For the past six months I have maintained (while in the short-term it's been painful) the reason we own mortgage real estate investment trusts (mREIT) stocks such as American Capital Agency (AGNC) is to collect income, or compound our investments. In that regard, I think they have a place in a portfolio, especially if you dollar cost average or pyramid down into the stocks and plan to hold for 10 to 30 years as a long-term investment. However, in light of recent events such as earnings and changes in the macro environment, I have had to reassess my position in AGNC.
AGNC's Approach To Make Money, Still Working?
At the most basic levels it really isn't that hard to understand how AGNC makes its money. AGNC pretty much borrows money and lends mortgage backed securities. So that takes us to the next question, what is a mortgage backed security? Well, mortgage backed securities are essentially debt obligations that represent claims to the cash flows from pools of mortgage loans. These can be residential or commercial in nature though AGNC tends to work primarily in the residential space, while its sister company American Capital Mortgage (MTGE) deals with commercial MBS. So if you want commercial exposure, you would be better off with MTGE. For residential, you would go with AGNC. If you want both, you might look for a hybrid REIT which does both.
OK, so back to making money. Mortgage loans are purchased from banks, mortgage companies, and other originators and then assembled into pools by a governmental, quasi-governmental or private entity that turns them into securities. Now, what feeds into their profits? The interest rate that AGNC borrows at and the interest rate they lend at is called the interest rate spread or simply the spread. As I have stated many times before, this metric is absolutely key and is one that you must assess and keep your eye on when owning AGNC or any other mREIT. AGNC then profits from this spread, but does so how? Well they borrow at the 2 year Treasury rate and lend to the 10 year Treasury rate. Or, they borrow the 10 year and lend to the 30 year. All they are really doing is borrowing at a lower cost item and then lending at a higher rate, generating a spread and pocketing the difference. If necessary, the company can and often does increase their exposure and risk when conditions call for it. This is called leverage, which allows potential profits to be magnified. This is when they lend further out on the curve, maybe to 15, 20 or even 30 years from funds borrowed at the lowest rates. This leverage is risky but can lead to huge profits. AGNC often carries a moderate to moderately high amount of leverage. So is this approach to making money still working?
What Impacts AGNC's Profit Potential?
The rate at which AGNC borrows is key. This is one reason to care about the Fed. Right now, they have a zero interest rate policy in place. This keeps shorter term interest rates between 0 and 0.3%, so AGNC can borrow money for almost nothing. This has been key to maintaining a favorable spread, borrowing at next to nothing, lending at some predetermined rate. Pressuring the spread has been Fed purchase programs which can keep the longer term interest rates low as they purchase longer term treasuries. Anything that changes the yield curve between the 2 year and 10 year or 10 and 30 year rates can cause either widening or flattening of the spread. We saw a spike in interest rates in shorter term debt from May to August, which put massive pressure on the mREITs. Many of the spreads narrowed. This led to these stocks getting crushed.
What is Prompting My Reevaluation? Earnings?
I have been silent since AGNC reported third quarter earnings last week on October 28, 2013, and presented mixed results.
Key Headline Statistics From Q3 Earnings
A lot of traders move stocks based on the headline numbers only. While I agree those are important as they give you an indication of the company's performance, they can be misleading at times. However, I would be remiss if I did not provide them. Overall, it was an improvement from Q2. In Q2 AGNC reported a $2.37 comprehensive loss per common share, comprised of $4.61 in net income per common share as well as a $6.98 other comprehensive loss per common share. This equates to an overall loss of $936 million for the quarter. Here in Q3, AGNC reported a $0.45 comprehensive income per common share, comprised of $1.80 in net losses per common share as well as a $2.25 other comprehensive income per common share. This equates to an overall income of $179 million for the quarter. Estimated taxable income was $0.29 per share, which is below the dividend that was paid out of $0.80 per share. As an investor, to assess the situation as a whole, I really need to dig deeper.
Spread on Interest Rates, A Critical Metric
With interest rates moving wildly in July and August then dropping off in September, I was unsure where the spread would land. I suspected it could go either way. The interest rate spread was essentially stagnant. To my surprise, when you include all of the factors affecting the spread, AGNC reported a net interest rate spread of 1.37%. which was an increase over the 1.24% reported in Q2. But we need to see where these numbers come from. AGNC's average net interest rate spread for the third quarter was 1.20%, a decrease of 29 bps from Q2 of 1.49%. Including estimated TBA dollar roll income/loss, AGNC's average net interest rate spread for the third quarter was 1.14%, a decrease of 72 bps from 1.86% during the second quarter. That's a massive loss to the spread. However, AGNC's average net spread income for the third quarter includes -6 bps of premium amortization cost due to changes in projected constant prepayment risk estimates, compared to 29 bps of premium amortization benefit during the second quarter. All in all, I was less than impressed, though pleased with the final spread of 1.37%.
Let's look at this a bit more to see where the assets and yields stand. First, AGNC's asset yield on its agency security portfolio for the quarter was 2.59%, compared to 2.92% for the second quarter. This is a huge drop of 33 bps. The annualized weighted average yield on the agency security portfolio was 2.64% for the current quarter, compared to 2.63% for the prior quarter, backing out the amortization catch-up payments, thus this number was stagnant. Overall, AGNC's average asset yield reported as of September 30, 2013, was 2.70%, a single bps decrease from 2.71% as of June 30, 2013. Furthermore, AGNC's average cost of funds decreased 4 basis points to 1.39% for the second quarter, from 1.43% for the first quarter, due to lower average swap costs as a result of a smaller hedge position.
To put this all into perspective, the cost to borrow declined but the average yields did as well. Thus, earnings potentials as a result from the interest rate spread alone have not changed. In fact, 1.37% for the overall spread is rather strong compared to competitors, though as investors we want this number to be as high as possible. This statistic is one of the first things I examine when looking at the performance of any mREIT. Overall, it wasn't that bad as it increased, but I was less than impressed with certain components of this figure (such as decline in yield on assets for the whole quarter). However, as the average yield on assets as of the end of the quarter was higher than the quarter average, this is a huge positive.
Issues With The Dividend
While we have known about the declared dividend for a while when it was announced in September, and it was a huge cut down to $0.80, I thought AGNC could easily afford it was a down 24% from the last dividend of $1.55 per share. This cut caught many by surprise as most were expecting a cut to around $0.90. Looking under the hood now, it would appear that the dividend was more than AGNC's estimated taxable income per share of $0.29. However, digging a little deeper, we see that AGNC still has $0.57 per share in undistributed taxable income. I believe that the quarter has started very strong for AGNC given that interest rates have declined about 40bps off their recent September highs. Thus, with undistributed taxable income and an improving quarter, I do not expect the dividend to be cut below $0.75 per share. It will likely be maintained.
I've Said it Before And I Will Say It Again. It Is All About Book Value
Book value is very significant for mREITs. Along with the net interest rate spread, it is absolutely critical. Why? Because you want to buy when the stock trades below book value and sell when stock price gets too far ahead of book value. This assumes that book value will remain relatively stable, or in the case of buying below book value, rising. With the extreme volatility over the last few months, book value was anyone's guess. But my readers know that I was adamant that AGNC was trading well below book value. Well, as it turns out, book value dropped as expected, but at the end of the quarter, the stock was indeed trading well below tangible book value.
On September 30, the stock was trading around $22.58. The book value was reported to be $25.27, which was a negligible $0.24 drop from the end of Q2, or a decline of less than one percent. However, it also meant that the stock was trading about 11% below book value, indicating at the time it was a great buy. Much of this decline was due to the volatility in mortgage-backed securities (MBS), and many investors and traders just dumping the stock for fear that it could have been much worse. Considering the stabilization in MBS prices, and interest rates over the last few weeks, we can likely safely conclude that book value at the moment has stabilized as well. In fact, as the fourth quarter has been off to such a great start, book value has likely increased. Therefore, it is likely we are still trading at a significant discount to book value, and thus we have a buying opportunity in the name still. What is keeping the price down? Mostly fear of a dividend cut, continued interest rate fears, and fear of the Fed.
Reassessing the Role of the Fed
I have been of the opinion that the Fed is not something to worry about. I need to reassess this. Rather, I should clarify. I think longer term they are not to worry about, but in the short-term, they could hurt AGNC. This is because the Fed does taper in December, we will see a moderate increase in the ten year from 2.5% to as high as 3.25%. This would steepen the yield curve however and lower mortgage prices, impact prepayments and could impact the interest rate spread. If the Fed delays tapering until say next summer, we could see the 10-year pull back to as low as 2.0%. This could flatten the yield curve and cause MBS value to rise. So, in the short term, I have reassessed. The Fed is something to be concerned with. Longer-term however, I think things will be fine.
What will happen when they actually do taper? The market may or may not overreact, but I for one will be glad when they are out of the asset buying picture so that stocks like AGNC and MTGE will no longer be hit by chatter over the next Fed move. For those just looking to get in, it is a pretty good time. You may get the chance to build a position at these low share prices. The stocks have been volatile thanks to the Fed, but they haven't really done anything different policy wise in the past 6 months. It's been business as usual. Much of the action has been a result of fear that they may do something. For now the Fed's accommodative stance remains in place because the economic data just isn't that strong. The Fed is not advocating for tapering of asset purchases ahead of stable economic news and/or meeting the economic goals laid out when the program was announced. So the fears of this issue that partially punished the mREITs are simply unfounded right now. Among the most important takeaways is that FOMC remains adamant that the low-rate situation, or the zero rate interest policy, will not change until the unemployment rate drops to 6.5%. This is not likely until late 2014 or 2015. The nomination of Janet Yellen indicates that the Dovish policies will continue. For now, we have a 14.5% yielding stock with a stabilizing environment.
After taking some time to absorb the most recent quarter and give myself a reality check with the Fed, I have several conclusions. First, I do believe right now the macro environment for mREITs is improving. AGNC had a rough first two months of the third quarter, but the evidence suggests it began improving in September. This improvement has certainly continued to this point. Book value is stabilizing and on the rise. The dividend yield is much lower than I am accustomed to with AGNC, now at 14.5% versus 20% or higher. That is a big negative. However, I see the dividend being maintained, or, if it is cut, no lower than $0.75. Net interest spreads are improving. Their hedging activities are working, and they have conservatively not levered up to over 8.0X. All eyes must be on the Fed in the short-term. Once tapering has occurred and is in the past, the private sector can absorb MBS trading. It will be an adjustment, but I believe management is taking the necessary steps to survive the transition.