We think the massacre that took place from October 5-October 15th with regards to the mREIT sector was a bit overdone. During that time period, mREITs as represented by the Market Vectors Mortgage REIT ETF (MORT), the iShares FTSE NAREIT Mortgage REITs Index ETF (REM) and industry leading mREIT American Capital Agency (AGNC). While the salad days of 15%+ annual yields and capital appreciation on mREITs aren't coming back, we have counseled investors interested in mREITs to view mREITs within a prudent perspective in that these are primarily income driven vehicles, you don't to pay too high a price for them, you don't want mREITs that have high prepayment rates on the portfolio and that you should not currently expect much in the way of dividend increases. We also believe that a prudent and disciplined investor should expect a median annual dividend yield of 10% though, which exceeds what they can get on bonds.
Source: Morningstar Direct
American Capital Agency used to be our favorite mREIT and is still one of our favorite mREITs, even though we do not currently have a position in the company. While we were cautious on the company's investment prospects during the summer due to its high market price premium to book value of 12-17% it was still one of our favorite mREITs because of the outstanding performance of CIO Gary Kain and his team. Based on the increase in book value that AGNC enjoys due to the unrealized gains on its MBS portfolio and due to the $.79/share spread income generated during the quarter, we can see that mREITs are capable of paying 10% annualized dividend yields even with the narrowing interest spreads due to central bank interventions. Considering that AGNC, American Mortgage Capital (MTGE) and Annaly Capital Management (NLY) have announced share buyback plans, we believe that this should establish a floor under mREIT shares.
Source: Company Press Releases
AGNC enjoyed such strong performance early in the year it issued 76M shares in March in a secondary public offering as well as 36.8M shares in July. Because of this rapid increase in outstanding shares going into the following periods as well as the steadily flattening yield curve, AGNC was facing a tough challenge in order to top its Q1 2012 results. Needless to say, AGNC's Q2 2012 and Q3 2012 results were not as good as its Q2 2011 results, Q3 2011 results or its Q1 2012 results. AGNC's Q3 2012 Net Interest Income declined by 78bp on a linked-quarter basis and increased by 64% year-over-year. Net Spread Income per Share was $0.79 for the period versus $0.94 in Q2 2012 and $1.15 in the prior year's comparable quarter. Factors that accounted for a sharp drop in Net Interest Income per Share include:
- A slight uptick in the actual CPR prepayment rate on its portfolio of mortgage-backed securities from 8% in Q3 2011 to 9% in Q3 2012 and on its forecasted CPR prepayment rate (from 13% in Q3 2011 to 14% in Q3 2012).
- The rapid increase in its total net premium amortization. During the Q3 2011-Q1 2012 periods, AGNC's total net premium amortization averaged $111M per quarter during those three quarters. In Q2 2012, total net premium amortization shot up to $196M and further climbed to $219M in Q3 2012.
- A 10% drop in the leverage ratio on an average during the period basis (from 7.9:1 in Q3 2011 to 7.1:1 in Q3 2012) and 9% on leverage utilized as of the end of the applicable fiscal period(s) (from 7.7:1 in Q3 2011 to 7:1 in Q3 2012).
- The continued decline in the average coupon of its MBS holdings narrowing the Average Net Interest Rate Spread enjoyed by the firm.
If AGNC was to cut its dividend in order to reduce its payout ratio, we are of the opinion that it won't cut dividends until next year. We believe it will continue to pay the $1.25 quarterly dividend for the rest of 2012 and before the Fed announced its QE Infinity program, we were expecting AGNC to cut its per share dividends by 10% or less beginning in 2013. Despite the weakness of the economy, our assumptions were that there was no fundamental reason for the Federal Reserve to resort to another round of money printing especially since it had already purchased $2.1T of assets during the first two rounds of money printing and instead of engaging in QE3 last year, implemented the Operation Twist program in which the Fed sold short term debt and bought long-dated debt.
We believe that the company can maintain its $1.25/share dividend for the rest of 2012 based on its level of net interest income per share and its strong level of realized and unrealized gains. We also believe that the open-ended monetary easing program will push the company to try to lock in its interest income received on its bonds and swaps in order to mitigate a likely decline in net interest income. The worst case scenario for AGNC is that it will cut its dividend by 33% and it would still yield over 10% and we believe that would still provide a satisfactory yield to reasonable and rational mREIT investors who are interested in a best-in-breed mREIT. We reiterate that investors who are looking for growth in Net Interest Income per Share and Dividend Growth from mREITs are looking for needles in haystacks.
Although the company's reported comprehensive income per share of $3.98 for the period was a 150% increase versus Q2 2012 and Q3 2011 levels, we don't have to waste much time identifying that this was due to the Fed's monetary easing activities pushing down bond yields, which pushes up bond prices. These actions helped contribute to $1.19B in unrealized gains on its mortgage portfolio, $51M in unrealized gains on derivatives, $210M in realized gains on the sale of MBS securities and $460M in realized losses on derivative instruments. AGNC is known for its below average CPR rates for its mortgage portfolio and its Q3 CPR was 9% Actual and 12% forecasted. We are pleasantly surprised that AGNC's actual CPR prepayment rates have been significantly lower than forecasted CPR rates and we are also pleased that its CPR prepayment rates are much lower than its larger mREIT competitor Annaly Capital Management.
98% of AGNC's $90B MBS portfolio is backed by fixed-rate securities, 1% by floating rate securities and 1% by CMOs with fixed and floating rate securities. AGNC's fixed-rate investment portfolio was comprised of $28.6 billion ≤15-year fixed-rate securities, $2.7 billion 20-year fixed-rate securities and $56.6 billion 30-year fixed-rate securities. 71% of AGNC's fixed rate investment portfolio was comprised of agency securities backed by lower loan balance mortgages and loans originated under the U.S. Government sponsored Home Affordable Refinance Program ("HARP"), which possess favorable prepayment attributes and a lower prepayment risks relative to generic agency securities. Considering that AGNC's weighted average cost basis on its investment portfolio was 105.4% as of September 30, 2012, we can see that the company will need to maintain its focus on minimizing prepayment risk.
In conclusion, we still see AGNC as one of our favorite mREITs especially because it is repurchasing $500M of its stock. With regards to mREITs managed by American Capital (ACAS), we would probably prefer American Capital Mortgage Investment. Our preference for American Capital Mortgage Investments is because its premium to book value is within 1% of American Capital Agency's, MTGE's ability to invest in non-agency MBS securities and MTGE has a lower asset size than AGNC. We also like the fact that the increase in MBS prices and the mREIT massacre that took place from October 5-15th has resulted in AGNC trading at around its book value rather than the 12-17% premiums we saw over the summer. We also like AGNC to Annaly because even though both have below average CPR rates, AGNC's Forecasted CPR rate of 14% is better than Annaly's Actual CPR rate of 19%.
Source: Company Press Releases and Saibus Research Estimates
Disclosure: I am long MTGE. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.
Additional disclosure: This article was written by an analyst at Saibus Research. Saibus Research has not received compensation directly or indirectly for expressing the recommendation in this article. We have no business relationship with any company whose stock is mentioned in this article. Under no circumstances must this report be considered an offer to buy, sell, subscribe for or trade securities or other instruments.