AGNC Investment Corp (AGNCN) 2013 Q2 法說會逐字稿

Good morning, and welcome to the American Capital Agency second quarter 2013 shareholders 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 that this event is being recorded.

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

Ms. Wisecarver, please go ahead.

Thank you, Keith, and thank you all for joining American Capital Agency's second quarter 2013 earnings call.

Before we begin, I would 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 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 August 13 by dialing 877-344-7529 or 412-317-0088.

And the conference ID number is 10031377.

To view the slide presentation, turn to our website, www.AGNC.com, and click on Q2 2013 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, 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.

Thinks, Katie, and good morning.

The second quarter was a very difficult period for everyone in the fixed income market and especially for agency MBS investors, both levered and unlevered.

The performance of mortgage REIT stocks has been even worse and in my opinion an overreaction to the current challenges we face.

Despite the reduction in earnings power due to a smaller portfolio and the cost of higher hedge ratios, there are some positives that should be considered once volatility subsides and we are comfortable increasing our risk posture.

Mortgages are cheaper today.

The yield curve is steeper.

Prepayments are continuing to slow.

The future supply of MBS is expected to be materially lower.

Our asset composition has evolved and the Fed continues to purchase a significant amount of MBS.

As a matter of fact, if we ignored the path that led us to get to this environment, many in the space would to probably characterize the current investment landscape as very attractive.

In other words, you certainly would not expect price-to-book ratios to be at historical lows.

That being said, I want to be clear that the current volatility is a negative and we take it very seriously.

Experience tell us that in such environments, the prudent course of action is to prioritize risk management over earnings maximization.

And that is the way I would characterize our actions.

Why?

First, the current environment could be a transitional one whereby we may be in the midst of a progression towards higher rates and wider mortgage spreads.

Secondly, given AGNC's current very low price-to-book ratio, investors are implicitly sending a signal that risk management is more important than risk management.

Overall, we believe this environment will be fine for us in the long run but as we have said in the past, management must be able to handle transitions and we do that through disciplined risk management.

On the other hand we are also cognizant of the reality that the current environment may simply be an over reaction on the part of the bond market to a change in expectations regarding Fed behavior and with respect to the duration of QE3.

If this is the case, rates should stabilize or decline from here and mortgage prices and spreads should recover some of the losses we have seen so far in 2013.

Now given the uncertainty as to which path we are on, prudent risk management is paramount.

Now before I discuss the second quarter results, I want to address head on some of the fears that many investors probably have about the current environment.

First and foremost, this is not 2008.

Financing has been readily available and liquidity in the fixed-rate MBS market continues to be very good.

Our interest rate risk is lower than it has been in years, and our exposure to MBS spread widening has also come down materially given the actions that we have taken.

With that as the introduction, let's turn to slide 4 so I can provide additional color around some of the second quarter results before handing the call over to Chris and Peter to discuss how the portfolio has changed in light of the current environment.

The second bullet point shows that our aggregate spread income, inclusive of dollar roll income, totaled $1.15 per common share in Q2 and was comprised of $0.66 per share of net spread income plus $0.49 per share of dollar roll income.

This figure includes approximately $0.14 per share in catch-up amortization benefit related to lower projected prepayment speeds.

Our taxable income totaled $1.04 per spare their share inclusive of realized gains and losses on asset sales and of settled TBA sales or dollar roll activity.

Since this figure is essential equal to the $1.05 common share dividend that we paid for Q2, our undistributed taxable income remained largely unchanged at $1.07 per share.

However, it is important to point out that our taxable income in the third quarter is currently biased materially lower given some of the actions we took in both the second quarter and third quarters which will be realized for taxable earnings purposes in Q3.

As we will discuss in a few minutes, the poor price performance of both generic and to a lesser extent specified MBS, outweighed the gains from our hedges and led to the 11.8% decline in our book value during the second quarter.

Our economic or mark-to-market return for the quarter was negative 8.2%.

Economic returns include our dividend plus the change in our book value.

Turning to slide 5, you can see that the market value of our investment portfolio, inclusive of TBA positions, decreased around $11 billion to $91.7 billion during the quarter.

As of June 30, our leverage was 8.5 times but has since declined to approximately 8 times as of July 26.

Turning to slide 6, I want to highlight what happens in the markets during the second quarter.

As you can see on the top, we saw a sizable sell-off in both treasury and swap rates.

While the quarter-over-quarter moves were large with yields on the 5- and 10-year treasuries increasing by over 60 basis points, the intra-quarter move in the tenure from the end of April to late June was closer to 100 basis points.

The MBS market, which has staged a meaningful recovery in April, traded very poorly in May and June as changing investor perceptions around the future of monetary policy combined with stronger economic data to push interest rates higher.

MBS had to contend with considerably higher volatility, convexity hedging needs, money manager redemptions, and concerns around changes to bank capital requirements, all of which contributed to the under-performance of MBS versus both swaps and treasuries.

As you can see on the bottom left chart, the lowest coupon 30-year fixed-rate mortgages dropped significantly in price with the 3% coupon dropping almost 5.5 points and the 3.5% coupon dropping a little over 4 points.

15-year MBS were also not immune to the negative market dynamics.

As shown on the bottom right, the 15-year 2.5% coupon dropped over 3.75 points.

Now surprisingly, higher coupon 15-year MBS such as 3.5%s, 4%s, and 4.5%s, which AGNC had a material position in, also dropped between 1.5 and 2 points despite the fact these securities are usually the least impacted by changes in interest rates.

As Chris will discuss shortly, we viewed this weakness as an opportunity to purchase these very defensive securities at attractive valuations during Q2 and early in Q3.

Turning to slide 7, I want to quickly review what happened to pay-ups during the quarter.

Given the large declines in MBS prices, pay-ups were sharply lower across the board.

However, the moves in pay-ups on lower coupons, 3%s and 3.5%s, were much more consistent with the interest rate moves this quarter than what we saw last quarter.

For example, the pay-up on lower loan balance 30-year 3.5%s dropped 69 basis points in price versus 73 basis points last quarter.

More importantly, however, 10-year rates moved over 60 basis points this quarter versus only 10 last quarter.

In contrast, the declines in pay-ups on 30-year 4% coupon were considerably larger this quarter versus what we saw last quarter with a lot of this move occurring in late June.

More specifically, the pay-up on lower loan balance 4%s dropped almost 2.5 points this quarter versus only 91 basis points last quarter.

Again, some of this larger move was clearly explained by the magnitude of the rate moves this quarter, but we certainly did not expect the aggregate price declines on specified 4%s to exceed that of specified 3.5%s.

And this is circled on the far left -- far right part of the table.

On an additional note, while not displayed on this slide, the pay-ups on higher coupon 15-years also performed poorly during the quarter.

The greater correlation between rates and pay-ups implies that our hedges were somewhat more effective this quarter at offsetting the pay-up declines compared to Q1.

More importantly, given current pay-up levels and our new portfolio composition, our pay-up exposure is no longer significant.

To be more precise, our average pay-up as of June 30 was only 0.25 of a point when measured across our TBA deliverable pass-through portfolio.

Hence, if pay-up were to decline to zero, our total hit to book value would be less than 2.5%.

With that, let me turn the call over to Chris to discuss the current portfolio.

Thanks, Gary.

During the quarter, we took meaningful steps to further adjust the composition of the investment portfolio for the current environment.

On slide 8, you'll notice that our total MBS portfolio was $92 billion as of June 30.

Net portfolio sales were approximately $5 billion, however this understates portfolio repositioning during the quarter as many of our sales were replaced with other securities better suited to the current environment.

There were three main themes to the portfolio composition changes we made during the quarter.

First, we materially reduced exposure to our lowest coupon 30-year MBS, more specifically, our exposure to 30- year 3%s declined by approximately $8 billion to 14% of the portfolio as of June 30, down from 20% at the end of the first quarter.

Secondly, we reduced our pay-up risk throughout the quarter via sales of approximately $6 billion in 4% coupon specified pools.

These sales were executed throughout the quarter at an average pay-up of close to two points.

Lastly, we increased the percentage of 15-year MBS in the portfolio by adding 15-year 3% and 3.5% pass-throughs.

These purchases, in combination with sales of lower coupon 30-years, increased our 15-year weighting to 42% as of June 30, up from 34% at the end of the first quarter.

As Gary mentioned, we have chosen to position the portfolio more defensively given elevated levels of spread in interest rate volatility.

We feel that this positioning gives us flexibility to proactively increase risk when we feel it is appropriate.

Before we turn to the next slide, I want to point out that while many are not concerned about prepayment risk in the current environment, that can change quickly and our portfolio composition remains well-balanced to perform at a low rate environment as well.

To this point, we still have a significant percent of our portfolio in specified MBS that could appreciate materially if rates decline significantly.

On the next slide, I want to spend a few minutes on the importance of asset composition in the context of market risk.

On slide 9, we have an example of two hypothetical positions and their exposure to a 25-basis-point widening in spreads.

Spread duration is the price sensitivity of a bond to a change in spread over other benchmark interest rates such as treasury and swap rates.

Spread duration differs from interest-rate duration, which measures the sensitivity of a bond to changes in benchmark interest rates assuming spreads remain constant.

As we have discussed in the past, both interest-rate duration and spread duration can affect the price of a fixed income asset.

On this slide, we compare the spread sensitivity of the 15-year MBS to the spread sensitivity of a 30-year MBS.

The analysis compares a 15-year 3% pass-through with a spread duration of 4.6 years to a 30-year 3.5% pass-through with a spread duration of 6.8%.

The lower spread duration of 15-year MBS means that price is less sensitive to a change in spreads, all else equal.

In the table, we show the net asset value sensitivity of the two hypothetical portfolios.

The NAV sensitivity is calculated by multiplying the spread duration by the change in spread, in this example 25 basis points, and then multiplying by the amounts of leverage assumed.

As you can see in the table, a 30-year MBS portfolio operating with six times leverage has roughly the same price exposure to 25 basis points of spread widening as a 15-year position operating at nine times leverage.

This is due to the spread duration of 15-year MBS being considerably less than the spread duration of 30-year MBS given a shorter amortization schedule.

It's also worth highlighting the fact that unlike other shorter duration MBS, such as ARMs and some REMIC securities, 15-year MBS trade in a liquid TBA market where prices are readily observable and also often benefit from favorable roll financing.

For all of these reasons in addition to attractive returns and favorable supply and demand technicals, we believe the greater share 15-year MBS in our current portfolio is appropriate in the current environment.

As I mentioned earlier, 15-year MBS comprised approximately 42% of the portfolio as of the end of the second quarter.

Now I'd like to turn the call over to Peter to discuss funding and risk management.

Thanks, Chris.

Today, I'll brief review our financing and hedging activity during the quarter.

I'll begin with our financing summary on slide 10.

Despite very volatile market conditions, our access to attractive funding remain uninterrupted throughout the quarter.

Moreover, despite the sharp rise in interest rates and the underperformance of mortgage-backed securities, our average repo haircut remained unchanged from the prior quarter at 4.8%.

Our repo balance increased to $70 billion at quarter-end and carried with it an average cost of 45 basis points, a 2-basis-point improvement from the prior quarter.

Our average original maturity remained relatively unchanged at just over six months.

Turning to slide 11, I'll briefly review our hedging activity during the quarter.

In aggregate, our hedge portfolio covered just over 100% of our repo and TBA positions.

With our hedge ratio at an all-time high of about 100%, we are obviously operating with a very defensive position.

As you can see from the table on the top right, our hedge portfolio totaled $88.7 billion at quarter-end, up about $1 billion from the previous quarter.

Even though our balance increased only slightly, the aggregate duration of our hedge portfolio increased significantly during the quarter.

During the quarter, we terminated shorter-term swaps and added longer-term swaps.

In addition, our swaption portfolio naturally extended as interest rates increased.

As a result, the overall duration of our hedge portfolio increased substantially to 4.9 years at quarter-end, up from 4.2 years the previous quarter.

Our hedge portfolio is designed to protect our book value against larger moves in interest rates.

Despite the sizable starting position as interest rates increased during the quarter, and the duration of our assets lengthened, we took frequent actions to increase our protection against further rate moves.

Our hedge portfolio performed as intended and offset a significant amount of the price decline in our assets.

In total, the market value of our hedge portfolio gained $2.2 billion in the second quarter, or about $5.50 per share.

Now let's look at our duration gap sensitivity on slide 12.

Given the actions we took with regard to both our assets and our hedges, our future extensure risk is now very limited.

As of quarter-end, our asset portfolio had an average duration of 5.6 years, a full two-year increase from what we disclosed as of April 30 on our Q1 call.

As the table shows, if interest rates were to increase another 200 basis points, our asset duration will likely extend less than a year to 6.4 years.

Thus the extension we experienced, we will experience over the next 200 basis points of rate change will be about half of the duration extension we experienced in May and June.

In the middle of the table, we show the offsetting duration benefit of our hedge portfolio.

As a reminder, the duration of our hedges in this table are expressed in units relative to our asset portfolio to make it easy to sum each column.

The duration of our liabilities, swaps, and treasury hedges was four years at quarter-end, up from 3.5 years as of April 30.

This increase was due to the fact that we added longer-term swaps and treasury hedges during the quarter and due to the decreased size of our asset portfolio.

At the bottom of the table we show the duration of our swaption portfolio.

At quarter-end, our swaption portfolio hedged one year of our asset duration, up significantly from the 0.3 years we reported as of April 30.

As interest rates rise, our swaptions will take on the interest-rate characteristics of the predominantly 7- to 15-year pay fixed swaps that underlie them.

The duration of our swaption portfolio will naturally increase as interest rates rise and as these options move into the money.

Said another way, our swaption portfolio will hedge a greater share of our asset duration as interest rates increase.

The protection provided by our swaption portfolio is easily seen in the up 200 basis point rate scenario.

In that scenario, our asset duration will likely extend about 0.8 years to 6.4 years.

At the same time, our swaption portfolio will likely extend about 0.7 years to 1.7 years in total, essentially offsetting all of the asset duration extension we will experience in that scenario.

As a result, our net duration gap remains relatively flat and very low at only 0.7 years, even in this extreme interest rate scenario and assuming no rebalancing actions.

Taken together, our actions with regard to our assets and our hedges have materially reduced our duration gap sensitivity.

Moreover, the stability of our duration gap in a rising rate scenario shows that the current size and composition of our hedge portfolio is sufficient to offset most of the remaining asset extension risk we face should rates increase further.

It also allows us to operate more safely with a larger duration gap at some point in the future.

The benefit of a low and stable duration gap can be seen on slide 13.

In the table at the top, we show the estimated change in our net asset value for a 50- and 100-basis point change in interest rates.

For example, if interest rates increased 100 basis points and no rebalancing actions are taken, our models indicate that we would likely suffer about a 6% loss to our book value assuming no change in mortgage spreads.

This estimated loss is down about 50% from 1Q due to the actions we took during the second quarter.

Finally, given the volatility of MBS spreads in the current environment, we thought it was important to provide you additional information regarding the sensitivity of our net asset value to changes in spreads between our assets and our hedges.

To that end, in the bottom table, we show you the estimated change in our net asset value for our 10- and 25-basis-point move in MBS spreads.

It is important to note that our sensitivity to mortgage spread changes does vary with interest rates, increasing as interest rates rise and decreasing as interest rates fall.

That said, the change in the spread duration is gradual, so you can use these numbers to estimate the amount impact of larger or smaller spread moves.

Finally, all of these sensitivities are model estimates and could differ materially from our actual results.

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

Thanks, Peter.

If you turn to page 14, I want to quickly summarize what you have heard from the team about how the portfolio has been repositioned for the current market.

First, asset composition has changed to be better suited to the evolving interest rate landscape.

Second, pay-up risk is now minimal as the weighted average pay-up on our portfolio as of the end of the quarter was less than 0.25 of a point.

As I said earlier, if the these pay-ups go to zero, the hit to book value would be less than 2.5%.

Third, we reduced both the interest rate and spread sensitivity of our assets.

We did this by reducing our exposure to the lowest coupon 30-year fixed-rate mortgages and materially increasing our exposure to 15-year MBS.

We have also stressed the importance of asset selection on every call over the past five years, and that is as true today as it has ever been.

Fourth, given the increased rate and basis volatility in today's market, we have kept our overall leverage in check.

As such, our leverage at quarter-end was only slightly higher than our average leverage over the past several years at 8.5 times and our current leverage is around 8 times.

While managing duration and convexity risk is considerably more important than having a low leverage number when it comes to managing risk, we have to date been hesitant to let leverage increase materially in the current environment.

Lastly, given the volatility we are experiencing, we have taken a defensive posture and we are more hedged today than we have ever been in the past.

The collective rebalancing actions that we took during the quarter have allowed us to keep our current duration gap in check and equally important have materially reduced our exposure to a further rate increases.

So what should investors take away from AGNC's actions this quarter and our current positioning?

That active portfolio management is just as instrumental in the current environment as it was in the generally falling rate environment that we saw over the past five years.

Yes, the asset selection process is different, but no less critical to our performance.

Additionally, we want to reiterate that experience teaches us that there are times when it is imperative to prioritize risk management over maximizing short-term returns, and this is one of those times.

That being said, over time we believe this market will evolve into a very attractive return environment.

It may even turn into a Goldilocks scenario once the current tail risk in the market subsides.

Importantly, we believe that the actions we have taken this quarter position us to take advantage of this future environment should it come to pass, and as importantly, gives us comfort in a week like this with a Fed meeting, a GDP release, and an unemployment report.

Before I open up the call to questions, I want to quickly point out one thing on the next slide that is probably surprising to many people on this call.

It is that despite a very disappointing first half of 2013, economic return or the combination of dividends plus the change in our book value over the past 12 months has actually been positive 3%.

I have heard some market participants point to QE3 and 2013 as evidence that the agency REIT model is flawed.

I think this conclusion is completely off-base as shown by our economic returns.

To the contrary, I believe the current environment proves that the agency REIT model is in fact durable.

In a really bad 12-month period, like we just experienced, where rates increased by close to 100 basis points and MBS prices tanked as a result of a dramatic shift in expectations regarding Fed policy, we still produced positive mark-to-market returns.

This after four years of 30% plus returns.

A more reasonable conclusion is that the model, when combined with active and disciplined risk management, has held up against the backdrop of a very challenging test.

So with that, let me open up the call to questions.

Thank you.

We will now begin the question-and-answer session.

(Operator Instructions)

Joel Houck with Wells Fargo.

Thank you.

The questions related to basis risk that is evident in all these models and you guys -- thank you for at least putting out the disclosure on page 13, but the question is, Gary, maybe you guys have been doing this a long time, talk about where we are in terms of MBS spreads relative to the short history say the last two to three years and then the long history, 10 to 15 years and then if you could maybe handicap where you see basis risk, or actually the inflection points both for wider basis happening as well as a tightening say over the next -- between now and the end of the year.

Sure, that is obviously a key question.

First off, to the first part of your question about where do we stand now, mortgage spreads, when you factor in the options associated or embedded in mortgage securities, are right now not very different from where they were before QE3 was announced back in September of last year.

And they are also not that different from, let's say, an average over the last few years.

But what is interesting about that, now that could tell you a couple of different things.

The way I read that is the Fed has just bought a considerable amount of mortgage securities, over $0.5 trillion worth.

Even in a tapering scenario, they are going to buy probably close to another $0.5 trillion worth of securities.

And against that backdrop, mortgage spreads are -- have not done anything.

And that -- when you combine that with the production of mortgage backed securities or new originations dropping pretty quickly, that sets up a very favorable technical backdrop for the product over the next six months to a year.

And so we do feel that the mortgage basis, so to speak, is well-positioned.

Now to the other part of your question where you talk about the longer run, if you go back before the crisis so to speak in 2008, mortgage spreads were and in particular option-adjusted spreads were considerably tighter than where they are now.

And so when you think about a full long-run scenario, there is no -- there isn't reason in a sense to call into question that view.

Big picture, we are despite our defensive tone on this call and defensive positioning, we still feel good about where the mortgage basis is.

In the short-run though, we recognize that the mortgage basis or mortgage spreads are going to be under more pressure if interest rates rise than they will be if interest rates fall.

And that is the driver of wanting to make sure we are more fully hedged right now because we do feel that there is and clearly has been a correlation between increases in interest rates and mortgage spreads.

That correlation should be very temporary and will go away over time but -- and even to the tune of a couple months, once you stabilize at a rate level, but it does exist in the near-term and so its something we have to keep in mind in our hedging decisions.

Was that helpful?

Yes.

That is extremely helpful.

On a different note Gary, you talked about leverage coming down from 8.5 to 8. Was that driven by more asset sales or is it driven more by recovery and value.

Walk us through what happened between June and now to take leverage down.

No.

That was a result of actions that we have taken in terms of repositioning the portfolio.

Obviously some asset sales and there has also been further repositioning of the portfolio during that period.

But it is driven by actions that we have taken versus recovery and book value.

Okay and lastly, how do you think about liquidity?

It looks like the space on your reports, you've got about almost $5 billion of unpledged assets and maybe $3 billion in cash.

That is about a little over 10% of total value of your assets and roughly almost, I guess it's almost 100% of the book equity.

Is that the right way to think about liquidity?

Is there any need to move it higher or is it just -- how would you characterize the liquidity position of the Company right now?

Hi, Joel.

This is Peter.

That is generally the way we think about it.

We think about our liquidity position as a function of our unencumbered assets and typically we operate with our unencumbered assets which means our percentage of assets relative to our NAV at around 50%.

50% of our NAV is about $5 billion and as you point out, we have been operating in this last quarter with a little greater share of that unencumbered assets in cash, $2 billion to $3 billion in cash is normal for us in this environment, $2 billion of unencumbered.

But it is typically right around 50% or 55% of unencumbered assets in terms of our liquidity.

So one thing, Joel, just at the highest level to add to that is that essentially our liquidity position is no different now than it was a year ago.

Again our leverages is essentially the same.

Despite the moves that we have seen and the impact on book value, there -- we have chosen to put ourselves in a position where our liquidity position is unchanged, and the logic there is that clearly spread risk and overall risk is higher, and against that backdrop we felt like that is the right thing to do.

At some point, we may be willing and we are certainly looking for the opportunity to increase our risk posture, and that could be either the a duration gap or leverage, but right now, just to reiterate, our liquidity position is exactly where it was prior to any of the things we have seen this year.

Great.

Thank you very much.

Douglas Harter with Credit Suisse.

Thanks.

Gary, just to dig into that last comment you made a little bit more, what are some of the signals you would be looking for to increase that risk posture?

I think it is clearly some stability and the combination of the rate market, and in mortgage spreads and which I think we may be approaching realistically.

And some -- and from the perspective of the current environment, it is not a problem with mortgage spreads or returns.

We are not waiting them for to tighten -- I mean to widen 10 more basis points so that we can take leverage up to 9.5.

I think realistically, it is more about the risk return equation right now is being driven in our minds more about the aggregate amount of risk that we think shareholders want to take in this environment.

And I want to reiterate something that I said in the prepared remarks, which I think is very relevant to this topic, which is in a way what the equity markets are telling us which is the REIT space in general is trading for the first time maybe ever outside of a couple brief instances, at a material discount to book and that's an indication that the focus is more on risk management prices or prices are being driven more by risk than return.

We are also cognizant of that trade-off as well.

But big picture I think we want to see the risk equation be more quantifiable and to calm down.

We are comfortable with the return environment at this point.

And then you guys did repurchase some shares during the quarter.

Can you just talk your appetite for continuing to do that as we move into the third quarter?

Yes.

Sure.

We have plenty of capacity in our authorized stock buyback programs.

We are extremely committed to using the stock buyback program when the -- when it meets kind of the combination of two thresholds.

One, when the price-to-book as measured by our best estimate at the time of our book value is at a material enough discount.

Second of all, we obviously have to operate within the typical constraints that any public company has to operate with respect to when you are allowed to be in the market and window periods and stuff like that.

Great, thank you, Gary.

Steven DeLaney with JMP Securities.

Good morning, everyone.

Gary, appreciate the detailed comments about how you guys actively manage the portfolio and hedges and I know you are not at all pleased with a 12% drop in book value but for what it is worth, we had minus 18% if we had simply marked your static March 31 portfolio.

I think you guys should be applauded for the steps you took for the quarter.

Looking -- trying to look a little bit into the TBAs and the dollar rolls, I noticed that the net TBA position dropped from about $26 billion at March 31 to 14.5 and it's not clear whether that's a net figure, are you seeing the dollar roll market now to be less special or less attractive or did you simply increase short TBAs during the quarter as a part of your hedging strategy.

Thanks, Steve.

This is Chris.

While there are a couple drivers between behind reducing our long roll position, while taxable income was one of the considerations, in the case of 15-year we actually took delivery of the roll in part because with the salt in rates and the steepness of the curve, seasoning can actually be worth quite a lot.

Roll implied financing rates, as you mentioned, also cheapened quite a bit relative to where they spent the first five months of the year.

The advantage of roll financing versus repo so that wasn't nearly as compelling as it had been.

And the other factor was that these positions also tended to be our longest spread duration positions and so in the process of managing leverage and spread duration, the position declined during the quarter.

The other thing, Steve, to keep in mind -- go ahead.

I was just going to say is what you are saying is the extreme specialness was in those 3% 30-year coupons that the Fed was gobbling up.

Exactly and 15-year 2.5%s, as well.

But one thing, Steve, I don't want to take away from those comments that dollar rolls aren't going to be useful in the future.

They have migrated to different coupons.

There are definitely higher coupon 15s are more special than they have been in ages.

Or even 15-year 2.5%s were in the past and 30-year 4%s are very favorable roll financing, as well.

There are opportunities in the roll markets but they are in different coupons.

And so embedded in this is a combination of recognizing that the return profile on the lowest coupon has changed as will is the risk posture as, Chris mentioned, and then moving to a position where we have the most flexibility.

Where we can let positions either season in the case of higher coupon 15s or if the roll economics are enough to -- and we feel that they will be around long enough that we can also have that choice, but everything is about flexibility in being able to manage through a variety of different environments at this point.

Okay.

Great.

I appreciate it, and I will leave you guys with an easy one since you tend to get complex questions on this call.

Lastly, I was just curious if you found the Freddie Mac stackers offering the M1 tranche, in particular, to be attractive and appropriate for the AGNC portfolio.

Thanks for the time this morning.

Sure.

We would not view that as an appropriate asset at this point for AGNC.

Thanks.

Mark DeVries from Barclays.

Thanks.

First question, Gary, getting back to your comment about the correlation you've seen between rates and spreads.

I am assuming part of that is due to different market participants having a sell duration as rates rally and in particular maybe market participants who may be more off sides in their duration gap than you are.

Are you seeing any signs that that risk is being mitigated?

That players are taking a more conservative approach to managing their duration gap?

What I would say is I think that you are absolutely right.

Clearly, if you look at any mortgage research reports, convexity hedging, and we mentioned it, is a factor.

Probably the largest factor our servicers who are managing positions, basically aisle-like assets, but hedge funds, REITs, obviously, and even unlevered accounts who are managing duration, banks.

We have seen a number of people want to rebalance their portfolio either from a duration perspective, as we've talked about, a spread perspective, or generally from a leverage perspective.

We have seen all of those types of rebalancing activities and the mortgage market clearly has had to contend with that.

We do believe that the bulk of that is done.

However, there are still a number of positions out there that from our perspective are pretty long interest rates.

And so what I would say is the majority is done, but I think if we were to see another 25 to 50 basis point backup you would certainly see more of it.

And we have to in our calculus, we can't view that as a 5% or 2% likelihood.

We've got to view that as being, again, it is not a 50% likelihood but it is something that is clearly a consideration for us in the market.

Okay, that's helpful.

We really appreciate all the great disclosure around the impact on your duration gap and NAV from movement to rates and spreads but I would appreciate hearing Peter's thoughts about how that map might change if we get a more gradual move in rates where you have more of an opportunity to build a hedge versus the more rapid simultaneous move reflected in those disclosures.

That is clearly a real key from a leverage portfolio perspective is that when you have gradual moves in interest rates, obviously, it is much easier to manage your portfolio, there's ample liquidity in the marketplace and you really can keep your risk in check.

In an environment like that, we can really keep our duration gap given the way our portfolio is structured today at very close to neutral if we so chose.

Obviously, it's the most significant challenge is when rates move very rapidly and you don't have an opportunity but that's really critical as to why we have the amount of options we have in our portfolio and the key message I was trying to communicate earlier at this point, given the composition of our asset portfolio, even if rates were to move suddenly in a very significant way, we really have very little ongoing rebalancing needs because our options will naturally extend and essentially offset almost all of the remaining extension we have in our assets.

This is the environment where we really feel it's important to have our option portfolio.

We always talked about our options giving us a lot of tail risk protection.

We've had a 100 basis point move and they have helped a lot.

They will help even more if rates moved another 100 or 200 basis points from here so it really gives us a lot of protection and makes our life easier from a rebalancing perspective should rates move further.

Okay.

Great, and just one last question.

Gary, during your comments you mentioned how some of the actions you have taken will end up biasing the taxable income lower in the third quarter.

Could you talk about within the context how you think about using that remaining $1.07 of undistributed income would you potentially look to use that maybe at least for a period of time to subsidize a dividend that might be above your core earnings run rate?

Look, I think in this environment what I would say is I think we've been pretty clear about our prioritization is around risk management and then secondly, that this is just an incredibly volatile period.

And so realistically, managing -- we are going to manage the portfolio in a manner that we feel is the best from a longer run, risk return framework.

With respect to the specifics on taxable income, it is clearly very difficult to forecast in a volatile environment.

And whether we're talking about book value, whether we're talking about core earnings, tax earnings, tax warnings, this is a period where I think we are going to be -- we want to just see how things develop and we will go from there.

Again, we feel like this over time could be a very good environment but right now, we are sort of playing defense.

Thank you.

Arren Cyganovich from Evercore.

Thanks.

Just on the last question, the TBA position has unrealized losses of close to $800 million.

Do those have to be realized or are you taking some of those on the portfolio?

How do we think about those embedded losses in the remaining TBA position?

In terms of -- in the TBA position, we can take in TBAs if they at a -- an unrealized loss position and mitigate that and I think so if you notice for this quarter, as an example, and what Chris was talking about, there are certainly some positions that we took in.

Now we could choose to sell out some of those positions and then get -- and then we would end up realizing be taxable hit, but the rolls don't have to -- they're not something that have to turn into a taxable loss.

It is sort of the same as a position -- as if you purchased a security out right.

In that if you choose to sell that security, then you are going to have to realize that any mark-to-market change.

But if you choose to hold that security, then it is not realized.

I don't think the dollar rolls per se are different in character so to speak.

I would just add, we talked about this on the last call.

Certain types of hedges that we have get recognized in a similar way if they are realized in roll like TBA.

For example, if we have TBAs hedged with treasuries and we roll out of the treasury or lift the treasury hedge at the same time we move our TBA position the gains or losses get recognized at the same time from a taxable earnings perspective.

Swaps and swaptions do not but treasury hedges do, and that's one of the reasons why we have operated with a higher treasury hedge ratio in the last quarter or two.

That is helpful, thanks.

In terms of the defensive posture, is this more now generally just a macro view or does QE3 even though it seems like it's mostly priced out of the MBA -- MBS market, is that still waiting on your defensive positioning commentary from the Fed, et cetera, as they look to taper ahead?

I wouldn't say it is a macro you.

We are not -- we are bullish on mortgages.

Generally, we feel that the rate move is probably overdone.

And so the reality is we are just being practical that the moves we are seeing in the market right now are large.

They are not only in the US, they involve currency moves.

We are seeing changes in activities of obviously other central banks, as well.

To your point, a lot of uncertainty about the Fed and how the market is going to react.

And if you go back to some of the other things I mentioned, bank capital, fixed income redemptions, there are a lot of big picture items which realistically we don't feel our shareholders want us to make a big bet on.

And so those other reasons for the defensive posture.

My gut is that what we have seen is an overreaction, and the mortgage basis should do better and rates should stabilize or maybe come down.

But on the other hand, the factors we are dealing with are very big picture issues that are hard to predict, and this is not something we want to draw a line in the sand on.

Great.

Thanks, and then lastly just when we look at the duration estimates that you have in your slide deck, the TBA has a duration of 5.5 years which seems a little bit low since it is mostly, it look's like I would imagine, current coupon 30- and 15-year, and how that relates to the pretty low projected CPRs for the lifetime of around 6% and 5%.

I'm just trying to get those together.

It seems like those don't make a lot of sense to me.

One thing to keep in mind, if you want to look at page 25, I think what you will find is a deeper or kind of -- what we did add this quarter was we added duration estimates by coupon out of our models with any adjustments or whatever, to by coupon, by 15s and 30s.

So if you want to see how those numbers were derived, that's probably the best way to look at it.

If we look at that, our aggregate for 15 years is about 4.5 years and keep in mind there are some positions there with a fair amount of seasoning.

Our aggregate for 30 years was around 6.5 years.

The problem is when you look at the TBA number, there is a lot of netting so there can be long and short positions.

That number could be very biased and it could actually be even outside the range of a typical duration.

It isn't in this case.

I would focus you on page 25 going forward and we will keep doing this in our efforts to be as transparent as we possibly can be.

I want to add to that, on page 25 we also added TBAs to those positions this quarter.

So it includes both on and off balance sheet assets.

Great.

Thank you.

Chris Donat with Sandler O'Neill.

Just wanted to dig a little bit on the portfolio construction as it was on June 30 and then maybe where it is going.

With the mix of the 15 years going up to 42% from 34%, I'm just wondering is there a target you have in mind?

Because It looks like you could in a quarter or two be over 50%, 15 years, if that's where you want it to be, obviously.

Just trying to get your thinking there.

I would say, look, we don't set a specific target for the 15-year versus 30-year.

It's going to be clearly a function of market prices, valuations, market conditions.

What I would say is our percentage of 15 years is higher than the 42%, that it was that at the end of the quarter.

But still well sub-50%.

If I had to look at things today, if prices and valuations are attractive our bias would for be 15 years to increase, given the environment and for all the reasons we talked about.

But it's something that we are very focused on, on valuation and we want assets to provide the best risk-adjusted returns.

And you can hedge 30 years to shorter durations but you recognize that you are taking longer spread duration and so forth so it's a major -- it's going to be a function of valuations over time.

Okay.

And then revisiting numbers you had in the past on the leverage ratio.

In the past, you said 6 to 11 times but the current ratio for the current environment, is that kind of the right way to think about leverage?

That you are not going to be stretching it in the -- assuming we don't have what you talked about before, things that would lead to less or more stability in the interest rate market.

I would say that, look, we continuously reevaluate our risk positions.

And I want to reiterate one thing that is really important.

I mentioned it in the prepared remarks but I think it's something that gets too little attention in the REIT space.

And that is, everyone when they think first about risk, the first thing that comes to mind is leverage.

I want to stress that leverage is not the best indicator of risk.

First and foremost, given the types of moves we have seen, duration and convexity positions will stand out as the more likely determinant of risk and a portfolio.

Second, it's the type of assets that you have.

Like use the chart that Chris went over 15s versus 30s as a great example.

If you asked a typical -- if you had a 100% -- a portfolio with 100% 15s at eight times leverage, that would be a lot less risky of a portfolio than 100%, 30 years at seven times leverage from a spread perspective.

Next is asset selection and then third on that list really comes down to leverage and so what I would say when we look at our risk posture, we are going to look at it from that perspective.

Now that being said, I think we've been a little sensitive to leverage in -- over the near term, but we view that as not necessarily something that will drive us continuously from here.

Okay.

Thanks very much.

Michael Widner from KBW.

Let me ask you two quick ones.

First, on the 15 year, I'm wondering how you would compare the relative attractiveness of actually holding them in the portfolio versus playing the dollar roll trade.

I shouldn't call it a trade but the dollar roll opportunity.

Maybe just talk about the effective yield as held versus rolled.

Great question and I am not going to give you a specific quantifiable answer to that, but I can give you a feel for how we think about it.

One of the big benefits that Chris mentioned in terms of 15s is the fact that they roll down the curve or shorten over time.

If you think about because of the amortization schedule, a two-year season, 15 year is materially shorter, more than 0.5 year shorter, closer probably to a 0.75 of a year shorter than a new 15 year.

And in a CPO curved environment, especially if rates are headed higher, that is really important.

And the thing when you dollar roll is that you are going to get extremely favorable financing, okay, maybe whatever like 40 basis points through, or negative, or 60 to 80 basis points through where you put something on repo.

At the end of that period, you are going to get -- you're going to take in probably a longer security and so you are balancing those types of trade-offs in that decision and so embedded in it is not -- it's not a return cutoff.

You are going to sit there and say how long do we -- are we going to get the roll benefits?

How important what is happening to the environment.

How different could our deliveries be six months from now or three months from now versus where we think they will be today.

And those all go into the equation.

But that's how we think about it, if that helps.

That helps a little bit.

Six-month delivery or how you think about six months from now seems like an eternity for you guys given how you manage the portfolio, but that's probably appropriate for the environment we are in.

Second one, let me just ask you about the fair value of the swaptions, where you are at today, and then I have a follow-up related to that.

Sure.

Specifically, the swaption market value at quarter-end was about $850 million.

It's on -- we moved the detailed pages to the appendix but it's in the back of our presentation on page 27.

From last quarter our swaption portfolio about doubled in value up to about $850 million.

Got it.

I appreciate that.

I saw the earlier page but lost it back there.

I guess the question I have on that is, you talk about the importance of that and hedging your book value risk and all that but a counterpoint to that is, they are options and as options, they basically decay all else being equal.

Is the yield curve does not move from here, all of your swaptions are basically out of the money and that is $850 million of book value that again all else equal tends to bleed off.

How do you think about that relative to the value they added and cheap insurance kind of way?

You are right to the extent that if the value was all time value, then certainly, over time if interest rates didn't move you wouldn't lose your premium and frankly, if we paid $400 million for these options and interest rates never moved and we lost $400 million that would be net-net a good thing for the portfolio because it would've been indicative of the environment but today what you have seen is our options and if you could track the coupon or the underlying coupon on the swaption portfolio, the $850 million of value isn't just time value.

In fact, many of our options have actually moved into the money so there are substantial intrinsic value in our portfolio as well and certainly from here, any further moves would be almost all intrinsic value so it would be very substantial and real monetary gain if the positions were expired or even terminated at some future point.

Okay, and I know last quarter you had a lot of the -- you had some more detail basically on option period and then term in the queue, and it looked to me last quarter like most of them were still pretty well out of the money but it sounds like what you're saying now is some of these may actually convert and be in the money swaps at some point.

Is at what I'm hearing?

That's what you're hearing.

Okay.

Thanks.

I appreciate it, guys, and definitely appreciate the added disclosures on the duration.

Great.

Glad you appreciate it.

Stephen Mead with Anchor Capital Advisors.

Yes, hi Gary.

Can you talk about the HARP portfolio in particular and just the fact that in terms of its price move fairly similar to what happened to 30-year regular mortgages?

Is that because they were selling at a higher premium going into this period?

And what has happened to those mortgages in terms of CPRs relative to the issue of default rates and other aspects of those mortgage pools?

With respect to the HARP securities, they like the lower loan balance and you can look on page 7. They have lost a fair amount of the price premium and some of that is to be expected given the interest rate move.

I think the thing that stood out the most was again the performance of let's say the 4% coupons versus the 3.5% coupons in both loan balance and HARP where I don't think any model would've told you that the 4% would have dropped more.

But from a prepayment perspective the HARP securities still maintain, with the exception of the lowest LTV, HARP positions where let's say the 80% to 90% bucket in particular, maybe the 90% to 95% LTV bucket, where you have seen enough house price appreciation on some of the older securities where they are starting or they were starting to prepay faster before this rate move as they became eligible for other types of refinances.

Those positions don't make up the lion share of our HARP security, so we do feel the prepayment protection is still there for the bulk of that position.

And you also have the benefits in some of the cases that they could turn over faster if we are in a rising rate environment as these borrowers have the first-time opportunity to either move up if we continue into a different house where as they have typically been announced for almost ten years at this point given that the original loan -- loans had to have been taken out in 2005 to 2008.

Big picture, we are comfortable with the performance of the assets.

I want to say one thing.

We've talked about the fact that our pay-up risk going forward is not significant but what I would say is we still have a lot of securities with very low pay-ups.

But if we were to rally back, we will get some of that pay-up back.

What we feel good about it is we have got decent prepayment protection, we've got securities that we like that are valued largely very close to generic securities that would have a lot of upside in a rally and again, we would still expect the performance to be good.

One thing I'd add is there is a lot of noise around [milwot] getting confirmed what could that do to the HARP date and will that make our securities -- could that create a problem even if we were to rally for our securities.

Based on the last numbers that I've seen, if the HARP eligibility date moved by one year, from May 2009 to May or June 2010, we would have less than 5% of our aggregate portfolio would be exposed to that change in the HARP date.

Big picture, we're very comfortable on the prepayment front.

We're comfortable even if we rally back.

We like our portfolio in that case and we are doing the steps that we -- that Chris talked about really to make sure that we are balanced for any of the scenarios that could unfold from here.

Was there much change in the net spread from the end of the quarter in terms of the portfolio on net spread?

If you looked at and I'll just quote you the numbers, if you look at on at the beginning on page 5, our net interest rate spread as of June 30, which is sometimes better to look at rather than the inter quarter one that is affected by catch-up am and so forth, was 159 basis points when you include dollar roll income and then in the prior quarter I believe that was 171.

Hold on.

Yes.

So that prior quarter, that was 171.

There was some decline in the quarter -- net quarter-end numbers and that's a function of higher hedge ratio, smaller portfolio and all the things that we discussed earlier.

Okay, thanks.

Jackie Earle with Compass Point.

Thanks for taking the question.

Just a real quick one.

I'm sorry if I missed this earlier.

I was wondering how much of your 30-year portfolio TBA deliverable?

It is almost the entire portfolio.

What we have said is that of our -- and this is a combined number what we disclosed of both 15-year and 30-year, 95% or more, so less than 5% of our fixed rate pass-throughs are non-deliverable into TBA.

That should give you a lot of comfort there.

Okay.

Thank you.

Daniel Furtado from Jefferies & Company.

Thanks for the opportunity, everybody.

I just had two kind of higher level questions and not necessarily rate related to AGNC, but potentially.

The first is, if one were to assume that the agency MBS asset class as a whole moves from a premium market to a discount market, what could an agency redo to not just reduce but to actually prevent decays in book value?

It just seems to me that in unless a manager was willing to go long rates and short MBS, falling MBS values for a levered vehicle would inevitably lead to lower NAVs.

Is there something that I'm missing there?

Yes, I mean just falling prices do not mean that NAVs will fall.

If you think about a fully hedged portfolio, even if spreads remain unchanged, should perform fine from and an NAV perspective and could appreciate even with dollar prices dropping.

And actually I think that's a very -- if we got a slow increase in interest rates from here, given our hedge portfolio, mortgages would probably tighten in a slow rising rate environment from here.

And you could see good performance, aggregate performance despite raising rates.

Again, that is dependent on hedges and so forth and movements and spreads.

But bigger picture, I think what -- the question really comes up as to if you want to look for different types of assets that will perform in that type of environment versus assets that really don't shorten over time and I think that is really the focus that we were talking, was taking that to the next level which was saying look, a higher rate environment is not really a problem even though that translates to lower prices.

It just means that you need -- you'll want to focus on different assets and it really does come down to hedging decisions.

Understood.

Thanks for the color there, Gary.

And then the second question is have you given any thoughts to potentially teaming with some of the servicing companies in the market today where you could own the MSR asset or some type of excess MSR relationship?

Yes.

We have given a lot of thought to that and I will probably -- it is something we have spent a lot of time and we focus a lot of attention on and I will leave it at that, at this point.

Understood.

Thank you for the insight.

I appreciate it, everybody.

Bill Carcache from Nomura.

Thanks.

I apologize.

I jumped on a little bit late from another call.

Sorry if you answered this already, but what percentage of your portfolio pays down each quarter?

Our CPR, if you just took the CPR and if you use -- I'm going to round to make it simple.

If you just assumed a 12% CPR, and then you essentially would divide that by 12.

Technically, you're actually supposed to take it to the 112 power but you can divide it by 12 and that tells you about -- that it is a little less than a 1% a month pay-down.

So then quarterly, it is a 2.5% to 3% at this point.

Okay.

And what I thought was interesting is I would have expected with generally speaking and I guess, all else equal, the rising rate environment to produced a more attractive reinvestment environment, but as you talked about earlier, we saw the average asset yield as a period end decline and I missed part of the call where I think you might have addressed this but is the change in asset composition the driver of that decline?

As we look forward from here, what direction will you -- assuming stable rate environment, what direction should we expect to happen with spreads from where we stand?

Well, I mean again, so yes, asset composition was a key driver.

And we don't worry about the number in terms of the higher or lower asset yield, we worry about kind of all in risk-adjusted returns.

And so I would say the biggest driver on spreads will be our decision on how much we want to hedge from here.

And whether we want to increase our duration gap which is something we, over time, we'll be very willing to do.

We don't want to and don't believe that we should be running at, we will call it, a 0.5 year duration gap when we have minimal extension risk in the assets that we are buying.

I think the biggest driver of spreads will be one of being willing to reduce our hedging activities and right now we have talked about why we have the position we do.

We feel it's the right thing to do.

I think its characterized by a week like this where you have a Fed meeting and you have GDP release and you have got an unemployment number, and I would be very, very uncomfortable walking in to a week like this with a large duration gap.

Okay, and then finally you guys have a very good track record of establishing whatever dividend level you believe will be sustainable at least for a period of, call it a few quarters or more.

Does the new dividend level that you said contemplate the change in asset composition and the environment as you see it from here such that it's reasonable to expect that the level you're going to sustain for at least the next few quarters?

Is that how you are thinking about it or is there some potential for that to move around given the volatile environment that we are in?

I think you were on an earlier call when I think I answered this the first time.

The reality is there's a lot of volatility in the market right now and so we respect the volatility.

We have to -- we are going to be focused on what happens in the market and how it affects the portfolio, how we are willing to position the portfolio going forward, and really that is the most I could say with respect to the dividend.

Okay.

Thanks very much, guys.

I appreciate it.

Ken Bruce from Bank of America Merrill Lynch.

First of all, I appreciate the disclosure.

It is very helpful.

I will add to the chorus of applause for that.

My question is ultimately, big picture, you and Peter and team have been involved in these markets for a long time and have seen a lot of different market moves.

Looking back at the first half of this year, what has been most surprising to you in terms of the way that the market has evolved?

What I would say is the consistent nature of the move in this direction.

So in the first quarter, it was a little less clear.

It was more mortgage market related, but what we have seen in the second quarter was -- I mean in other periods where rates have backed up or mortgage spreads have widened, so to speak, or there has been less liquidity and a lot of volatility, what we have seen is three bad days, a good day, then four bad days, then three good days.

When I would say was more surprising about the second quarter was the very few number of good days from -- in May and June.

That would be the one thing that stands out on that front.

The other thing is the movement against the backdrop of the massive Fed purchases of those treasuries and MBS but in particular of MBS, and it definitely confused the decision-making, so to speak to some degree, because what we have said in the past is you've got to be really open-minded about these things.

Whether it is the new prepayment picture or the new market landscape.

The Fed's massive role in the process is something that you have to respect and you don't always know exactly how things are going to materialize.

I will let Peter add something.

I would just add I think one of the things that was a little bit surprising to us was the use of agency MBS to hedge other products which appear to be the case later in the quarter as the global market started to unravel a little bit and investors or money managers or hedge funds begin to get worried about other illiquid spreaded products they had.

The logical hedge is agency MBS because they are liquid and they have a spread component.

I think we saw additional pressure this cycle on MBS prices because people were using them to hedge other more illiquid spread positions and I think that was a little bit unique toward the end of the quarter.

And to Peter's point, just one last thing is that I would throw in markets that can be shorted tend to react very quickly and that's true of the treasury market.

It's true of the fixed-rate bond market, or mortgage market, I am sorry.

And so they tend to react quickly and can over adjust or overreact, whereas markets that are not shortable tend to move all at once essentially people can sit there and think that they would love to sell something but it's not that liquid.

They can't short it.

Someone who doesn't have a position can't come in and take a position.

You do tend to see a difference in shortable markets versus non shortable markets and they tend to move first and that is certainly what we saw in May with the agency mortgage market relative to other spread products that really struggled in June.

And considering the technical backdrop, do think that this is at the change?

Trying to zero in on one of the earlier questions as to what it will take or what you will be looking for for the market to be safe to reenter or take a more aggressive stance in and I realize it may not be today but just trying to understand what some of those stability factors are that you are looking for.

We talked about this little earlier but what I would just reiterate is we are bullish on the mortgage market.

We do feel that mortgage spreads are likely to do well from here.

The biggest dynamic, positive dynamic, for the mortgage market is that gross issuance has gone from $150 billion plus where it was in the first quarter to -- it's probably in Q4 -- it's sub $100 million now, or will be sub $100 million could be $80 billion to $90 billion in Q4 at the rate we are going and so the Fed's purchases are bigger than origination at this point in the same landscape versus being a much smaller percentage given the pickup we saw in origination in Q1.

That's an incredibly strong technical and it will dominate in an environment where rates are not moving and other factors such as convexity hedging aren't driving the markets.

As we said earlier, in terms of convexity hedging, most of that is done assuming we don't head closer -- up another 25 basis points.

But if we do head there, there will be a period -- there will be another period of weakness, probably albeit not as bad as the last time around, but I think those are the kinds of things we are looking at but I also want to reiterate that we are looking for other market signals as well.

It is not -- we would like to see the Fed and walking into the idiosyncratic risk right now versus fundamental risk is pretty high and so I think we don't -- we feel much better about dealing with evolving fundamentals versus digital changes based on one speech or one announcement.

I wasn't clear -- I didn't understand if you were making a comment to an earlier question about whether the OAS which had been pre-crisis much tighter if it was biased higher from here given the backed up they just laid out?

No.

We would believe -- we think that it is biased lower in the absence of another sell-off but I would reiterate even if we sold off, we would think the movement wider would be temporary.

So we are relatively bullish on mortgage spreads, and the challenge is though that the drivers of volatility in the market right now are very idiosyncratic.

And so it is not a return picture that we're dealing with right now.

It is an idiosyncratic risk picture that we want to see some comfort level on before we take a more aggressive posture, so to speak.

Understood.

I think you did a very good job of managing risk in the second quarter.

My last question and I want to set this up properly.

I recognize that there was no issue in terms of repo and funding in the first half of the year but there has been some changes in terms of bank capital specifically relating to securities and that looks like it may have an impact on repo markets.

I don't know if you've got a view on that one way or the other or how you are thinking about what changes may come as bank capital and some of the pressure from regulators on wholesale funding markets really pressure on banks relating to wholesale funding markets may impact your business?

Yes, at a high level, I would say that we are not particularly concerned about it.

You are right that all other things equal there are incremental pressures.

What we have seen with our larger counterparts, and we've obviously talked to a lot of our counterparties about this and we really haven't gotten any direct feedback from them that they thought it would have a material change to their repo business or their capacity.

Our larger European counterparties have -- we believe already positioned for the environment they are in and obviously there is still some ambiguity as to how the largest US financial institutions will be treated but those rules are not yet final.

They're going to be out for comment for a couple more months.

And how exactly the leverage ratio will be defined, I think will be critical to whether or not it impacts capacity but again this is going to be phased in over the next five years.

It may have sort of an incremental negative impact on capacity or on the cost, but in the end I think it will be small and I think it's in the best interest of the regulators and the industry and for all interested parties to ensure that repo stays liquid because it's obviously a critical component to the financial markets.

Great, thank you for your comments and nice recovery in the second quarter.

Thanks everyone for joining us on this call and we look forward to talking to you next quarter.

Thanks Keith.

You can now close the call.

The conference is 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 August 13 by dialing 877-344-7529 using the conference ID 10031377.

Thank you for joining today's call.

You may now disconnect.

(End of Transcript)