Gary Kain - President and Chief Investment Officer
Mark DeVries - Barclays Capital
American Capital Agency Corp. (AGNC) Barclays Capital Americas Select Franchise Conference May 16, 2012 10:00 AM ET
Mark DeVries - Barclays Capital
All right. Thanks for joining. We are very happy to have Gary Kain join us now from American Capital Agency. Gary is the President and CIO of the company. Joined in 2009 where previously he ran the retained portfolio of Freddie Mac. For those of you who aren’t familiar with AGNC, they have had some very impressive returns over the last three years. Book value has gone up 11 out of the last 12 quarters. Very impressive considering the volatility we have seen in the mortgage and rates markets over that period. And their total economic returns have averaged over 40% for each of the last three years. So looking forward to hearing Gary’s comments and his explanation for how they generate these returns.
Well, thanks, Mark, and thanks for everyone for joining us and for your interest in AGNC. For those of you that are new to the company, I do want to just start by giving a really quick overview. AGNC is an agency mortgage REIT. As such we invest only in Freddie Mac, Fannie Mae or Ginnie Mae mortgage securities. Now obviously, all three of those guys have significant government support as such credit risk is minimal for us in our business. We do have to focus a lot on interest rate risk and prepayment risk are the real things we are trying to manage, and liquidity risk.
Our objective is to provide attractive risk adjusted returns to our shareholders and that’s through both dividends and share price appreciation or capital appreciation. Quickly, as Mark mentioned, we went public in May of 2008 and an IPO price of $20 a share. Dividends paid since then are $20.11, so yes we have -- in four years we have paid out in cash dividends basically something just above the share price.
But importantly we have grown book value as well to $29.06. So thinking about that the question we always get is, well, did you guys -- how you have been covering that dividend. And the short answer is, yeah, we have covered the dividend but we have also added incremental returns through book value growth over the same period. So that’s incremental value on top of that. And then lastly, we have grown our market cap. AGNC started as a $300 million market cap company, literally four years ago. And market cap now is just shy of $10 billion. And that’s helped both our operating efficiencies in terms of cost structure as well as obviously the liquidity in the stock.
So turning quickly to the next page. If we take a look at the residential mortgage market in the U.S., it’s $11 trillion market and roughly half of that market is Fannie, Freddie and Ginnie securities with the bulk of that being Fannie and Freddie. The other biggest group is whole loans that are on bank balance sheets. Those are not agency whole loans. And then maybe $1 trillion of that is also non-agency securities. But clearly the agency securities market and for that matter the mortgage market is a gigantic market.
As importantly though, on the next slide you can see that in addition to being very large, the market is extremely liquid. It is the second most liquid bond market in the world. So if you look at the top line or average daily trading volumes for U.S. treasuries, below that in the light blue are the average daily trading volumes for agency mortgages. And normally the bottom, the red lines are trading volumes in the U.S. corporate market, all of U.S. corporates. So I think that gives you a feel for just how liquid and how much trading volume there is in the agency space.
So when you combine the government backing with the tremendous liquidity, it’s both very feasible and prudent to be able to lever agency mortgage securities because of this liquidity. When you couple that with the fact that the GSEs are under conservative ship and if that is willing to lend to any banks in unlimited quantities against U.S. mortgage backed securities like they were treasuries, then you could argue that the financing capabilities or the liquidity of the product is better than ever at this point.
And when you, to that point actually if you look at 2008 and 2009, kind of the dark days of market liquidity, you could see that agency mortgage liquidity relative to treasuries were sort of at an all time high. So even in that kind of the weakest period you were still able to trade a product and you still had substantial liquidity.
So turning to slide six, what's important here is kind of just to understand the backdrop of the agency mortgage market. Fannie and Freddie essentially were the largest levered investors in mortgage securities. And they were for profit for the last few decades or most of the last few decades until the last couple of years. And they had massive comparative advantages versus other investors and I worked there so I think I understood that, both at that time and now. The GSEs could lever 50:1 in their retained portfolios, in their purchases of securities versus kind of 10:1 for maybe the average REIT. They could also fund their positions with agency debt, longer-term debt at levels that were more attractive than where other investors could fund.
So when you combine the ability to lever and the funding advantages, the GSEs, really it wasn’t a fair fight. They had comparative, core large comparative advantages in the mortgage investing business. And with their portfolios now set to shrink over the next seven years, then the opportunities for other levered investors such as the mortgage REITs are considerably better. Now throw in too what you are seeing on the regulatory front with banks. So dealer arb desks or proprietary trading desks have obviously shrunk as well. They were another smart competitor that enjoyed substantial leverage.
So when you put kind of those two things together, I think it’s a really important kind of thing to keep in mind that the landscape on the investment is side is considerably more favorable then it’s been in a while. And you really can’t overstate this. Because not only is this is a key reason why we are able to generate the returns that we are able to generate today. But I think it’s also really important to realize that this probably increases the staying power of these returns.
I think one of the biggest misconceptions about the REIT space is that it can only be profitable when the yield curve is very steep and interest rates are very low. That was sort of the case in the past, but the landscape is different now with the GSEs out of the picture. So we are actually very confident that if we can transition to new environments with the right hedges in place, we will be able to make money even in a flatter yield curve environment. And I think that will be a big surprise for the market.
So just quickly, how do we make money? The bottom line is we make money by earning a spread on our assets and we also make money on capital appreciation and repositioning our assets as well. But the yield on our portfolio at the end of March was 3.06% and again that’s yield on agency mortgages at cost basis, given the projected prepayment speed. Our cost of funds was 99 basis points and that includes our repo borrowing costs plus our interest rate swaps. It excludes some of our supplemental hedges such as interest rate swaptions, TBA short positions and mortgages and short treasury positions, for example.
But the net interest spread is 2.07%. Multiply that by leverage and add the asset yields and you get a gross ROE of 20%. Our fixed costs are a total of about 1.5% which brings a net ROE, an indicated net ROE in the 18.5% to 19% area. That’s something obviously that’s pretty attractive. It’s consistent with the dividend that we are currently paying. And if you look in the past, we have also added significant value both in terms of other income via trading gains and/or realized gains on the portfolio or on hedges. So that has added significant ROE in the past, but even if you just look at kind of what many people term core earnings, you can see how you can get to high-double digit yields.
So with that background let me move on to the next slide which tends to be my favorite slide in these presentations for someone out new. But if you look at the top right, what we have really -- this is what we term economic returns. And if you look over the last several years, AGNC were -- you can see the -- you get the combination of dividends, the light blue, and book value growth which is the dark blue. And then you can see our peers, the other agency mortgage REITs in green and you see the same breakdown.
So what you see over the past three plus years is that in each year agency has been able to outperform the average of our peers and by a pretty substantial margin. And so one thing to keep in mind, why do we feel that this is kind of the most important of best measure. Well, this is the way other financial, kind of professionally managed investment companies would be evaluated. In that, you look at the cash we pay out which is a dividend. And you look at the where the net asset value of our remaining securities and where they are marked. And so this is in essence not a function of whether you took realized gains or you left gains as unrealized. It’s not a bad accounting assumption that one firm is making versus another. This is kind of -- in our minds, true economic returns.
As you can see the stock price on the bottom there has followed a similar trend. And one thing, if you look at kind of the total stock return of AGNC. If you do look at it in terms of dividends reinvested since IPO, it’s close to 260%. It’s 170% even without dividends reinvested. So the numbers are pretty striking. And then it’s also interesting that -- I mean those are a little over double where the average of our peers has been. So one thing that I bet is on a number of people’s minds that are new to AGNC is, you are probably surprised that within the agency mortgage space, that you could see such a different returns from different companies. If you are all buying government guaranteed securities and the real key is, what I mentioned earlier, while we have very little credit risk, prepayment performance and prepayment risk is absolutely critical to your returns as is interest rate risk management. And agency mortgage securities are very different when it comes to prepayment risk. So -- and we will go into this in a little bit, but that is a key driver of the return differentials.
So turning to slide nine. What you can see is that not only have we produced strong returns kind of over that period but on the top we show you kind of what's happened with interest rates, the blue line is the ten year treasury rate. We have seen obviously a fair amount of volatility with QE1 or QE2 operation twist. You know different, multiple European debt crises, a U.S. debt downgrade. On the prepayment front you have seen multiple HARP programs. You have seen buyouts of delinquent loans on the part of the GSEs. You have seen a couple of different refi waves.
So while this is clearly been a very good environment for agency mortgage REITs, it has by no means been a one way street. And on the bottom, which you can see are the quarterly changes in AGNC’s book value, again after paying the dividend, and you can see that it’s very consistent performance in terms of your not seeing the gains in one quarter and then big losses in another quarter and things averaging out. So you are seeing more of a hedged out performance. It’s a function of finding relative value in the marketplace which is certainly our objective.
And one thing that’s key to consistent returns though, is obviously asset selection, but it’s also the ability to reposition your portfolio as interest rates change or as the different prepayment kind of impact or challenges kind of evolve. Whether it’s new government programs or things along those lines. The securities that worked for us in one quarter or a year ago don’t necessarily work in the next quarter. And being willing to reposition your portfolio is absolutely critical to consistent returns.
So I think what's interesting on this slide is that what it shows you is really that point of how different, different types of mortgage securities or different type of mortgage attributes can be. And how they can be priced in the marketplace. So if you look at what we have graphed here, there is three lines. The green line are generic or what are called TBA mortgage securities. So this is if you don’t specify what you want in the mortgage market and you buy a 30-year fixed rate with a 4.5% coupon, it will be priced kind of along the green line.
The blue line are securities that are backed by loans that have already gone through the HARP program. Now the HARP program is a program set up -- was set up in 2009 by the GSEs, so that if a borrowed had a GSE guaranteed loan with a high LTV, they would not in general have been able to refinance. Under the HARP program, LTV is essentially ignored and if the borrowers current on their loan, they can refinance if it’s a GSE loan. But they can only do in one time. It’s a onetime program. So once they have refinanced one of these higher LTV loans, it’s extremely difficult for them to refinance. They will have to put up a lot of cash to get their LTV down. So you don’t see that happen.
By definition, these have prepaid very slowly over the last couple of years, and I mean the prepayments have been more like around 5 CPR whereas many generic mortgages have been 25 or higher CPR. The sliver line are lower loan balance or smaller balance -- securities backed by smaller balance loans. And these prepay slowly because there are a lot of fixed costs associated with the refinance process. And if you have a small loan, you need a much bigger incentive -- rate incentive, to make it worthwhile to refinance. You know if you are saving $50 a month it takes a long time to make up for $1000 in closing costs.
That’s important, but that’s another advantage. If you are Wells Fargo or if you are JPMorgan Chase and you are making money of these originations but you are working around capacity, are you going to want to refinance $100,000 loans or $500,000 loans. When you get, let's say, paid 2% on a loan. On $100,000 loan you are going to make $2000, on a $500,000 loan you are going to make $10,000, you are going to focus on the larger loans. And so the busier originators get the more they kind of put smaller loans on the backburner and the more they focus on large loans.
So you have two sources of protection in a sense in smaller loans. But with that as the background, again those are the silver lines here or gray lines. Back in early 2011 and really almost through August, people rates -- you know rates were higher, prepayment speeds were not that big of an issue, and you could buy all of these mortgages at around the same price. You could actually buy that HARP security slightly below at TBA security because it was less liquid. But look what's happened really since September. As interest rates plummeted, the investors became more and more concerned about prepayments and more focused on finding a way to avoid them. And so you can see a kind of continuous divergence in the prices amongst these different security.
So back in September, if you look at this chart, you could have -- the HARP security was now one point above a generic security. Whereas today or as of the end of April, that price differential was now 2.8 points. And so when you think about that, generic mortgages, the green line, after rates kind of got to their first bottom in September, the generic mortgages have done very little in price whereas the HARP and lower loan balance loans have appreciated quite a bit. Well, this has driven both strong returns, very strong book value results for AGNC because roughly 80% of our portfolio was in lower loan balance or HARP securities at the end of September. We have been cutting that down a little given this tremendous price performance, but it still makes up the majority of our portfolio even today.
So moving to slide 11, and I think this is sort of a very simple stylized example for why people are willing to pay so much now, in an environment like today, for better prepaying securities. So at 10% CPR, which is a measure of prepayment speed, and 10% CPR means 10% of the borrowers would refinance over an annual period or in a yield. So the yield on a 4% coupon, 30-year fixed rate, is 3.01% at a 10% CPR. If you assume a sample cost of funds of 80 basis points again which would kind of close to today's market, you would repo and funding costs or swap costs. That gives you a net margin in this hypothetical example of 2.21%.
Now let's apply 8 times leverage and create a hypothetical REIT portfolio with a gross ROE of 20.5% or 20.69%. But look what happens if prepayments are faster. Let's look at the 30 CPR column. The yield, because of these are at a premium, and the premium is amortized very quickly, drops to only 163 basis points. The margin drops to 83 basis points and your ROE is only 8.27%. Now to put this in perspective, the HARP securities are prepaying right now at 5 CPR, the lower loan balance tend to be around 10 CPR. But generic mortgages have -- there are plenty of parts of the mortgage market that have been at or around 30 plus CPR. The average of the whole mortgage market has been in the low 20s the last couple of months and expected to pick-up a little bit from there.
So across this spectrum of speeds, depending on the circumstances, anything here is certainly possible. But again, when you think about the different ROEs for different types of portfolios when you see this chart, you can see why ROEs can be very different across the space.
So again, this is a quick view of our portfolio at the end of March. And I am not going to go through this. You can see the prepayment speeds on our portfolio have hovered around the 10 CPR, that’s on the top right. And what you will see is in the last month, they came down to 9. You can see a breakdown of how much is low loan balance and where the prepayments are and both in 15-year and 30-year mortgages. And just quickly, if you look on the second column from the end on the 30-year side, you could see the lower loan balance of 7 CPR, the HARP securities at 4 CPR. Whereas our generic mortgage surface certainly have been noticeably faster than that.
But in the interest of time, let me move on to slide 13, because I think this is kind of what's more on people’s minds today. And this is really three scenarios that we are thinking about, as we try to position the portfolio now to be able to perform kind of against a range of different scenarios.
So first, we have the scenario that up until pretty recently, I think a lot of people viewed as the most likely one, and that’s where moderate growth in the U.S. continues. Inflation kind of remains at or near the FED’s target, what would probably happen in this scenario is we would not get a QE3. The yield curve would steepen some. Maybe the 10-year goes up to 2.25%, 2.5% from here. Again, it’s not an overheating of the U.S. economy. It’s just strong enough to keep the FED at bay. This is sort of a Goldilocks scenario for most mortgage investors in that prepayments generally slow down because interest rates are going up. The yield curve gets a little steeper. New purchases are attractive and the performance of existing mortgages are probably pretty good. So again, pretty good scenario for the space.
The next scenario is one is that is, again, starting to look a little more likely nowadays, and that’s one where the U.S. weakens, inflation expectations fall or remain in check. And that could occur obviously due to weakness in Europe or just domestic issues in the U.S. But certainly the employment numbers have looked weaker of late and if you look at the bond market, it’s certainly telling you that this a more likely outcome at this point. In this case we would likely get a QE3 from the FED. The FED is likely to buy a large amount of agency mortgage securities. So many of you might be thinking, well this sounds like a good scenario. You are a levered holder of agency mortgage securities and the FED is basically going to want to buy a large percentage of agency securities so your portfolio should do very well. And the answer is, yes, if we own the right things and we will come back to that.
But from a cash flow perspective and a return perspective this is actually a very challenging scenario. So it is not a Goldilocks scenario for us. And the reason is, is that the FED will buy low coupon fix rates, drive the mortgage rate in the U.S. lower, put more pressure on prepayments so prepayment speeds will generally go up. The yield curve is likely to be flatter, and more importantly, our new purchases will be considerably less attractive because we are going to be fighting with the FED who is going to be buying 300 or 400 billion of agency mortgage security. So unattractive new purchases, faster prepayments on the existing book, certainly not things we are looking for. But on the other hand, the prices of certain parts of your portfolio if you own the right instruments will go up quite a bit. And so -- and we will touch on those in a second, you can generate very attractive returns in a QE3 scenario but you really have to be positioned for it.
The last scenario is one that I think almost everyone in this room would agree is a relatively low probability. That is where the growth in the U.S. picks up dramatically and inflation expectations pickup. You could get this scenario in another way that’s probably a little more likely which would be loss of confidence in the U.S., whether it’s with the government or the FED. And what happens is you get a very large and rapid increase in interest rates. Let's say 100-200 basis point increase. While it’s a low probability, we can't just dismiss this out of hand unfortunately, because the outcome is pretty severe. I mean if you are not well hedged and if you haven’t you know kind of thought through this scenario then you are going to be looking at significant book value declines. And you are not going to be able to take advantage of the wide spreads on new purchases and things like that.
So again, while it’s a low probability, it’s really important and we talk a lot about this in our earnings presentation about our hedges, but if you position yourself and think about it before it occurs, you really can get caught off sides. Again, even if it’s a pretty low probability outcome.
But if we turn to the next page, I would like to conclude by saying -- that’s the backdrop -- you know, what are you doing? And the short answer is, we view it as -- we have to think about all these. But on the asset selection side, we feel it’s imperative to have most of our portfolio in either HARP or loan balance mortgage securities because they have good prepayment characteristics. And so they will perform both in today's environment like we’ve seen and they will hold up well in a QE3 environment.
The other segment that we have increased is low coupon generic fixed rate. They will perform a little better if rates go up, but the challenge -- but if we do get a QE3 from the FED, then that is what the FED will buy. They have only in the past bought low coupon fixed rates because they are trying to drive down the mortgage rate to have borrowers be able to refinance or buy houses at lower interest rates. So they will buy low coupon fix. So we would be able to sell those assets to the FED at obviously very inflated prices and then reinvest in other part in the market.
So those are -- on the asset selection we are also avoiding the loans and are eligible for HARP 2.0, the new program -- the HARP program that was enhanced at the end of last year. Because we feel we are seeing speeds pickup but we think that can continue and that could continue under a couple of different scenarios. On the hedging side we have increased our hedge ratios, we are generally paying on longer-term swaps. But more importantly we have bought a lot of out of the money options. So we talk a lot again in our earnings presentation about it. We had $10 billion in swaptions which basically give us the option to enter into 7 and 10-year swaps for the most part. And that gives us particular protection against kind of scenario three where interest rates could shoot up. And so we are spending some of our book value gains and performance to buy some component of insurance for that scenario.
Now lastly I want to talk about leverage, because leverage is really important. In both the first and the second scenario that we talked about, the two likely scenarios, mortgages should do relatively well. And so we are better off with higher leverage. The last thing you would want in QE3 scenario is to have put off purchases and to have to do more purchases once you are competing with the FED. So low leverage in our minds, right now we are operating in the neighborhood of eight times leverage which is where we have been. You know low leverage is not, in our minds a good strategy. So you can make an argument for higher leverage, mortgage prices have done very well of late. So I think that we have kind of backed away from kind of being at higher leverage, but on the other hand moving to much lower leverage for us a non-starter given the potential for a QE. And given that in a normal kind of that scenario one, mortgages would do pretty well and we don’t think we could buy mortgages hedge adjusted cheaper even in that scenario either.
So with that I would like to conclude and take questions. We do hope that summarizing kind of our thought process in this kind of fashion makes it easier for investors to think about the tradeoffs that we have to make in managing the portfolio to be able to do reasonably well across a set of scenarios like this. But the other thing is that really it’s the ability to kind of reposition the portfolio over time as the market and the landscape changes, that’s really been the key to kind of market leading performance. And so while current positioning is important, what you do going forward is going to be the key driver.
So with that let me take questions.
Mark DeVries - Barclays Capital
Just a quick question on the industry as a whole. So with the proliferation of mortgage REITs over the last couple of three years, and all the issuance that we have seen and obviously you guys are one of the larger providers and players in space. What's the secret sauce? What's the competitive advantage that you guys have versus some of the others and how as investors can we differentiate others versus your organization?
Great question. Look, we have issued a fair amount of equity as the space as a whole. I think I would start by, in the end you are -- what's important is picking a management company. And the management team is what's going to differentiate one REIT from another. And so you know where we feel we stand out is that the management team, the average for our top five people, the average years of experience is like almost 20 years in the business. Not only that, our people come from kind of very -- places such as, we have a number of people from Freddie Mac where the GSEs were the center of the mortgage market. And a lot of the decisions we have to make relate to what are the GSEs going to do, how do they make decisions, even though they are under conservative ship. How does the government or FHFA look at things?
And I think we have an edge in understanding kind of things from the GSE perspective and trying to choose assets. But in the end it is -- if you look at the track record and you look at it based on the economic returns that we talked about, I think that’s kind of in the end it’s about results. It’s not about what each of us say. I think agency has been as successful as it has been both in terms of growing, but more importantly on the return side. Because we have kind of been very upfront about our thought process about the positions, the types of assets we are buying. The things we are concerned about. And then we back that up with performance. And I think the combination of the disclosure and kind of continuous performance both on the prepayment side and just in total returns, has I think the -- is the biggest reason why the company has flourished.
Hi, clearly your performance has been very good. Can you talk about or can you describe your investment team? How many people, how they are split, how they go about that business? And also how you incentivize and pay them to make sure you can keep them looking for you.
Sure. All very good questions. So first off, I mean it is a very scalable business so our front office is not a large staff and it’s in the five to ten person area. But what we do to get leverage there is that we use BlackRock Solutions as our risk management tool so we have all of their models, we don’t need people building models and running risks, developing risks systems. We are using a state of the art third-party provider. Our back office and clearing is done with JPMorgan actually, and we have Bank of America as well as a backup clearer so to speak. So when you put that together we have kind of outsourced some components of -- the less fun components of the business to external vendors. So we focus day in and day out on the mortgage market and finding the best assets, finding the right hedges.
And so in terms of the particular team, there is kind of two -- there are three key people that report to me. One Chris Kuehl, who is our SVP and Head of Agency Mortgage Investments. He worked at Freddie Mac for a long time. He ran structures like CMO purchases and mortgage pass-through repurchases at one of the largest investors in the agency mortgage market. His day-to-day focus is, do we have the right assets? He is doing things so that every month when a prepayment report comes out, he is not only looking at did low loan balance do well, but he is looking at within our low loan balance loan portfolio, what coupons is he concerned about. Whether certain issuers, JPMorgan Chase, were they faster than Wells? Are there any trends he need to worry about within those kind of levels. And day to day is there something that we should be selling. So that’s what he is looking at.
Peter Federico, who is our Chief Risk Officer, is looking at our interest rate risk on a daily basis. And he is looking at the swaption strategies and our swaps and treasury positions and how the portfolio is hedged. He is looking at different yield curve shifts. He is looking at changes in volatility and option prices. So he is kind of -- he is looking at what can happen to us in a 200 and 300 basis points movement. We get those runs every single day out of the model. And then Jason Campbell, who leads our financing efforts. So he is doing repo and financing the book. So that gives you a feel for the team.
Now you asked another question which I won't pretend that I forgot, which comes to the, how we incent people, how do we pay people? And compensation is important. And there are metrics that specifically go to all of the key employees and the manager. And those are the economic return, which is combination of dividends and book value, but not just absolute. It’s relative to the space. Okay. So if it’s just a great environment for our mortgage REITs then that’s not something, it’s something that’s nice but it’s not something that people directly get paid for. Where it’s most important is for them to outperform their peers.
Another one is the price to book of the company. And why? Because if you are taking excessive risk or if you are doing things that you don’t communicate or investors are worried about, then it will show up there. And so we look at that versus our peers as well. We also look at total stock returns versus our peers. So when you put those together, those are numerical objectives that are kind of in -- that are going to affect people’s compensation. Yes, and I want to be clear. The growth in assets under management or equity under management is also important.
And then lastly, a good percentage of the management team’s compensation comes in the form of stock in the two REITs that we manage, AGNC and MTGE. So I think you will find that as well as can be expected we are kind of very aware of the compensation trends in the industry and there is really an attempt to kind of, to make this kind of structure work well against that backdrop.
Thanks, Gary. Even though we are still in the relatively early stages of implementation of HARP 2.0, we already have the administration further pushing their mass refi efforts, talking about, I guess what would equate to HARP 3.0. Have you discounted the risks around that? That maybe we have re-HARPing and further risk to some of the buy-ups that you have been purchasing?
So, great question. And goes it around further policy risk on the prepay front. And honestly, we feel that the policy risk right now is actually at a relatively low point and is actually much lower than it was maybe a year or two ago, despite the noise about the Menendez box or bill and things like that. As an example, even just six months ago the discussion was mass refi, refi everyone. Now if you look at one of the bills, the main bill out there the Menendez Bill, which based on what most people have said is not going to pass and has a very very low probability of passing. But even there it really is about fine tuning the HARP program one more time. And the actual recommendation in there is to move the HARP eligibility date from May 2009, as an example, to May 2010. So move it ahead one year.
While actually a very small percentage of our portfolio, right now less than 5% of our portfolio, is even exposed to the current HARP date. In other words it was not originated before May of 2009. Even if they moved it out in a year, that would still only capture another incremental part of our portfolio. Maybe that would go up to 10% or 11%,12%, something like that. So the vast majority of our portfolio including HARP securities that we have are originated after even that date. So the bottom line is, we don’t -- we feel that you are going to hear a lot of noise because it’s an election year. You are going to hear people talking about new programs. We feel from all the intelligence out there, things are not likely to pass. But more importantly even the things that are being talked about are so much more benign than the things people were talking about six months to a year ago.
So I think we feel more confident on that front than we have felt in a long time. The other thing I just want to quickly say about the pay-ups and the premiums on the lower loan balance and HARP securities. We have tried to reduce our exposure to those premiums. We have benefitted quite a bit in the last two quarters and we talked a lot about this on our earnings call as those pay-ups have gone up, and that’s driven a good chunk of our book value growth over the last couple of quarters where we have been able to grow book value about 8%, while most of our peers have been relatively flat of a couple of percent. And what we have done is we have sold our highest pay-up and highest prices HARP securities and we have kind of reinvested in either lower coupon generic fixed or lower coupon other HARP or loan balance securities.
So if you actually looked at our portfolio at today's pay-ups, you get a portfolio from six months ago. We have really cut our exposure to pay-ups almost in half from where they were. So we still feel good about our prepayment positioning. And we still feel like we are well positioned going into a QE3 if we would happen to end up there. But we don’t want to give back all those gains and we do view it as a benefit to reduce some of that idiosyncratic risk given the tremendous performance of those securities.
Mark DeVries - Barclays Capital
I think we are going to conclude with that. Please join me in thanking Gary for his time.