AGNC Investment Corp (AGNCO) 2014 Q2 法說會逐字稿

Good morning and welcome to the American Capital Agency second quarter 2014 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 Andrew.

Thank you all for joining American Capital Agency's second quarter 2014 earnings call.

Before we begin I'd like to review the safe harbor statement.

This conference call and corresponding slide presentation contains statements that, to the extent they are not recitations of historical fact, constitute forward-looking statements within the meaning of the Private Securities Litigation Reform Act of 1995.

All such forward-looking statements are intended to be subject to the safe harbor protection provided by the reform act.

Actual outcomes and results 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 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 12, by dialing 877-344-7529 or 412-317-0088 and the conference ID number is 10049322.

To view the slide presentation turn to our website agnc.com and click on the Q2, 2014 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, Chair and Chief Executive Officer; Sam Flax, Director, Executive Vice President, and Secretary; John Erikson; 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 will turn the call over to Gary Kain.

Thanks Katie.

And thank you all for your interest in AGNC.

I am pleased with the very strong performance of our portfolio during the second quarter.

In Q2 our economic return totaled 10% making it one of our best quarters since the end of 2009.

Through the first half of 2014, our economic return totaled 15% and was bolstered by our decision, late last year, to shift away from a risk off mentality and back toward a more normal balance between risk and return.

Our performance in the second quarter was driven primarily by the significant tightening of agency MBS spreads which is now widely attributed to the supply and demand equation that we laid out in detail on our Q1 earnings call.

These favorable dynamics gave us the confidence to maintain our leverage at 7.6 times, as we entered quarter and to significantly increase our holdings of 30 year MBS early in Q2, both of which enhanced our returns.

We also benefited from actively managing our hedge positions.

To this point, late last year we made the strategic decision to run a larger duration gap verses where we had operated over the past several years.

The driver of this change was a combination of our comfort with the US interest rate outlook and the relatively limited extension risk inherent in our portfolio.

We also made significant adjustments to the size of our Swaption portfolio.

Again, this was the result of an explicit strategy that Peter discussed on our last two earnings calls.

Following the increase in rates and changes to our portfolio in 2013, option based hedges became less important to us as a negative convexity in our portfolio declined.

This was also true for the entire MBS market and when combined with low realized interest rate volatility the reduced demand for options led to a significant decline in most Swaption prices.

As such had we maintained the same Swaption portfolio as we carried for most of 2013, it would have resulted in a significant drag on our financial performance.

With that as the introduction, let me quickly touch on a few highlights on slide 4. Comprehensive income totaled $2.43 per share comprised of $0.08 of net income and $2.35 of other comprehensive income.

Net spread income, inclusive of dollar roll income totaled $0.87 per share.

It was comprised of $0.48 from our Repo funded on balance sheet portfolio, and $0.39 per share from dollar roll or drop income associated with our forward TBA positions.

Our TBA position averaged around 14 billion in size during the quarter.

Given the relatively large contribution of dollar roll income to our aggregate NIM, we added a couple slides to the presentation this quarter to provide investors with a more detailed analysis of how our net spread income is impacted by forward TBA position.

We will also specifically address the sensitivity of these results to changes in the amount of specialness inherent in dollar roll levels.

As I discussed in our Q1 earnings call, dollar roll income nets against our capital loss carry forward and is therefore not included in our current period taxable earnings.

This is an additional positive of the dollar roll strategy, in that it allows us to utilize the capital loss carry forward and, in essence, pass that benefit onto our taxable shareholders.

We will continue to base our dividends on the true earnings of the portfolio, and this decision will not be impacted by the use of the capital loss carry forward.

This will result in a portion of the dividend being treated as a return of capital verses ordinary income.

For shareholders that hold AGNC stock in taxable accounts this will likely result in an improvement in their after tax returns.

Book value per share increased over 7% to $26.26 per share.

This coupled with our $0.65 dividend drove economic earnings for the quarter to approximately 10% or almost 40% on an annualized basis.

Turning to slide 5, our portfolio increased slightly to just under $72 billion during the quarter.

Our at risk leverage declined to 6.9 times at the end of the second quarter, verses 7.6 times as of Q1.

If you turn to slide 6, I want to quickly point out something that may be surprising to many of you.

Our strong 2014 economic returns have offset almost all of the economic losses incurred during 2013.

Additionally, as you can see on the circled portion of the bar chart, the aggregate economic return on our portfolio is now positive 15% for the two years since the beginning of QE3.

Our performance over this period, while excluding changes in our price to book ratio, fully reflects the mark to market of AGNC's portfolio and is quite respectable given the roughly 100 basis point increase in rates and the tremendous spread volatility we experienced over the past couple years.

It also provides further evidence that our business model is durable across a wide range of economic environments including one where interest rates rise significantly.

Turning to slide 7, I want to follow up on my earlier comments on how dollar roll income impacts our aggregate net spread income.

If you remember on slide 4, we discussed that our total net spread income was $0.87 consisting of $0.48 for the on balance sheet or Repo funded portfolio, and $0.39 from the dollar roll funded position.

It is important to understand that this dollar roll contribution is a gross number and does not include any allocation of swap hedge cost, operating expenses, or preferred stock dividends.

The corollary of this, is that all of those costs are allocated solely to the Repo funded portfolio which unfairly depresses its contribution to the aggregate net spread income.

If you look at the table on the top of the slide, you can see how this picture changes when we allocate the various cost to each portfolio on a duration weighted basis.

The second line from the bottom shows the adjusted break down of our net spread income after this expense allocation.

As you can see, the revised net interest margin for each portfolio is materially different from where we started.

The contribution from the Repo funded assets increases from $0.48 to $0.61 per share while the dollar roll component drops to $0.26 from $0.39.

The last row on the table shows what this translates to in terms of NIM.

Importantly, the Repo funded assets show 152 basis point margin and despite the adding costs, NIM on the dollar roll portfolio remains extremely attractive at 259 basis points.

In its entirety the net interest margin on the portfolio averaged 173 basis points in Q2.

Now, it's important to understand that the difference in the margin between the two sub portfolios is a result of both TBA specialness and portfolio composition.

More specifically the Repo funded portfolio contains the bulk of our seasoned 15-year positions which are lower risk but also generate lower margins.

On the other hand our TBA position is concentrated in newer 30-year MBS which generate higher yields and margins.

Now, if we turn to the next slide I want to address the earning sensitivity of our TBA position with respect to changes in dollar roll financing levels.

As you can see on the second line, if TBA financing rates improve, that is get more special, our dollar roll income will obviously benefit.

Conversely to the right of the table, if TBA financing rates increase our dollar roll income will deteriorate.

Let's look at the extreme scenario where dollar roll financing rates increase by 100 basis points.

In this scenario our dollar roll income will decrease by $0.10 per share assuming we had the same $14 billion TBA position.

This is a very unlikely scenario.

100 basis point increase in dollar roll funding would essentially wipe out all specialness in the TBA market.

Said another way, if financing rates converged on rolls and Repo, and we went back to Repo financing our entire portfolio I would expect our adjusted spread income to drop by about $0.10 per share.

A more likely scenario is one where dollar roll specialness deteriorates by 50 basis points.

In this case our net spread income would drop by about $0.05, again assuming TBA position identical in size and scope to the average Q2 portfolio.

I should reiterate something that we have stressed many times in the past.

Net spread income, or what many analysts refer to as core income, has shortcomings and is not a complete measure of the true earnings power of our portfolio.

It is based on the historical cost basis of assets and swaps, and not their current yields.

It also does not capture cost associated with treasuries, Swaptions, or forward looking swaps.

As such we continue to be more focused on economic earnings.

With that let me turn the call over to Chris.

Thanks Gary.

Turning to slide 9 I will briefly review what happened in the markets during the second quarter.

As you can see in the top two panels, both Treasury and Swap Rate curves continued to rally and flatten.

Ten year treasury notes rallied 19 basis points, while five year notes fell only 9 basis points.

Mortgages performed extremely well versus rates during the quarter, and while hedge positioning on the curve was again a factor, mortgages meaningfully outperformed both the long end and the front end of the curve.

For perspective, 15-year 2.5's were up nearly 1.75 points whereas the five year treasury note which has a similar duration, increased in price by less than 0.5 of a point.

30-year MBS also did well during the quarter, but 3.5's and 4's up over 2 points, despite having a duration in between that of 5-year and 10-year treasuries.

Which on average were up just a little over 1 point.

So far into the third quarter, mortgages have given up some ground against backdrop of heightened geopolitical concerns and the curve has continued to flatten with 10-year rates rallying 6 to 7 basis points and 5-year rates selling off 6 basis points.

Let's now turn to slide 10 to review our investment portfolio composition.

As Gary mentioned, at risk leverage was down from 7.6 at the start of the quarter to 6.9 as of June 30th despite slight increase in our investment portfolio to $71.9 billion.

The decrease in leverage is due to a combination of net asset value appreciation, sales of other rate equity positions and the preferred stock issuance in May.

As we discussed on our last call it was our expectation that MBS, and in particular 30 year MBS would benefit from strong technical factors, lower levels of implied volatility, the flattening yield curve, and attractive role financing.

Given the continued bull flat movement rates during the month of April, we added 30 year MBS both out right as well as verses selling 15 year MBS.

Quarter-over-quarter our 15 year position was down from 48% at the start of the quarter to 38% as of June 30th.

With the majority of the repositioning done in the first half of the quarter.

While 15 year MBS will remain a core position for the reasons we have discussed at length on prior calls, valuations on higher coupon 15's became stretched at the same time that the benefits of those positions became less meaningful in today's lower rate, lower volatility, and flatter yield curve environment.

Our net TBA roll position as of June 30th was $18.4 billion up from $14.1 billion at the start of the quarter.

Role implied financing rates were very attractive throughout the second quarter, and while rolls have weakened since start of the third quarter, implied financing rates on 30 year 3's through 4's as well as 15 year 2.5's are significantly lower than on balance sheet Repo rates and the financing advantages still materially exceed the need for holding certain specified pools.

With that I will turn the call over to Peter to discuss funding and risk management.

Thanks Chris.

I'll begin with a brief review of our financing activity on slide 11.

At quarter end, our weighted average Repo cost was 41 basis points, down 2 basis points from the previous quarter.

Our liquidity position continues to be strong marked by significant excess capacity and good access to attractive longer term funding opportunities.

On slide 12, we provide a summary of our hedge portfolio.

Given the stability of interest rates during the quarter our total hedge portfolio remained relatively unchanged.

Turning to slide 13, we show our duration gap and duration gap sensitivity.

At quarter end, our duration gap fell slightly to 1 year down from 1.2 years the prior quarter.

As we have discussed before when measuring interest rate risk, it is important to consider both the starting duration gap as well as how the duration gap will change as interest rates change.

This incremental change in duration gap is due to the negative convexity in mortgage assets.

We refer to the increase in duration, as rates rise, as extension risk.

On slide 14 we show our reported duration gap as well as what our duration gap would be if interest rates increased by 200 basis points and no rebalancing actions are taken.

The distance between the two lines is the extension risk we face due to negative convexity.

By estimating our duration gap in this up 200 basis points scenario, we fully capture the dual effects of our starting duration gap.

Plus the change in our duration gap due to the negative convexity in our mortgage assets.

As you can see when extension risk is high, we operated with a low or negative duration gap.

During the middle part of 2013, when the fixed income markets were unstable and idiosyncratic risks were high, we carefully managed our overall interest rate risk position through a combination of managing our starting duration gap, the greater use of option based hedges, and through asset selection.

More recently with both extension risk and mortgage spread risk diminished we proactively increased our spot duration gap.

This larger duration gap however, does not mean that we are operating at a higher level of aggregate interest rate risk.

In fact as you can see from the graph, over the last two quarters, despite having a larger duration gap, our aggregate interest rate risk position, in the up 200 basis point scenario, has been relatively constant at about 1.8 years and below where it has been for most of the last 2 years.

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

Thanks, Peter.

Before I open up the call to questions, I want to close with some thoughts on the MBS market as we look ahead.

The bottom line is that we remain reasonably constructive on agency mortgages especially in light of the recent weakness early this quarter that Chris referred to.

We also continue to believe interest rate volatility will remain relatively low.

While agency MBS tightened materially in the second quarter their performance over the last few years has still lagged other fixed income alternatives, such as investment grade bonds, high yield, European and Japanese sovereign debt, and even most emerging market debt.

More over, agency MBS have some significant advantages that are not enjoyed by these competing products.

These include tremendous liquidity which is only surpassed by that of the US treasuries and they also benefit from favorable financing rates, and this relates to both Repo rates and obviously via dollar rolls.

Lastly reduced interest rate volatility is also a much bigger positive for MBS given the embedded prepayment option.

Now when it comes to the supply and demand equation, and this doesn't get enough attention, even after the end of the QE3 program, the Fed will still almost certainly be the largest buyer of MBS in the market as they continue to reinvest sizeable pay downs on their $1.8 trillion portfolio.

To this point the Fed is likely to continue to buy around 20% of the gross issuance of mortgages well into 2015.

Lastly, as we mentioned last quarter, Fed ownership of almost one-third of the agency market has forced investors to significantly reduce their own footprint in the market and most remain under weight or under invested in the product.

As such short term periods of wider spreads are likely to be met with significant buying, a technical that other products are less likely to enjoy.

Lastly, as we discuss last quarter, the supply picture for agency mortgages remains extremely favorable with gross and net supply continuing to be anemic.

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

We will now begin the question and answer session.

[Operator Instructions]

The first question comes from Mark DeVries of Barclays.

Please go ahead.

Thank you.

Let me be the first to thank you for the new slides and the presentation, very helpful.

First question is on the interest rate risk management slide, for Peter.

I think as you mentioned you took down your duration gap a little bit on the sequential quarter basis.

But with rate volatility at very low levels here, and swap rates quite low.

Could you just talk about how you think about the appeal, how you assess appeal of taking your duration gap even lower here?

Yes, sure Mark.

Certainly there is a lot of factors that go into where we position our duration gap and obviously we positioned a larger duration gap at the beginning of the year when rates were higher.

As we continue to drop to lower levels you could expect our duration gap also to shrink.

But it will also depend on the composition of our portfolio, the mix between 30 years and 15 years, the coupon concentration in our portfolio.

And also, as rates stabilize and interest rate volatility also continues to stabilize and drop, we will likely continue to shift the composition of our portfolio from swaps to potentially more option-based hedges, if the interest rate environment warrants that sort of protection.

So we will continue to move our position around dynamically and looking importantly at our extension risk and contraction risk.

As rates move around we'll actually have a little bit more contraction risk in our portfolio than we do have extension risk, which is why we've added some receiver Swaptions to our portfolio over the last quarter.

The one thing Mark I'd like to add quickly is that, we've always, and I think everyone in the industry tends to talk about duration gap.

The other thing is that when you look at what's going on with the yield curve right now.

A bigger source of interest rate risk really is curve risk.

And that's something that we're watching very carefully and that we've moved hedges around.

You can run a, quote, short duration gap with a lot of three-year hedges.

Or you can run a short duration gap with all 20-year hedges, obviously less of them, and get to the same duration gap.

And I think you know at this point, for us, it's much more about factoring in protection throughout the curve and making sure that's consistent as our portfolio evolves.

And that's really a another key factor that I think everyone in the industry's got to be focused on.

Just to add to Gary's point, although our overall pay-fixed swap portfolio didn't change very much.

We did add about $3 billion of pay fixed swaps, terminated some other shorter swaps, so we did move some of our hedges during the quarter to more in the 10-year to 15-year part of the curve to give us a little bit more protection against longer term rates, particularly against our increased 30-year portfolio.

Okay.

That's helpful.

Next question.

Just trying to better understand the allocation to 30 years.

It sounds like you're generally a little bit more constructive on that, than the 15 years.

Even leaving aside the specialness that we are still trading.

If that specialness disappears or at least drops 50 basis points, as you talked about being more realistic scenario would you see yourself reallocating back towards 15 years here?

It's possible.

If you remember last quarter, I talked a little bit about higher coupon season 15s as being a relatively inexpensive option on rates.

And had the market sold off, you know the position would have performed extremely well verses newer production 15s or 30-year MBS.

Instead we rallied and flattened, and the position still performed very well, and in fact better than we would have expected.

So you know, that's why I described valuations as being somewhat stretched in my prepared remarks.

And we took the opportunity to take advantage of that to reduce our 15-year position by around 7 billion in lieu of 30's, just given the move in rates.

Okay.

Got it.

Finally, it looks like you took up your TBA position even more since the end of the quarter.

I guess you are at $18 billion today.

Is that an indication that it is trading even more special and you are getting more spread today?

Or you are just feeling more comfortable with that trade and taking up a little bit more leverage there?

It's not an indication.

Things are not more special today than they were throughout the second quarter or most of the second quarter.

The TBA allocation -- but again, while they're not more special they're still special in the returns on dollar roll funded TBAs are still better than specified.

But I think an important piece of this, and this is really something that's just very different inherently than where we were two years ago, is the need for a specified pool is very different today.

First off, if interest rates go up from here, the specified pool we want is one that's seasoned, maybe 50 months, 60 months old because it will pre pay a little faster and will be shorter.

That's in 30-year, and it's even more important in 15 years.

On the other hand we have rallied down to almost 240 on tens and if we were to rally another 25 basis points the overriding concern in some coupon, to the mortgage market is going to be prepayments.

And you are going to want a low loan balance pool or some other prepaid protected pool, again in a few different coupons.

So right now when you think about the equation, prepayments are generally benign, still expected to be relatively benign.

But yes that could change.

On the other hand if rates went up the type of pool you would want is the more seasoned, less prepayment protected pool.

So again there is even uncertainty as to what type of pool you would like.

The other thing is just given the fact that the prices between different coupons and mortgages, what we call coupon swaps, are such that if you really want to protect yourself from prepayments, the most cost effective way is probably to go into a lower coupon and not to go the specified coupon route at this point.

Obviously these things can change.

So, again, I just want to build back to the crux of your question.

Our TBAs verses specified pools are a combination of the funding advantages for TBAs and what you expect for them going forward, weighed against the need or benefit from having a particular pool.

And both of those things are telling us to go in one particular way.

Hopefully that helps.

Yes it does.

Just one last question.

Gary I think you mentioned that you don't see the specialness completely disappearing as a realistic scenario.

Would you just give us some context for what's a normalized level when the Fed is not an active participant in the market?

It varies.

I think you have to be cognizant that there isn't, like -- it depends on coupons.

It depends on a lot of different factors.

What I think is absolutely safe to say is that the Fed's stock effect, the fact that they own a third of the market and that those coupons are not, those positions are not going to come back into the market probably impacts the dynamics of dollar rolls for a pretty extended period of time.

So, in the end we expect dollar rolls to remain special certainly through this year and well into next year.

And when I say special, it doesn't mean that they necessarily get back to the levels where they were last quarter.

We don't need them to.

I mean let's face it in this market, in any market, a 50 basis point funding advantage is tremendous, again, especially against the backdrop of not having a huge need for a specific pool.

Okay.

Thank you.

The next question comes from Douglas Harter of Credit Suisse.

Please go ahead.

Thanks.

I was hoping you could talk about some of the factors that will move around the specialness of the roll market.

And in the coming quarters as the Fed sort of eventually ends their purchases the ex the RE investment?

Sure.

Look, there are a lot of factors, and so we can't isolate any one.

But I think what's important to keep in mind is that in general what creates specialness, what creates specialness is because of the tremendous liquidity of the mortgage market, people can have short positions and always have short positions in TBA coupons.

Originators, Wells Fargo or JPMorgan Chase, are taking locks from borrowers two and three months into the future.

In order to hedge their exposure they're selling TBAs into the market for future settlement dates.

And those pools will be delivered, you know again, over the course of a couple months and then the whole process will be repeated.

Against the backdrop of that, you have an outstanding float of mortgage securities that have already been created that can be delivered against other against other trades.

And to the extent again that the Fed's holdings of those other securities are so large and then it's not only the Fed.

I mean as the Fed's been buying over this period also other buy and hold investors, banks, insurance companies, REITs, have also bought pools that aren't likely to come out to the market, as well.

And so against that backdrop while short positions may have some temporary factors, like short positions may have dropped as people have gotten a little more comfortable with the mortgage market, these other factors are likely to stay in place.

So, it's a difficult question to answer, but what I would reiterate is there is always poor shorts in the market.

The thing that really allows for rolls to be on the low end of specialness is an outstanding amount of float, a larger outstanding amount of float of pools that people don't want to keep.

And we don't think we are fundamentally in that kind of environment.

Got it.

Thank you.

And then you know when you are thinking about sizing the position of the TBAs or the rolls, how are you thinking about that from here, given the clear financial advantage that those are offering you do?

Look, we obviously agree that there are advantages, that there are substantial advantages.

I think there are practical considerations, and what I would say is that, it's in the area of 25%, 30% of your portfolio they're probably considerable rolling.

Let's be practical that you don't want to be in a situation where -- again we operate with the assumption we can take these in at any point in time.

And so, we want to make sure whether it is whole pool coverage, or a variety of different coupons of different types of assets in our portfolio, you do want to manage concentration risk.

Just to add to Gary's point on making sure that we carry enough excess capacity from a Repo perspective to take on the entire position is really critical from a risk management perspective.

That's why in the beginning of my remarks I mentioned the fact that we currently have about as large of excess capacity position as we have ever carried.

So we have lots of capacity to bring all the position on balance sheet if we needed to and still have excess room.

It's interesting actually.

People ask about the Repo availability sometimes.

One other hindrance is that, our Repo position has contracted obviously over the last year or so, and especially as we have built up the TBA position.

We have increased the amount of lenders pretty substantially and just keeping enough securities to kind of maintain that relationship with people is another factor that we have to think about.

It's sort of the opposite of what people might think.

Thank you.

Thanks Doug.

Next question from Arren Cyganovich of Evercore.

Please go ahead.

Thanks.

Just a question about the increase to the 30-year MBS from the 15-year.

What specifically was in those assets that you purchased that was able to keep the duration of the assets roughly the same as the prior quarter?

Normally I would think increasing 30-year relative to 15-year should increase your asset duration risk.

And it would have.

We didn't have any magical dust to sprinkle on them.

They are longer duration assets.

But remember we had a pretty healthy rally in the mortgage market, which sort of brought down the durations of everything.

And so what you're seeing is, and this is where it all makes sense, right, is that as the market was rallying and dollar prices are going up and your durations are shortening, and at the same time we, actually maybe a little bit before that, we were allocating more toward 30-year.

But what you are seeing is a confluence of few different factors.

That's why you didn't see the durations change.

Got it.

Thank you.

That's helpful.

In terms of, I guess it was slide 8, where you talk about sensitivity of the dollar roll.

Could you talk about that relative to your historical dollar roll income?

That has bounced around a little bit more.

I think in the second half of 2013 you actually had negative dollar roll income, I suppose because you actually had short positions in TBA.

That's a pretty big delta from $138 million to zero, and what are the mitigating factors of that, maybe taking some of that on balance sheet or something of that nature.

Sure.

Look, that's really all related to just the size and the position.

To your point in the second half of, or around the middle of 2013, we were actually net short TBAs.

And so, given the symmetry of how we report we were reporting a negative drag in income due to those short TBA positions.

I think to your point and something we tried to stress in going through this table is obviously one key variable for aggregate dollar roll income is the amount of specialness, which is what we've outlined here and the other is the size and composition of the portfolio.

To the extent that the portfolio gets larger, obviously, that's going to help the aggregate amount of dollar roll income, to the extent that it moves to 30-year coupons or coupons with more specialness, then that can help as well.

To your point, there really are three factors.

There are more, but the three biggest are going to be size of the position, composition of the portfolio, and then the specialness that we have outlined here.

Okay.

Thank you.

The next question comes from Matt Howlett of UBS.

Please go ahead.

Thanks Gary.

Thanks for taking my question.

Just on, the comments around volatility said you expect volatility to be rather low going forward.

Just getting back to what happened last May, I think that surprised a lot of people, then looking at today.

I think you still see consensus in terms of economists out there having some of the 10-year over 3% by year end.

Which would imply, obviously, a pretty big move from now until then.

But I guess just, what do you see around volatility?

Why do you expect it to remain fairly benign going forward?

I will start and I will let Peter chime in, as well.

I think the most important thing to keep in mind on the volatility piece is really within the mortgage market what's happened is durations have extended.

There isn't the same aggregate negative convexity in the market, especially when you think about the fact that the Fed has absorbed a third of the market.

So the aggregate negative convexity in the mortgage market as a whole is already lower than it normally is before Fed tightening cycle.

Second of all, a third of that is held by the Fed.

Some people might think, well, in the past that was held or a lot of that was held by GSEs.

The difference is GSEs would have hedged that, and they were actually the most active hedger in the market.

So there is a massive difference there.

So I think one thing you really have to keep in mind is just the fact that the mortgage market's impact on volatility and for that matter even spread volatility last year was just completely different than it is today.

And so even if we had a quick increase in interest rates it won't feed on itself the way it did and it won't create the same stress within mortgages.

Originators have already sold originations, tiny at this point.

Just across the board none of those factors are really in place.

I would just add two other points that I think are really important, and that will show the difference between the environment we face today verses the environment we're in.

The first one is correlation between rates and mortgage spreads.

You know last summer what we saw was a positive correlation between rates and mortgage spread.

So rates were going up mortgage spreads were widening.

That was compounding the sort of rebalancing needs and pressure on the MBS market.

So you really couldn't afford to have any duration exposure because you had the incremental exposure of spreads widening as rates were moving.

And then the final point that I think makes today's environment different is how fragile the economy is.

Now it is growing and the economy seems to be doing well.

But we saw a 100 basis point move in rates last year, what that did to mortgage rates and what ultimately that did to the housing market.

The housing market even today, even though the economy is showing signs of strength, the housing market itself is sort of neutral and perhaps even slowing.

Those are some other factors that I think will mute the rate increase.

Thanks for pointing that out.

There are clear differences.

On that note when you look at the potential to raise capital, you did the preferred deal this quarter.

It looks like that market is over for you.

I am sure there is room for you to go on that end.

But when you look at it, I think you thought that -- I think you said earlier that the [agency] market was fairly valued.

I don't want to put words in your mouth, but not rich.

Not really cheap.

You think about maybe buying more MBS, and yet you hear on our end, how sort of the real money investors, the [temples] aren't buying for a while.

What would make you decide in terms of adding more MBS at this level?

Where could you take leverage?

I know that is a question you get all the time.

Would you look to issue more preferred?

If you do get above book would you look potentially to raise additional equity?

Look, I think there are so many different factors that go into the equity equation.

We certainly, to your point will consider all options around raising capital and we'll evaluate them the way that we have in the past.

Obviously going back a few years ago we issued a lot of common.

We demonstrated the symmetry of that equation by being willing to buy back a lot of our shares when we thought that made sense.

And we'll continue to be opportunistic around the aggregate landscape.

Again with respect to mortgage market and our view on spreads, that is a factor.

But, we understand there will always be different opinions around which way spreads are headed or what the market looks like.

But we'll continue to focus on what we really feel makes the best, most sense for shareholders.

Great.

Thanks, Gary.

The next question comes from Joel Houck of Wells Fargo.

Please go ahead.

Thanks and good morning.

This question is on the mortgage REIT holdings you had at the end of June.

Obviously you took some exposure down.

Do you have the total IRR of the overall investment and mortgage REITs?

And how comfortable are you looking forward with what you still hold?

Look we, in terms of the IRR, I am not going to go into specific numbers.

If you look at the returns we have quoted, they made about $50 million in Q1 on an all in basis.

$24 million.

$24 million this quarter.

So you're looking at $75 million in profits as of the end of the quarter.

How you judge the specific size of that portfolio at different times will determine the end result.

What I would say is with respect to go forward, I am really not going to comment at this point.

It's not a huge portfolio.

And big picture, I think we have given people enough direction over the last couple calls as to how we're thinking about it.

We'll continue to be opportunistic with respect to the holdings and make decisions as the markets change.

Okay.

I guess a conceptual question.

If, in fact, the financing in the dollar roll market gets more expensive, is it necessarily -- is there a tight correlation with overall spread widening, in a sense that, you know, if you had less contribution from that portfolio you would find more attractive opportunities in a regular market where you are funding with Repo, or is there not really a correlation at all and as we think about modeling spread earnings going forward, we shouldn't necessarily make that connection?

What I would say is, if you are trying to think about the overall earnings power of the portfolio, I almost don't think it matters that much for you to try to model the aggregate or the size of one portfolio verses the other.

I mean so if we, it's more a matter of making an assumption on dollar roll specialness and then, yes you can make an assumption.

You'd have to make an assumption on size.

But in a sense, whether it turned out that we move stuff one way or the other.

Again I think what we're showing you is that you can allocate those, the returns between the two portfolios and you can think about them in aggregate.

And then the key variable is going to be what happens with specialness in terms of the aggregate net spread income.

So, I know that realistically if rolls become less special, I think there is a likelihood that we will move some more to TBA or to specified pools.

But I just want to be very clear that you have to be practical, like even a 30-basis point funding advantage is extremely valuable over time.

And so, I think the mistakes some people make is they focus on little too much on the direction of the move.

Did it get 40 basis points worse or 25 basis points worse as opposed to, is the difference still compelling.

And the file I guess on the dividend, you clearly have -- there is a lot run rate in terms of core earnings versus the $0.65 dividend, and for reasons that you already pointed out that the taxable earnings are lower.

But that is a good thing for shareholders.

How should, for income oriented investors, should they focus more on the core earnings in terms of dividend direction and ignore taxable earnings at the moment?

How should we think about that?

Sure.

First off, what we have told you is yes, to ignore or not to focus on the taxable earnings numbers.

The dollar roll income is essentially left out.

We are able to use that against the tax loss carry forward, which is a big advantage.

So yes you should basically ignore the taxable income number and we will be essentially ignoring it as we set the dividends.

Okay.

Great.

Thanks, Gary.

The next question comes from Mike Widner of KBW.

Please go ahead.

Good morning.

I had a couple follow up questions.

Let me start by beating this horse with specialness, with just a real simple.

The chart you showed on page 8. You know that 0% column represents where the specialness was in 2Q on average.

If you had to characterize that as of today, four weeks after quarter end, would you say we are around the 50% column, or is it a little more or less than that, again, point in time.

It's hard to get the exact, we'll say this was a quarterly number across multiple coupons.

And where we are now is, we have different coupons, but yes dollar roll funding levels have worsened over where they averaged.

I would say that you're not that far off, maybe in between the plus 25 and the plus 50.

Maybe it's a little closer to the plus 50.

I think it would be really hard for me to try to point an exact picture at it.

That makes the follow up harder, because I was going to ask you to forecast the rest of the quarter.

But I guess I will just skip that.

Maybe hold that for after the quarter.

We might be better at it then.

That sounds good.

A different follow up question I had.

I think it was Matthew that asked the question about sort of volatility and rates.

And you answered that with I think, you know, very good and elaborate answers on why this year and the current situation probably isn't set up as badly as it was a year ago, or maybe a little over a year ago.

But as far as absolute rates go, we are sitting here with the 10-year back below 2 1/2%, and you know without getting too precise about it, four out of the last five years we've seen pretty strong rally in yields and rates have risen pretty sharply somewhere around year end as we get that seasonal enthusiasm cycle that, this spring is going to be the year the economy really breaks out.

One of the set ups that's a little worse this year is actually the Fed has been tapering and Fed has been talking about moving forward in terms of when it actually raises rates.

The reality of whether it happens or not is pretty clearly secondary to the market emotion around that expectation of spring time economic all is good.

Does that concern you at all as you set up portfolio?

And again against the back drop that mortgage spreads have actually tightened up quite a bit?

It doesn't matter about the fundamentals, as long as the market is going to trade on sentiment and expectation.

So, I am just wondering, you did a lot of repositioning this time last year.

It doesn't sound like you are really thinking about it in the same way now and I am just curious.

I think what's important is that we react to the change in the environment, and it's not just like the absolute level of rates.

Let's go back to, you are absolutely right.

If you looked back a year ago in the second quarter we shrank our portfolio, we repositioned from 30's to 15 years.

We reduced our duration gap and, to Peter's earlier point, we reduced our exposure to negative convexity, as well, because we were very worried about idiosyncratic spread risk.

If you had then described, over the next year, to your point, at that point there was quite a bit of concern that for technical reasons, fundamental reasons, everything, that people weren't able to distinguish between a tapering or the potential of a tapering and a rate increase, everything was sort of lumped together.

We felt that really defensive position made sense.

But if you think, look at 15's for an example which is a difference.

If you look at what's happened over the last year, we have seen a tremendous flattening of the yield curve, a massive decline in interest rate volatility.

These aren't things that you necessarily would have expected at that point.

Yes, they were kind of the best case scenario in the end for the mortgage market, but they're not the reasons why you were buying 15 years last year, right?

You weren't buying them to be the best performer and a massive flattener and where rates came back down and where volatility dropped.

And so we have to be practical.

The other thing I would say is your point about, we have rallied back and the 10-year is back to 2.50%.

It's still 75 to 100 basis points higher than where it was.

But more importantly, I think what you are describing is the dynamics of the yield curve, which I mentioned earlier in response to another question.

The 5-year treasury has actually sold off this quarter.

The 3-year and 5-year have been significantly effected already by changing expectations around where the funds rate will be.

I do think, I want to just stress what we mentioned earlier.

Active management is just really important all the time, but it's more important when all these kind of factors are moving around.

And so I would say our focus right now is, at least, is probably more on the yield curve and what might happen there as much as just the absolute level of 10-year rates.

To build on that point, even though our overall hedge portfolio hasn't moved a lot, we have done a lot of rebalancing within the hedge portfolio.

Take our payer swaption portfolio for example.

while it looks relatively small at about $8 billion, we have actually shifted that composition, to Gary's point, to be much more concentrated on the intermediate part of the yield curve.

For example the four-to-seven year part report of the yield curve where we think interest rates really could become volatile in a scenario where the Fed backs up, and not maybe so much on the back end of the yield curve where the position had been last year.

So there has been pretty significant changes in where our hedges are located along the yield curve.

Well that makes sense.

What I am really taking away from it, in summary, is you are not feeling nearly as defensive, and probably appropriately so, as you were at this point a year ago.

But you are still going to maintain AGNC philosophy of active rebalancing any time the wind suggested that's the thing do.

The other point I want to go back to I made at the beginning is we are looking at exposure for both interest rates up and interest rates down now.

We have much more two-sided risk today than we had a year ago.

A year ago we had a lot of extension risk.

Everybody had a lot of extension risk.

Today we have to also look at our duration gap, and what will happen to our interest rate risk if rates rally by 100 basis points.

So that is not out of the question.

Although not likely, out of the question.

We have could be cognizant about our position in both directions.

Mike, first I want to thank you for the question, because this is obviously a really important topic.

And the other piece, the other thing to keep in mind, is the other side of this is everyone wants to look at what's happened in the past when the Fed has tightened.

Okay.

And I think you have got to be a little careful with looking at those as examples.

Because, in the past when Fed has tightened it has never just gone through a tapering process.

And the tapering process still created a more than 100 basis point increase in rates.

It's already done things like crushed the supply of mortgages and have people go through a complete hedging process.

And all of that occurred like a year ago and, maybe, potentially a year and a half before an actual tightening process occurred.

So I think that there are a lot of dynamics here that we feel that have to be taken into account, and just looking back to 2004 or 1994 is very short-sighted, given what we have already gone through and the fact that we have never had a situation where the Fed had already tapered.

And so that's another key thing to keep in mind.

Yes, certainly appreciate it.

Nice quarter.

Thanks.

We have time for one more question.

And our last question comes from the line of Eric Beardsley with Goldman Sachs.

Please go ahead.

Hi.

Thank you.

Just perhaps a follow up to that, in terms of what is your current outlook for rates?

Are you positioned for how you expect the yield curve to develop as it is today?

Or are you more bullish on where rates go and where spreads go on MBS?

And then if we were to progress along the yield curve as it is today, how would you change your positioning?

I think big picture, we don't have a strong view right now on interest rates.

And you can see that by our duration gap coming down to sort of a more normal position at a year.

Again, a year with still, even though it's increased a little, still very little extension risk in an environment where we don't think extension risk, and so if interest rates go up we actually think that's the better environment for mortgage spreads right now, not a worse environment.

So, a key factor for us is the comfort level that, and I think others share this, the comfort level that if interest rates went up, and we were 2.75% to 3%, that would bring in a lot of buyers for both treasuries and rates, and likely hold mortgage spreads in very well.

In that kind of environment, this is a pretty neutral duration gap given the amount of extension risk we're in.

So, we're not making, we'll say, we don't have a big eternal call on interest rates.

And around the curve I think that the one thing, the curve is flattened a lot and while that's logical outcome in assuming the Fed remains on their current course, then it is logical to assume the curve is even flatter six months to a year from now.

But we have to be cognizant that something could change that.

And to Peter's point the housing market has been a weak link recently.

So, what I would say is, if there is room for one really big move in the market it's probably if the market were to start to price out a Fed tightening or push that back that you could, again not counting on it, but a scenario we have to watch for is one where the 5-year rallies significantly.

Because that's a place where there are very, very deep shorts in the market place.

Great.

Thank you.

Thanks.

We have now completed the question and answer session.

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

I'd like to thank everyone for their interest in AGNC.

We look forward to joining you next quarter.

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 August 12th by dialing 1-877-344-7529, using the conference ID10049322.

Again, 10049322.

Thank you for joining today's call.

You may now disconnect.