AGNC Investment Corp (AGNCO) 2019 Q1 法說會逐字稿

Good morning, and welcome to the AGNC Investment Corp.

First Quarter 2019 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, Carrie, and thank you all for joining AGNC Investment Corp.'s First Quarter 2019 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 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 May 9, by dialing (877) 344-7529 or (412) 317-0088, and the conference ID number is 10129960.

To view the slide presentation, turn to our website, agnc.com, and click on the Q1 2019 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 the call include Gary Kain, Chief Executive Officer; Bernie Bell, Senior Vice President and Chief Financial Officer; Chris Kuehl, Executive Vice President; Peter Federico, President and Chief Operating Officer; and Aaron Pas, Senior Vice President.

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

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

In the first quarter, we saw a near complete reversal of the volatility and risk off mindset that characterized Q4.

U.S. equities recovered the vast majority of the losses suffered late last year.

Interest rate volatility also declined materially while credit spreads retraced the bulk of Q4's widening.

Agency MBS spreads benefited from the decline in interest rate volatility but generally lagged the tightening in credit-centric sectors.

The one major difference in markets over the 2 prior quarters was on the interest rate front.

As you recall, interest rates fell materially in the fourth quarter in conjunction with the overall risk off move in equities and credit-sensitive assets.

In Q1, however, instead of interest rates reversing the fourth quarter move, like we saw in other financial assets, rates continued to decline with the yield from the 10-year treasury falling 28 basis points to 2.41%, a long way from the 3.24% yield we saw during the fourth quarter of 2018.

The rally in interest rates over the last 2 quarters has been substantial and reflects both the reduction in the global growth and inflation outlook as well as the change in the Fed's monetary policy stance.

At the end of Q3, the Fed was fully committed to continuing to raise interest rates with more than 4 rate hikes priced in over the next 18 months.

By early in the first quarter, the Fed had reversed course and communicated an essentially neutral stance on interest rates.

The Fed also decided to end the reduction of its balance sheet in September of 2019, significantly sooner than what market participants were expecting.

Lastly, almost all of the other major foreign central banks also acknowledged the deteriorating global economic landscape and correspondingly adopted considerably more dovish stances on monetary policy.

The way I would summarize the change in the economic landscape over the past 2 quarters is this: the equity and credit markets have recovered as the more dovish central bank policies and corresponding lower interest rates have offset rather than eliminated the current concerns around the global economy; the markets are clearly not expecting more accommodative monetary policy to produce a return to global synchronous growth or materially impact the inflation picture.

This is evidenced by both the flat yield curve and the fact that longer-term interest rates are much lower than they were 6 months ago.

So what does this mean for AGNC as we look ahead?

A stable interest rate environment with a Fed that is unlikely to either raise or cut interest rates is clearly a major positive for our portfolio.

It significantly reduces the likelihood that our economic returns will be negatively impacted by interest rate volatility.

Said another way, the probability of us achieving attractive economic returns is materially higher when interest rates are stable and convexity costs are minimal.

With that said, it would be wrong to characterize the current market conditions as a Goldilocks environment.

The incredibly flat yield curve clearly limits our net interest margin potential; and recently, our repo funding levels relative to LIBOR have deteriorated.

In addition, the rebound in Agency MBS spreads during the quarter while boosting book value also reduces the potential returns on new investments.

Consistent with this backdrop, we were able to generate very strong economic returns of 7.3% during the first quarter as book value increased 4%.

That said, our expectation for go-forward ROE unlevered MBS investments is approximately 2% lower than what we anticipated at the beginning of the year, predominantly as a function of the unexpected compression between our funding levels and 3-month LIBOR.

Additionally, tighter spreads on Agency MBS and higher spec valuations are also factors.

Consistent with this new return outlook, we anticipate lowering our dividend to $0.16 per month or a $0.48 per quarter run rate, beginning with the dividend we declare in May.

The $0.48 still translates to around a 10.5% dividend yield at our current stock price.

We believe these returns are still attractive against the backdrop of the rebounded valuations for all financial assets coupled with lower interest rates and reduced risk premiums.

It is also important to reiterate in light of the expected stability in the interest rate picture, convexity costs should be low, and we believe the probability of achieving these returns is higher.

At this point, I'd like to turn the call over to Bernie to review the results for the quarter.

Thank you, Gary.

Turning to Slide 4, we had total comprehensive income of $1.22 per share for the quarter.

Net spread and dollar roll income excluding catch-up am was $0.52 per share or $0.01 lower than the fourth quarter.

As Peter will discuss shortly, our net spread and dollar roll income was negatively impacted during the quarter by higher funding costs.

Our net spread and dollar roll income was also negatively impacted by faster prepayment expectations as a function of lower rates.

Our accounting yields are based on forward prepayment assumptions rather than actual CPRs for the quarter.

Consequently, although, our actual prepayments declined during the first quarter by over 100 basis points to 6.3%, our forward-looking CPRs increased to 2.5 -- excuse me to 10.5% as of the end of the first quarter from just under 8% last quarter and had the effect of lowering our current asset yield even after adjusting for catch-up am by approximately 4 basis points.

Together, these 2 factors, higher funding cost and faster prepayment expectations as well as lower dollar roll returns, acted as a drag on net spread and dollar roll income and more than offset our higher operating leverage during the quarter.

Tangible net book value increased 4% to $17.23 per share at the end of the quarter as Agency MBS spreads recovered some of their fourth quarter widening.

Our net book value further benefited from our specified pool holdings, which significantly outperformed generic pools during the quarter, given their greater prepayment protection in the current low interest rate environment.

Including the increase in our tangible net book value and $0.54 of dividends declared per common share, AGNC generated a positive economic return of 7.3% for the quarter.

Moving to Slide 5. With our fourth quarter capital raised fully deployed, we operated with an average at-risk leverage ratio of 9.3x our tangible net equity for the first quarter versus 8.4x for the prior quarter, ending the first quarter at 9.4x.

Lastly, during the first quarter, we issued $235 million of 6.875% fixed-to-floating rate preferred equity, which we believe will be accretive to earnings available to common shareholders given the sizable spread between the yield and the preferred stock and expected ROEs on new assets.

With that, I will turn the call over to Chris to discuss the agency market.

Thanks, Bernie.

Turning to Slide 6, you can see that rates continued to rally during the first quarter but despite lower rates, which elevated prepayment risk for certain sectors of the market, mortgages performed well, reversing a good portion of the weakness experienced during the fourth quarter.

The move lower [in] market implied rate volatility was the primary driver of MBS outperformance versus hedges.

Again, with the large change in volatility, various spread measures of MBS performance including OAS and static spread changes did not provide a comprehensive picture of MBS performance.

As you can see on Slide 6, option-adjusted spreads widened during the first quarter, however, this is due to lower model option cost given lower market implied volatility.

When we run option-adjusted spreads without adjusting volatility inputs, results are very different with spreads roughly 5 to 10 basis points tighter during the quarter.

Given that we don't hedge a significant portion of our exposure to implied volatility, the 5 to 10 basis points of tightening is consistent with the improvement in our net asset value.

Another important driver of our net asset value improvement during the quarter was the outperformance of specified pools.

As I mentioned, on the call last quarter, we expected that higher quality specified pools would perform well, given the deterioration in the quality of TBA float, lower interest rates and weaker roll implied financing rates.

Specified pools did in fact outperform TBA during the first quarter with the average pay up on our portfolio increasing just under half of a point, which significantly exceeded hedge ratio implied price changes.

While valuations on specified pools are likely to be well supported going forward, given favorable supply and demand technicals, at current levels, the trade-offs versus prepayment benefits in pricing are now more balanced.

Turning to Slide 7. You can see in the chart on the top left of the page that the investment portfolio increased to $102 billion as of March 31, consistent with the increase in leverage and deployment of preferred capital raised during the quarter.

The size of our TBA position was relatively unchanged over the quarter at $7 billion however, we did shift the composition in favor of lower coupon 30-year MBS.

Roll implied financing rates for production coupon 30-year MBS traded with little to no specialness versus repo during the quarter, and over the near term, we expect that will continue to be the case.

Lastly, as Bernie mentioned, you'll notice that our long-term prepayment estimates increased as of quarter end, largely due to the move lower in rates, however, given the composition of our specified pool holdings, we expect that prepayment speeds on our portfolio will remain well contained.

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

Thanks, Chris.

I'll start with a brief review of our financing activity.

Our average repo funding cost in the first quarter was 2.64%, up 25 basis points from the prior quarter as the financing markets fully reflected the Fed's December rate hike.

Also contributing to the higher funding cost was the persistent period end funding pressure that caused both January and March funding levels to be somewhat higher than expected.

Repo rates over quarter end were again elevated but to a lesser extent than what we experienced at year-end.

With the Fed on hold for the foreseeable future, I expect our average repo cost to trend lower by 5 to 10 basis points over the next several quarters.

If the Fed does in fact remain on hold, it would be logical for most short-term rates to compress toward 2.5%.

Our aggregate funding cost was also adversely impacted by the significant decline in 3-month LIBOR during the quarter, which dropped from around 2.8% at the beginning of the quarter to just under 2.6% at the end of the quarter.

The drop in LIBOR reflected both the dramatic shift in the Fed short-term rate forecast as well as greater global bank liquidity following the ECB's announcement to reopen its long-term lending facility.

With 3-month LIBOR falling throughout the quarter, the received leg on our swap portfolio was lower than expected.

On Slide 9, you can see how the spread between our repo funding rate and 3-month LIBOR has shifted significantly over the last 5 or 6 quarters.

With repo rates being higher in the first quarter and 3-month LIBOR rates being lower, the favorable funding dynamic that we have enjoyed in 2018 has reversed and in fact, turned negative in the first quarter.

Looking ahead, consistent with the Fed and ECB likely being on hold throughout 2019, I would expect the spread between our repo cost and 3-month LIBOR to stabilize somewhere closer to 0 or slightly negative as compared to the 15 basis point positive differential we were expecting to earn as we began the year.

As Gary alluded to earlier, this deterioration in the funding equation negatively impacts our expectation of go-forward ROEs by a little over 1%, all else equal.

It is also important to note that while government funding levels have deteriorated somewhat, repo availability continues to be extremely favorable and margin requirements continue to decline.

These positive developments are supportive of our business over the long term.

Turning to Slide 10, we provide a summary of our hedge position.

Our hedge portfolio totaled $72 billion at quarter end and covered 77% of our funding liabilities.

The reduction in balance and hedge ratio is consistent with the shortening in our asset duration due to the rally in rates and with the more benign short-term rate outlook given the Fed's neutral policy stance.

On Slide 11, we show our duration gap and duration gap sensitivity.

Given the changes we made to our hedge portfolio, our duration gap remained unchanged quarter-over-quarter at 0.2 years.

In the current environment, we continue to believe that maintaining a positive duration gap has important aggregate risk management benefits given our view that mortgage spreads will likely widen in a rally and tighten in a selloff.

With that, I'll turn the call over to Aaron, to discuss our nonagency portfolio.

Thanks, Peter.

I'll provide a quick update on our credit investments.

Our credit portfolio totaled $1.8 billion at the end of the first quarter or roughly 4% of capital, up modestly from $1.6 billion in Q4 of 2018.

The majority of our credit portfolio is comprised of investments in GSE credit risk transfer securities, which referenced the performance of loans guaranteed by the GSEs.

The balance of the portfolio is split between the selective mix of post crisis RMBS and CMBS.

With the pace of household formation outstripping new construction, we expect housing fundamentals to remain strong.

In addition, some of the concerns that cropped up late last year regarding affordability are now somewhat mitigated by the significant decline in mortgage rates.

Within the housing market, we believe credit exposures tied to lower priced homes is attractive relative to jumbo balances as the favorable fundamentals driven by the supply demand imbalance favors lower-priced homes.

After a significant widening in Q4 2018, credit spreads largely reversed in the first quarter with investment grade CDX retracing the Q4 move completely and high yield retracing about 75% as of quarter end and roughly 100% through yesterday's close.

Residential and commercial AAA and credit spreads also tightened meaningfully in Q1.

But not on pace with corporate credit spreads.

As the equity markets generally sit at their peaks, we are approaching the tights in some credit sectors.

So where does that leave us today with respect to ROEs in the residential credit space?

Generic on the run, CRT M2s, which typically carry a B, BB rating are currently trading around LIBOR plus 200 basis points with repo rates just inside of LIBOR plus 100 and a 20% haircut.

This translates to a gross ROE of about 7.5% to 8% using 2.5 to 3 turns of leverage.

New issue RMBS subordinate tranches trade in the same levered return context, using leverage amounts that we view as appropriate given the longer spread duration and inherent idiosyncratic risk in these smaller pools.

For AGNC, given our small credit allocation, we're able to use corporate level funding almost exclusively for this portion of the portfolio by putting more Agency MBS out on repo.

This can increase ROEs available to us by about 200 to 250 basis points depending on leverage amounts and prevailing nonagency repo rates.

Given current spread levels on Agency MBS and opportunities available in credit, we'd expect the growth in this part of the portfolio to be measured in the near term.

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

Thanks, Aaron.

And at this point, I'd like to open up the call to questions.

(Operator Instructions) First question will come from Doug Harter of Crédit Suisse.

Gary, can you talk about how this environment, while volatility is low today it seems like you have bouts of spikes of volatility.

How do you look to manage the portfolio on this type of environment?

To your point, volatility is low today.

It was high last quarter.

We always have to manage a levered portfolio, assuming conditions can change.

That said, I mean, we are -- we also have to be practical that there's been a fundamental kind of change in, let's say, base expectations on interest rates.

For the last 3 or 4 years, there's been this consistent we'll say hawkish Fed with the intention of taking back accommodation, raising rates, shrinking the balance sheet.

And it's been a question of how quickly and concerns about how high interest rates can go.

I mean this is -- we again, we need to be practical.

This is the first time in 3 or 4 years where the Fed is essentially, neutral.

So I think that's a big picture change.

But I do want to get back to your question and get back to the -- to how do you manage the portfolio kind of with the mindset that the base cases that rates aren't changing but always knowing that the base case doesn't always work out.

And I think one thing that's important is -- we always say when we manage the portfolio, we always have to assume something, rates can change.

And so when we look at an environment that's going to create falling rates, for example, in -- at this point, if rates were to fall materially, it is probably another flight to quality, a weaker global growth picture, weaker U.S, potentially a recession and in an environment like that we're going to see equity prices falling, we're going to see credit spreads widening.

It's logical to assume agency mortgage spreads while they'll outperform other credit-centric products, they are still going to widen in that kind of environment.

Vice versa, if we see interest rates go up 50 basis points, retrace kind of the move that we've seen over the last couple quarters, we're probably looking at a very kind of healthy economic environment, one where credit spreads are going to tighten further, equities are doing very well.

We would expect Agency MBS spreads to do well especially, at some of the kind of the prepayment risk that -- I mean it's not -- it hasn't -- it's not at a peak or anything like that now, but it's clearly ticked up a little bit.

We get to a very, very comfortable prepayment environment.

So in that scenario, we would expect mortgage spreads to perform well and tighten.

The reason I went through those 2 examples is to get to the one thing that's different in terms of how we're looking at hedging the portfolio and as you can tell from our disclosures, we've essentially reduced our hedge ratios.

We've made an effort to stay -- to have a little bit of a positive duration gap that would not have happened without rebalancing activities, both last quarter and this quarter.

And we did that because of that spread sensitivity.

Again, we're trying to look at kind of maintaining the value of the portfolio in both directions.

If you think spreads are going to widen when rates fall then that argues for somewhat of a positive duration gap, again, especially, when you think that spreads could tighten if we sell-off.

But that's really the main difference.

But I can -- I do want to reiterate just that, again, this has been a period of -- we've gone through a period of 4 years where the question was, how fast and how far the Feds are going to go and in the last basically 3 or 4 months that period has likely come to an end.

The next question will come from Bose George of KBW.

I just wanted to ask about incremental spreads and ROEs.

I mean, you guys said ROEs are down 2% from year-end.

Just, can you sort of narrow it down where sort of current returns are?

Sure.

I mean, if you think about a logical mix of assets that we could -- we would be purchasing, you're probably looking at a spread of, let's say, give or -- conservative maybe 90 basis points.

When you look at 9 to 10x leverage and then you add just under a 3.5% yield, you get to an ROE we'll say around 12.

And then I think what's important is, you have to look at AGNC's expense ratio, which is only 80 basis points.

You're looking at a net ROE that's still 11 or slightly better.

And look, we feel that's attractive.

One thing that people should really think about with respect to agency mortgage product versus credit.

Why do you get paid?

Why is there with a completely flat yield curve and with a government-backed product, why is there still 90 basis points of spread?

The reason is, it's because when you buy agency mortgages, you're short a prepayment option, right?

And if you think about that, you're short -- that prepayment option is tied to interest rates.

And so when you think about a Fed that's on hold and when you think about a stable interest rate environment, Agency MBS are a major beneficiary of low volatility because we're short these options.

And so in a world where volatility is expected to -- interest rate movements are expected to be low and where -- and all central banks are essentially saying that and there's a high hurdle to kind of -- to change things, a major beneficiary is the Agency MBS market and in particular, a levered portfolio of Agency MBS.

And so I think from our perspective, yes, we were benefiting from this kind of funding advantage, which was repo relative to 3-month LIBOR and as we've discussed, that's deteriorated to back to kind of average levels, certainly much better than where it was still in like 2014 and '15, but not where it was last year.

But big picture, the environment that we're describing makes sense, which is the potential returns in a low volatility environment, where spreads on other products are contained, the potential return is a little lower but the probability of an attractive return is higher.

I mean the thing that hurts our x -- our realized returns is lots of volatility in interest rates and convexity cost.

And we're probably looking at a world where those are going to be to the very low end of kind of historical norms.

So I think that's the best way to characterize the environment we're in.

Okay, great.

That's helpful.

And actually just on leverage, the current run rate, is that good or what -- where do you see that, [because you did] take it up or down from here?

Look, I think the bias is still higher for leverage.

First off, I think we're pleased with the timing of -- we had talked about raising leverage probably for a 1.5 years and said it was a logical thing, we should be doing it.

We feel pretty good about the timing of when we did that because returns are definitely lower than where they were once spreads were wider 3 months ago.

But I think the bias is still higher, but we'll be opportunistic with respect to moving leverage.

And the one thing that's interesting is when -- we're not an outlier in any way, shape, or form on this front.

If you look at our hybrid peers on average or many of them if you just look at their capital committed to the agency sector, they're already running higher leverage and have been for a while higher leverage than where we are.

So I think, again, just from our perspective, the bias is still higher but we'll be opportunistic.

Okay.

I could just, one more on funding through Bethesda.

The changes -- proposed rules changes through FICC, does that change anything in terms of how much you could fund through there?

No.

Our expectation Bose is that we're going to continue to run.

Right now, we actually increased it to around 40%.

It could go up another 5% or 10% but I think in this 40% range, that's a safe place to be for right now.

Your next question will come from Rick Shane of JPMorgan.

Thanks for taking my questions.

They were on leverage and trajectory, and you've answered them.

Thank you.

Your next question will come from Trevor Cranston of JMP Securities.

Question on the repo rate and the spread versus LIBOR.

Peter talked a bit about the longer-term expectations for how you think that'll trend versus where it was at the end of the quarter.

But can you talk a bit more about what you think has been causing the increased sort of pressure in the repo market around quarter end.

And also, maybe provide some additional cover if there's been any sort of downward trend in it so far this quarter?

Sure, Trevor.

First off, there was pressure at the end of the first quarter, and I think the market was generally spooked by what happened at year-end and of course year-end has its unique challenges with respect to balance sheet and concerns around composition of balance sheet.

So year-end was unique, and it did sort of carryover a little bit into January.

We saw markets stabilize in February.

But the turn itself for the end of the first quarter traded throughout the quarter at an implied rate of 4%.

On the actual day, the average funding rate was closer to 3.25%.

And that's -- so that's down from the average at year-end, which was 4.25%, thereabouts, give or take.

So there was 100 basis point improvement but still 3.25% is well above the average that we had.

I do expect it to sort of stabilize more going forward.

One of the issues that the market faced in the first quarter was the technicals weren't great.

There was about $150 billion of net treasury supply that the market had to absorb and that used up capacity following the issues around year-end.

That technical should change actually from a seasonal perspective almost completely in the opposite direction in the second quarter.

So far this quarter, we've seen some improvement.

There was a big outflow, about $50 billion of money, out of money funds ahead of April 15.

So that caused repo rates around the tax payment date for the week or 2 before, and the week or 2 after to be higher.

But I do expect them to continue to trend lower.

Just sort of order of magnitude, I said 5 to 10 basis points over the next several quarters, I think in the second quarter, we're looking at a couple of basis point, 2 to 3 basis point improvement and then gradually move lower into the 150s.

The repo curve itself from 3 months out to 12 months is actually already inverted in light of the Fed expectations.

So 3 months marginal funding is around 255, 256 and '12 months is around 250.

So that's why I do think that repo rates, LIBOR rates, they're all going to sort of compress around 250 as we go forward.

Okay.

Got it.

That's helpful color.

And then one question on the hedge portfolio.

The hedge ratio, looking at the chart on Page 20 dropped pretty meaningfully.

And you guys -- you talked about wanting to have some positive duration, to offset potential spread widening if rates rally.

But then I was looking at the rate sensitivity table, and it shows that the book value declined for the down rate scenarios is a bit higher than it was prior quarter.

So just kind of wondering how to reconcile that with the lower hedge ratio and how to think about that?

Yes.

Obviously, we had to do a lot in the quarter to change the aggregate size of the portfolio and the duration, as Gary has mentioned.

We lowered our duration of our hedge portfolio by about a year, full year, which offset the contraction in our asset portfolio, and we ended the quarter at 0.2 years.

Subsequently, we have increased it further and right now, we're operating with a duration gap of about 0.5 years so that gives us, just, sort of going back to Gary's point, we do have more protection than what's implied in this graph on Page 11.

As you can see on Page 11, at the end of the quarter, the contraction and the expansion in the portfolio was almost identical at about 2 years.

So in a sense it's saying that, that was sort of the peak negative convexity point of our portfolio, and we moved significantly in both directions.

But we have shifted it a little further longer to give us a little more protection for the down rate scenario at the expense of the up rate scenario.

But just keep in mind, those moves, the negative convexity in the portfolio is nonlinear and so at higher interest rates, you're down -- you have a bigger cushion in terms of a down rate scenario.

Where we ended the quarter was sort of made the 100 basis point decline in interest rate to take you to a lower level of interest rates where essentially the whole portfolio is refinanceable and the duration of the portfolio is as short as it can get.

So some of what you're seeing is just nonlinearity in that calculation.

But to Peter's point, the key there is, we're sort of splitting the difference from a risk perspective with a slight bias toward being a little longer and that bias comes from the spread directionality that I talked about earlier.

Your next question will come from Matthew Howlett of Nomura.

Gary, just on the comments of volatility.

You look at the Fed and their reduction in the balance sheet.

Do I -- can I read through your comments as sort of implying that volatility has already been priced in or the Fed's been effectively priced in or is there still some uncertainty with the path of their reduction going forward and the impact that could have on volatility going forward?

No.

I think it's priced in at this point.

They've announced that they're going to end the shrinkage of the balance sheet in September.

I think that, that's largely a known quantity.

I mean they're going to continue to run off the mortgage portfolio.

Again, these things could change if there is a big move -- if they have to cut rates and if they implement some type of QE or something like that.

But I think practically speaking, the balance sheet move is priced in.

Where it may not be fully priced in is in, let's say, the money markets and on the funding side.

I mean when they stop running down the treasury portfolio that could help the repo piece a little bit.

And that wouldn't be kind of priced in on a forward basis.

So that's really the main area where I think there that we could see some benefits toward the second half of the year.

Okay.

And just real quickly on the creditors.

I appreciate you going over on the ROE's on the levered credit piece.

There's been more of the GSE selling down the capital stack and making it more sort of REIT friendly.

What are your thoughts of adding more subordinate credit risk to the portfolio?

This is Aaron.

So -- like you pointed out, we referenced ROEs for the M2 part of the capital structure.

Your point's valid.

The GSEs in particular, one of them has been selling further down the capital structure.

So those opportunities are out there.

We think when we combine kind of, while we do have a positive outlook on housing and the borrower, when you couple that with where we are in the credit cycle and where other credit products are trading, we think a lot of those instruments further down the capital stack are probably trading through where they should.

So we don't [love] them from a risk return perspective.

Your next question will come from Mark DeVries of Barclays.

Could you talk about where you think your realized returns could end up compared to kind of your expected returns?

If we saw further rate rally with the 10-year maybe getting to 2% or even 1.5%, and just talk about kind of where you think returns would be impacted in that type of scenario?

Sure.

I mean -- so first off, that's a realistic or it's a potential scenario if people are calling for recessions in 2020 or whatever.

I mean that obviously could be a driver of a move like that.

And so obviously, it depends on the specifics of the drivers to get you there.

It would also depend on the Fed's response function.

But practically speaking -- and I guess the biggest variable might be what happens with the yield curve in that scenario.

But big picture, I think investors should expect mortgage spread widening in that scenario and -- but on the other hand, if you look at kind of the hits to the portfolio just implied by interest rates alone in our table on Slide 26, I think that will overstate the negative impact on the portfolio because of rebalancing activities.

We would do significant rebalancing activities, which would keep that number lower there.

So all in, people -- what you would expect to see is that we would be entering a very good earnings environment because mortgage spreads would be wider, prepayments risk would be priced to kind of the max.

And so I think we would be looking at a mid-double-digit earnings environment.

But there would be some manageable hit to book value in that transition.

That would be my expectation, kind of based on the caveats that it would -- how the rally unfolds will certainly affect things.

Okay.

But when you talk about kind of a mid-double-digit earnings potential.

Is that -- are you referring to just the potential on new investments as opposed to the existing portfolio?

I'm really talking about -- in a way, it would be -- if you assume let's say there is a 5% hit to book value than you've now repriced your existing portfolio to market.

So I'm essentially, talking -- yes, it applies to new investments but from the book value level, I think that it would -- you could apply it to the whole portfolio.

So said another way, I think what would be a logical market response would be that, and I'm going to use an example, book value drops 5%, okay?

And that's just an example, stock price in theory might not really decline, in that the stock price is sitting there and saying the earnings environment now looks really good.

This is -- we're in the midst of a credit event or a recession.

This is a sector that can produce excellent returns.

So what -- so that would be a very logical response that price-to-book ratios expand in that kind of environment because of the expectations of higher go-forward returns.

Okay, understood.

And then, sorry if I missed this, but could you talk about what kind of fundamentally [or typically] caused the evaporation and the specialness that you saw in TBAs and why you kind of expect that to persist?

Yes.

Sure.

This is Chris.

So we think rolls are likely to remain cheap to historical averages for, I mean a, couple of reasons, given the quality of the cheapest to deliver float, which has deteriorated without a Fed reinvestment bid.

It's certainly been a factor.

And the marginal bid for mortgage supply has also been generally from TBA, rolling investors and so that's been a negative supply/demand technical.

Over time, I think these factors will balance back out but over the near term, I don't see the dynamics changing all that much.

I do think long run, we still expect that eventually, we'll get back to levels of specialness, consistent with call it 20-plus year historical levels of around 10 basis points or so through repo.

For the -- just given the structural dynamics of the TBA market that we've discussed at length in the past.

Okay.

And how much did that contribute to the deterioration and the return you expected versus the issues that Peter talked about with repo and LIBOR?

So TBA specialness was a small component -- in a way, we had reduced our TBA position to a much smaller position.

So TBA specialness was a component of it.

But in terms of -- it was not as unexpected.

I think the 2 things that were -- that developed kind of we'll say later in the quarter.

I mean, actually if you look at Slide 9, what you can see on the funding side is actually in January, the funding picture was improving sort of as we expected and then really over the second 2 months of the quarter, you see the straight line down where it deteriorated.

So that was sort of a big kind of a big change we'll say in the second half or second 2/3 of the quarter.

And then with the further rally in rates, the amortization expense picked up as well.

So those were more intra-quarter change -- more of the intra-quarter kind of dynamics than the specialness on rolls.

We've 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 participation on our first quarter 2019 earnings call.

And we look forward to speaking with you again next quarter.

Thank you.

The conference has now concluded.

Thank you for attending today's presentation.

You may now disconnect your lines.

Have a great day.