Ready Capital Corporation

Joseph Lloyd McAdams - Chairman, Chief Executive Officer and President Joseph E. McAdams - Chief Investment Officer, Executive Vice President and Director

Steven C. Delaney - JMP Securities LLC, Research Division Daniel K.

Altscher - FBR Capital Markets & Co., Research Division Douglas Harter - Crédit Suisse AG, Research Division Michael R. Widner - Keefe, Bruyette, & Woods, Inc., Research Division Jason Stewart - Compass Point Research & Trading, LLC, Research Division

Thank you very much, Mike, and good morning, and -- or good afternoon, ladies and gentlemen. I'm Lloyd McAdams, and I welcome you to this conference call.

We will summarize the company's recent activities and answer your questions about both the past and the future during our question-and-answer session. First, I will briefly describe the recent results and events, and then I will comment on our current plans.

During the second quarter of 2013, Anworth earned net income to common stockholders of $21.6 million, which is $0.15 per diluted share. This amount includes a realized gain of approximately $2.1 million, which is also approximately $0.01 per share.

Stockholder equity available to our common stockholders at quarter end was approximately $856 million, which equates to a book value of $6.01 per share. This is all based on 142 million shares of common stock outstanding at June 30.

This also represents a decrease of 14.6% from our book value at March 31, 2013, which was $7.04 per share. The unrealized loss component of our book value is, at quarter end, $82.7 million, which means that the balance of our book value, excluding this component, was $938 million.

Based on what I've recently read and heard, I believe that it is fair to say that for many investors, the increase in mortgage interest rates during the second quarter and the decline in the value of agency MBS and then the declines in the book value of many mortgage REITs was a significant event, resulting in the sharp decline in the prices of these stocks. As you know, Anworth's stock price declined from $6.33 on March 31 to $5.60 on June 30.

This is a decline of approximately 11.5%. This morning, our stock was trading at $5.10, a further decline.

Later in the call, we look forward to answering your questions relative to these recent events. Next, I would like to talk a bit about our mortgage-backed security portfolio at quarter end.

The fair value of our portfolio of agency mortgage-backed securities at quarter end was approximately $9.45 billion, which we can assign to 4 broad categories of agency mortgage-backed securities. One thing I would like to point out, and I think most of you are aware, we historically, during our earnings call, have described 3 or 4 categories, but in our financial reports, the press release which we issued yesterday and financial statements which we've filed, we do have 9 categories of mortgage-backed securities, 2 of which are fixed rate and 7 of which are adjustable rate mortgage categories.

But for the purpose of our discussion here, we'll talk about the 4 broader categories. And they are ARMs whose interest rate will reset within one year, that is 18% of our portfolio.

The second category is hybrid ARMs whose interest rates will reset between 1 and 3 years. That's 19% of our portfolio.

And then we have the hybrid ARMs whose interest rates will reset after 3 years. That's 42% of our portfolio.

And finally, the fourth category is both the 15-year and 30-year fixed-rate mortgage-backed securities, whose interest rates never reset. That's 21% of our portfolio.

One observation I might make about this last category, it consists of 2 very different parts: 15-year mortgage-backed securities which we have purchased in the last few years; and some 30-year mortgage-backed securities which we purchased many, many years ago, a smaller part, but the average coupon is in excess of 5%, which makes them quite unique. Before these -- because these 4 broad categories have quite different price sensitivities to interest rate changes, the declines in prices that resulted in the declines in book value during the second quarter were quite different amongst the categories.

As an example, the one-year ARM category declined in value by about 0.5%. Another category, the hybrid ARMs whose interest rates reset between 1 and 3 years, declined by about 0.8%.

On the other hand, hybrid ARMs whose interest rates reset between 5 and 10 years, declined in value by almost 3%. And the 15-year fixed-rate category, it also declined in value by about 3%.

I might note that the 30-year smaller fixed-rate category declined considerably less. The observation about these different price changes for these ARM mortgages largely rest in the fact that the price changes of all mortgage-backed securities change differently in different yield curve environments.

Whether it's steep, flat, inverted, the price changes of these various ARM categories will definitely be different. Specifically, shorter ARMs relate more to the short end of the yield curve, which of course, is controlled by monetary policy, and longer reset ARMs relate more to the longer end of the yield curve, which is controlled by global inflationary expectations.

As these 2 forces move in different directions and different degrees, the yield curve moves in changes that would not be called parallel shifts, but we call shifts that torque. So as these 2 forces torque the yield curve into various degrees of flatness and steepness, this large -- this torquing largely determines which group of ARMs will perform the best during that environment and which group of ARMs will perform the worst.

Interestingly, I don't know if any of these ARM categories that we list that would be expected to perform the best in all yield curve scenarios. And I guess, for that reason, I don't know any that would be expected to perform the worst.

Talk a bit about our financing of our mortgage-backed securities. The purchase agreement financing of our agency mortgage-backed security assets, which was approximately $8.4 billion at June 30, was 8.7x our total equity, which consists of common stockholders' equity, plus our preferred stock equity, plus our long-term junior subordinated notes.

Floating to fixed rate interest rate swaps of $3.435 billion, which represents approximately 41% of our outstanding repurchase agreement balance as of June 30. Anworth, as I believe most of you know, is, largely because of its age, different from most mortgage REITs in that a significant portion of our portfolio, the approximately 18% that I mentioned earlier, consists of seasoned adjustable rate mortgages which we have owned for many years and whose coupons contractually reset over the next 12 months to about 1.5% above LIBOR, give or take.

This rather unique interest rate adjustment feature results in these securities having materially less price sensitivity to changes in interest rates than either the fixed-rate mortgage-backed securities or most newly issued hybrid adjustable rate mortgage-backed securities, whose interest rates are, of course, fixed for several years until they start to reset. Equally important is this part of our portfolio, when we acquired it many years ago, we acquired it at dramatically lesser prices than they trade today in the market.

So when we evaluate the interest rate sensitivity of the price of our total portfolio, this short ARM category contributes very little to the portfolio's overall duration or price sensitivity to interest rates. This is why we have, for a long time, in these calls also presented the ratio of our interest rate hedge position to repurchase agreement balance, both with and without these presently low-duration categories.

At June 30, without including these assets, the ratio increases from 41% mentioned earlier to 51%. From my risk management perspective, I view this latter statistic as probably the most informative.

And while the size of our swap position is a significant component of our balance sheet, I believe that the position's average remaining term of 36 months is more significant. Management's current decision about the size and term of this swap position has and will probably continue to be the major determinant of the return on equity that we achieve, and therefore, the dividend yield that we're able to pay to our stockholders.

We should note that when determining the nature of what we believe is a prudent interest rate hedging strategy, we take into account specific characteristics of the assets which we've invested in, in each of these categories of agency mortgage-backed securities. I might point out that during the second quarter, our 3.4 billion of swaps increased in value by $41 million.

This is approximately 4.2% of our beginning common equity at the beginning of the quarter. Now I would like to take a few minutes discussing the company's income earned during the second quarter of 2013.

Our net interest rate spread increased from 0.89% in March 31 to 1.00% at June 30. This increase occurred mostly because of the recent rise in interest rates, which has materially reduced the probability of prepayments from almost every 15-year fixed-rate and hybrid mortgage-backed security that we have purchased during the past 18 months.

This significant change reduces our FASB 91 prepayment assumption, and this reduces our premium amortization expense. The weighted average coupon of our portfolio of agency mortgage-backed securities was 2.72% at quarter end, a slight decline.

The weighted average term to reset on Anworth's adjustable rate Agency MBS was 44 months at quarter end. After adjusting for interest rates paid through our swap transactions, the average interest rate of our repurchase agreement liabilities was 0.95%.

And the average term to interest reset of our liabilities was 471 days. The prepayment rate of Anworth's portfolio of agency mortgage-backed securities that occurred during the quarter was approximately 24%.

The average amortized cost of our portfolio was 103.16%, which is a slight increase from the previous quarter. I'd like to take a few more minutes before we take questions and discuss our current plans.

In this regard, I believe there are several important subjects. First, as I've described in my letter to shareholders published in our annual report for more than a decade, our earnings level is determined by 3 primary variables: interest earned, interest paid and the amortization of mortgage-backed security premium purchase.

As we reported yesterday for the second quarter of 2013, our net income relative to first quarter was, in various degrees, influenced by each of these 3 variables. Interest rates received declined as ARMs continued to reset during the quarter to lower yields, and new investment yields were lower than our portfolio average.

I might point out that there is a probability -- a better probability that next quarter, this may not be the case. But there was a 14 basis points decline in interest received.

Number 2, our borrowing costs decreased mostly from our higher swap agreements being replaced at lower cost. The decline was approximately 8 basis points.

Most significant, however, was the dollar amount of premium amortized decreased as our assumptions for ongoing rates of refinancing decreased, largely due to a significant increase in longer-term mortgage interest rates during the quarter. That was a 17 basis points decline.

Secondly, as I have said during our last call in April, if the Federal Reserve System's program of aggressive acquisition of mortgage-backed securities and its multiple quantitative easing programs reduce for whatever reason, or even a hint that they will be reduced, I believe there are parts of the yield curve which would be more vulnerable to a larger correction in price than others. Maybe the first time I've ever quoted myself in an earnings call from the last quarter, but this hinting actually did occur during the second quarter of 2013.

And the markets did respond by raising interest rates as discussed previously. Since Mr.

Bernanke quickly hinted again that he didn't intend to raise rates too soon, interest rates are now more stable. But from my perspective, the cat is out of the bag.

And I think that it is likely that the only words of a new Federal Reserve Chairman could allay the fear now felt by the bond market bulls, who have supported the markets rally in 2012 and 2013. When that day comes, I'm not sure.

But from what I understand, it's sooner rather than later. These comments indicate to me that most mortgage rates will not likely go up or down a lot during the balance of the Bernanke era.

For us, the best part of price stability is always that each passing month will likely result in our existing hybrid ARM portfolio having less interest rate sensitivity as each of these securities moves closer to its interest rate resetting date. As a third item, about plans for the future, is that the growing scarcity of 5-year hybrid ARMs experienced during the past year may actually dissipate, as rising 30-year mortgage fixed rates, which clearly occurred in a significant way during the second quarter, with, I believe, the Freddie Mac rate increasing by almost 100 basis points -- that the spread between this new mortgage rate, 30-year fixed and the 5-year hybrid ARM, whose mortgage rate did not -- whose rate did not go up as much, will widen to the degree that for a homeowner, there may be more inclination to take out a 5/1 ARM, so maybe the originations will increase to some significant degree.

If this occurs, I see this as a favorable prospect for our strategy. That's the end of my prepared remarks.

I would now like to turn the floor over to Mike, our moderator. And for the question-and-answer session, with me here today is Joe McAdams, our company's Chief Investment Officer and a Director of the company; Thad Brown, our Chief Financial Officer; and Chuck Siegel, our Senior Vice President of Finance.

Thank you, Mike.

The first thing I'd like to ask you about is leverage. It's understandable why it increased to 8.7x.

You were able to and elected to maintain your portfolio at $9.5 billion. But I wonder, as we sit here today and we're moving into the new quarter, can you talk about how long that the team is comfortable operating with leverage, say, 8.5 or higher.

And if conditions don't change, meaning bond prices improve or whatever, to bring that down the way you'd like for the leverage to come down, should we assume that you'll be selectively paring the portfolio back over the next 1 to 2 quarters?

Steve, this is Joe McAdams. As you pointed out, our leverage at June 30 was 8.7x our long-term capital.

I think we have talked for a while that, given the type of portfolio we have, it's expected both absolute and hedged price volatility, as well as liquidity and haircuts that we see in the repo marketplace, that 8x leverage is a comfortable point for us. It is something that we view to a certain degree as a target.

We've operated below that target level for a while, partly because our portfolio was appreciating, and the unrealized gains in the portfolio was pushing the leverage down. And also, because in some cases spreads on the types of investments we were looking to make were narrower than we would have liked to see in the marketplace, and over the past several quarters had realized some gains on certain parts of our portfolio as a result.

But with the decline in the net value of our portfolio, over that 8x target, we certainly are looking to move back towards our target leverage of about 8. Subsequent to June 30, I made reference to it in the press release, we have had some sales of securities.

About -- close to $250 million of securities were sold. In addition, we've utilized paydowns that have come in, in the month of July to pay off some repo.

So where we stand now, our leverage is approximately 8.3. And I think it's important to have some good discipline in this process of managing your liquidity.

As you're aware, the market did trade off fairly significantly on July 5, after June 30, but has stabilized some since then. So we have looked to pare some portions of the portfolio.

One of the benefits of having a portfolio that has a relatively high organic prepayment rate from our ARMs is that we do have significant cash flows coming back on a monthly basis.

Well, it sounds like you've already made good progress. And, possibly, the payoffs over the next couple of months could get you close to where you'd be comfortable, at your 8% target.

I was curious, the -- everyone who -- we've had earnings reports thus far that's involved with hybrids indicated, they called it disorderly or severe dislocation. I'd just be curious to your thoughts as to, have you seen hybrid Z-spreads and I guess, bid-ask spreads as bad as -- at any time in recent history?

I guess, maybe not since 2008. Just any color you can give us, Joe, on where you saw the market in late June?

And, I guess, more importantly, are you seeing any signs of an improvement in liquidity or firming up of prices since June 30?

Sure. Obviously, there was underperformance in the hybrid sector.

As you've pointed out, Steve, it's been discussed here and elsewhere. There was I think -- even though these were securities that have shorter durations than the fixed-rate portion of the marketplace, I think the underperformance, on an absolute basis, was pretty similar.

We saw about 5/8 of 1 point of underperformance from our ARMs -- I'm sorry, price decline in our ARMs, which given the move in rates, we wouldn't have expected any price change. Whereas with some of our longer hybrids, they were down, in some cases, close to 1 point more than we would've expected purely based on duration.

So I think that underperformance has stabilized this quarter. I think that our book value and the marks we get from brokers at quarter end were reflective of the underperformance in the ARMs market at June 30.

And I think spreads have been more stable since, certainly since the first week of July, whereas, we still have seen some additional underperformance in the fixed rate sector so far this month. Relative to your liquidity question, the ARM, the hybrid market is not -- clearly is not as liquid on a day-to-day or a minute-to-minute basis as the fixed-rate market.

I don't -- it's clearly, there were some parts where there were wider bid-offer spreads than we would have expected or been used to seeing. But at the same time, I think the relative level of underperformance of ARMs was in line with the broad mortgage market as a whole.

I do think that given the smaller niche of the market that we operate in, I did feel towards the end of June and then in the first week of July, that there was some more disorderly widening in ARMs than maybe we saw in the fixed-rate market. But in terms of the overall level of liquidity, I feel it's sort of what we would have expected to see given what went on in the marketplace.

I think that given that you have a smaller section of the market, maybe a less -- I'm sorry, more concentrated group of traditional buyers, many of whom were probably not actively buying new securities during late June and early July, I think that explains some of the maybe additional underperformance versus our expectations going in.

And that traditional buyer would be banks primarily, right?

Right, banks. And again, on the margin, I think mortgage rates are still a marginal buyer of hybrids, but I think an important marginal buyer.

Something that I thought was noticeable in the quarter was the share buybacks. And I think there's maybe been some hesitation in the past few quarters to actively go out and buy stock.

So I guess, what changed this quarter beyond the share price dipping lower? Was there anything that sparked and said "Hey, this is -- we have to buy back stock at this price?"

This is Lloyd. No, there was nothing that said, you had to buy back stock.

We make a decision based on the amount of free equity capital that we need to have to meet margin calls, the attractiveness of securities that we might buy, and then, lastly, the price of the stock. I guess you could always say in hindsight, when you buy your stock back at a discount to book and then the book goes down, as with all investments, there's always hindsight to think about.

We are no different than anybody else. We usually want to buy back stock when we think we can sell it later and, hopefully, sooner at a higher price.

We're no different than any other investor. Obviously, we thought we could sell the stock at a higher price, and we bought it during the second quarter.

Clearly, we cannot sell it at a higher price. But that's pretty much the way we think of it as making a decision.

If the stock had gone up, it would have been a much more attractive investment.

Got you. And, yes, that make sense.

Now second question is on prepayment rates. I'm wondering, because I thought your comment about scarcity of 5/1 ARMs going to be -- not occurring as much anymore.

But when we think about CPRs with rates higher on an absolute basis, how do you think that's going to move the CPRs for ARMs? Are folks going to go out there and see an opportunity to -- or get scared and refinance their ARMs into a fixed-rate mortgage because they're scared of rates moving materially higher from here?

Or do you think it's the exact opposite, that CPRs are going to start to come down materially because absolute rates are higher?

This is Joe. In terms of -- it's a very good question, to think about the dynamic of CPRs on our ARM portfolio versus the fixed-rate portion.

Our overall premium amortization expense, as Lloyd pointed out, is a function of the actual prepayment experience of the quarter, as well as our long-term prepayment assumptions that drive a level yield. So while the significant increase in mortgage rates we saw during the quarter led us to, on the margin, reduce some of those long-term prepayment expectations, which resulted in lower premium amortization expense for the quarter, we still saw, for the most part, unchanged prepayments during the quarter.

And we may still see similar levels of prepayments for the next month or 2, simply as the effect of this increase in rates works its way through the pipeline of new mortgage refinance applications. So I don't think we'll see an immediate decline in prepayment rates due to this move up in rates.

It generally works with a several-month lag. That's my -- I guess, my first point.

My second, maybe getting more to the nuance of your question, is we have, in the past, seen a case where ARM -- some ARM burrowers who are looking to refinance, either into a new, longer hybrid or into a fixed rate, sometimes a higher move in that headline 30-year mortgage rate can push him them off the fence to perhaps refinance, if they've been putting it off or waiting. So sometimes with ARMs, we see a short-term increase in prepayment rates before we see the expected decline.

But in general, what we would expect to see would be lower prepayments on our hybrid ARM portfolio during the fixed rate -- the fixed coupon, fixed interest rate period on those mortgages. And then if there is going to be a significantly higher reset, once it moves into the ARM portion, to see an acceleration of prepayments potentially into fixed or potentially into new hybrids around that point.

So there are some different dynamics of prepayments. But overall, we would expect lower longer-term prepayment rates on our ARM portfolio.

It's Lloyd. I'll just make one -- I made reference, too, that we got lower interest income during the quarter.

A part of that is if you had an ARM that's fully resetting, and the margins can differ, this summer, you just had your ARM reset at 2 5/8% or 2 3/4%, that's your new mortgage rate for the next quarter. And during the quarter, the interest rate on the new 30-year fixed-rate mortgage, the Fannie Mae commitment rate that I made reference to, or maybe it was the Freddie Mac commitment rate, is almost 4.1/2%.

So the decision to give up a 2 3/4% ARM and go into a 4 1/2% 30-year fixed, probably has a lot to do with your conviction as how high interest rates will go, and can you afford to make a higher monthly payment for the next few years? Some people will make a decision that they don't want to deal with that higher monthly payment and will stick with their ARMs.

And that leads to the type of consumer behavior, I think, that we're referencing here. There's a good chance that the prepayment rates will stay lower, simply because the mortgage rates are quite low on fully indexed ARM investors right now.

Fully indexed investors, but also, homeowners.

Just a clarification on the premium amortization change. Was the benefit you saw this quarter, was that a sort of a one-time catch-up?

Or was that kind of the rate that we should expect going forward?

This is Joe. There's -- anytime we have shifts in our long-term prepayment assumptions, there is a component of that shift that, I guess, you could describe as a one-time or catch-up component.

We did not make significant changes in our long-term prepayment rates. So, certainly, some component of that, you could say, would be viewed as not repeating.

I would say, if our prepayment -- actual prepayment rates don't decline, we would expect to see our quarterly premium amortization rate rise from that $16 million. I wouldn't expect it to rise back to the point it was before without prepayment rates increasing.

Great. And then can you just talk about what you think -- where you see your duration gap today?

And how that might have changed since March 31?

Sure. At March 31, our duration of our assets was a little over 1.5 years, and we had a duration gap of approximately 0.25 year.

Given the backup in interest rates, and particularly the backup in -- the increase in mortgage rates, which drive our expectations of prepayments during the quarter, the duration on our assets extended by approximately 1.25 years. So out to about a 2- and 3-quarter year duration at June 30.

So our gap widened from a 0.25 year to almost 1.5 years. That is based on our history of fairly significant interest rate gap.

It's one that we will be looking to take steps to narrow. As we mentioned, subsequent to June 30, we've had some asset sales which have reduced that duration somewhat, as well as entering in some -- some additional swap agreements that have narrowed that gap down by about 0.25 year since June 30.

One of the factors we take into account given our portfolio, is looking where specifically that -- the drivers of that widening interest rate gap our. In fact, the largest contributors to the increase in duration of our portfolio was, actually, from the 3/1 and 5/1 hybrid portions of our portfolio.

Partly because they're significant parts of our portfolio, and also that we had fairly high prepayment expectations, and therefore, very low, in fact in some cases, close to 0 duration expectations at March 31. So while we have a gap that's wider than our typical range of 1 year or less, and we will be continuing to move our portfolio through new acquisitions, new hedges, as well as any additional potential dispositions in the portfolio, to bring that gap more in line.

We do take some comfort in the fact that the areas where this additional duration has come in the portfolio is generally on the shorter part of the yield curve. Cash flows that'll be benchmarked off of 5 year and shorter rates, where given our view on interest rates and potential for any additional curve steepening should be parts of the curve that we wouldn't expect to see hit significantly hard.

And in fact, that have performed pretty well even with the pretty large selloff in the marketplace.

Let me just first follow up on the premium amortization, because I'm not sure if I've become more clear or more confused. But let's assume a base case of rates kind of being stable here, which would imply, I think, that prepayment speeds slow, not necessarily this month but over the coming quarters.

Is that $16 million premium amortization indicative of what you guys would sort of expect, given that stable rate scenario? Or were you -- I mean, I thought maybe I just heard you suggest that they would actually go back up from there?

And then that's where, I guess, I'm sort of getting confused.

Yes, this is Joe. Sorry if I wasn't clear on my answer before.

We had a situation during the second quarter, where our actual prepayment rate was little changed from the quarter before, yet we had a decline in the amortization expense during the quarter. So the reason that happened was predominantly because of the change in our long-term expectations.

So in that sense, there's this idea that it is a -- that it's related to the change in your prepayment assumption. If next quarter, we left our new lower prepayment assumption unchanged and prepayment rates continued to be unchanged, we would see a slightly higher rate of premium amortization than that $16 million.

So in that sense, there is a sort of one-time event to that change. But what I was hoping to make clear when I answered the question the first time was that we do expect to see the near-term speeds start to slow.

And when we see lower actual speeds, that will, in our opinion, support that level of premium amortization on a quarterly basis going forward.

Okay. And I mean, I guess, the piece that I was a little confused on, is there's -- I think as you said, there's a one-time component in that as well.

So what I think I'm hearing you say now, and correct me if I'm wrong, is if things progress as you would expect, as I think most of us would expect, that prepayment speeds do start to trend lower from here, that the $16 million is, roughly speaking, all else equal, kind of what we should think of as the run rate, because that embeds your expectation of sort of slowing prepays.

That's correct. That's our expectation.

Okay, great. And then, I guess, I just want to follow up on the hedging a little bit.

I mean, with the book value down pretty significantly, I think it's fair to say that the hedges weren't completely effective, and yet, your duration gap by your own estimates has kind of widened out. And then as I look at the actual composition of the swap portfolio, it looks like as you get out into the over -- well, even the 3 to 4 years, as well as the over-4-years bucket, you're relatively light there, where as you've been kind of adding to the longer-duration hybrids as well as the 15-year fixed bucket, which would kind of suggest you're adding duration.

And so, I guess, I'm sort of struggling with what to make of it, if they were ineffective to begin with, or less effective than I think you'd probably like to begin with and durations have extended, then where does that leave us now in terms of exposure relative to where we were at the beginning of the quarter?

Sure. Well, when we think about the -- let's just talk about the past quarter first.

Our overall portfolio, price change was -- went from 104.75% to about 103% on average across all of the sectors of the portfolio. So a 1.75 point price decline.

If you simply looked at our measures of duration, which there was a fairly narrow gap at the beginning of the quarter, and even taking into account that the rates moved about 70 basis points higher if you look at swap rates during the quarter, and taking into account that due to the negative convexity, the change in prepayment assumptions that the marketplace would have that, that gap would widen out some over a move of 70 basis points. We would have expected our -- given what happened with the move in rates, we would have expected the decline in our portfolio to have been about 1 point, or maybe even slightly less than that.

Given our swap portfolio, we would have expected that swap position to offset about 0.5 point of that price decline. So starting with a fairly narrow asset/liability mismatch, having a pretty significant rate move that would open that mismatch up some during that period of time, we would have expected to see some decline in the net value of our portfolio.

The fact that our portfolio was down 1.75 points instead of 1 point or less is a function of this -- the fact that mortgage rates, in general, whether you want to call it basis widening or underperformance, the fact is that a mortgage rate for a similar duration security went up by more than a similar expected maturity treasury or swap rate. And that is a portion of what isn't hedged by our swap position.

So there's that, and that was the significant driver of the decline in price. That if there was 1.25 points of net price -- net unhedged price decline, I would say 0.75 of it was related to things that we weren't, at the end of the day, expecting to be caught by -- captured or offset by our hedges, and about 0.5 point that was related simply to the move in interest rates.

So that drove a quarter. We haven't seen this subsequent to June 30, as getting back to your question.

We haven't seen in our portfolio anything like that similar magnitude of underperformance relative to benchmarks. Rates haven't moved a lot since June 30.

We have seen some underperformance in fixed-rate securities. And so we see that 15-year pass-throughs are down about 0.75 point in the quarter, even though swap rates are relatively unchanged.

So there still are some aspects of spread widening that we've seen so far. So that's sort of where we -- how we've gotten to where we are.

In terms of looking forward, you're right, we do have an interest rate gap. I do feel a significant portion of that gap is still cash flows that are 2 to 5 years, in that range, where we do still have some significant hedges.

We still do have a large portion of our portfolio prepaying at a -- still in the 20s. So declines in prepayments based on these increases in rates will still result in fairly significant prepayment rates, even if they're shorter than they are before.

So I feel comfortable in where our overall hedge position is. You are correct that, given this backup in rates, and we do have more of a gap.

We have put on some additional swaps this quarter, out in that 4- to 5-year sector. And I think we'll continue to do so on a margin to manage that duration gap.

I do -- just to sort of summarize, I think a significant amount of our duration and our duration gap, given our portfolio, are in relatively short-term cash flows, cash flows off of hybrids that have 5 years or less. I think that is an area where I -- we are not as concerned about the potential for the sort of key interest rates, whether it's treasuries or swaps, on those parts of the yield curve to be significantly higher in the short term.

One of the things we like about 15-year mortgages is that they do not see nearly the degree of duration extension when the market trades off, as you do in the 30-year section of the market because of its -- the accelerated amortization schedule of 15-year pass-through. So that's actually been an area where, while we have a significant amount of absolute duration there, that has not been a major source of our interest rate mismatch.

So we're looking to narrow that gap. We like to keep our target duration gap inside of 1 year against -- to protect against significant increases in rates.

But I think given where we're positioned and where that -- where those -- the duration mismatches of our portfolio, where along the yield curve they are, we are comfortable where we are and the progress we're making towards getting back down to that target range.

I appreciate it. If I could ask just one follow-up that might be a little more philosophical.

We're at the end of a pretty long bond bull market, as I think you guys have sort of alluded to, where the propensity, I think, for durations was to kind of move -- for any given bond as long as rates are falling, the propensity is to move lower, not higher. And that mechanically, I guess, suggests that you can afford, or a mortgage REIT in general can afford to be a little slower in the response.

Your book value is going to improve if you're late on reacting to duration contraction versus now, we had a quarter of kind of unprecedented, or at least unprecedented in the last 20 years volatility in MBS. And while I definitely appreciate what you're saying about reducing the duration gap and kind of moving in the right direction, I guess, the question is sort of one of pace.

And, I guess, what I'd submit is that during the quarter, as you started out with only a 25 basis point gap, it was really -- the day the rates moved for first time, that gap widened. And any gingerliness in sort of hedging that out just increases your every incremental bit of exposure to the next 10 basis points in rate.

So, I guess, the philosophical question is, are we in a different time here, where it calls for much more active management, faster management of that swap side of the portfolio, in particular. I would think that most people would agree that the likelihood of rates is to -- is no longer a 50-50 chance of moving lower versus higher, but there's probably a skew toward moving higher, and that hurts a mortgage REIT a lot more than you generally tend to get help by rates moving lower, if that make sense.

So I don't know, it's a long question, but does it change your thinking about how to hedge kind of dynamically and actively going forward?

Well, I guess, I'll give you the -- maybe a philosophical answer in terms of how we've approached this business, even though a lot of our history has been in a bull market. I think over the long run, the attractiveness of the mortgage REIT model comes from the fact of -- that the securities we buy yield more than other similar-duration instruments.

In, large part, because of the uncertainty over prepayments and prepayment expectations, the idea that there is negative convexity in the securities. And that allows for a good measure of our net interest spread that we earn over the long run.

The fact that the yield curve is steep is also a contributor. The fact that, especially, when rates come down, having an asset/liability mismatch may contribute to that.

These are also factors. But over the long run, I think we viewed our approach towards investing in our portfolio and managing its risks was this idea that there was a sense of volatility in the short run being a negative for our portfolio and wanting to be positioned in a way that, while we could earn an attractive spread, but also be in a good position if there were to be periods of significant and especially prolonged volatility or one-way moves in the market, which would neither the shorten the portfolio through expected rapid prepayments, or lengthen the portfolio and you're worried about declines in the market value of your portfolio and your adequacy to meet margin calls, et cetera.

That's one of the reasons why we have had a primary focus in ARMs and hybrid ARMs and in recent years, the majority of our fixed-rate purchases have been in 15 years, is that while there is this potential for extension, a 5/1 hybrid, a 7/1 hybrid, if your rates move higher for a quarter or 2, there is certainly a potential for that gap between your assets and liabilities to widen as the portfolio generation extends. There's also built-in features in those securities, namely the reset date moving closer, where you're going to be in -- over a period of years, but still much sooner than the majority of the bond, move yourself back to that sort of position.

So we've always liked the fact that our portfolio, to some degree, had some built-in protection. That there was negative convexity in the shorter run, but over the intermediate term, there was sort of built-in protection in ARMs and with 15-year mortgages, the shorter maturity, and pretty significant amortization schedule is important as well.

So from the big picture, that's how we've looked to construct our portfolio. Had we had a portfolio that had a significant amount of 30-year mortgages, we clearly would have undertaken different hedging strategies.

You would want some other portion of the strategy, whether it was on the liability side through options or some other way to protect yourself against the really big moves or prolonged directionality in the marketplace, so -- but given our portfolio, we've always been comfortable with the fact that, that's one of the real benefits of having ARMs and hybrid ARMs. In the short run, you're correct.

In a given quarter, when we have a significant move in interest rates, that gap does widen out. And that is one of the challenges we face is looking to maintain our risk parameters within the tolerances we want and, at the same time, not winding up in a position where we're delta hedging on a daily basis and losing money by having -- by selling low and buying high.

So I do think, clearly, we're vigilant in this marketplace. We want to have some discipline.

When our leverage did start to move higher during the third quarter, we did actively take steps to bring that back down and to narrow our gap. But that is one of the challenges we face in terms of how tight a band we want to operate around those risk parameters.

I want to go back to July 5, when the markets were liquid, and we had difficulty with price discovery. And then you had factor day thrown in the mix.

Are your margin calls based upon last trade, or is there some mechanism that values collateral on a smooth basis?

This is Joe. It can vary from -- we have a number of different counterparties that we have repo transactions on with.

They all have their own procedures. But at the end of the day, it is our lenders who determine the value of the collateral.

They're going to base it on their best assessment of the marketplace and make margin calls based on that. Given the -- our portfolio composition, given that we have securities that the majority of the -- the portfolio on average was down 1.75 points over the entire second quarter of the year.

There clearly were price declines. Whether there were a lot of trading activity or not, you would have expected there to be some price declines on July 5.

But far and away, the vast majority of the margin call that was made during that period of time was relative to the approximately 2% or 3% paydown that we saw based on the month. And it was not really driven by a lot of uncertainty over where the prices in the marketplace were.

But the answer to your question is, the dealers who lend us money on repo borrowings, they -- they have to -- they will value our securities based on their good faith estimate of where they think the marketplace is.

Okay. And I noticed at June 30 that the unpledged MBSs were about 6% of the pledged, as opposed to more like 8% to 10% in other quarter ends.

Is there some range you target for comfort on that?

We discussed it a little bit on an earlier question. We would view an 8x leverage area as a target level that we are comfortable with given our portfolio composition and it's -- our expectations and its potential price volatility.

If you sort of work through the math of 8x leverage, it would mean that 1/9 of your portfolio was, in effect, purchased with equity. And we have approximately 5% haircuts that are required in the marketplace.

So you'd say, at that target level of leverage, you would have 5% of your securities pledged as collateral on top of the portion that was purchased with borrowed capital. And then you'd have about 1/9 right.

So 1/9 is 11%; you have that 6% unencumbered assets. That sort of falls out of the math of where our target leverage would be.

The fact that the quarter before, we were below our target level leverage meant that we had more unencumbered capital than we would have had at that target level. And the fact that at June 30, we were at approximately 8.7x leverage meant that we had less than we would at that target.

But that's an area that we're operating around.

Okay, yes, sure. That's just another way of looking at leverage.

And then a final question is, your dividend policy is driven by core EPS, which included realized gains in the first half. Now the third quarter is going to have a realized loss.

So is it fair to say that the third quarter dividend will reflect this loss?

We have -- this is Lloyd. We haven't made any decision on that, and there's a reasonable likelihood that it will not reflect that loss.

We will discuss it with the Board of Directors, and they will make that decision.

With regard to the sale of 15 years, post quarter end, could you give us an idea where the realized or unrealized gain and loss position stood at June 30 relative to what you've disclosed here on the final sale?

I'm sorry, the question would be where -- for the specific securities sold, what would have been the unrealized gain or loss at June 30?

Well, I think you mentioned in your comments we've seen some underperformance out of 15 years. And on this specific $235 million that you sold and realized the $5.6 million loss, if you could give us an idea of where perhaps that unrealized loss stood at the end of the quarter, so we can get a sense for the decline in 3Q.

I'm sorry, I don't think I have that specific information available for those pools. I will say that the underperformance that I was -- the securities were sold at levels that were slightly lower, but close to where they were marked at quarter end.

And so I would estimate that the majority of that unrealized loss was in the net portfolio number at June 30. The market is -- the sales were made in the first half of July.

We've seen prices decline since then in 15-year, so I'm -- sorry I don't have more specificity for you there, but there was, for example, for 15 year 2.5s were at par and 1/2 at the end of June. And they're at 99 22 right now.

The sale price would have been within that range, maybe a little closer to the June 30 level.

Okay, that's a good color. And then when you're thinking about repurchasing stock, maybe just to give us a sense for how you look at the stock price versus book value, could give us an idea of the timing during the quarter that you're repurchasing stock?

And at what price you felt book was at or what discount to book you were purchasing the stock at?

This is Lloyd. You're referring to the second quarter and that we repurchased shares in the $5.60 range, and you're saying where was a book value when we were thinking like that, is that -- that's your question?

That's correct, yes.

The book value was, obviously, in the $7 a share and just-under range. The purchases took place when we had a perception that the book value was still at a fairly high level.

Once the book value started coming down, clearly the target prices that we would wanted to have sold the stock would have also gone down. The -- however, the principal driver -- as I mentioned earlier, that was the third driver.

The first driver is the need for equity capital, the need for the right leverage ratio. All the things that we need to provide the safety and protection from never having to be in the situation of having to -- having insufficient unencumbered capital.

We would then be forced into, obviously, selling stock at a very low price, if such a situation occurred. So we measure all 3 of those items to determine where we would purchase -- where we would use our equity capital to purchase stock as opposed as part of our risk management program.

So it's 1 of 3 items that takes it into account. Unfortunately, the stock doesn't tend to trade above book value when things are doing very, very well.

It trades above book value at that time.

Well, Mike, thank you very much. And everybody who attended the meeting today, we thank you for your attendance in today's conference call.

And most specifically, we appreciate your interest and support of Anworth. And if you need additional information about the company, you can please visit our website at www.anworth.com and contact our Investor Relations.

So thanks again for your participation. We look forward to visiting with you either in person, or again during our conference call about this same time next quarter.

Good day.