how extended you think the assets are at this point? And how much incremental extension risk remains if we were to see a further move-up in the 10-year, say, to something like the 3.50% range?

Sure. So again, I would point you all to Page 18 in terms of what the interest rate position is relative to the sensitivity of the book for rates going up. All right? One thing on extension, again, it depends on the coupon mix that you have. Our book is -- consists of 3s, 4s and 4.5s. And your view on extension, obviously, is very much predicated upon your view of the curve and your view of rates. So at this point, it would be hard for me to say that something is fully extended or not fully extended because we've all lived through scenarios where one of the par coupon ends up trading at $92 price. So I'm very, very cautious when I say something is or isn't fully extended. I can tell you very clearly that, right now, the durations of our book, we think about them as maybe the 3.5 coupon as being fair relative to its long-term duration, maybe 4s and 4.5s, with some extension potential still to go, and 3s almost fully extended. I can tell you that, that's sort of our view on the durations. But again, I can't tell you, Trevor, that, that means that we're just going to stop hedging or anything like that. But that's just our view on the durations at this point.

And then I'll remind you, Trevor, that a large part of our book has no extension risk with the CMBS portfolio. That's why we keep that CMBS portfolio in place. It is not to totally expose us to extension risk. And I think in our last quarterly presentation, we actually showed some examples to show you what a portfolio would look like if we were all 30-year Agency RMBS versus us having a diversification of the commercial paper within the book of business. And then, Trevor, let me just point out one other thing because I've read this in different -- by different mortgage analysts. And they'll say, "Well, these -- the mortgage securities are fully extended." You know what, no. The answer is no. It can vary. The durations are -- still can vary. I would say that the -- where most people will get surprised, they'll get surprised on their 3s, I think, more than you'll get surprised on your 4s and 4.5s or even your 3.5s, potentially. But I think the lower coupons is they -- if you really did have a move-up in rates, and they dropped below 90, that's where I think most people will be surprised, more than you will on the 4 and 4.5s. So that's based on my experience of fast rate movements.

Especially if you own a large amount of specified pools with high pay-ups. That can vary...

The pay-ups will really -- that's a great point Smriti [make]. This is really important, Trevor, because I see this in the marketplace. And I don't want to sound like an old baby boomer, but I am. I started trading the securities in '87. And I -- the place where the biggest surprise and especially in 1994, was really that low -- those lower coupons and how much they took another level of extension at the highest point of the rate movement.

There are no further questions at this time. Byron Boston, I'll turn the call back to you.

Thanks, everyone, for joining us. And again, we emphasized a couple of points here. We really do want to be transparent and very straightforward. We believe we have a simplistic strategy, if you ask us. You don't have to add back any funky tax adjustments or any exult to our results just so you can understand what the book value is or whether we've made money or not. That's important to us. We appreciate you joining us, and we look forward to you joining us at the end of the year. Thank you.

This concludes today's conference call. You may now disconnect.