Orchid Island Capital Inc (ORC) 2017 Q3 法說會逐字稿

Good morning, and welcome to the Third Quarter 2017 Earnings Conference Call for Orchid Island Capital. This call is being recorded today, October 27, 2017.

At this time, the company would like to remind the listeners that statements made during today's conference call relating to matters that are not historical facts are forward-looking statements, subject to the safe harbor provisions of the Private Securities Litigation Reform Act of 1995. Listeners are cautioned that such forward-looking statements are based on information currently available on the management's good faith, belief with respect to the future events and are subject to risks and uncertainties that could cause actual performance or results to differ materially from those expressed in such forward-looking statements.

Important factors that could cause such differences are described in the company's filings with the Securities and Exchange Commission, including the company's most recent annual report on Form 10-K. The company assumes no obligation to update such forward-looking statements to reflect actual results, changes in assumptions or changes in other factors affecting forward-looking statements.

Now, I would like to turn the conference over to the company's Chairman and Chief Executive Officer, Mr. Robert Cauley. Please go ahead, sir.

Thank you, operator, and as we've done in the past, I would request that our readers, if you haven't already done so, to download our slide deck off of our website. I'll be flipping through the slide deck as we go through the call and obviously, it will be much easier for everybody to follow along if possible.

If you're ready to go, I'm going to start on page 4. This is just a table of contents, just going to go over the order of the conversation today. The first thing we'll do is we'll give a brief overview of our financial highlights for the quarter. Then I'll spend some time talking about market developments, basically going over the developments during the quarter that provided the backdrop for our performance. I'll then go through our financial results, basically to go through how these developments affected us, and then finally, I'll go through the portfolio characteristic, credit counterparties, and hedge positions. This gives us a chance to describe how we manage the portfolio through this backdrop and how we're positioned for what we see going forward.

Turning now to Slide 5, just quickly go over this. We reported GAAP net income of $0.33 per share. As you're well aware, if you've been following the company for long, we use what is known as available for sale accounting. What that means is that any mark to market realized or unrealized gains and losses in both our assets and all of our derivative or hedge instruments. Go through the earnings and income statement. We reported losses of $0.18 per share over this quarter. Excluding those mark to market gains and losses, we reported $0.52 of earnings. We'll go through in a little more detail later on as I said.

Our book value declined $0.08 from June 30 from $9.23 to $915. We reported -- we declared and paid $0.42 of dividends and we had an economic return of $0.34 per share or 3.7 cents un-annualized, which is 14.7% annualized.

Book value per share was $9.23 at June 30, 2017. That's a decrease of $0.52 or 5.3% from $9.75 at March 31, 2017.

Dividend declared for the quarter, there were 3 for $0.42. And the result of all this is an economic return of negative 10 basis points or 1% annualized -- 1% un-annualized, 4.1% annualized. Year to date, our book value is down $0.95. Our dividends paid us $1.26, which results in a gain of $0.31 or 3.1% not annualized.

Now, let's talk about market development, if you'll turn to slide 7. The first thing we show here, what happened to the rates market. We have the 10-year Treasury note and the 10-year swap rate. As you can see over the course of the quarter, we had a dip both cases yields declining going into early September and then reversing. The end of the quarter we basically finished unchanged. In the case of the 10-year treasury, we were up slightly less than 3 basis points. In the case of the 10 year swap, even less, 1 basis point.

Basically, if you look at these results, you really see the tale of 2 stories. In the first case, the rally that we saw into early September was basically a continuation of things that had developed or started to develop in late March or late Q1. There were 3 basic drivers of what was going on.

The first had to do with the current administration and Congress. As you recall last year, when republicans swept the White House and Congress, there was a tremendous amount of optimism surrounding what was viewed as a very pro-business administration. There was talk of repeal and replacement of the Affordable Care Act, tax reform, infrastructure spending and the like. And over time, this optimism quickly turned to pessimism.

There were several failed attempts to change the Affordable Care Act. The process really exposed a tremendous amount of infighting between the administration and republicans and amongst republicans themselves, and there was even turmoil in the White House itself. And all of these seemed to undermine confidence in the eyes of the market that anything meaningful would get done.

We also had geopolitical events, particularly in North Korea and the Korean peninsula. There was a talk about a nuclear test and the administration was very outspoken through Twitter and otherwise, which raised tensions quite a bit and as we all know, there were tests -- missile launches over Japan and so forth.

And then finally, the third leg of that was just inflation. Realized inflation is reported in either personal consumption expenditures or the consumer price index. Starting in March, reported figures were 0.1% of a month, which drove year-over-year figures down sharply from the low to mid 2s down towards the mid-1.5, 1.7 range. And this was persistent but then things seemed to change.

In early September, they changed quite remarkably and ironically, the biggest event was the government passing -- increasing the debt ceiling and basically, avoiding a shutdown in the government at least into December. And ironically, just to show how much discord there was amongst republicans, this deal was the result of a compromise between the Trump administration and leading democrats in congress.

But it did stem the tide. We also had finally 0.2% of a percent increase in inflation on the CPI level, which seemed like maybe the decline in inflation had bottomed. It might turn around. And then finally, on the 20th of September, the FOMC met and at the conclusion of the meeting, Chair Yellen in her press release was quite hawkish. Made it quite clear that they intended to hike in December absent material erosion in the economic data and probably more importantly that they thought that most of the factors driving inflation down of late were temporary and transitory, that they thought they would reverse and move back towards their 2% target.

Since the end of the quarter, we've seen a continuation of very strong economic data, especially even today with the [GD Print] for the third quarter at 3%. There's renewed optimism that maybe they can get -- Congress can get a tax reform package through. Tensions in North Korea abated somewhat. While inflation is still low and the Fed is quite hawkish, the risk assets have done extremely well and as we all know that various stock indices were hitting all time new highs several times over the last few weeks.

With respect on slide 8 to the shape of the curve, we show the U.S. Treasury curve and the dollar swap curve. There was modest flattening that took place throughout the quarter. But it's noteworthy that if you look at the slope of the curve between the 5 and 30-year point, it's been flattening for quite some time. And in fact, on November 17th, or October 17th, got into the lowest point that it's been since late 2007. This was driven by the combination of the fact that we had low realized inflation and very hawkish leaning Fed, which meant that inflation risk premium was quite low so the long end of the curve rallied and the front end of the curve moved modestly higher to price in the hawkish Fed rhetoric.

Since then, we have backed off from those lows but they are still quite remarkable given where we've been. As far as Fed hikes, it appears quite evident now that December, the market is pricing at a very high probability, although it's quite remarkable that at the end of the second quarter, the market was barely pricing in one more hike through well into 2018.

Turning to Slide 9, we can go through the developments in the mortgage market and we have 4 tables here. The first on the top left shows the price performance of 30 year fixed rate mortgages, 4%, and 4.5% coupons. These are kind of our core holdings in addition to 20-year 4% securities, and as you can see, these prices were relatively flat over the quarter. In the case of 3 or 4s, they were only up a little over a tick and 4.5s were up about 3.5, which is in stark contrast to lower coupon mortgages, which tightened quite remarkably into the end of the quarter and beyond. This was really a result of the fact that most other spread products were trading at equally tight spreads and with low volatility in the market, the mortgages seemed very attractive. And it just seemed that the tightening could go on forever. Since then, we've backed off some but in the low vol environment, mortgages have appeared very, very attractive and done very well.

Bottom left shows the roll market for these coupons. Generally, people that do a lot of dollar rolling don't do so in these coupons, tend to be in lower coupons. But the takeaway from this slide is the fact that the drops were not particularly appealing during the quarter and that has relevance for how investors view the specified pull market. To the extent dollar rolls are week, specified pulls look attractive.

If you look to the right side of the page, this is the pay up for 2 of the typical specified pulls you can be. In the top right, this is the low loan balance pull or 85,00 KMAX pull. It's probably the highest quality form of call protection you can buy. These tend to trade at the highest premiums and at the other end of the spectrum, which we just call new production. These tend to pay at more modest pay-ups simply because their call protection is quite fleeting. It really only lasts for the first several months after the loans originated and then disappears quite quickly.

But if you look at the performance of those payouts, particularly the 85K payouts throughout the course of the year, you can see they were quite strong. Obviously, the market was rallying most of the year but even in September, when the market started to back off, they did not fall much at all and that's consistent with the fact that mortgages have done very well, but we've also seen capital rates in the REIT space by our peers and so there's a lot of buying that tends to take place of late. So that's also helped these pay ups stay quite strong.

Now, I'll start to talk about our financial results. If you turn to slide 11, we have 2 tables here. The first one on the left is our income statement and as I mentioned, we used the fair value of accounting, so that means that as I said, all realized and unrealized gains and losses go through our income statement and that can introduce a large amount of volatility on a reported headline earnings per share.

So what we've done here is basically decompose the income statement into these realized and unrealized gains and then the other more core components, if you will, just the headline interest income expense and our G&A expenses. I don't want to dwell on these too much other than to point out that it's important for you to recognize that the one on the left is more reflective of what we tend to earn over time.

The headline number interest income is just -- we don't -- fair value accounting does not require us to amortize premium through interest income so that number is quite large. It actually represents approximately $0.86 for the quarter. The second line is interest expense. We do not use hedge accounting so this just reflects absolute interest expense, and of course with the Fed raising rates that number has been creeping up.

And then finally, with respect to our G&A expenses, most of these are non-variable. We are externally managed so our management fee does vary as the size of the company grows, as does our allocated overhead. But the rest of those are relatively stable and in fact, on a year-over-year basis are flat to lower than they were last year. This column foots to $0.52 earnings for the quarter and we'll talk about that a little bit more.

On the right side of the page, we show the returns for the portfolio broken down by sector. We run both a pass through portfolio and then also a structured securities portfolio comprised of interest only and inverse interest only securities. The pass through portfolio generated a very strong return. The portfolio has a very high concentration of 30-year fixed rate mortgages with high coupon mortgages. So we have a lot of premium exposure and we generated a very strong for the quarter. That's even inclusive of realized and unrealized gains and losses, and derivative losses, which are basically our funding hedges, TBA shorts.

So the returns for the quarter were quite strong, which is what you would expect for a quarter such as we had in the third quarter. With respect to the structured securities, the returns were slightly negative. The biggest driver of that was mostly unrealized gains, in this case losses on the interest only securities, and that was driven primarily by fees that were a little on the high side and resulted in negative fair value adjustments. Combined, we generated a 4.5% return for the quarter un-annualized for the portfolio.

Page 12, as I mentioned, this shows our decomposition of our earnings per share. The yellow line represents our earnings absent fair value adjustments and the blue line is our head line or reported earnings per share. And I'll say a little bit about this more in a few moments but as you see in the last few quarters, there's been some slight erosion in our kind of run rate earnings. As I said, we don't use hedge accounting so any interest expense is not adjusted for any impact of our hedges. This is really just capturing the Fed increases, as I mentioned earlier. So there's been some slight erosion.

Turning to slide 13, this slide shows our capital allocation between the 2 strategies. The right side is a roll forward of the portfolios. Nothing much remarkable to point out here. I will point out that the allocation of the capital between the 2 strategy was basically unchanged for the quarter. The pass-throughs went from 63.6% to 63.7% so essentially unchanged. With respect to the roll forward of the portfolios, you can see that the pass-through portfolio grew. As I mentioned, we did not change the capital allocation so this just reflected a slight increase in leverage (inaudible) talk about in more detail in a few moments.

With respect to the interest only and inverse interest only, they were flat. We did have some additions to that portfolio. When we get into a little greater discussion of the details of the portfolio, we'll go into what we bought. But for now, you can just see that it was relatively flat. So the purchases were just offset by return of principal and mark to market losses. And that's basically it for that slide.

So now, we can go into our professional characteristics. Slide 15 is a snapshot of the portfolio at the end of the quarter. We show our allocation to the various sectors. I would point out that in the third column from the right is the allocation of the percentage of the portfolio. Before we get into the details of this, I think it's probably better to go to Slide 16 to just give you some context on how we are not just positioned now but have been in the past.

And there are 4 tables here. The top left just shows the size of the portfolio. So over time we've grown and the portfolio is approximately $4 billion. As you can see, in the last quarter it grew but as I mentioned that was just because of slight increase in leverage. We did not increase the shares outstanding during the quarter. The bottom left shows the allocation between the 2 sectors. As you can see, the pass through strategy is at the high end of the range and it probably will continue to be at least in the short-term. Hunter will have a few words to say in terms of relative value and attractiveness. The short answer now is that the structure portfolio, this security is just not necessarily appealing right now.

The top right shows the allocation within the fixed rate portfolio. The only notable thing here is that over the last few quarters, we've taken our allocation to 20 year securities down slightly, although it is up slightly this quarter. And most of those fixed rate securities we own, as I mentioned, are higher coupon and we've been buying generally lower pay out forms of call protection. We've shied away from the highest forms of call protection. Those levels are quite rich and unattractive at the moment.

And then finally on the bottom we show our allocation to our structured securities. Hunter, do you want to say a few words about what we like and don't like in that space of late?

Sure. So as I think Bob alluded to earlier, we saw rates go down in the mid-third quarter and start to sort of rebound. We've seen a continuation of that and inverse IOs and IOs in particular performed very well. We spent quite a bit of time and capital in the first quarter buying IO positions off of mortgage backed securities that had jumbo balances, so balances in the $500,000, $600,000, $700,000 per loan area. Those performed, as you would expect, relatively poorly as rates came down in the first 7 or 8 months of the year and then have since started to rebound quite nicely.

So it doesn't leave us with a lot of options with respect to the IO and inverse IO book. Inverse IOs have sort of run out of what we would call negative duration. So the ones we had with the exception of maybe some of the more generic inverse IOs we have off of Fannie and Freddie, 4.5 have really sort of their IO component has extended to sort of maximum or at least there's not much room left for it to extend at this point.

So that puts us in a position where those particular assets had more LIBOR risk than we like on the books with the mix of hedges that we have in place. So we either need to increase hedges or decrease or let run off the number of those assets. And we're still sort of noodling on which of those paths we're going to take. Most likely, we'll continue to have more or less the same position. We might let some of it run off or sell a few pieces, which just don't fit our risk profile at this point in time.

If we can find IOs, we'll certainly add them but they are hard to find and relatively tight in price. So we probably won't do too much there either. So I wouldn't be surprised if our allocation towards specified pools grew in the coming months, particularly if we raise any capital.

Thanks. If we can go back to Slide 15 for a moment, this -- as I said -- this is a snapshot of the portfolio. There's not a whole lot to talk about here. The third column from the right is the allocation. As you can see, the fixed rate, we're at 95.17%. Of course, that's of assets. We don't apply leverage to the structured securities. So in terms of the percentage of our assets allocated to that portfolio, it's very small. Capital is of course much higher. I also mentioned that we had been deemphasizing the 20 year securities. This particular quarter they grew slightly and as a result, our weighted average coupon of the fixed rates was 4.37%. It had been 4.42% last quarter. But as Hunter said, we'll probably continue to run the fixed rate portfolio at or above these levels.

And I think it makes a lot of sense given where we are for several reasons. One, we're at the end of the calendar year, so we typically tend to see a slowdown in prepayment speeds. For the most recent month, our fixed rate portfolio prepaid at 8.1 CPR. That's actually probably slightly on the high end of the range of what we realized for the year but we would expect that to moderate, and then also with the backup in the rates that should put even more downward pressure on speeds. And that as a result, allows to feel more comfortable owning those securities. And as Hunter mentioned, the ability of IOs to extend meaningful and provide operating protection is somewhat limited in this space. So they're really not an attractive asset to add to the mix.

Turning now to Slide 17, I did -- we have 2 things on this table. The first is just our repo counterparties and our weighted average borrowing rates. We have extended our weighted average maturity some. We've started to put on some longer-term repo. As you can see, the weighted average maturity in some cases is over 100 days. That's somewhat of a departure from the past where we've tended to fund pretty much exclusively in the 30 to 60 or 30 to 90 day range.

On the right hand, we show our leverage. As you can see, we took it up so question might be why would we do so and the reason is the following. As we mentioned, the market was rallying for the first almost 9 months of the year and our portfolio, because of its high concentration of high coupon fixed rate mortgages in IO demonstrated a very negative empirical duration. So in other words, book value was going down in the face of a rallying market.

When we got to the point in early September when the 10 year was on the verge of going through 2%, and the market had priced out all future Fed hikes, it seemed to us that absent a real rolling over of the economy to the point where the market was going to start to price in easing that rates on the long end of the curve could not go sustainably below 10% or 2% and that the market had pretty much priced out all of the Fed.

So given the fact that we thought it was hard to believe that the market could sustain below 2% for any length of time and our year dollar shorts and so forth had been hit so hard with the market pricing on some much Fed hiking that the risk profile going forward was very asymmetric and we were compatible taking up the leverage back to around 9%. And we'll say few words about how we view things today, but at that point that seemed like a very easy decision to make.

Slide 18, this is the last slide we'll talk about and this just shows our hedge positions. And this is a good point, once I go through this slide, we can talk about how we see the world today and going forward, and what that means for our not just positioning but earnings probably going forward.

So we basically show all of our various hedge instruments that we use here on the top left, or our Eurodollar positions and you can see we have $1 billion on each contract out through the December of '20 contract and we show our average rates. Below that, we have our treasury futures short, which is the TY contract. We took that down quite a bit. It used to be around $465 million and just like we had the same discussion about the change in the leverage. When we got into late August, the curve had flattened quite a bit and the market was pricing up so much Fed hiking that we just thought it was not the appropriate point to hedge and we moved our hedges out of the 10 year part of the curve. We put on more of a 30-year TBA short, which is shown in the top right corner because mortgages had tightened and it seemed like the market was just headed on an ever flattening trend. So we were not getting any satisfaction. In fact, the mark to markets on those were quite negative on the long end of the curve.

Below that, we show our swap agreements. We have slightly over $1 billion in place mostly between the one and 5 year parts of the curve. Our weighted average pay -- these are pay fixed swaps and our weighted average rate on those is 1.43. And then finally, below that swaption. We added to our swaption positions in August. As I mentioned, we had changed our TY hedges. We had seen the belly of the curve lead the rally as the market priced out the Fed and vol was at extremely low levels. It just seemed like the perfect opportunity to put on some swaptions whereby we have the option, in both cases, for a year, to enter into some form of a pay fixed swap. We entered into 2 of these, one on the 5 year part of the curve, the other on the 7 year part of the curve.

Those of course have done well and even beyond the end of the quarter have done well. But this is an interesting point. So that's basically our positions. And so one of the things that's been a challenge for us and all of our peers, frankly, has been the fact that we've been in a long period of time where we've faced an upward sloping yield curve and the need to hedge. So what does that mean? We're either paying fixed on swap or entering into Eurodollar shorts or the like and in all cases, these are done at levels above current funding levels and the forward curve has not been realized at all up until recently, which meant that these hedges tended to end and be closed out, out of the money, and they added to our interest expense.

And so in our Q, we show extensive tables that show what we call our economic interest expense, which is just our absolute interest expense adjusted for the fact that our hedges were expiring out of the money. And that number has been increasing. But as we look into 2018 that's not so much the case anymore. The Fed has started to raise rates and we're not sure how much that will continue to do so, but now, these hedges aren't necessarily out of the money. Our swaps have a picked fix rate of 1.43% and a receive rate of 1.31%. Now, that?s 3 month LIBOR. Today, 3 month LIBOR is 1.37%, 1.38% and there's the prospects that the Fed may hike in December, in which case it would likely go higher.

Our Eurodollar contracts had a strike rate of 1.62 for this year, at least for the December contract, which is out of the money, but for 2018, that rate is 1.845%. So at least on the Eurodollars and the swaps, our blended rate for 2018 is 1.64%. So what does that mean? Well, today, as I said, 3 month LIBOR is at about 1.37, 1.38. But as the markets priced in a very high probability of a December hike, it's been moving up every day.

If in fact, the Fed were to hike in December, and given what we've observed in prior periods, in other words if you go back to before the period where the Fed was being priced in, 3 month LIBOR was trading about 16 basis points above effective Fed funds. So effective Fed funds right now is about 116 basis points. If the Fed were to hike another 25, it takes you up to about 1.41. If the relationship between 3 month and LIBOR and Fed Funds holds, in other words about a 16 basis point cushion, that would get 3 month LIBOR at the end of the year in the mid to high 1.50s, which would put our swap positions in place and our Eurodollar positions close to in place.

Now, we don't hedge 100% of our liabilities. We -- using all of the various Eurodollar swaps and swaptions and so forth, we're at about 63% of our funding and inclusive of our TBA hedges about 71%. So it's not 100% but at least for the first time in a while, these hedges have gotten to the point where they're actually going to actually help us on the interest expense side.

Now, as we all know, the Fed has been raising rates and there's prospect for doing so in the future, and it's put some pressure on earnings. We publish in our queue, as I mentioned -- and I apologize it's not available for you yet today but I can give you some information that's in there -- but our economic interest expense and our economic yield on our assets for the first 9 months of the year has moved from 114 basis points for the first 9 months of 2016 to 162 basis points for this year. So it's gone up by 48 basis points.

Our yield on average MBS assets has moved from 380 basis points to 404 so a 24 basis point increase, which reflects a change in the mix of the assets, a slightly higher coupon. So on net, our net interest margin has declined by 24 basis points, basically reflecting the fact that the Fed has been hiking.

But what does that mean going forward? Well, based on our positioning, we talked about how the asset mix is unlikely to change. We are taking a lot of prepayment risk. We think that's prudent given the back up in rates and the seasonal slowdown that we typically see. But also, we now have hedges that are going to start to help us some, at least to the extent we're hedged. And what that means for our NIM going forward is really just a question of what happens to the shape of the curve. As we mentioned, it had been flattening for some time but that has changed somewhat of late.

Last Thursday night, the Senate passed a budget resolution, which cleared the way for potential tax reform. Yesterday, the House did so as well. So now, there is the potential -- who knows what we'll actually get -- but there is a potential for meaningful tax reform, which could be stimulative to the economy. As we mentioned, we've seen some acceleration in the growth of the economy, 2 straight quarters of growth over 3%. All of the economic data of late has been very, very strong. So it is possible that the market starts to re-steepen the curve and that allows our asset yields to creep up slightly higher. That would be both positive from an earnings perspective as well. So we could see that offset the effect of the Fed.

We don't know what will be the case. Time will tell but we think that our positioning is appropriate for this environment and glad to see that our hedges are finally going to have some benefit for us. That's pretty much it. Operator, we're done with the presentation. We can open up the call for questions.

(Operator Instructions) And our first question comes from the line of Christopher Nolan with Ladenburg Thalmann.

In your comments, is it correct to say that if the Fed tightens in December, the Eurodollar futures start to go in the money? Did I hear you correctly?

Not quite. Weighted average strike for next year is 1.845% and that of course is -- the underlying index is 3 month LIBOR. It would be still out of the money with that one hike but it's getting closer. We don't know what the Fed will do next year, whether it's 1, 2, 3, hikes or none. But the breakeven given where we -- assuming the Fed hikes in December would be 1.5 hikes, more than 1.5 hikes would get those in the money.

So it would start not the December hike, but if we get a normal pace of hikes, say quarterly or something to that effect. It doesn't take very many before it turns and goes the other way. Also worth noting that while we're not hedged 100% on the front end of the curve, we have taken some steps to pretty dramatically increase our longer-term funding positions.

So we put a, I don't know, it was a few hundred million dollars' worth of 1-year REPO on at sort of the rate lows. That struck in, I think, in a blended basis in the low 1.50s. So those should be in the money and we've also been rolling -- the December hike is still not 100% priced in and so we've been taking advantage of that by locking in rates in the 1.39, 1.40 area, which take us well into January.

It looks like you guys really slowed down your equity issuances this quarter. Is that correct and is there a particular reason?

Yes, that was a product of 2 things. One was just the fact of the stock price through most of the quarter was not trading a very high premium to book value and then secondly, we had a lot of tightening in the mortgage space. As you know, some of our peers were active in raising capital and that put some pressure on pay up premiums for the very specified pools. So the combination of a modest premium to book on a stock price and not necessarily really attractive investment opportunities due to the crowding out of our peers, we just didn't think it was a good time to do so.

We've seen a widening in mortgage assets since quarter end and also we have -- our stock price is trading relatively well, especially today. If that holds in then I think that creates a different dynamic than we saw in the third quarter.

You have a fair amount of cash liquidity on the balance sheet. Should we expect that to be deployed?

Not necessarily. That could just affect timing, unsettled security purchases, and the like. It might be just modestly above where we would typically keep it. So no, absolute capital raising, I would not read too much into that unless there is capital raising. The cash balance today is not reflective of imminent new investments as of 9/30.

Final question. In your questions, you indicated that you were decreasing your exposure to 10 year futures and going to 30 year TBA shorts. So is it fair to say that you're expecting a steeper yield curve in the 30 year part of the curve?

Well, we did until about 2 -- a week or so ago. What had happened, we had such a pronounced flattening of the curve, those shorts were getting very painful to own and mortgages were tightening at the same time. So it just really didn't make sense to continue to have that hedge to that magnitude at that point in the curve. And by the way, when we're shorting TBAs, we're shorting the 3% coupon, which is a pretty long duration asset. So to the extent the market turned around and went meaningfully the other way, you would expect that 30 year mortgage, the 3% to extend coupled with the fact that it had been tightening.

So we just thought it was a more effective hedge instrument and we also entered into some swaptions. The belly of the curve had led the rally into early September. The middle of the curve is the most sensitive to long-term Fed expectations and as the market was pricing out the Fed, the belly of the curve was rallying quite strongly and volatility was very, very low. And so that was just an easy trade to get into. So they just seemed to be more effective hedges at the time.

If I could just chime in on that, I think that what we saw was as rates were coming down toward the low 2%, the convexity in our portfolio gets pretty bad. So I mentioned we had some of those jumbo IOs. We have alto of prepaid sensitive assets. So if rates were to break through sort of the psychological barrier of 2% hedging with treasury note futures is not necessarily what we want to be in.

So mortgages have tightened like crazy into the third quarter. Volatility was very low so that was really more a function of -- that was really more where our hedges were than anything. One, we wanted to shift, reduce our basis of exposure by getting out of treasury backed MBS product or treasury backed product and into MBS hedges. Vol was very low so we thought it was a good opportunity to maybe buy a little bit back so we bought the 5 and the 7 year swaptions and then that also allows us to improve the convexity of the portfolio because now as on more MBS portfolio, we would expect underperform into more of a rally. But into a selloff, we would have a little bit of positive convexity through purchasing option hedges or conditional hedges as opposed to just straight duration.

And they were out of money too. So they were sort of in the position where their durations could go from being relatively low to sort of increasing at an increasing rate as we sold off.

(Operator Instructions) Our next question comes from the line of David Walrod from Jones Trading.

You mentioned there wasn't a lot of capital raising done in the third quarter but there was quite a bit in the second quarter. Was there any leftover hangover or earnings drag that was realized in the third quarter from deploying the capital that was raised in the second quarter?

That's typically the case especially because it was so back loaded. Most of that capital raising was May and June. There's slight downward pressure. It only lasts a month but effects July basically.

And then your speeds ticked up and I know you said you think that will move back down but have you attempted to quantify the impact on this quarter's results of the uptick in speeds?

Well, we have the one number, which as you know, we call it premium loss due to pay down because under the fair value option. We don't amortize premium directly and interest income and that was probably about $1 million, slightly above last quarter. And maybe $1.5 million at the moment. And in terms of we bought 45 million shares outstanding so 2% or 3% increase. So it definitely had some slight modest impact and as I said, we do expect that to reverse in this quarter. But don't forget, we hit the year to date low in rates in early September, 2.04% on the close. So and the speeds in the past, structured (inaudible). So I would expect that to be the high watermark for the year and so premium amortization, even using our methodology, was probably peaked for the year.

We're pretty bullish on speeds going into the next 3, 4, 5 months. So especially at today's rates levels, 2.40 or above 10 year treasury levels hold. One, we're going to have the seasonal slowdown, which has been more -- the seasonal nature of pay downs has been more pronounced it seems like in recent years than was previously the case and also we've had -- we've moved over a very [cuspy] part of our portfolio in terms of where mortgage rates are now and we would expect to see declines in speeds, particularly in the IO book.

We have a follow-up question from Christopher Nolan from Ladenburg Thalmann.

Dave just answer my question.

And I am showing no further questions at this time. I would now like to turn the call back to Mr. Robert Cauley for any closing remarks.

To the extent anybody has any questions that pop up later or they don't listen to the call live and listen to the replay and have questions, feel free to call us in the office. The number is 772-231-1400. Always willing to take any questions. Otherwise, we look forward to speaking to you at the end of the year. Thank you.

Ladies and gentlemen, thank you for participating in today's conference. This does conclude the program and you may all disconnect. Everyone have a great day.