After the bell on Monday, American Capital Agency (NASDAQ:AGNC) reported results that could only be described as fantastic. Adjusted EPS of $0.87 beat estimates by $0.19 (all financial and operating data available here). This quarter confirmed that AGNC is one of the better performing mortgage REITs around, and it is certainly an attractive value, but there are medium term risks, which could pressure the dividend. On the whole, I do believe the rock-bottom valuation does outweigh these risks, but investors should not expect dividends to return to 2012 levels of $1.25 (versus $0.65 now) any time soon.
Let's begin by exploring the quarterly results. Investors should focus on net spread and dollar roll income as this is indicative of AGNC's funds available for dividends, and $0.87 was very strong. This result was comfortably in excess of the $0.65 quarterly dividend. Mortgage backed securities performed well during the quarter thanks to the drop in interest rates and Fed buying, though they have been giving back some gains in the month of July, which may prove to be a slight headwind for Q3 results. The company generated a strong 9.9% economic return on equity, which pushed book value to $26.26
At the end of the quarter, AGNC held $71.9 billion in agency MBS. $69.3 billion of this portfolio is fixed rate. AGNC used the rally during the quarter to pare back leverage from 7.6x to 6.9x. Booking gains was a wise strategy, and if rates rise, AGNC has the capacity to re-lever and add bonds at a more attractive interest rate. Prepayment rates were a bit higher than AGNC's long term forecast of 8%, coming in at 9%. As interest rates rise, refinancing and principal payments should slow, which should bring prepayment rates lower. With many MBS trading above par, it is important not to underestimate CPR as this shortens the maturity, effectively cutting the annual yield as the premium to par is amortized over a shorter period of time.
AGNC levers its balance sheet by borrowing on the repo market. With overnight rates at 0-0.25%, this funding is very cheap. Net interest spread was a solid 1.84% in the quarter, up nicely from 1.43% in the prior quarter. Simply put, this quarter was strong. AGNC took some profits and reduced leverage at precisely the right moment and continues to do a good job generating shareholder value. It is no wonder shares rallied 1.5% after-hours on this report, especially as we can be certain the dividend is safe for at least another quarter and likely through the end of the year.
AGNC currently trades at a 10% discount to book value and yields a fantastic 11%. Investors would pay more to build a similar MBS portfolio than buying AGNC; that is a compelling valuation. Additionally, that dividend provides fantastic current income, which is especially attractive given low interest rates. With strong operating performance and a great entry price, it is hard not to like AGNC. However, I do think investors should beware one concern: a flattening yield curve.
At its core, AGNC make money by collecting the spread between short term and long term bonds. It buys MBS, which typically have a duration in excess of 5 years, while borrowing on the repo market, typically inside of 30 days. As the yield curve steepens (the spread between long and short term bonds increases), AGNC makes more money. Conversely, a flattening curve cuts AGNC's interest rate spread and dividend capacity. Over the next 12-18 months, I expect the yield curve to flatten.
In about a year, the Federal Reserve will likely begin raising overnight rates, which will push up the yields at the short end of the curve. This increases AGNC's interest expense, and it will need long term rates to rise equally to maintain its interest rate spread. However, I expect the Fed to continue reinvesting principal in the MBS and treasury markets into 2016, which puts downward pressure on long term rates. While it raises short term rates, it will still be a buyer of long term bonds, helping to flatten the curve. Technical factors also support the flattening thesis. With mortgage activity slower than previous years, MBS supply will be relatively low. Interest rates in much of Europe and Japan are also lower than in the US, which will lead some of those investors to buy US bonds, further capping long term rates.
By the end of 2015, I expect the yield curve to flatten by about 50bp, which will pressure AGNC's dividend capacity, though it has the ability to raise leverage to compensate for lower spreads. As a consequence, I do not foresee AGNC being forced to cut the dividend unless we see the curve flatten by closer to 75bp. Still in this environment, a dividend increase is unlikely. At a discount to book value and 11% current yield, I believe investors are being compensated fairly for the risk and would be cautiously long shares.
Disclosure: The author has no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. The author wrote this article themselves, and it expresses their own opinions. The author is not receiving compensation for it (other than from Seeking Alpha). The author has no business relationship with any company whose stock is mentioned in this article.