AGNC Investment Corp (AGNCO) 2012 Q3 法說會逐字稿

Good morning, and welcome to the American Capital Agency third-quarter 2012 shareholder call.

All participants will be in listen-only mode.

(Operator Instructions).

Please note this event is being recorded.

I would now like to turn the conference over to Katie Wisecarver in Investor Relations.

Please go ahead, Katie.

Thanks, Frank.

Thank you for joining American Capital Agency's third-quarter 2012 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 such forward-looking statements are intended to be subject to the Safe Harbor protection provided by the Reform Act.

Actual outcomes and results to 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 statement are included in the risk factors section of AGNC's periodic reports filed with the Securities and Exchange Commission.

Copies are available on the SEC's website at www.SEC.gov.

We disclaim any obligation to update our forward-looking statements unless required by law.

An archive of this presentation will be available on our website, and the telephone recording can be accessed through November 16 by dialing 877-344-7529 or 412-317-0088, and the conference ID number is 10019151.

To view the slide presentation, turn to our website, AGNC.com, and click on the Q3 2012 Earnings Presentation link in the upper right corner.

Select the webcast option for both slides and audio, or click on the link in the conference call section to view the streaming slide presentation during the call.

Participants on the call today include Malon Wilkus, Chairman and Chief Executive Officer; Sam Flax, Director, Executive Vice President and Secretary; John Erickson, Chief Financial Officer and Executive Vice President; Gary Kain, President and Chief Investment Officer; Chris Kuehl, Senior Vice President of Mortgage Investments; Peter Federico, Senior Vice President and Chief Risk Officer; Bernie Bell, Vice President and Controller; and Jason Campbell, Senior Vice President and Head of Asset and Liability Management.

With that, I will turn the call over to Gary Kain.

Thanks, Katie, and thanks to all of you for joining us today.

We recognize that there is a lot of uncertainty right now surrounding the Fed's latest round of mortgage purchases, and its impact on the mortgage REIT space.

Therefore, we do plan to dedicate a substantial amount of time on this call to QE3 and the implications for our business.

But let me say up front that we continue to be very comfortable with how our portfolio is positioned, and we believe that prepayments on our specific mortgage assets will remain muted, despite a more challenging prepayment landscape.

And yes, while spreads have compressed, we believe risk is also lower in today's environment.

Therefore, nothing has changed our view that an actively managed portfolio can continue to generate attractive, risk-adjusted returns.

So with that, let's review the highlights for the third quarter on slide 4. Our net comprehensive income, which incorporates both realized and unrealized gains and losses on assets and hedges, was $3.98 per share or $1.3 billion for the quarter.

As a reminder, net comprehensive income is a complete mark-to-market earnings measure.

Net spread income, which is what many people refer to as core income, dropped to $0.86 per share when we exclude $0.07 per share for the one-time premium catch-up amortization attributed to the increase in our CPR projections.

The decline in this figure from last quarter was largely a function of faster prepayment projections produced by our models, which increased from 12% to 14% CPR.

This projection is now around 50% higher than the actual Q3 CPR of 9%.

Lower average leverage versus Q2, partially as a result of our share offering in July, was also a factor.

Now taxable net income, which is not impacted by the changes in projected prepayment speeds or by unrealized gains and losses on derivatives, was $1.36 per share.

Our undistributed taxable income closed the quarter at $1.52 per share, and this despite the higher share count.

Book value grew by more than $3 per share, an increase of just over 10%, to $32.49 per share, as our efforts to position the portfolio with lower coupons and prepayment protected mortgages paid off.

The combination of book value growth, coupled with our dividend of $1.25 per share, led to total shareholder value creation or economic returns during Q3 of $4.33 per share, or 59% on an annualized basis.

Now turning to slide 5, our mortgage portfolio increased to $90 billion by the end of the quarter.

Leverage during the quarter averaged 7.1 times and ended the quarter at 7 times.

Our portfolio CPR actually dropped to 9% during the quarter from 10% last quarter.

Additionally, our latest monthly CPR released in October was also 9%.

Our quarter-and net interest spread was 150 basis points as of September 30.

For comparative purposes if we had used a CPR projection of 9%, which is equal to our Q3 actual speeds, our net interest margin would have been just over 170 basis points.

We also raised $1.2 billion of new accretive equity in July, the proceeds of which allowed us to enhance our portfolio, especially in light of QE3.

Lastly, given the recent stock weakness we have witnessed across the space, our Board approved up to $500 million in share repurchases.

Similar to equity issuance, share repurchases can be an important way to enhance economic returns, especially if our shares are trading at a meaningful discount to book.

But it is important to realize that there are costs to both issuing and repurchasing shares, and so buybacks are unlikely at relatively small discounts to our real-time estimates of book value.

Management and our Board are committed to utilizing all tools at our disposal to enhance long-term shareholder value.

Now turning to slide 6, we can see AGNC's economic returns since 2009.

As a reminder, economic returns include both book value changes and our dividend.

Now there are two really important takeaways.

First, despite significant volatility in both interest rates and in the prepayment landscape, AGNC has been able to generate consistently strong performance over this period.

The second point is that last quarter's performance was our strongest since early 2009, and we feel very good about this achievement.

Now if we turn to slide 7, we can see what happened in the market during the quarter.

As you can see on the top left, Treasury rates rallied, but these moves were relatively small with the 5-year yield falling 9 basis point and the 10-year dropping only around two.

Declines in swap yields were a little more pronounced as swap spreads contracted on the back of a more risk-on mindset during the quarter.

Even so, changes in swap rates were still relatively muted compared to other quarters.

The prices of mortgage-backed securities were very strong during the quarter, largely as a result of QE3.

As the table on the bottom left shows, the price of 30-year 3% coupons increased over 3 percentage points.

In contrast, the price increased notably less as we go up to higher coupons.

For example, 4.5% coupons increased less than 1 point in price.

The same trend can be seen with respect to 15 years with 15-year 2.5% coupons up just over 2 points, while 15-year 3.5%s were up only around three-quarters of a point.

Now importantly, these funds have actually become even more pronounced since quarter end, and we will review that in a few minutes.

So now let's turn to slide 8 and focus directly on QE3.

As everyone has probably heard by now, the Fed announced a third round of quantitative easing and extended their low-grade guidance through mid-2015 at the September 13 meeting.

The new program involves large-scale agency mortgage bond purchases and is open-ended in nature.

The Fed is buying around $70 billion in agency MBS every month, which is comprised of $40 billion a month for QE3 and close to $30 billion related to reinvesting paydowns on their existing portfolio of mortgages.

It is also important to understand what the Fed is buying and why.

As they should, the Fed is purchasing the lowest coupon fixed rate MBS, because these securities have the greatest impact on the rates offered to borrowers.

As such, in the absence of a material change in interest rates, the Fed's future purchases are likely to be focused on mainly 30-year 2.5% and 3% MBS, and 15-year 2% and 2.5% coupons.

Now turning to slide 9, we discuss the implications of QE3 on the mortgage market.

Bottom line is that QE3 is likely to affect investors in two major ways.

First, prepayments on existing mortgages will be faster than they otherwise would have been.

Secondly, QE3 will clearly increase the prices of lower coupon mortgage securities, which will reduce the yields and ROEs of new purchases in the sector.

Participants can debate the magnitude of these impacts, but the direction is clear in both cases.

As we turn to slide 10, you can see what has happened to the prices of various types of mortgage securities from the day before QE3 through last Tuesday, October 23.

Notice how stark the difference is between the price performance of the lower coupon and the higher coupon MBS, now that the market has had a chance to more fully digest the implications of QE3.

For example, 30-year 3% coupons have rallied 1.36 points, while 30-year 4.5%s are essentially unchanged.

The contrast in 15 years is even greater, with 15-year 2.5% coupons up two-thirds of a point, while the next lowest coupon, 3%s, are unchanged, and 15-year 3.5%s and above are actually down in price.

Again, this differs from the prior market information slide that we went over a few minutes ago, in that it is not a quarter-over-quarter comparison but a post-QE3 comparison.

Treasury and rate changes were small, so this table pretty clearly shows how important it was to have been positioned in the right coupons.

Interestingly, as of last week the prices of the majority of outstanding MBS were essentially unchanged from beginning of QE3, with some even lower in price.

Now in contrast, 30-year MBS backed by lower loan balance and HARP securities have appreciated meaningfully relative to TBA mortgages in most coupons, as prepayment fears have increased post-QE3.

For example, if you look at the table on the bottom left, pools backed by lower loan balance 4% mortgages have appreciated almost a full percentage point over their generic counterparts, which were up less than $0.20 since the start of QE3.

I chose this example because 30-year 4% MBS backed by lower loan balance and higher LTV HARP mortgages were our single largest position going into the third quarter, totaling close to $20 billion.

So now that we have an early read as to how QE3 has impacted the mortgage market, let's turn to slide 11 and quickly review a slide many of you may remember from our first-quarter earnings call.

On this slide we describe three different environments we were focused on toward the end of Q1.

The middle scenario highlighted the potential impact of a hypothetical QE3.

Back then, we described this as anything other than a Goldilocks scenario.

If you look back at how we presented the potential implications on the call, the main takeaway from this slide is that it is possible to forecast the market reaction to certain events.

Furthermore, you can often position a portfolio to reduce exposure to key risks, which in this environment is clearly prepayments.

And if successful, shareholder returns are enhanced while aggregate risk is lower.

So now let me turn the call over to Chris to cover more about the prepayment landscape and how our portfolio is currently positioned.

Thank you, Gary.

Managing prepayment risk is even more critical now than it was in the past, given today's low interest rate, high dollar price environment.

On slide 12, we have a graph comparing the prepayment performance of several types of 30-year 4% pass-throughs.

As clearly evident on this slide, prepayment performance continues to vary significantly across various types of mortgages, even of the same vintage and coupon.

And as we will review in a minute, even relatively small differences in prepayment speeds can materially impact the returns on a levered mortgage position.

But with respect to the graph, just look at the two extremes.

On the slower side, we see that lower loan balances -- in this case, loans between $85,000 and $110,000 -- as well as higher LTV HARP securities continue to perform well, prepaying around 10 CPR.

Contrast that with the ARMs line, which represents pools backed by conforming jumbo mortgages, which generally have loan balances averaging over $500,000.

Despite the dip in speeds last month, these securities are still prepaying right around 50 CPR.

Now the generic or TBA securities shown in blue on the graph are also quite fast, with speeds in the mid-30s.

Now keep in mind that the October speeds in this graph are a result of the environment prior to QE3.

The 30-year mortgage rate today is approximately 25 basis points lower than rates driving the October speeds in this chart.

We expect that today's low rates, coupled with the tremendous media attention given to QE3, should continue to drive prepayment speeds higher in the coming months on more generic securities.

On the other hand, we expect that prepayment speeds on lower loan balance and higher LTV HARP securities will remain well-behaved.

Let's turn to slide 13 to review the price performance of a few of these difference sets about pool strategies.

Look at the performance of jumbo 30-year 4%s represented by the orange line.

Over the last quarter, they are unchanged in dollar price despite TBA 4%s rallying more than 1.25 points, and 30-year 4%s backed by 110K maxed loan balances which are up in price by more than 3 points.

The performance of pools with favorable prepayment characteristics is striking, but it shouldn't be that surprising.

At current dollar prices even a 5 CPR differential is very material to returns, and a 20 CPR differential is massive as we can see on the next page.

Turning to slide 14, we have two hypothetical yield tables.

In the table on the top-half of the page, we have yields on generic 30-year 3.5%s at the TBA price as of October 23.

To help put this table into perspective, the 2011 Fannie Mae 30-year 3.5% universe, which is now on average only 12-month seasoned, paid at 31.6 CPR last month.

Even at a 25 CPR, these pass-throughs yield approximately 1.43%.

And if we assume a funding cost of 75 basis points, that leaves just 68 basis points of net spread or a hypothetical gross ROE of just 6.19% with 7 times leverage.

Now if we look at 30 CPR, which is more likely over the next year or so, the net margin falls to only 28 basis points and the ROE drops to just 3%.

The picture clearly gets much worse on generic mortgages with speeds above 30 CPR.

Now on the table in the lower-half of the page, we again have 30-year 3.5%s.

However, here we are calculating yields at a price 1.5 points above the TBA price, which is approximately where MLB or 110K max loan balance pools were trading as of last week.

Clearly, the returns on specified pools where speeds could be expected to remain slow are still compelling versus holding more generic pulls.

On these types of assets, low double-digit returns are still achievable in the absence of a prepayment surprise.

Let's turn to slide 15 to review the composition of the investment portfolio.

As we've discussed previously, asset selection is critical to performing in this environment.

During the quarter we continued to maintain a large percentage of holdings in positions with favorable prepayment characteristics.

The majority of our other non-loan balance and HARP positions were in new production, lower coupons.

In the case of 15-year, mostly 2.5%s, and in the case of 30-year pools on our balance sheet mostly 3.5%s as of quarter end.

However, given the yield table that we just went over on the prior page, generic 30-year 3.5%s now comprise a very small percentage of our position as this exposure has largely shifted to 3%s.

I'd also like to highlight the fact that the portfolio's prepayment speeds remain well contained as Gary mentioned earlier, coming in at approximately 9 CPR for the most recent October factor release.

That was down from 11% the prior month, as we continue to fine-tune our holdings given evolving market conditions.

Now I'd like to turn the call over to Peter to discuss our funding and hedging activities.

Thanks, Chris.

Today I will briefly review our financing and hedging activity for the quarter.

I will start by reviewing our financing summary on slide 16.

At the end of the quarter, our repo portfolio totaled $79 billion.

The cost of our repo funding increased 4 basis points to 46 basis points at quarter-end.

This increase was due primarily to generally higher costs across the repo market.

Another driver of the cost increase was the continuation of a strategy that I have talked about for several quarters, namely the maturity extension of our repo funding.

Again this quarter, we increased the average maturity of our repo funding by 20 days, bringing the average original days to maturity to 141 days.

As a result of this strategy, we have reduced our reliance on one-month funding and increased our ability to access longer-term funding, with maturities extending out as far as three years.

We are very pleased with the progress we have made to date with regard to our funding, and we are encouraged by the growing number of counterparties who are now willing to provide us repo funding with maturities of one year or longer.

Over the last year, we have meaningfully reduced our repo rollover risk and significantly enhanced our liquidity position.

Looking ahead, we will continue to source longer-term funding in an opportunistic way.

Turning to slides 17 and 18, I'll briefly review our hedging activity.

Our pay fixed swap portfolio increased slightly during the quarter to $48.9 billion.

During the quarter, $3.2 billion of pay fixed swaps either matured or were terminated.

We chose to terminate some of our shorter maturity swaps this past quarter because these swaps were no longer an effective hedge of the market value of our mortgage assets.

We replaced these terminated swaps with swaps that we believe will more effectively hedge the market value fluctuations in our mortgage assets.

The average maturity of the pay fixed swaps we entered into during the quarter was 5.6 years.

Our swaption portfolio remained roughly unchanged from the previous quarter at $8.6 billion.

During the quarter we purchased $2 billion of payer swaptions at a cost of $44 million.

We focused our swaption purchases on structures with longer option terms and longer underlying swap terms.

The average option term of the swaptions we purchased was 3.3 years, and the average underlying swap term was 9.1 years.

Our swaption purchases occurred late in the quarter when the cost of interest rate options such as these fell meaningfully.

The recent decline in the cost of swaptions is yet another benefit of the Fed's monetary policy position.

In short, the Fed's programs have led to lower interest rates and lower interest rate volatility.

The decline in interest rate volatility has led to a decrease in the price of swaptions.

This price decline in turn has led us to favor swaptions in the current environment over alternative hedging instruments.

Taken together, our swaps, swaptions, and Treasury hedges covered 81% of our liabilities at the end of the third quarter, down slightly from 86% the previous quarter.

Turning to slide 19, I'll review our duration gap information.

Given the drop in mortgage rates and the increase in mortgage prices experienced during the quarter, the effective duration of our asset portfolio fell to 2.3 years from 2.7 years the previous quarter.

The aggregate duration of our liabilities and hedges decreased only slightly to 3 years.

As a result, our net duration gap at the end of the quarter was negative 0.7 years.

As I mentioned last quarter, given the current low level of mortgage rates, there is a significant amount of extension risk inherent in mortgage assets.

Given the composition of our portfolio at 2.3 years, our asset duration is near its lowest point.

If rates rise, mortgage durations will extend significantly.

Additionally, we believe that our models are somewhat understating our asset durations in today's unique environment.

For these reasons we believe it is prudent from a risk management perspective to position the portfolio with a negative duration gap.

With that, I will turn the call back over to Gary.

Thanks, Peter.

Now if we turn to slide 20, we can look quickly at the business economics slide.

As we usually do, let's focus on the first column in the middle of the page.

The asset yield declined 20 basis points as a function of faster prepayment projections and intra-quarter activity.

But, and this is a critical point for comparative purposes, if we had used a CPR projection of 9%, which would have been equal to our actual prepayments for the third quarter, our asset yield as of September 30 would have been around 284; yes, above where it was at the end of Q2.

So funding costs declined by about 8 basis points during the quarter, leading to a 12 basis point decline in net margin.

The lower margin and reduced leverage results in an approximately 2% reduction in the ROE from the prior quarter.

Now as a reminder, this table is based on our cost basis for assets in the original cost of the swap portfolio.

Footnote one gives you the same figures on a mark-to-market basis.

Now before I open up the call to questions, let me quickly summarize how we see the current landscape.

The bottom line is that we remain confident that we can continue to produce attractive, albeit lower, risk-adjusted returns.

Spreads on new purchases are generally around 10 to 20 basis points tighter.

That change is meaningful, but returns can still be in the low double digits if you buy the right assets.

And yes, the prepayment landscape is considerably more challenging.

But as Chris mentioned, given the composition of our portfolio, we remain confident fees on our portfolio will be well-behaved.

This should not only allow us to maintain reasonable yields on our existing portfolio, but also to minimize our reinvestment needs.

But investors should not ignore the positives associated with the current environment.

We believe that the risk to significant book value declines is also lower.

Why?

Because the Fed's large purchases should reduce the likelihood that agency MBS will significantly underperform our swaps, swaption and Treasury hedges.

In addition, liquidity and funding risks should be lower as the Fed's purchases not only directly improve market liquidity, but also remove substantial amounts of securities from the hands of private holders.

A meaningful percentage of those securities would otherwise have been put on repo by private buyers.

So when you put all this together, we believe returns can still be attractive, especially when you factor in the lower risk quotient and where realistic returns are on other types of investments.

However, we would expect the composition of our income to be much more total return in nature.

As such, this is absolutely not the time to worry about earnings geography, as it is imperative to be willing to reposition your portfolio and capitalize on opportunities that emerge as a result of the Fed's involvement in the market.

So with that, let's move to the most important part of the call, your questions.

(Operator Instructions) Steve Delaney.

Another excellent total return performance.

Do you have me?

Hello?

Yes, we have you now, Steve.

Okay, great.

I'm sorry.

I said congrats on another great total return, Gary.

Look, you guys have built sort of a reputation and a track record of being able to look ahead and anticipate the next big shift in the market, whether it was GSE delinquent loan buyouts in the first half of 2011, or HARP 2.0 or QE3.

I mean it has been one thing after another.

So thinking where we are sitting here with the election and who knows how that turns out, but I've seen a lot of posts recently on Bloomberg from various strategists who are trying to -- at least doing a little gaming theory on what the outlook on mortgages might be one way or another.

Apart from the politics of this, I would just be curious if you could share with us kind of as you look out over the next six to nine months what do you see as the greatest policy risk out there for a mortgage investor, whether that be GSE policy, shifting Fed policy?

Would just love any thoughts you have there.

And specifically if there is any view that you might have towards the FHFA and the job they are doing, and the decisions they have to make with respect to the GSEs.

Thanks.

Sorry for the long question.

Thanks, Steve.

Look, we actually feel, now that QE3 is not a question but a reality, that we feel that within the mortgage market most of the information that we are going to get, we already have.

So what I would say is I think that kind of changes whether they come from Washington or somewhere else right now, in terms of kind of structure or policy, are going to be very much at the margin.

And interestingly, I think when you asked specifically about FHFA, obviously there have been rumors of the potential for a change at the top of that organization.

And we applaud DeMarco for the job he has done in a very difficult seat, and we would not want to see that.

But what I think what we would like to stress is that with respect to kind of the HARP program and some of the biggest impacts on the agency space, we think there is kind of a broad recognition, whether it is Treasury, the Fed, whether it is the GSEs, whether it's HUD, of where we are in terms of refinance policy.

And I think the bigger differences relate to obviously things like principal write-downs and so forth, which are not really that big of an issue from the agency perspective, their big picture market issue.

So what I would like to say more in conclusion to this question is we feel more comfortable right now than we have in ages with respect to policy risk.

And we feel like we know the landscape that we are facing, and we think that the moving parts at this point are going to be very marginal relative to some of the things we've seen in the past.

So instead of sort of a repositioning tied to a single event, it sounds like your approach now looking forward will be just good, old-fashioned risk management with respect to managing duration risks, being well hedged to protect against extension risks, that type of thing.

So to that end, you didn't make any comments on your call specifically about IO or MSRs, but are those products that you continue to sort of have in your toolkit as you look forward to the eventual world of possibly higher rates?

Yes.

I mean look, we are going to always consider every instrument that is out there that we can source.

And rather than go into anything specific around kind of how we see the market and what transactions or what specific strategies we have at this point, what I would just say is asset selection, even though the policy landscape is relatively straightforward, asset selection is going to be incredibly important as is hedging, as you mentioned.

And we actually feel like there are going to be a lot of really good opportunities to continue to position the portfolio well as the landscape changes.

So, again, not policy-oriented landscape changes, but probably more rate and pricing-oriented changes.

And so we feel that this is a relatively comfortable environment right now.

Yes, it is a lower return environment, but I think we know the risks and we know how to manage them.

Very good.

Thanks so much for the color, Gary.

Bill Carcache, Nomura.

Good morning.

Gary, you talked about the decrease in leverage and how that was partly influenced by your offering, but I was hoping you could elaborate on how much the decrease in leverage was driven by the fact that we are now in an environment where the Fed has bid up prices, and you don't really want to take leverage up into that.

And so along those same lines, can we expect or should we expect leverage to actually move a bit lower from here; and if so, by how much?

If you could just give some color around that.

Sure.

What I would say is when I made the comment about the offering, that was really only related to the average quarterly leverage, and that is really important.

In other words, you do an offering, clearly no matter what, your on-balance sheet leverage drops that day a lot, right?

And so the average leverage number was the only thing impacted by the offering.

However, leverage was then built up again.

One of the reasons actually, the biggest reason why the leverage at the end of the quarter ended lower than the average leverage was just the fact that book value was so much higher.

And so if you adjust it for that, we actually ended the quarter with a significantly larger portfolio, so to speak.

So I think you need to -- this was a quarter where leverage kind of moved a fair amount intra-quarter, and that you also have to factor in the impact on book value on the end-of-quarter leverage number.

Now with respect to go-forward leverage, look, over -- what I would like to say at a high level is if mortgages, if QE lasts a long time and if mortgages continue to tighten from here, we certainly see a world where our leverage will be lower six to nine months from now.

We will not take leverage up to try to offset spread tightening if we feel that their risk-adjusted returns still make sense.

However, in the short run I think that it is very unclear and it is dependent on prices and opportunities in the market.

And we would be willing to take leverage up in the short run, kind of given that we know the environment that we are in.

Again, over the long run in a QE3 environment, the trend is probably toward lower leverage.

Okay, that's really helpful.

And you also mentioned the 170 basis point spread, along with the 284 basis point asset yield when adjusting for your actual CPR experience.

Can you talk about, is that kind of fair with where we are today?

And I guess how, in terms of the sustainability of the dividend going forward, I guess you've got a significant buildup in your UTI.

How comfortable are you drawing that down?

If you could just kind of give us some perspective around that, that would be helpful.

Sure.

So let me break that into two questions.

First, the numbers I threw out there, look, we obviously have confidence in our projections.

We feel that making projections makes sense.

We also know that is not the only way people in the space report returns.

So what we are trying to do, and that is why I stress for comparative purposes, trying to give you a feel that if we look at things kind of more or like on an actual basis what kind of returns you would see, in the sense we don't give ourselves to some degree credit for our CPRs right now is the way you could think about it.

We obviously make that money, but it just shows up directly in book value and in OCI, not through kind of the interest income line, because we amortize faster.

So that is what I was trying to address by giving out the numbers at a projection basically of around 9%.

Now that one put aside, let's look at the dividend.

What I want to stress with respect to the dividend is that we have quite a bit of flexibility.

Our two biggest constraints, and we talked about this at length last quarter, are obviously taxable income and undistributed taxable income.

And clearly, we have a large amount of undistributed taxable income.

What does that mean?

It means taxable income is unlikely to be a constraint anytime soon with respect to the dividend.

The other key thing is that we don't want to put book value at risk by paying a larger dividend, even if we have taxable income to do that.

When you look at what has happened with book value over the last year, we are actually since the beginning of 2012, not just this quarter, book value is up like 20% despite paying a healthy dividend.

So we feel we have a lot of flexibility with respect to book value.

Said another way, we could maintain the current dividend for an extended period of time.

We have to evaluate, though, against the environment that we discussed, one which is clearly characterized by somewhat lower returns but also lower risk, what the right strategy is for the dividend in 2013 and beyond.

And we don't really -- and I think it is too early to kind of give any more detailed discussion of it.

But that is how we are thinking about the world, and we know it is obviously an important issue to investors.

That is a really helpful perspective.

I have one last one I would like to squeeze in for Peter, if I may.

Peter, can you talk about some of the risks that you guys face in the swaptions market, and how you -- just how you get comfortable with counterparty credit risk in particular?

I think there has been some commentary about some of the risks there, and I was hoping you guys could give some perspective on how you get comfortable from your perspective with the exposures there.

Sure, Bill, I appreciate the question.

There has been some discussion about the counterparty credit risk difference between swaps and swaptions.

And at the end of the day, there is absolutely no difference between the credit risk created by a swaption versus a swaption, and there's really two reasons for that.

One is that the underlying instrument in a swaption is simply a swap, so the underlying instrument is exactly the same.

And the second is that the counterparty has the exact same obligation to perform on a swap as they do on a swaption.

So there is indistinguishable difference between the credit risk created by a swap in a swaption.

Now, from our perspective, the real issue is how do you manage overall counterparty credit risk with regard to derivatives.

There is really three things that we do.

The first thing is, obviously, we choose our counterparties carefully and monitor the credit risk of our counterparties.

The second, which is an important one, is that we size our derivative position with counterparties very carefully because at the end of the day if there is the unlikely event of a counterparty problem, we have to move our derivative position from one counterparty to another.

So we want to be sure that we are not overexposed in terms of the notional amount that we have out with any one counterparty.

Then the final key point is that with all of our swaps and swaptions, we mark to market those positions on a daily basis and exchange collateral with the counterparty on a daily basis, usually in the form of cash or mortgage security.

So we go home essentially every night with a position that has been fully collateralized.

The risk then is that the counterparty has a problem overnight and that we have to move that position to another counterparty.

And during that time period where you move a position from one counterparty to another, you are exposed.

That is why I mentioned the point about making sure that our positions with any one counterparty are not too large.

So those are really the three things that we do from a credit perspective with regard to our derivatives, and it makes us very comfortable with the positions that we have.

I just want to quickly reiterate -- Peter mentioned this -- but again, both swaps and swaptions are -- basically, they are with the same counterparties.

They are both collateralized daily with the exact same process.

So there really can be no difference in terms of the kind of credit risk, essentially, between -- our counterparty risk between the two instruments.

But to Peter's point, the bigger question is just how you manage that globally.

That's really helpful.

Thanks, guys.

Arren Cyganovich, Evercore.

Thank you.

You mentioned in your remarks that -- I believe you said you decreased the 3.5% 30-year exposure, which I think was a little under one-third of the portfolio, to 3% coupons.

Can you kind of confirm that I heard that correctly?

And also, how are you managing the extension risk on a 3% 30-year fixed period?

I know you kind of touched on it in Peter's comments about the duration being a little bit lower on assets with the models.

It just seems like you have to be very careful on the liability side with that asset.

Sure, and it's a good question and thanks for asking.

First off, I think the comment was that we are less comfortable with generic 30-year 3.5% positions, okay?

And we give you a lot of information on page 30, which allows you to distinguish between the generic positions and let's say loan balance or HARP securities which make up a lot of our 3.5%s.

So absolutely one should not infer from that comment that we are moving like one-third of our position to 3%s.

It is actually -- that position that Chris alluded to having reduced the size of is actually a relatively small position at this point.

It is not immaterial.

It is just nowhere near the sizes you were alluding to.

So keep that in mind, and that is -- I really will evaluate those trade-offs.

And we certainly understand the difference in duration of the different instruments.

Now with respect to hedging extension risk, I want to be very clear what Peter -- I want to just highlight what Peter said earlier and I think it's a critical point.

Swaption vol or the price for options right now is pretty much at a record low level.

And so you can buy options to protect from a rising rate scenario, which most likely probably isn't going to occur anytime soon, but you can buy that protection at a very, very reasonable price right now.

You can also pay on 5 and 10-year swaps, which as we've kind of pointed out to you we've been doing quite a bit.

So hedging duration and to some extent hedging extension risk and convexity, as we call it, is cheaper than it has ever been to do.

So it gives you more flexibility to participate in lower coupons.

As we've told you over the last few quarters in being concerned about being at higher coupons, we felt it has been imperative to increase our hedge ratios, and we've been doing exactly that.

Okay, thanks.

That's helpful.

And I guess back to the 3.5% 30 years, I guess 31% of those you have is lower loan balance and HARP.

So it's still a pretty big nut that would be --.

No, I think you need to look on page 30.

And if you look at the bottom in the 3.5% coupons, what you will see on the table is two-thirds of those are HARP and low loan balance.

I'll talk to you offline because I'm not seeing the numbers correctly there.

Okay.

You know what, I think there may have been -- one presentation may have been put up earlier that had numbers, so make sure you reload that presentation.

There may have been an issue with the first one that went up and if you haven't updated it, that might be explaining the difference.

I'll take a look.

But the real number, just for anyone listening, of that position two-thirds of that position is HARP or lower loan balance.

And again, the aggregate of that whole line item given the specific more generic securities that we have, the actuals last month CPR for that entire line item was 4%.

So there are some places that clearly it is not an immediate concern from the prepayment side.

Great, thanks.

And then just lastly, it seems as though the larger banks continue to control capacity by keeping rates a little bit higher.

Maybe you could give your thoughts on how much you think that is impacting rates and what the potential is, I guess, for another leg lower once they kind of get through this current influx of refi activity for pressure on those -- on rates going forward.

And then maybe the impact of the higher g-fees that are coming through on rates in general.

Sure.

I mean, look, obviously the higher g-fees net increase rates to borrowers by a small amount.

But to your point, capacity constraints are a big issue in the mortgage market right now.

And even though mortgage rates have dropped post-QE3, there is more room for them to drop from here.

That being said, I think that what -- the actual rates that are going to be available to borrowers are still also going to be a function of what happens with interest rates.

But I think you just have to be concerned -- you can't manage your position hoping that capacity constraints in a sense bail you out for an extended period of time.

They really help over the short period of time in that rates don't drop as much or as quickly because, as you say, there is -- a lot of volume originators can keep rates higher.

But over time, they will react with pricing.

So we do feel that it is very possible that mortgage rates to borrowers continue to drop over the next three to six months in the absence of a shock to interest rates.

So when you put that together, it really just reiterates what we keep trying to tell people; this is an environment where you do have to hedge and you do have to protect against extension risk.

But you just have to be careful on the prepayment side or your returns are going to be very weak.

Thank you very much.

Bose George, KBW.

First, just a follow-up on the capacity question.

We've seen the MBA refi index tick down now for several weeks.

Is that consistent with your expectations, and is that sort of the capacity issue?

I think it is partly a capacity issue.

It is partly that mortgage rates have also ticked up post-QE3.

We talked about mortgage prices weakening a little bit.

So I think what you've seen is that you had a major dip in rates post QE3.

Rates have ticked back up.

One factor is just mortgage prices.

Another factor is kind of capacity and the fact that the refi index spiked for a couple weeks to kind of -- to almost 6000.

But I think people should be very -- shouldn't be quick to like get a lot of comfort around the decline in the refi index.

It is still high, and as soon as it drops a lot more, you will see originators lower rates and that will bring the refi index back up.

So our mindset is this is not an environment where you are necessarily going to see massive increases in prepayments.

But you are going to see noticeable increases in prepayments, but they are going to be sustained longer than they otherwise would be because of the fact that there is plenty of room for originators to continue to lower rates as necessary to maintain volume.

So look, you are going to see movement in the refi index like you always do.

But I think the challenge in this environment is maybe a little less of what you used to see 10 years ago, which is massive spikes, and more of extended but still very fast prepayments.

And all you need to do is look at the yield table that Chris went over, and you can see the impact of that on generic mortgage positions.

Taking a step back, though, we now have two or three years of really, really good information about how both lower loan balance and HARP securities can perform in these kind of environments.

So I want to reiterate my comfort with respect to our specific positions.

And I'm basing that both on our knowledge of the market but even more so on just what we've seen over the past few years.

Okay, great.

Thanks for that.

Then just in terms of incremental spreads, just core spreads on your portfolio, should we just look at slide 6 and look between the 12.5 and 15 and say range is somewhere between -- I guess it is 130 to 110?

Look, that slide is designed to be a hypothetical kind of picture.

There could be differences on the specific -- there will be differences over time in the cost of funds how much we choose to hedge.

The prices of those securities move around.

But that is a component, the specified 30-year mortgages, that is a reasonable representation of the yields on a specified 30-year mortgage.

We obviously do buy 15 years as well.

Those are lower returns and lower spreads, but lower risk.

So I would put that in a context -- in a larger context, but you could certainly use that as a building block.

Great, thanks.

Then just one final thing.

Just the 5-point difference between actual prepayments and your modeled number, is there a lot of flexibility in that number, and could you discuss that difference?

There really isn't.

As I've talked about in other calls, we have a very kind of defined process for coming up with the prepayment projections.

And they are based on a third-party vendor who is a leader in the marketplace, BlackRock Solutions.

I want to be very clear we don't -- like the process isn't one where Peter, Chris, Bernie, and I get in a room and say it looks like 14 for this quarter where there is flexibility with respect to that.

It is very much kind of model driven, and it is a very kind of rigid process, so to speak.

And that is really where those numbers come from.

And let me take that one step further.

The big inputs into the model are the mortgage rate or an imputed mortgage rate, and also kind of mortgage prices are by definition kind of a driver of those.

So if you look at kind of mortgage rates which are based on general interest rates and mortgage prices, that is the single biggest driver of the prepayment projections.

Over time what happens is we, coupled with BlackRock, will evaluate the performance of the model versus what we've seen in terms of prepayments.

If there are differences, then we may recommend changes or discuss areas where the model is over or underpredicting speeds.

Or BlackRock on their own may come and say, hey, our model is not doing that well on this and we are going to update it.

That happens as well under a kind of pretty controlled environment.

But those are -- that's the part that is different, let's say, that the pure interest rate or mortgage price-driven component.

And I would add to that if BlackRock were to make a change, it would be a change not unique to AGNC but a change that would apply to all of the users of their model.

The only other thing to keep in mind is that the CPRs that we are forecasting are lifetime CPRs.

So when you compare that to our one-month actual CPR in the case of 30-year, that is on a nine [WALO] on average population.

So as loans season, CPRs tend to increase somewhat.

So that is also a factor when you compare those two numbers side-by-side.

Okay, great.

Thanks for that.

Daniel Furtado, Jefferies.

Good morning, and thank you for the opportunity.

Gary, why do you think the GPs have risen at the GSEs, and how much higher do you think they will go in 2013?

I think they have risen because realistically, the GSEs have a lot of pricing power, and arguably had probably been working for a lot less than they probably should have been.

So irrespective of what you think the actual credit cost is and whether they could have been making money before, 95% of all volume is going through the GSEs.

They have no competitive -- there is no other alternative if they were -- they should be raising prices, and you are seeing to that.

Obviously, they also care about mortgage rates and they're in conservatorship, so they are not being run as purely private companies.

But I think that they are trying to achieve, and I think they are going in the right direction toward achieving the right balance.

I don't think there is significantly -- there is more than another 10 to 25 basis points the GSEs are going to go up.

And I think that the GSEs are probably already quite profitable on their new business, given where GSEs are currently.

So I think when you think about it, there is still an upward bias in GSEs.

What FHFA and the GSEs are doing makes total sense.

But I wouldn't expect more than another 10 to 25 basis points in today's environment.

I guess just so I understand you clearly, you view it as more of an attempt to improve profitability at the GSEs, as opposed to putting more parity in the origination market between the GSEs and other nongovernmental originators.

I think it accomplishes both goals, but in a sense what I would say is, look, I think people -- the government should make sure that the business is profitable and that they are pricing it correctly, even though in a sense they have a monopoly in the GSE space.

So I think that is what you are seeing.

So I think you get -- I mean the good thing about this model is that the government should be able to run the GSEs in a very profitable manner because government-backed mortgage securities add a lot of value to the housing system.

It allows leverage from not only REITs but hedge funds and banks.

And other investors can leverage their investments in mortgage securities and can provide the $10 trillion that is necessary for the housing market.

Without the leverage in the system, you couldn't get there.

So the reality is the GSEs can run that business in a very profitable manner while still being pretty dominant.

But if they raise the g-fees in the direction that they are going, they will allow private capital on the credit side to come in in an increased way.

So I think it is just -- it makes sense kind of all the way around.

Great.

Thanks for the commentary, and nice quarter.

Thank you.

Mark DeVries, Barclays.

Gary, could you talk about the outlook for bank demand for MBS in this kind of lower return environment?

Are we reaching a point where spreads are tight enough that you may start to see them step away?

I think that it is going to be difficult for banks to step away in a large kind of way because realistically, they have incredible amounts of cash on hand.

They have plenty of liquidity, and there aren't other alternatives for them in terms of investing capital.

And I think one of the reasons you are seeing prices and mortgages, let's say, over the last three weeks kind of weekend is that a number of investors have said, let me wait and hope something gets better.

But I think there is even more cash on the sidelines.

And again, it also comes from prepayments.

Anyone who has a decent sized mortgage portfolio is going to be getting reasonable amounts of prepayments back, and so banks are going to be even more flush with cash.

And so I think what you will see is a hesitancy, but the reality is I think people need to be invested generally, and there are no better alternatives.

Okay.

Given the strong move we've already seen in price performance for specified pools, do you still see more upside there?

Could we see convergence on returns to specified pools with generic collateral?

We do think there is more upside.

We still find value in call-protected specified pools.

If you think back to slide 14, there is still a lot of relative value in call-protected specified pools versus more generic securities.

The absolute valuations are also really attractive despite the higher payouts, despite the fact that the Fed's buying a decent percentage of issuance.

On certain segments, we still think that low double-digit growth ROEs are still achievable.

And also as Gary mentioned in the pullback in spreads over the last few weeks, we also find value in lower coupons; in the case of 30-year, 3%s; in the case of 15-year, 2.5%s.

One thing, just to reiterate what Chris just said, just look at slide 14 and the returns on generic mortgages where you don't have prepayment protection.

And this is a pretty recent slide of a week ago.

You are -- I mean on generic 3.5%s you can expect over the next year over a 30 CPR if something doesn't change.

And the returns there are nothing special.

And it is a very different picture if you look at the bottom of the page and you look at the returns there.

So I think practically speaking, as you see people who have to own pools coming into the marketplace, I mean it is actually hard to decide to buy the top.

The payoff isn't even close, or the generic one on the top yield table.

And just one more quick question.

Was the 4 bp increase in the repo cost Q over Q, was that all related to the extension of the terms of your repo?

No, it was actually more just -- repo costs were just kind of generally higher throughout the quarter across the market.

Maybe about 1 basis point was due to the extension.

So it was more just the general elevation in repo costs in the marketplace.

We've started to see those levels come down a little bit, so we're kind of optimistic that we might see a little bit improvement in the fourth quarter.

Got it.

Thank you.

Edward Friedman, McLean & Partners.

Congratulations on a great quarter.

I was wondering in light of the sharp increase in the book value and the non-movement of your hedge portfolio, previously when interest rate declined sharply, so you saw a big jump in book and the hedges sort of offset this.

(inaudible) didn't happen and I understand that it's mostly because the increase in -- sorry, the decline in rates was concentrated in a specific pool in a specific part of the market.

I was wondering how do you plan to protect these gains in the future with the hedges not performing probably as a hedge on that part of the market?

Sure.

Look, it's a good question.

And to your point -- and we had talked about this on our second-quarter call -- we think the bond market in a sense had more fully priced in QE3.

And QE3's impact on the bond market would be less than QE3's impact on the mortgage market.

And I think that is exactly what you saw this quarter, which was a rally in interest rates but it was relatively small.

And so the fact that we were hedged more didn't really impact us that much.

And going forward, we absolutely understand the point that if the Fed were to disappear tomorrow, then mortgages would cheapen up very quickly and you would get -- and you would unwind this price performance.

But let's be practical.

I think that's not going to happen realistically.

I think the shortest timeframe that the Fed could exit the market would probably be in Q2 realistically.

In theory it could be sooner than that, but I think it is unlikely.

And realistically, what I think people should have learned from the last QE1 and even in Treasury's QE2 is that the Fed's impact on the market is more about how much they've absorbed, and they call this the stock effect, versus the flow effect which is what they are doing on a monthly basis.

And so I go into this just to say that in the near-term, given that mortgages have cheapened up from -- they are still at higher prices in the lower coupons than where they were before QE3, but those price increases are not ridiculous at all, given the new dynamics in the market.

So we feel pretty good about kind of the support for the mortgage market over the near-term and actually feel that risk to book value, while clearly always there, are not as high as they've been in the past.

Now over time, that will clearly change.

And it is our job -- I mean that's what you are relying on us to do -- is to manage the portfolio through changing market environments.

And hopefully, we've demonstrated over the years the fact that as markets change, we'll adjust our positions accordingly.

Thank you very much.

Ken Bruce, Bank of America Merrill Lynch.

My question picks up from where you just left off.

When you look at what has been some modest improvement in the housing market, does that give you any either reason for pause, or how do you consider that in the context of what that may do to prepay speeds over time?

And maybe even more so, importantly how it changes the Fed's willingness to pursue QE3 and maybe the broader rate complex?

Sure.

Look, as we've stressed to investors on almost every call is we think about a range of scenarios.

If you think about QE3, on the two kind of endpoints there is one scenario which is one where the housing market gains a lot of steam, the economy turns around, the fiscal cliff is avoided, Europe and China both start to -- or the global economy gains momentum and the Fed backs off pretty quickly from QE3, we should see higher rates in that environment.

We can't discount that scenario.

It certainly could happen and we understand that, and we absolutely are evaluating that as a realistic possibility.

On the other hand, there are plenty of scenarios.

A QE forever is the one people throw out there where the global economy doesn't improve much from here or gets worse.

The US economy continues to muddle through despite a stronger housing market, in which case QE could go well into 2014.

And so what we are trying to do as we've done in the past is recognize that both of those are real possibilities, and try to manage our portfolio accordingly.

We could very easily decide tomorrow that we want to sell the portfolio and go into cash because we are worried about the first scenario I went into.

And in that case if that happened really, really quickly, we would be well-positioned.

On the other hand, if the second scenario happened, we would be toast, so to speak.

So in the reality, we've got to balance those two scenarios.

If you look at leverage, if you look at coupon selection, if you look at hedging, we feel that it is those tools as a whole that you use to kind of toggle through that.

And you've done a very good job of managing through what has been a very volatile market backdrop.

So certainly congratulations are in order for that.

My next question is really more of a subtlety, but I'm hoping you might give us a little bit of sense as to how you think about relatives when you put that in the context of attractive returns.

Is that being defined in terms of a benchmark rate that you are looking at or relative to other levered MBS strategies?

How do you think about how you define attractive in this market?

Look, I think you have to look at it from a big picture.

I mean we've been very clear that -- and I think it's obvious that overall kind of projected ROEs, the trend is lower.

But I think you have to be practical about where returns are in other investments.

You've got the 10-year Treasury at 170 basis points.

You've got corporate debt spreads in a lot.

You've got yields on most non-agency mortgages in a kind of we'll call it around 5%, depending on the type.

So you've got to be practical about where the return environment is.

What are yields on what, quote, people call high yield.

So what I'm saying there is that even at considerably tighter spreads, this sector can still produce reasonable risk-adjusted returns from a big picture perspective.

And I think that is what people have to keep in mind.

Agreed.

Well, again, nice quarter.

Very good job in a very difficult environment.

And thank you for your comments, always appreciate it.

Thank you, Ken.

Jasper Burch, Macquarie.

Good morning, guys.

I guess just starting off with -- I appreciate all the color on your CPR projections, how they are imputed.

I was just wondering if sort of broader market do you have any views on regulatory changes?

I know you said that most of the surprises are behind us, but things like HARP changes and FHFA, rep and warrant relief, and how they might impact the prepayment landscape.

The bottom line is we think that changes going forward will be very much on the margin.

Again, the rep and warrant relief is a small incremental change.

We really don't expect to see anything noticeable in terms of prepayment characteristics from that.

I think it is more of a long-run kind of issue that will allow people to change their business model somewhat.

But it is not going to have any kind of material impact in the near-term.

With respect to changes to HARP 2.0, first off, HARP 2.0 is working.

Second of all, there is a Menendez-Boxer Bill out there that is being discussed which would change HARP 2.0, and let's call it HARP 3.0.

But what is really interesting is that one part of that, the extension of the HARP eligibility date which would probably have the biggest impact on the market was dropped from the first kind of submission to the second one after the bill was discussed.

And so all that bill does at this point is kind of make HARP 2.0 more effective for the universe it touches, so to speak, which are loans originated before May of 2009.

And so from our perspective, that is well less than 3% of AGNC's portfolio.

So it is really irrelevant in the grand scheme of things.

So again, when you think about this really bad refi bill that is out there and if it passes, which it probably won't -- it would affect less than 3% of our portfolio -- it is really hard for us to be overly concerned on that front.

All right, that's fair.

And again, that is on your asset selection which has obviously been your strength and is really commendable.

I was just wondering with a $90 billion portfolio now, how large can you get before you start to lose some of that flexibility?

We get this question every quarter and understand why.

But I think what we continue to demonstrate is that we can manage the portfolio as it gets larger.

That isn't to say that the size at this point is an incremental positive.

But just take a step back and look at what the Fed is doing.

The Fed is buying $70 billion in agency MBS every month.

What other market -- there is no other market really outside of maybe Treasuries where the Fed can do that.

And so just put that in perspective when you worry about a $90 billion portfolio, that the Fed is buying $70 billion a month.

That's fair.

And I guess on that point, you know, with increased liquidity we would normally expect to see sort of declining in cost of funds or on the repo side.

I was just wondering can you guys comment on what are the dynamics going on in the background where despite massive Fed buying, increased liquidity, we are still seeing an increase in repo rates across the agency MBS space?

There obviously has been a lot of shifting, I would say, in balance sheets, really over the last nine to 12 months.

I think we are still seeing that in some of our counterparties where there is some asset and balance sheet reallocation going on.

I think that is one of the reasons why rates have been somewhat elevated for the last couple months.

And we continue to see that, but at the same time we also see other counterparties increasing their capacity to this sector.

And where we've been successful at finding new counterparties, I think we will be able to add some new counterparties in the fourth quarter.

So overall, I would say it has been pretty balanced, but at the end of the day there has been some capital being reallocated, which I think is ultimately causing mortgage repo rates to stay where they are.

Do you think those come back down early next year?

I think it's possible.

Obviously, global stability will matter and how the European situation evolves.

But from our perspective, the key counterparties that we interface with, I think have generally seen some stabilization.

There's still a little bit going on, but overall I think that it is fairly stable outlook.

And we are pretty positive about our capacity and our fund dividend.

And as I mentioned earlier, we are really pleased with the fact that a year ago there was very little in the mortgage repo market being done beyond three months.

And now we have counterparties who are willing to lend us three years or longer.

So that has been a very positive development and I think it is indicative of the fact that the mortgage repo market is very strong.

That's helpful.

Then I guess just one last question.

Gary, you said that you see sort of a low risk to book value decline.

You went through a little bit talking about how asset values are holding up.

And I think that we, along with probably most people, think that as long as you and Fed are buying asset values, they are going to stay elevated.

But I was just wondering in terms of the actual readthrough on book value, shouldn't we just mathematically expect book value to come down as the unrealized appreciation runs off as the portfolio recycles?

In our case, the portfolio recycles incredibly slowly unless we proactively, which we probably will, choose to sell certain things and buy other things.

But if you just think -- and we will only do that and that will hurt, maybe can hurt accounting yields, but that is not real.

That's accounting.

But in terms of the portfolio, in our case recycling because of prepayments at our numbers, it's going to take a really long time.

Now that is not going to necessarily be the case equally throughout the industry.

But I think you really need to kind of think about that when you talk about something like a 10% or even a 14%, whatever the number is.

When you think about those kind of numbers that recycling is an incredibly slow process.

And so we are not saying that book values aren't going to move around, and there will be definitely periods where book value doesn't just go up, and investors should completely understand that.

But what we are really talking about when we talk about kind of the risk is to really big moves in book value relative to hedges.

And we are not saying that is a two-year statement that risk is low, but in the current environment right now, it would be really hard to see book values or mortgages significantly underperform other instruments.

And that is really kind of the only point we are trying to make.

Yes, the environment is going to be different a year from now and we are not sure exactly how, but it will be different.

And it is our job to make the right decisions to continue to position the portfolio correctly, both on the asset side and on the hedging side.

So with that, thank you guys for taking the time to call in, and really feel free to reach out to the team with any further questions.

We have now completed our question-and-answer session.

I'd like to turn the call back over to Gary Kain for his concluding remarks.

Thanks again for calling in.

I know we changed the time of the call, obviously related to the weather, and we appreciate people working with us.

And again, feel free to reach out to us if you need to.

The conference has now concluded.

An archive of this presentation will be available on the AGNC's website.

A telephone recording of this call can be accessed through November 16 by dialing 1-877-344-7529, using the conference ID 10019151.

Thank you for joining today's call.

You may now disconnect.