AGNC Investment Corp (AGNCL) 2012 Q4 法說會逐字稿

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

All participants will be in listen-only mode.

(Operator Instructions).

After today's presentation, there will be an opportunity to ask questions.

(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.

Thank you, Denise.

Thank you for joining American Capital Agency's fourth-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 facts, constitute forward-looking statements within the meaning of the Private Securities Litigation Reform Act of 1995.

All such forward-looking statements are intended to be subject to the Safe Harbor protection provided by the Reform Act.

Actual outcomes and results could differ materially from those forecast due to the impact of many factors beyond the control of AGNC.

All forward-looking statements included in this presentation are made only as of the date of this presentation and are subject to change without notice.

Certain factors that could cause actual results to differ materially from those contained in the forward-looking statements are included in 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 February 22 by dialing 877-344-7529 or 412-317-0088 and the conference ID is 10024358.

To view the slide presentation, turn to our website, AGNC.com, and click on the Q4 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 today's call include Malon Wilkus, Chairman and Chief Executive Officer; Sam Flax, Director, Executive Vice President, and Secretary; John Erickson, Director, Chief Financial Officer, and Executive Vice President; Gary Kain, President and Chief Investment Officer; Chris Kuehl, Senior Vice President, Mortgage Investments; Peter Federico, Senior Vice President and Chief Risk Officer; and Bernie Bell, Vice President and Controller.

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

Thanks, Katie.

Good morning, everyone, and thanks again for your interest in AGNC.

As we're going to discuss on this call, our portfolio continues to perform well despite the challenges posed by QE3.

We also believe our near-term earnings outlook is actually stronger now than it was at the end of the third quarter.

During the fourth quarter, mortgages did cheapen and favorable financing opportunities presented themselves in the TBA market.

Moreover, since these financing dynamics are a function of QE3, they are likely to remain very attractive for most of 2013 and possibly beyond as the stock effect of the Fed's purchases dominates the market.

With that introduction, let's turn to page 4 and quickly review AGNC's 2012 full-year performance.

Most importantly, we generated economic or mark-to-market returns of 32% for the year.

This was comprised of the combination of $5 per share in dividends and $3.93 of book value growth.

[We will call it] a 32% return is a worthy accomplishment in any environment, we are especially pleased with AGNC's performance against the backdrop of QE3 and even more so because this is the fourth year in a row where AGNC has been able to produce economic returns in excess of 30%.

Against the backdrop of strong multiyear economic returns, a very sizable amount of undistributed taxable income and our confidence in our investment strategies, we continue to believe AGNC's current dividend remains appropriate at this point.

However, investors should remember that future dividends are, by definition, uncertain and will be a function of market conditions and our actual performance.

Now turning to slides 5 and 6, I want to highlight the most important takeaways from AGNC's Q4 performance.

I also want to apologize upfront as I will be bouncing back and forth between these two slides.

So first and foremost, when thinking about our true earnings power and our risk, it is now important to focus on both our on-balance sheet leverage and our net forward purchases of TBA mortgages.

As you can see on the top of page 6, our average on-balance sheet leverage for the quarter was only 6.8 times.

However, that number was 7.8 times when you include our net TBA position.

At the end of the quarter, our on-balance sheet leverage was 7 times while what we call our at-risk or true effective leverage was 8.2 times inclusive of our $13 billion in forward TBAs.

I started with a discussion of the TBA positions because it is imperative now to the understanding of our results both for this quarter and probably for most of 2013, assuming we continue to roll a sizable amount of MBS.

So now turning back to slide 5, we had $0.89 per share of net spread income during the quarter of which approximately $0.11 related to catch-up am.

However, our estimate of the implied carry related to rolling TBAs or purchasing them for forward settlement dates totaled an incremental $0.29 per share.

If you add that to the $0.89 per share of net spread income, you get a total of $1.18 per share or $1.07 excluding the catch-up am component.

So if you were wondering about the relative contributions of the TBAs versus the on-balance sheet assets given the sizes of the positions, you do have to remember that currently the entire swap cost is essentially allocated to the regular net interest spread on the on-balance sheet portion, which therefore understates the true economics there.

If we look at the bottom of page 6, the snapshot of our net interest spread as of December 31 was 140 basis points, without including the TBA positions.

However, this estimate increases to 161 basis points when we include the economics implied by year-end dollar roll levels.

This gives a more balanced perspective on the relative contributions to earnings power.

We have always stressed that our investment strategies must evolve with market conditions and that dollar rolls are really an excellent example of this.

Additionally, AGNC will always prioritize true economics over earnings geography.

Now before moving on, I want to point out a few other highlights related to the quarter.

First, book value did decline 2.6% as agency mortgages gave back some of their large gains achieved during the third quarter after QE3 was announced.

If we look at the two quarters together, AGNC grew book value by 7.5% in addition to paying $2.50 per share in dividends and we feel really good about that performance both in relative and absolute terms.

Secondly, taxable income remained very strong at $1.93 per share, driving undistributed taxable income to $2.18 per share, our highest level ever.

Lastly, we did repurchase some stock during the brief period in November when our shares traded at a significant discount to book and we remain committed to repurchasing shares again in the future when conditions warrant.

With that, let me quickly turn to slide 7 and discuss what happened in the markets during Q4.

As you can see on the top left, the 5 and 10-year treasury and swap rates increased between 9 and 14 basis points.

Prices of mortgage-backed securities clearly weakened during the fourth quarter as well.

Interestingly, the highest coupons where AGNC has almost no exposure were the worst performers despite the fact that they should have been the least impacted by higher interest rates.

The weakness in these coupons was widely attributed to faster prepayments driven by the continued success of HARP 2.0.

The prices of lower coupons also dropped more than what would have been implied by the changes in treasury or swap rates and this was the key contributor to the modest decline in book value during the quarter.

With that, I'd like to turn the call over to Chris to discuss our asset selection and the portfolio.

Thank you, Gary.

QE3 has driven outperformance in lower coupon MBS, but as Gary just mentioned, it's also responsible for extremely attractive implied financing rates through the dollar roll market.

A dollar roll is simply a transaction where you simultaneously sell and agree to repurchase a TBA mortgage with the same term, coupon, and issuer but for a later settlement date.

And the price difference between the current month's settlement date and the next month's settlement date is referred to as the price drop, which is the economic equivalent of the net interest income earned for holding a mortgage security on balance sheet and financing it with repo.

At the bottom of the slide, we have a table that illustrates the financing advantage of rolling a position versus funding a mortgage security with repo.

In this case, as of December 31, the price drop from January to February on a 30-year 3% TBA was a little less than 0.25 point.

In other words, the market was willing to pay $0.24 more for January's settlement versus February's settlement.

Embedded in this price drop is an implied financing rate and you can solve for this rate given a dollar price and a prepayment speed.

In this case, the implied financing rate is negative 20 basis points at 2 CPR, which is approximately 60 basis points lower than where the pool could have been financed on balance sheet via repo.

Clearly, these financing differences are worthy of significant consideration if they're likely to persist and given that the Fed's mortgage purchase program is a key driver of these dynamics, we believe that these financing opportunities should remain in place for most of 2013, which makes them very difficult to ignore.

To this point, the implied financing rate on dollar rolls since year-end has actually improved significantly and is now even more compelling today than it was during the fourth quarter.

For instance, the implied financing rate on 30-year 3%s is now around negative 50 basis points, which again is approximately 90 basis points through one-month repo.

So to summarize, QE3 has led to tighter spreads on lower coupon MBS versus interest rate swaps, but when you overlay the extremely attractive financing rates available through the roll market, current margins are actually wider.

Let's now turn to the next slide to discuss our portfolio composition.

On slide 9, you'll notice that we have added another portfolio composition chart in the middle of the page, which includes our net-forward TBA position.

As of quarter-end, it was approximately $13 billion.

Now moving down to the tables at the bottom of the page, you can see that the portfolio remains well-positioned against prepayment risk with the vast majority of these positions backed by loans with lower loan balances or loans that were originated through the HARP program.

While the aggregate percentages within each category haven't changed that much over the past few quarters, we have been upgrading some of these positions to subsectors of the market with better attributes.

For example, within the HARP category, which includes loans with LTVs over 80%, we've sold a significant amount of the holdings with LTVs between 80% and 95% versus adding some higher LTV pools.

This is important because the higher LTV loans will remain very difficult to refinance even against the backdrop of an improving housing market.

I would also like to highlight that our prepayment speeds remain well-contained post-QE3 coming in at 11 CPR as of January and slowing to 10 CPR for the most recent release this past Wednesday evening.

It's also important to recognize that these speeds would be even lower if our TBA position was on balance sheet since it's largely comprised of low coupon, 30-year 3%s and 15-year 2.5%s.

Turning to slide 10, we have the familiar graph comparing prepayment performance of several types of 30-year 4%s versus more generic MBS.

We did further break out the 80% to 90% LTV HARP subset to show how the lower LTV pools have been ramping up.

And as I mentioned earlier, we have been reducing our exposure to some of the lower LTV HARP categories in both 3.5%s and 4%s.

I am not going to spend much time on this slide, but you can clearly see that prepayment speeds remain contained on lower loan balance and higher LTV HARP securities and we expect that this will continue.

Now I will turn the call over to Peter to discuss our funding and hedging activities.

Thanks, Chris.

Today, I will briefly review our financing and hedging activities during the fourth quarter.

I will start with our financing summary on slide 11.

Our repo funding costs totaled 51 basis points at quarter-end, a slight increase from 46 basis points the previous quarter.

The higher funding cost was due to both normal year-end balance sheet pressure, as well as our continuing effort to extend the maturity profile of our repo funding.

The average original maturity of our repo funding increased to 181 days, up from 141 days the previous quarter.

Importantly, we continue to have good access to longer-term repo funding.

By quarter-end, about 7% of our funding extended beyond a year, up from 2% the previous quarter.

Looking ahead, we see encouraging signs in the mortgage repo market that we believe will lead to slightly lower funding rates in 2013.

These factors include a general easing of balance sheet constraints following year-end, the end of Operation Twist, which pressured dealer balance sheets, and finally, the Fed's ongoing purchases of mortgage-backed securities.

Turning to slides 11 -- excuse me -- 12 and 13, I will review our hedging activity.

On slide 12, we provide a breakdown of our swap and swaptions portfolio.

Our pay fixed swap portfolio totaled $46.9 billion at quarter-end, a decline of $2 billion from the previous quarter-end.

During the quarter, we terminated $3.7 billion of short maturity swaps that were of little value in terms of their ability to offset fluctuations in the market value of our assets.

We replaced these swaps with $1.7 billion of new pay fixed swaps, which had an average maturity of 8.7 years, considerably longer than the one and two-year maturity swaps that we terminated.

We also significantly increased the size of our swaption portfolio during the quarter.

At quarter-end, our swaption portfolio totaled $14.5 billion, an increase of 70% or $6 billion since September 30.

The swaptions we purchased had an average option term of 1.7 years at an average underlying swap term of almost eight years.

Largely as a result of QE3, the price of these options was extremely low by historical standards, making it an attractive time to add to our portfolio and to purchase the protection for a longer period of time.

Since year-end, prices have remained low and we have grown our swaption portfolio to its current size of just over $21 billion.

On slide 13, we've summarized our treasury and TBA positions.

At quarter-end, our treasury position totaled $11.8 billion, up from $7.3 billion the previous quarter.

Finally, as Chris mentioned, our TBA position was long, $12.9 billion at quarter-end.

Taken together, our hedge position, including the effect of our long TBAs, covered 80% of our liabilities, up from 76% the previous quarter.

Lastly, I will review our duration gap on slide 14.

Our duration gap at year-end was negative 0.2 years.

During the quarter, the duration of our assets increased to 3 years from 2.3 years the previous quarter as prices fell and mortgage yields rose.

Mitigating some of this extension, the aggregate duration of our liabilities and hedges increased to 3.2 years from 2.9 years the previous quarter.

Given the generally low level of interest rates, improving economic fundamentals, and the extension risk inherent in mortgage assets, we believe our current hedging activities are prudent from an interest rate risk management perspective.

Looking ahead, we will continue to actively manage our hedge portfolio, adjusting both the size and composition as market conditions warrant.

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

Thanks, Peter.

Before I turn the call over to questions, let me just quickly say a few things about the business economic slide on page 15.

First and most importantly, given the earlier discussion about our off-balance-sheet TBA positions, this slide is considerably less instructive than it has been in the past.

The effective asset yield, the cost of funds, and our leverage, which drive most of the calculations, ignore any impact from the TBAs.

Additionally, as we have mentioned consistently in the past, the cost of funds number excludes the effect of our swaption and treasury positions as well.

So with market conditions and our current investment strategies remain in place for an extended period of time, we will look at revamping this page to make it more fulsome.

But for now, we thought it was preferable to remain consistent with the prior quarters.

So with that, let me ask the operator to open up the lines for questions.

(Operator Instructions) Joel Houck, Wells Fargo.

Thank you, good morning.

I guess the first question is related to the spread differential from the TBA positions.

At the end of the quarter, it's 22 basis points.

How should we, for modeling purposes, think about that in 2013?

I think you also made comments that you believe the strategy is sustainable; however, the relative advantage has also improved.

So is it one of those things where 20 to 22 basis points is kind of a starting point and we build from there?

How should we think about it?

Look, Joel, it's a good question and obviously relevant.

I think what you are referring to -- are you referring to the difference in the kind of net interest -- quarter-end net interest spread?

Yes, the 139 versus the 161 including the TBA?

Right, right.

So, look, what's going to happen there is it's going to depend on a number of variables, right?

You obviously alluded to one, which is how attractive is the dollar roll financing assuming the dollar roll position remains the same size, right?

The other thing, the other key variable is going to be just how big is the TBA position, let's say, relative to the on-balance sheet position, and let's be practical.

Both of those are going to be -- could vary meaningfully during the course of the year.

But what we would say, and I just want to reiterate what Chris mentioned, is that we do believe when it comes to the dollar roll situation -- dollar rolls have been special in the past and plenty of times you look at that and you ignore them, but what's fundamentally different here is the driver is very straightforward.

It's QE3 and the Fed is having a very material impact.

The dollar roll levels could remain special even after that program is done given the amount that the Fed is purchasing and it is going to be absorbing in the lowest coupons.

So we feel that those -- that these favorable financing levels are really likely to be in place.

Now they are going to bounce around.

Could they be negative 20, could they be zero?

Could they be negative 50, where they are right now or could they go to negative 70?

They could be in all of those places, but I think we are pretty confident that if you look at them over the course of this year, they are going to be very special using the term that people use in the market and the financing is going to be attractive.

So I would say, look, you always have noise when you model anything.

We don't know where our repo rates are going to be and there is variability there.

There's clearly more variability in the case of where the dollar roll levels are going to be, but we do believe they are likely to be again well through repo rates for the entire year.

Okay, that's helpful.

And if I could, just one more and then I'll jump off.

I can't help but notice the substantial increase in the swaptions portfolio, almost triple currently from where it was at 9/30.

That's -- you guys have always been obviously -- use swaptions where others haven't.

But is this -- how should we look at this?

Is it a huge signaling effect that you think that is a substantial risk that rates could move materially higher over the course of the year?

How should we think about that?

Joel, this is Peter.

No, I wouldn't look at it as any sort of signal.

It's consistent with the way we've approached our portfolio in the past.

It's just a combination of what hedges do we like in what environments?

In this particular environment, we are particularly attracted to swaptions given the cost profile of the options.

A real benefit from the Fed is that they have reduced interest rate volatility to the point where options that we are buying today are at historically low prices.

So we just think it's a really good time to buy options.

And on top of that, we continue to point out that there is a lot of extension risk in mortgages and we see some improving fundamentals in the economy, so rates have backed up some, but we certainly want to be prudent about hedging for higher interest rates and we think options would be a valuable tool if rates back up anything meaningfully.

Okay, that's helpful.

Thank you very much.

Bill Carcache, Nomura.

Thanks very much.

Good morning, guys.

So it sounds like the opportunity that the dollar rolls are providing is certainly very attractive.

Just try to understand, Gary, whether it's reasonable to expect a scenario where dollar rolls can kind of help you carry you through this kind of period of where artificially -- spreads are being kept artificially tight up until the point where the Fed exits.

Let's say then, at that point, maybe the 10-year gets back up to kind of, just to pick a range, call it 2.25, 2.50 level and now does that get us back into potentially kind of a Goldilocks environment that you've talked about in the past where you could potentially -- there's a reasonable scenario where you kind of get through all of this without having to actually cut the dividend?

I guess could you just talk about that thought process and I guess talk about the reasonableness of something like that being possible?

Look, our future performance is obviously going to be a function of the market conditions and how they unfold.

What we would say to your first point is -- and really just to reiterate what Chris said in his comments, mortgages have tightened 10 to 20 basis points depending on where you look at how you want to -- how you evaluate it.

But the financing opportunities have improved if you go to the TBA market more than 50 basis points across some of the coupons.

And so when you look at that, to your point about kind of navigating through QE3, if QE3 lasts a long time, mortgage spreads will tighten.

It will elevate prepayments.

It will keep -- it would generally keep -- hurt your returns, but, on the other hand, the financing opportunities should remain in place longer too, in which case you have lost -- if they were to last for 10 years, you would have lost 10 or 20 basis points, but made back 50 in financing.

That's unlikely, but to your point, over the -- during the period of QE3, there is a very good offset, and you could argue even kind of an improvement if these dynamics stay in place.

What we have to navigate and what we have to keep in mind is -- again, it goes back to the question about hedging.

We are cognizant that there are two possible scenarios over the course of 2013 or over the next six to nine months, let's say.

One is that the economy doesn't really pick up.

This is another kind of false start and the Fed stays with QE3 well into 2014.

That's a very, very realistic possibility.

We do need to be cognizant that there's a potential for that.

And then the other one goes to the hedging side of it, which is there's also a possibility that the economy does continue to pick up momentum.

It picks up significant enough momentum that, toward the end of 2013, QE3 is a thing of the past.

We have got to be positioned to be able to navigate both of those positions -- those outcomes, and I think that's what you're seeing us do.

Okay, thank you.

That's really helpful.

And switching gears, can you share your thoughts on the mortgage refi legislation that Boxer and Menendez introduced yesterday?

It seems like some of the controversial provisions were dropped last year and they remain out of this version, but I was just hoping you could share any thoughts there.

Really nothing new.

I think we've talked a lot about it.

We really just don't view policy risk as being a material issue at this point.

Look, there's always risk on that front, but we would say again that when you sit there and think about the risks, either on the prepayment front or kind of in aggregate, policy risk at this point is pretty low down on our kind of radar screen.

Okay, thanks very much.

Douglas Harter, Credit Suisse.

Thanks.

I was hoping you could talk about sort of how you get comfortable with the TBA position.

Should we kind of go down the scenario where the economy doesn't improve, rates fall, and prepays pick back up?

Hey, great question.

What I would say there is that is something that's always led us in the past to want specified collateral to make sure we are protected in that environment.

But in the environment where rates stay very low or go lower from here, we are highly confident that QE3 will stay in place for a longer period of time, which will mean these financing differences will stay in place and those coupons will be extremely well-supported and perform better than they theoretically should given their prepayment protection.

So this is exactly what, in a sense, our investors are paying us for, which is to figure out the differences between this environment and last year's environment and the year before that.

And so in our minds, when we think it through, the equation and the trade-offs are different this time and they are different because of QE3 and the communication we have had from the Fed about how long it will stay in place.

So we see in those lowest coupons we see very, very few scenarios where in 3% coupons where specified prepayment protection will outperform generic mortgages.

Great.

And then I guess just staying on the TBAs, how do you think about sort of the sizing of that position relative to on-balance sheet, if this is more attractive, why not bigger -- how do you think about the size constraints?

That's a good question and the short answer is there's plenty of room for it to increase over where it was at the -- we will say at the end of the year.

We also have to keep in mind, while going back to your first question, there are scenarios where we might want to go back to specified pools.

Again, we are not sure in those coupons, but maybe in other coupons and so you want to have a balance between taking advantage of these opportunities, but also being able to have other positions and not be in a situation where you have to do really big kind of allocations.

So there definitely is much more capacity in the dollar roll market, our positions are a tiny fraction of what's going on there.

So there's considerably more capacity, but I don't think you should sit there and expect us to have more than half of our positioned dollar roll even if the numbers were attractive.

Great.

Thanks, Gary.

Steve DeLaney, JMP Securities.

Thanks.

Good morning, Gary.

There are two things I noticed in the press release I would just like you to probably comment on.

First, I noticed you did use your buyback plan and spent about $80 million and acquired the stock, it looks like about 92% of year-end book value, which I think is appropriate -- certainly an appropriate use of capital in an uncertain mortgage market.

The other thing I noticed, and harking back to last year's fourth-quarter press release when you were in a position to give some guidance as to the 2012 dividend rate, this year, that was a -- you did not do that, so we were somewhat expecting some possible comment there.

Now noting that, you certainly have an ample reserve and UTI, but tying those two points together, it seems to me this year has the potential to be fairly volatile regarding the uncertainty around QE3 and the sort of backing off -- the fallout from any backoff of the Fed on the agency mortgage market and directly then impacting the agency mortgage REITs.

As the Board considers dividend policy, is the idea of having a warchest of excess capital or let's say excess distributable income to have that available for opportunistic stock purchases if we were to get the stocks trading down below 90% of book or something like that?

Is that something that is in the discussion as you weigh whether to pay out a cash yield that's already well above average or whether you hold onto a warchest and reward your shareholders with very accretive buybacks?

Thank you.

Sure, look, I think when -- just first on your first point on its own, we announced a buyback.

We were serious about it and we put a plan in place that was executed on and we would do the same thing in the future if the conditions warranted.

We look at that as we have issued accretive equity in the past and we should buy back our stock if it's at a significant discount to book in most scenarios and I think investors should count on us doing that.

With respect to the dividend and undistributed taxable income, what I would say is there's no one factor with respect to the dividend that management and the Board look at.

The reality is the undistributed taxable income we feel does give us quite a bit of flexibility, but we also have to consider what's going on in the market, what our earnings potential looks like, what our leverage is, and what's happening to book value in particular.

So we look at all of those factors, and when we put all that together, what I would say is what I said in my prepared remarks, which is we didn't feel like any other announcement was necessary at this point, but that's something that we have to continuously reevaluate.

Got it, got it.

Okay, so we will find out about the first-quarter dividend when you normally would announce that in the middle part of March I think is what you're suggesting?

Okay.

Just one, if you could, one follow-up.

Could you give us a little color on sort of market pricing changes since year-end?

We sort of see the generic 3.5%, 4%s down a point or better or more and it seems, especially recently, these payups -- a lot of people -- the market is kind of saying maybe prepaid protection isn't worth as much or is not as necessary as it was a couple months back.

So any color you could give on the market would be helpful.

Payups have come off a little bit since the beginning of the year, but, generally speaking, prepayment risk is still very much a material risk in this rate environment and so specified pools for us, as you can see in our portfolio composition, are going to continue to serve an important purpose in the context of both risk management and returns.

But there has been -- mortgages have weakened somewhat, to Gary's point earlier, and payups have come off a little bit as well.

Just going back to what I said earlier, it's very dependent on the coupons, where you see that value, and I think that our mindset on specifieds is somewhat different than it has been.

And there are places where we feel it's very, very important, you can't be in higher coupons where there is substantial prepayment risk, even irrespective of 25 basis points one or the other in rate.

But on the lower coupons, it's actually hard to pay for that protection.

And so I think what you have is these markets change and we have always stressed to investors that we need to be flexible and whatever we did a year ago or did two years ago or three years ago is not necessarily what is appropriate for the new market.

I think there are parts of the specified market that we're not very interested in and there are other parts we are extremely interested in.

So hopefully that answers --.

Yes, that's helpful and I appreciate the comments.

Thanks, guys.

Mark DeVries, Barclays.

Yes, thanks, I have a follow-up on those questions.

I would be interested to get your perspective on the near and intermediate term outlook for rates and mortgage spreads and how you think your portfolio may perform in that.

I think the short answer is that, given QE3, and given kind of the overall economic environment, we actually -- our base expectation is we are not going to see a huge amount of movement in rates at this point.

But as Peter talked about and as we have always said in the past, we are not going to know before there's going to be -- two weeks before there's a big movement.

So we're going out of our way to make sure we feel like our portfolio is positioned to be able to handle sizable moves in either direction, so that's the starting point.

Do we really think something is going to happen?

Probably not, but again it's our job to position the portfolio to manage through either of those scenarios.

On the mortgage front, the base assumption is, unless something changes, which is the economy really heating up, the Fed is likely to -- the absorption from the Fed's QE3 program is very likely to be driving mortgage spreads tighter from here.

And so there is obviously a chance, if the economy really heats up quickly, that they will exit earlier and mortgage spreads will be under pressure in that environment, but the more likely scenario is that the Fed's absorption right now is so high that mortgage spreads could be materially tighter three to six months from now and we are as worried about being or maybe more worried about kind of being positioned for that environment right now.

Okay, that's helpful.

I also have a follow-up on your comments around your forward settling TBA position.

Is it reasonable for us to assume if dollar rolls trade even more special than they are right now that you would look to size that up?

Yes, I think that that's reasonable.

We have plenty of capacity on that front and what I would just reiterate what Chris said, they are materially more special today or they have been over the last month, let's say, than they were at year-end.

Okay, got it.

Just one last thing.

Are there any complications under the REIT rules that we need to be aware of for participating in the dollar roll market?

No, basically, you do have to obviously keep track of your whole pool percentages and TBA forward positions really don't contribute to the whole pool percentage.

But from the perspective it's good REIT income and -- so there are things you have to keep in mind, but the short answer is there's plenty of capacity to be involved in the dollar roll market.

Okay, got it.

Thanks.

Bose George, KBW.

Good morning, just a couple of follow-ups on the dollar roll market.

First, just on the accounting, when you recognize the gain, is that a marked gain and then it's trued up when you settle the purchase?

I was just curious how that works.

I'm sorry, can you repeat that?

Just when you enter into the transaction, the gain that you are recognizing in any given quarter, is that a mark-to-market gain that's trued up later?

I am just curious what is it -- what's flowing through the income statement?

Sure, so the TBAs are just treated as derivatives and marked to market.

So the TBA position is basically marked to market on a monthly or daily basis and quarterly.

And so you will get a price change and that's what flows through the other income line is that these are basically just fully marked to market.

On the other hand, what we have cut out in this dollar roll estimation is just really the price difference between the current settlement and the future settlement, so it's only what we've singled out in that 29 basis points is -- or $0.29 -- is just the difference between the current month and the future settlement, and it's not a full mark-to-market.

Does that make sense?

Yes, that makes sense, thanks.

Actually then switching to the capital side of doing this, should we think of it from a capital perspective that you are holding on-balance sheet capital for this, so we should now look at your, whatever, pro forma leverage?

Well, absolutely.

I mean, look, it's interesting because not only are you getting better financing, but the haircuts or margin requirements for TBAs are actually lower than putting something on repo.

Plus you don't have the drag in terms of prepayment timing differences, so it's actually much -- being involved from a TBA perspective actually has lower counterparty risk and in a sense a lower capital requirement, but absolutely the reason we discussed, and I want to be very clear about this, the reason we highlighted what we call our at-risk leverage or our leverage including the TBA positions is we don't want people to understate kind of the aggregate exposure.

The only thing I would add to that is just with the exception of the one-month roll of cash flow, which you've reduced that risk, but you have all the same risk exposure that you would ordinarily have to the underlying MBS with an on-balance sheet finance position, spread risk, duration, convexity.

So you should think about it as the effective leverage.

Okay, great.

Thanks a lot.

Daniel Furtado, Jefferies.

Good morning, Gary.

Congratulations on another good or perhaps I should say more special quarter by you and your team.

The first question I had is, in regards to the potential issuance of GSE credit securities, I know there are a lot of unknowns at this point, but would owning these securities be within your charter and if so, do you imagine you would participate in this program in the AGNC vehicle?

The short answer is we don't believe that we would participate in purchasing at this stage of the game in AGNC purchasing credit tranches from -- or kind of a security that has kind of the embedded credit exposure from the GSEs.

This vehicle, AGNC, that's not kind of in its -- it's not in its wheelhouse.

Understood, thank you.

My second question is another hypothetical.

In looking further out to the Fed exit from their MBS purchase programs, one of my concerns is that most mortgage REITs are predominately hedged by interest rate derivatives.

However, the technical selling pressure from a Fed exit of MBS purchase activities could potentially result in a situation where asset prices fell off meaningfully where rates take longer to move commensurately higher, leaving portfolios unhedged for a period of time.

Do you think this is a material concern?

If so, what strategies do you envision hedging against that risk?

I will start and I'll see if Peter wants to add anything, but what I would say is, look, mortgages are going to outperform hedges at some point in time.

They will underperform hedges at some point in time.

I think there are two things to consider with respect to your scenario where interest rates or the expectation is the Fed is going to stop QE.

One is that the Fed is buying both treasuries and mortgages and they are buying a lot of duration, so when it's clear the Fed is going to exit the mortgage market, you're likely to see a pretty sizable move in the rates market as well.

So I don't think they are likely to be uncoupled.

That being said, mortgages still could fall a lot quicker than treasuries and swaps, and the way we are trying to deal with that potential is by at least being, as we say, much closer to fully hedged.

So you are always exposed to spread widening and that's the nature of our business.

We really recognize we can't lay that off, right, but we can lay off the component that is correlated with rates and we can manage our leverage such that we can handle very large spread underperformance by mortgages and that's really the position that we are in.

But I do want to overlay one thing and what we have learned in the mortgage market from actual experience is that when the Fed exits and when they exited QE1, it's -- a lot of the impact is not their flow.

It's not what they are buying today, it's how much they have absorbed and the Fed has absorbed an extremely large percentage of the lowest coupon, longest mortgages.

So I think there's another world or another reality, which is after a period of a few weeks of kind of dislocations, you actually could see mortgages perform very well as the Fed has absorbed so much of kind of the incremental duration that's in the mortgage market.

I would just add to it, I think your question is a good one, and you're right, rates may or may not keep pace in the environment that you described, but what we feel pretty confident about is that the interest rate volatility in the marketplace or the market's expectation for future interest rate volatility in that environment will certainly increase and that's one of the reasons why we like option-based derivatives because those derivatives will gain value as the market starts to anticipate a lot of interest rate volatility.

And I think that is going to be the case when the market starts to price in an exiting of the Fed.

So we think the swaption portfolio will gain incremental value in that environment relative to swaps.

Great, thank you for the insight.

I really appreciate it.

Good luck again this quarter.

Arren Cyganovich, Evercore.

Thanks.

Just taking a really basic view of the TBA positions, given that they are forward settling transactions, will these 3%, $13 billion 3% eventually make their way to the balance sheet or is there some sort of hedge position that actually treats it more like an arbitrage?

They could.

Certainly we can choose to take delivery and have them enter our balance sheet and we have plenty of repo capacity to then turn them to on-balance sheet assets.

We could also swap them for other coupons, and so the reality is they could end up on the balance sheet.

They certainly don't have to in their current form or in just TBA or generic form, but that's sort of -- I think you have a lot of flexibility within the mortgage market to adjust things over time.

You can again change coupon.

You can change -- you can upgrade it to specifieds if we wanted to in the future and so it's hard to say exactly what will happen.

So is it more so than more of a kind of arbitrage on taking a long and a short position at the same time and you are just getting that spread there over that specific period?

What's the risk in that trade I guess is what I'm thinking about?

Well, what I want to be clear is that we are net long.

It is that we've purchased these assets.

That's why you should look at the TBA position as increasing our effective or at-risk leverage, as we mentioned earlier, but the financing piece is what you should think about.

You should think about this as we could choose to take in those 3% coupon mortgages and put them on repo at 40 basis points and earn a spread like the one in the example or alternatively we could delay taking delivery by rolling them out continuously or for a certain period of time and get the equivalent of financing them at minus 20 to 50 basis points kind of over the last few months.

What we are saying is that it clearly makes sense to go that route if you believe that those roll levels kind of will persist for any period time.

And so that's the way you really should think about the dollar roll piece of it is you have exposure to the mortgage market.

We would have exposure to mortgage market anyway and you really should look at it as deciding which way you finance it and because of the Fed's outsized involvement in the market and their targeted purchases and specific coupons, there are technical factors that should persist.

That's helpful.

I guess in terms of the Fed's activity, once they decide to stop purchases, how long of a kind of delayed impact?

It sounded like there would be some sort of a lag effect in the TBA market.

How long would that last?

It's hard to say.

It actually could last a while.

It could be a month or two, but, look, I think the important thing -- that's around the financing piece.

The bigger question in that environment will be what's happening to the price of the underlying and that goes back to the discussion on hedging and spread risk and so forth.

So the financing has a decent chance of hanging around for a little while after that, but again the bigger issue in our minds is how we manage the price risk of the underlying in that environment.

Okay, and then just coming back on the maybe accounting of this and the tax, is this drop income considered taxable income for a dividend distribution?

Yes, so it is considered taxable income.

It is good REIT income and essentially it's recognized at the time when you settle a transaction.

So if you would originally purchase something for February, you get to February and you've rolled that to March, then you are recognizing it in the month that it would have -- the original transaction would have settled.

And so actually it ends up within a month or so or a couple months being all realized taxable income.

Okay, and then lastly, the example that you used in slide 8 has a CPR for the assumed life of 7% for those 3% at 30 years.

That seems pretty low.

I guess that's just prices and assumption of extension risk given where rates are today?

Yes, you have to remember that these are 30-year 3% -- exactly, these are 30-year 3% passthroughs with an average coupon or note rate of around probably 360, so the 7 CPR just is based off of our model to a forward curve.

So there's not a lot of -- to your point, in today's rate environment, there's not a lot of incentive for these borrowers to refinance.

Great, thank you.

Ken Bruce, Bank of America.

Thank you, good morning.

First, thank you for the discussion around the TBA activity.

It's been very helpful and your disclosure is always excellent, so I look forward to the enhancements there.

I guess one follow-on question, just as we hear a lot of discussion around some of the changing collateral requirements, does that impact the dollar roll market or is that impacted by the dollar roll market?

Could you discuss those assets please?

We don't anticipate any changes to kind of collateralization, any material changes to collateralization of the forward TBA positions.

Again, their margined as it is right now and those margins, and it's on basically the equivalent of an exchange, and so what ends up happening is that could those margin requirements go up by a percent from 3% to 4% or something like that, yes, but realistically they are still noticeably below where repo haircuts are, for a couple month repo and again, you don't have that prepayment delay, which essentially increases your effective repo haircut.

So there's plenty of room if they were to go up, but we don't envision any kind of material changes there.

Okay, and just maybe you could -- maybe you said it earlier and I just missed it, but as you look at what the changing hedge requirements would be because of a larger TBA or a larger forward purchase part of your portfolio, how would you envision that?

Would that be more through swaptions or how would you just manage the different risk of a TBA, an ultimate TBA versus your traditional spec pool investments?

Well, it's like Chris said and Gary said earlier, we view the TBA position as having the same risk profile from an interest rate risk position, whether it be on-balance sheet or off-balance sheet, so we treat the two similarly from a hedge perspective and we will continuously and dynamically change the way we hedge the portfolio in terms of the hedge composition really just based on market conditions and the price of different options and hedge vehicles.

So we look at them similarly and we look at it comprehensively from an interest rate risk management perspective.

Okay, lastly, there's been some recent chatter about trying to have DeMarco replaced yet again.

Any thoughts around that, notwithstanding your earlier comments about just the general rate risk that you see?

I think, look, it is certainly a possibility.

I think the odds of him being replaced have probably come down versus where they were a few months ago, but I don't think we have any great insight as to the odds.

What I would just say is we think that it's really not something that we view as being that big of a risk to the prepayment landscape and certainly to our portfolio.

Again, the main things that are kind of being discussed wouldn't have that material of an impact.

It really wouldn't have a material impact on the performance of agency securities, things like the principal write-downs or more delinquent loans that have already been pulled out of pools for the most part.

So there's really not a lot of risk on that front and again, we really don't have the HARP 2.0 exposure, so things like the Menendez bill, if they came through, really wouldn't impact our portfolio.

So the bottom line is irrespective of -- we would like to see him remain in office because I think he's doing a really good job in a very difficult position, but from the perspective of managing the business, we really don't see it as being a huge picture issue.

Okay, well, that is it and for what it's worth, I thought your discussion around the impact of the TBAs on your traditional metrics was very helpful, so thank you.

Thank you.

I appreciate that.

Stephen Laws, Deutsche Bank.

Congratulations on the 30% plus economic return last year and I appreciate the color on the dividend outlook, as well as the TBAs.

I guess I just want to make sure I get the -- have grasped the different comments on the TBAs and have these correct as most of my other questions have already been addressed.

But to recap, it was not -- doesn't qualify against the whole loan pool.

It is good REIT income, lower haircuts and margin requirements, it does qualify as taxable income when it's settled if rolled forward and otherwise, if you take delivery on the balance sheet, is it then simply an unrealized positive mark and not considered taxable distributable income until that gain is realized?

Yes, that's correct.

If you end up taking delivery, then you're going to have an on-balance sheet asset with a basis.

Right, so until you realize that gain, it would not fall into the taxable income.

Outside of those things, is there anything else we need to consider from the income or asset or REIT requirements that might be of note?

Not that I can think of.

Okay, great.

Thanks for letting me on the call.

I appreciate it.

We have time for one more question and it will come from Chris Donat of Sandler O'Neill.

Hi, thanks for taking my question.

Just one last one on the roll market here.

Can you talk a little bit about the competitive landscape and the barriers to entry of others getting in because I would think whether hedge funds or other mortgage REITs or other players in the market see the kind of returns and at the risk you're doing it at that that might sort of take away some of the profit opportunity.

But could you just give us some color there, especially given the size of the market?

The size of the market is really huge in the coupons that we are participating in, especially like the main one, 30-year 3%s.

Look, no one knows the specific [holding] interest in the coupons that get rolled.

It's like hundreds of billions of dollars is -- it's clearly in that ZIP code.

So realistically, when you look at the size of the Fed's activities, this is a market that -- it's huge.

You are looking at a monthly issuance in those coupons across Freddie, Fannie, and Ginnie, like $70 billion-ish type or more.

So realistically we think those markets are extremely deep.

Just to give you a couple examples, you can do $5 billion to $10 billion in a day of (inaudible) where you -- we have done a $10 billion roll trade where you've got -- were we've gotten favorable financing in one particular transaction.

There are plenty of times where you can do $5 billion or more in a day without moving the market.

So I think the reality is that this market is much deeper than we are and I think a lot of people are already participating in it, and we are not -- we are focused on making the right decisions and communicating correctly, appropriately with shareholders and not really concerned about the competitive dynamics.

Okay, thanks very much.

We have completed the question-and-answer session.

I would now like to turn the conference over to Gary Kain for concluding remarks.

I just want to thank everyone for participating on the call and we look forward to talking to you next quarter.

Thank you.

The conference has now concluded.

An archive of this presentation will be available on AGNC's website and a telephone recording of this call can be accessed through February 22 by dialing 877-344-7529 using the conference ID 10024358.

Thank you for joining today's call.

You may now disconnect.