AGNC Investment Corp (AGNCN) 2017 Q2 法說會逐字稿

Good morning, and welcome to the AGNC Investment Corp.

Second Quarter 2017 Shareholder Call.

(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, Phil, and thank you all for joining AGNC Investment Corp.

Second Quarter 2017 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 forecasts 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 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 10 by dialing (877) 344-7529 or (412) 317-0088, and the conference ID number is 10110075.

To view the slide presentation, turn to our website, agnc.com, and click on the Q2 2017 Earnings Presentation link in the lower-right corner.

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

Participants on today's call today include: Gary Kain, Chief Executive Officer; Peter Federico, Executive Vice President and Chief Financial Officer; Chris Kuehl, Executive Vice President; Aaron Pas, Senior Vice President; and Bernie Bell, Senior Vice President and Chief Accounting Officer.

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

Thanks, Katie, and thanks to all of you for your interest in AGNC.

We were pleased with AGNC's solid performance during the second quarter and continue to have confidence in the outlook for our business going forward.

During the second quarter, the key themes from Q1 remains in place with sustained strength in equity and credit-centric fixed income products and very limited interest rate volatility.

As we noted on last quarter's call, we believe the diminished probability of near-term fiscal stimulus, coupled with benign inflation and ongoing geopolitical concerns, was likely to keep rates within a reasonable band.

The year-to-date decline in interest rates is consistent with this view, and market participants today are clearly pricing in a lower probability of any meaningful increase in bond yields.

Agency MBS spreads were relatively stable during the quarter despite clear communication from the Fed that they plan to begin tapering their MBS and treasury purchases at some point in 2017.

The stability of MBS valuations in the face of relatively hawkish communication from the Fed is clear corroboration of our view that the MBS market had already largely priced in the reduced support from the Fed.

Against this backdrop, levered returns in Agency MBS continue to look attractive as the funding picture remains supportive with agency repo spreads to LIBOR remaining on the low end of the postcrisis range.

Additionally, given the relatively flat curve, investment returns are considerably less dependent on a large duration gap.

With this in mind and given the year-to-date decline in interest rates and implied volatility, we chose to further reduce our aggregate interest rate exposure largely through the purchase of swaptions.

Peter will discuss this in greater detail in a few minutes.

With that as the introduction, let me turn to Slide 4 and quickly review our results for the quarter.

Comprehensive income totaled $0.40 per share.

Net spread income, which includes dollar roll income and excludes $0.04 per share of catch-up amortization, increased to $0.67 per share from $0.64 in the first quarter.

Tangible book value per share was essentially unchanged at $19.25 as of June 30.

Economic return on tangible book was positive 2.5% for the second quarter.

Turning to Slide 5, our at-risk leverage remains essentially unchanged at 8.1x tangible book, with our portfolio increasing to almost $64 billion as of June 30.

Additionally, we raised approximately $500 million in new common equity during the quarter.

Importantly, this capital was accretive to tangible book, enhanced our scale and common stock liquidity and further improved our already industry-leading operating cost structure.

Inclusive of the fee income for managing MTGE, AGNC's all-in operating expenses are expected to be approximately 70 basis points of equity.

For comparison, the 70 basis point figure is about 1/3 of the average operating expense for the 5 other mortgage REITs with over $3 billion in equity.

Alternatively, on a gross asset basis, AGNC's go-forward expense structure is expected to be less than 10 basis points, which is in line with low-cost bond ETFs.

It is important to note that these ETFs are passive, unlevered and completely unhedged, and they clearly do not have access to the broker-dealer financing that AGNC enjoys.

At this point, I would like to turn the call over to Chris to discuss the market and our agency portfolio.

Thanks, Gary.

As you can see on Slide 6, both interest rates and Agency MBS spreads were relatively stable quarter-over-quarter with 5- and 10-year swap rates ending the quarter lower by 11 and 12 basis points, respectively.

Option-adjusted spreads on Agency MBS were slightly wider during the quarter despite continued tightening in residential credit, CMBS and investment-grade corporate debt.

Over the last 3 quarters, Agency MBS have materially underperformed other fixed income sectors in anticipation of the Fed tapering reinvestments later this year.

For perspective, Agency MBS option-adjusted spreads have widened around 20 basis points, while investment-grade corporates and CMBS have tightened 15 to 20 basis points.

The underperformance of Agency MBS versus residential credit is even more dramatic with spreads on CRT more than 100 basis points tighter over the same period of time.

As such, levered returns and other fixed income sectors are being squeezed, while spreads on Agency MBS remain attractive and are supported by favorable fundamentals, including: low interest rate volatility, which reduces the cost of hedging an MBS portfolio; a benign prepayment risk environment, which will likely remain so unless we rally to new lows in rates, in which case, Fed tapering may be pushed further out; and finally, the funding backdrop for levered investors in Agency MBS has improved materially over the last several years.

And so against this backdrop, while we do expect that the Fed will begin tapering reinvestments later this year, the schedule, as described, following the June meeting, provides for a well-controlled and gradual process that, in our view, makes it unlikely that spreads will widen materially.

Let's now turn to Slide 7. Our investment portfolio increased to $63.8 billion as of June 30, while leverage was more or less unchanged, consistent with the $500 million increase in equity during the quarter.

The capital raise during the second quarter was invested primarily in TBA MBS as the combination of the benign prepayment environment and favorable dollar roll financing continues to drive attractive risk-adjusted returns.

Moving to the chart on the top right of the page, you can see that prepayment speeds on the portfolio remain well contained given the composition of our holdings as well as the low percentage of borrowers facing a compelling incentive to refinance at today's rate levels.

I'll now turn the call over to Peter to discuss funding and risk management.

Thanks, Chris.

I'll start with our financing summary on Slide 8. The 22 basis point increase in our repo cost over the quarter reflects the Fed's rate increase on June 14.

This higher cost, however, was largely offset by a lower cost on our pay fixed swap portfolio.

As a result, our aggregate cost of funds rose only marginally, while our net interest margin actually increased.

Here are the numbers for both those measures.

Our aggregate cost of funds, which includes the implied funding cost on our TBA position and the net interest expense on our pay fixed swap portfolio, increased modestly during the quarter to 131 basis points, up from 126 basis points in the prior quarter.

This increase was driven by 2 factors: first, repo rates relative to LIBOR deteriorated somewhat during the quarter given the Fed's rate increase and as dealer balance sheets were a little more constrained this quarter; second, as I'll discuss shortly, we increased the size of our swap position during the quarter, which resulted in slightly higher cost of funds.

Turning to our net interest margin.

Despite the Fed raising short-term rates, 50 basis points so far this year, our net interest margin actually improved to 155 basis points in the second quarter, up from 151 basis points in the first quarter.

The stability in our cost of funds and net interest margin reflects the benefit of operating with a very high hedge ratio in an environment when short-term rates are increasing.

In fact, since December 2016, the Fed has increased the federal funds rate by a total of 100 basis points.

Over that same time period, our aggregate cost of funds increased just 16 basis points, while our net interest margin improved 10 basis points.

For your reference, we provide a history of these numbers on Slide 20 of the presentation.

Turning to Slide 9, I'll highlight a few changes to our hedge portfolio.

As Gary mentioned, we reduced our interest rate exposure during the second quarter by increasing the size of our pay fixed swap and swaption portfolios.

In total, these changes increased our overall hedge ratio to 98% of our funding liabilities, up from 90% the prior quarter.

This high hedge ratio reduces our net asset value sensitivity to interest rate fluctuations and provides stability to our net interest margin.

We increased the size of our swaption portfolio in the second quarter as the price of these hedges declined to levels experienced on only a few occasions over the last 20 years.

The price of these options and the low absolute level of interest rates made it an ideal time to add more swaptions to our hedge mix as an insurance policy against an uptick in volatility or an unexpected increase in interest rates.

While we don't necessarily see these moves occurring in the near term, the low-cost of this additional production allows us to continue to generate very attractive returns despite meaningful reduction in our aggregate interest rate risk profile.

The impact of these changes can be seen on Slide 10.

As shown in the center column of the table, we reduced our current duration gap to 0.4 years as of June 30 from 1.1 years at the end of the prior quarter.

The column on the right side of the table shows our expected duration gap after an immediate 100 basis point increase in rates.

Given the swaptions that we added during the quarter, our expected duration gap in this up-rate scenario is more muted at only 1.4 years.

So to summarize, given our high hedge ratio and the changes we made to our hedge mix, our cost of funds has remained relatively stable, and our need for significant rebalancing, should interest rates increase, is relatively low.

In fact, it is quite possible that we would not need to do any duration rebalancing even in a scenario where interest rates increased by a full 100 basis points.

And with that, I'll turn the call back over to Gary.

Thanks, Peter.

And at this point, I'd like to ask the operator to open up the lines to questions.

(Operator Instructions) The first question comes from Steve Delaney with JMP Securities.

Gary, it's kind of hard to overlook the increasing contribution of dollar roll income, so $0.27, 40% of total earnings.

Help us understand the practical constraints there.

Obviously, there's relative attractiveness in dollar rolls, and I think I heard Chris or Peter indicate that the -- much of the $500 million was actually put into dollar roll trades.

So help me understand sort of your internal band of TBAs to settle the MBS or to the total portfolio.

And as part of that, does the '40 Act exemption and the whole-pool test, does that tie into your thinking as you choose between those 2 asset types?

Thanks, Steven, and it's a very good question.

The first thing I want to point out just with respect to dollar rolls and the "contribution to aggregate net spread income" is that in some ways, the way we report it overstates the dollar roll contribution because all of the hedges, of which some of them relate to the dollar roll position, are attributed to the on-balance sheet assets, whereas the dollar roll income is really just the net spread or income from the asset less the implied funding costs.

So one -- that's one thing you ought to keep in mind.

The other thing to keep in mind is that dollar roll specialness has really been, let's say, 15 to 20 basis points over the last couple of quarters, it's not outsized.

So there isn't this great exposure on the part of -- on our -- on the income front to a decrease there.

I mean, that's literally the floor that you can lose because, at some point, if dollar roll levels were to get worse, you're just going to take them on and add them, you'll take in assets and add them to the on-balance sheet portfolio.

Now let me address your other point, which was the size of the dollar roll position.

And what I would say is we absolutely do -- there are a number of considerations that go into how we size that position.

I would not say that it's at a max at this point, but I would say it's on the higher end of the range in terms of the percentage of the total assets.

And it is less, let's say, the whole-pool test certainly not in today's environment, but we have to consider the need to reposition our assets over a range of different scenarios: one, where interest rates fall 50 to 100 basis points; where interest rates rise 50 to 100.

And against that backdrop, if we have too large of a TBA position, then it puts a lot of pressure on pool selection if you were to take that in over a relatively short period of time, one of the factors being, potentially in that scenario, whole pools.

So the short answer is I think that the TBA position is on the higher end of the range we would look for on a percentage of the portfolio.

It's not at a max yet.

But -- and there's a range of things that factor into it, and it's really the ongoing position management across a range of scenarios that we're thinking about.

The next question comes from the line of Rick Shane with JPMorgan.

Hey, can we talk a little bit about the impacts, both in terms of swaps gains and losses and NIM, on a flattening of the curve?

And when we think about this in terms of our own models, where in the curve should we be focused in terms of looking at benchmarks?

Well, yes.

This is Peter.

Just in general, in terms of our hedging activity, our hedges are focused more from the 3- to 10- to 15-year part of the curve.

Certainly, we think about the overall curve exposure from an all-in business perspective and often use our hedge portfolio as a hedge against a flattening of the yield curve.

Said another way, to create a hedge portfolio that will likely benefit somewhat in a flattening of a yield curve environment to offset some of the earnings impact.

But the other, I think, key point is that in general, a curve flattening is not that significant to us from an overall business perspective because we essentially hedge out our exposure across the yield curve, and that's evidenced not only by our flat duration gap but just by the fact that our hedge ratio is so high and spread out across the curve.

So in general, we don't have very much exposure.

Gary, do you want to add to that?

No, I think that's a good point.

I would say that most -- when you think about changes in the curve, a flattening will have much less impact on book value than a steepening of the curve, generally, is what we've seen and what's consistent with our modeling of the portfolio.

And as you've seen with a flattening yield curve, when you are relatively fully hedged, you don't see the earnings impact that everyone kind of expects when you think about a portfolio that's just short-funded or whatever.

So I think those are the key considerations.

Got it.

And what about from a book value perspective?

Again, just from a book value perspective, a flattening, generally, is less impactful to book value than a steepening.

So our hedges do tend to be a little frontloaded, and so just in either direction, whether it's a bull flattener or a bear flattener, flatteners tend to be less of a factor.

And our last question comes from the line of Doug Harter with Crédit Suisse.

I was hoping you could talk about the relative cost of adding the swaptions in this environment.

Sure, Doug.

This is Peter.

Yes, what I mentioned it to be -- in my prepared remarks was that what we've really observed over the last several quarters, and in particular, in the second quarter, is a fairly pronounced drop in the price of interest rate options, in part because of the low-volatility environment that we're in.

And as I mentioned, really, the price of these options have hit levels that we really have only seen on really probably, 2 occasions over the last 20 years.

So if you think about it, we're buying options right now, low prices, historically.

From a carry perspective, if you think about it, we're buying options 1, 2 and 3 years predominantly on 10-year swaps.

If you thought about the cost of that premium amortized over the option period, it's very similar to the cost of carry on a 10-year swap if you just did that trade outright.

So there's not a lot of difference in the cost profile, but obviously, the payoff profile is very different on the option.

If you put a 10-year swap on today, it might have the same cost.

But obviously, it has a very different exposure to interest rate increases and interest rate decreases, whereas, in the swaption, we obviously have a limited downside should rates fall.

So we like the idea of adding that optionality to the portfolio.

It's not meaningfully more costly than putting on other hedges.

In fact, historically, it's at very attractive levels.

We like the strike price of them, and it essentially gives us a lot of protection against big upright moves in rates.

And again, it's a fairly low probability, but we thought it was prudent to add that sort of out-of-the-money option protection to our portfolio given the level of prices.

What I would just add is that if you think about our business, we used to use a lot of swaptions kind of go -- and one of the things that helped us going into 2013 was the size of our swaption portfolio.

Over the course of the last 4 or 5 -- 4 years, we really haven't felt the need, and we thought we were not -- we didn't think volatility or implied volatility was that fairly priced.

So we've kind of let that portfolio shrink to the point where it hasn't been that material.

And given the declines in volatility across the board, we've chosen to kind of build that portfolio back up some.

But what I want to make sure we -- want to make sure we point out is we still think we're in a low-volatility environment.

We still think that's going to help our earnings as we're going to have low convexity costs.

But you should think of this as our general business is to be short volatility, and that's been a good position to be in.

And maybe what we're doing right now is covering 20% of our short is in volatility, and that's a good way to think about it.

It's not a bet that we think volatility is going to pick up.

It's a realization that it's come down a lot and that our net business model tend -- is really one where we're always short, and this is a time, given pricing, where we should be short less of it.

It helps us with convexity management.

It's also -- would help us to the extent that we wanted to increase leverage over time and helps, again, allow us to manage kind of shocks in that environment.

I guess on that last point about leverage, you -- sort of your commentary of [spine in] the market, relatively attractive given the backdrop, what are your thoughts about willingness to increase leverage or look to raise some other form of additional capital to continue to take advantage of the opportunity?

I'd put the 2 separate.

I mean, capital raising and capital activities, we tend to look at from the perspective of we can adjust leverage in the agency mortgage market very quickly.

So whether we -- so in a sense, our leverage targeting, so to speak, is generally outside of a week or 2, independent of capital transactions.

But what I would say around the leverage issue is consistent with what I've said over the last couple of quarters.

As we've brought more of our financing into Bethesda Securities, our internal broker dealer, our haircuts have declined pretty meaningfully.

We expect to continue to do more business there.

We actually expect haircuts probably to decline kind of over the next few years as more cleared repo solutions come into the market.

So I mean, practically speaking, we could increase leverage a turn or 2 without reducing, meaningfully reducing the amount of unencumbered equity we have.

And so I think, practically speaking, I think the industry as a whole will move to higher leverage.

Now what I would say is we are going to be opportunistic.

I think the trend is again, over the next few years, is going to be towards operating with noticeably higher leverage.

But on the other hand, we're going to be opportunistic.

We understand we're in a world where kind of most spread product is pretty tight.

Agency MBS are sort of an exception, but we do expect some spread volatility over the next year given kind of the Fed's kind of balance sheet trajectory.

And if we get some of that, then we'll view those opportunistically.

So again, trend is to higher leverage, but we're looking for a catalyst to kind of implement that position.

And I guess, Gary, in that world where leverage trends higher, do you think about -- do you think you need to tighten the duration gap or run less interest rate risk to sort of manage the overall risk profile?

Or how do you -- how does the -- that risk side change with higher leverage?

I mean, look, you could -- you have to manage risk appropriately with 8x leverage, and you -- the onus goes up a little more if you're at 10x leverage.

So that's absolutely the case.

I would say right now, we are running with a low-enough interest rate risk position that we'd be comfortable moving higher in leverage without necessarily adjusting that.

But yes, I think it's very -- I think it's true that when you -- as you start moving to higher leverage positions, running things like 2- and 3-year duration gaps like some people in this space do, I think is not prudent.

But we're usually not in that position.

So I don't think from our perspective, we don't feel like we would have to tighten up our interest rate risk from where it is today to run lower -- to run higher leverage.

And our next question comes from the line of Bose George with KBW.

Actually, can you just talk about incremental spreads and ROEs on 30-year fixed that you're buying now?

And also, just in terms of incremental spreads, can you just talk about OAS versus nominal spreads, just how we should think about that?

Sure.

Thanks, Bose.

So yields on 30 year are 3.5%, for example, right now are around 3%; 5-year swap rates, around 190 basis points.

So yes, the spread differential between those 2 was about 110 basis points.

Now we wouldn't actually hedge them quite that long, and that also doesn't include any specialness or financing advantage versus LIBOR.

And so gross spreads on 30s are closer to 125 basis points.

And so with a 3% asset equity funded yield and a spread of 125 basis points, that gets you to gross ROEs still in the low double digits before convexity costs.

Okay, great.

And then just the OAS versus nominal spreads.

I mean, you guys said OAS widened about 20 basis points.

So just how should we think about how that translates into the incremental investment opportunity since, I guess, we have to think about the level with -- of which you're hedging as well?

I think if -- when you think about the OAS widening, option-adjusted spread, by its definition, is the spread after you take out the option costs associated with prepayments.

And so one of the key inputs into the option-adjusted spread is the level of volatility.

And so as volatility drops, that tends to help option-adjusted spreads or it tends to -- at the same static spread, an option-adjusted spread would be wider.

And so I think what we're -- what you can -- actually, if you wanted to think about putting together this discussion with the one earlier on swaptions, what we're doing right now is more in a low-volatility world, we're taking out a little of that sensitivity to volatility by hedging more of the option component.

But big picture, Chris gave you the static environment, which is more to what we hedge in terms of the spreads, and they're just not that different what from what we've been talking about 6 months ago.

And again, I think the key distinction is that's not the case in other products where you're seeing dramatic spread tightening.

So when you think about fundamentals, fundamentals for the agency mortgage market, it's not about credit, obviously, but it is about volatility.

And volatility coming down and feeling like it's going to stay here is a big fundamental positive for the agency market.

And I think sometimes that gets lost in terms of how favorable this backdrop is.

Okay, it makes sense.

And then, if you just -- going back to your comments earlier on the returns on the TBA dollar roll and where the hedge cost was being allocated, just how should we think about the incremental benefit from the TBA dollar roll position just on a kind of an economic basis?

So rolls are trading currently in the -- on production coupons in the, call it, 10 to 20 basis points through a mortgage repo, so that's kind of how I would think about the advantage.

It's not that different than where they averaged during the second quarter.

So -- and again, as Gary mentioned earlier, these are not extremely rich levels of specialness by historical measures.

So we think they're going to continue to trade in this range for the foreseeable future.

Okay.

And then, actually, one more, just to follow up on the -- your earlier discussion on leverage.

Just when we think about leverage going forward, is it fair given your commentary that leverage probably ticks up a little bit over time on your portfolio?

Yes.

I think that's a fair assumption.

I think that's definitely the message we're trying to send.

But I wouldn't -- I just -- I wouldn't think of it as a straight line.

I think it's going to be more opportunistic.

There may be periods where -- quarters where it could come down.

But like if you -- as we see the environment 2 or 3 years from now, I think we see a world where we're operating at higher leverage, and generally -- and that's the general trend.

But again, I wouldn't just expect to see it 0.25 turn higher every quarter or something like that.

The next question comes from the line of Dan Kelsh with UBS.

I wanted to see if you guys had given any thought to, I guess, in terms of like raising more capital even going back out to the preferred equity market.

We've just seen a few other names sort of in the broad space come in, whether it's Hanley, Two Harbors, et cetera.

And just seeing if you guys had given thought to that being just sort of another pocket of capital you could go to.

Particularly, it would cheapen up some of the rate you have in your existing preferred that I think it's callable now.

Yes.

Look, we certainly look at the activity that occurs in the space on all fronts, and I would say, up until maybe the last, say, few months, I don't think the preferred market was that interesting to us.

But levels have improved, and so it is something we would -- we certainly will consider.

And look, we have a fundamental, a key fundamental advantage in terms of as we raise equity, there is no incremental internal cost to -- operating cost to shareholders or it's negligible.

There's no incremental management fee versus the other people that you mentioned.

So for us, the equation is generally better than it is for our peers, both in the preferred market and the common market.

And so that's the aggregate picture of something that we certainly will look at.

And just to clarify, I guess, you're saying that you guys are not an externally managed REIT, which is why you don't have that incremental cost to shareholders that's punitive.

That's correct.

And again, given the nature of our business, if we were to raise equity, we're not generally hiring more people in our operating expenses.

And even like transaction costs and clearing and things like that, the changes are negligible to the bottom line.

So yes, essentially, you can think about it as if we raise new capital, there's essentially no management fee or no operating cost on the new capital, so it brings down your overall cost of capital.

Sure.

And I'm sorry to have bothered you with one last tidbit and that is with some of those preferreds that have come to market, there has been a change in structure, whereas a lot of folks, I think before probably 6 months ago, you could still get sort of fixed for life, and then it just continues to be a shift to a lot more sort of fixed to float.

Does that influence how you guys would think about accessing the market, just knowing that after a 5-, 7-, 10-year period, you might have a floating exposure as opposed to just kind of having you fixed for life?

Yes.

I mean, we obviously understand that the change in structure and -- but our business is managing interest rate exposure and converting fixed to floating and managing those kinds of exposures, generally.

So it's something that's relatively straightforward for us to price and factor into our capital structure, especially to the tune -- when it's a very small percentage of your overall capital.

And AGNC has one of the lowest percentages of preferred, kind of existing preferred, to our peers and so there's plenty of room to kind of to manage that exposure.

And your last question comes from the line of George Bahamondes with Deutsche Bank.

Just a question on repo costs.

Was wondering if repo costs have remained relatively unchanged in the third quarter to date.

Well, George, this is Peter.

They actually have improved so far this quarter from where they were at the end of the last quarter.

As I mentioned, the second quarter end was fairly tight in terms of repo funding, and there was some deterioration in the spread of repo costs relative to LIBOR.

Just to give you a sense, regular deliverable repo toward the end of the quarter, it was probably LIBOR plus 5 to 10 basis points.

For the repo that we were doing through our broker-dealer, it was close to LIBOR flat.

Those levels have improved 5 to 10 basis points since quarter end.

So we're funding through LIBOR by about 5 to 10 basis points through the FICC into the third quarter here; and about LIBOR flat to plus 5 or so for regular delivery repo.

So we expect it to continue to improve as we go through the quarter but not meaningfully from here, maybe 5-or-so basis points.

Okay.

Great.

And just a high-level question here.

How do you expect the Fed's balance sheet unwinding to impact the TBA market and dollar roll specialness?

I think that's a good question.

We've addressed it kind of on some prior calls as well as the topic of the balance sheet has come up.

I think a couple of things to keep in mind.

One is that well before the Fed even had a mortgage portfolio, dollar roll specialness existed.

And the #1 driver of dollar roll specialness is the fact that mortgages are originated a couple bonds out, and the originators, Wells Fargo and Chase and Quicken, they're hedging their pipeline by selling mortgages forward.

And because they're being produced a couple of months forward, then there's a natural discount associated with kind of that hedging process.

And that's the #1 driver of why there's consistently specialness.

Now I don't want to discount the value of the Fed.

The value that the Fed has brought is they're a current month bid and a large one.

But the bigger value has been that they absorb some of the cheapest-to-deliver pools or some of the faster prepaying pools.

And where that becomes very relevant is in a low-rate, higher-prepayment environment.

So where the Fed not being a player, let's say 2 years from now, would be a bigger issue on dollar roll specialness, would be in a very low-rate, high-prepayment environment.

Now one might argue that if we got in that environment, the Fed may actually not continue to roll off its balance sheet or might actually institute another QE, so they might not disappear in that environment.

But in the other environments where rates stay around here or go up, the prepayment environment is pretty benign, and the Fed's influence, therefore, is muted.

And then over the next year, it's important to point out that what Chris mentioned, that even in a tapering of reinvestments environment, they're still going to be buying over $100 billion in mortgages, probably more like $150 billion.

And then -- and again -- and in that kind of environment, for the next year, they're still going to be absorbing a fair amount of weaker pools out there.

So short answer is, it's a factor.

There are a couple of scenarios in a low-rate environment where it would be a much bigger factor.

But I think it's very manageable and not likely to have that big of an influence in most other scenarios.

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

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

I'd like to thank everyone for your interest in AGNC, and we look forward to talking to you again 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 10 by dialing (877) 344-7529 or (412) 317-0088, and the conference ID number is 10110075.

Thank you for joining today's call.

You may now disconnect.