In my continuing series that is NOT ADVICE, I’ll share my impression of MFA, another of the original round of amREITs that were launched in the late 1990’s.
It sports just over $2 billion in current market cap, and has an indicated yield using last quarter’s dividend of 13.3% trading at today’s price just under $7.60 per share.
This is one of my core mREIT holdings dating to the great washout last year (when everything was unbelievably cheap). I haven’t been moved to add any since that time.
MFA has a bit of an identity problem, because it can’t decide whether to be an Agency mortgage REIT or non-Agency. Or maybe that’s how they differentiate themselves.
On the positive side, they do not have the issues CIM has as a pure non-Agency mREIT that achieves more return through structured deal leverage. On the negative side, they never quite get the respect the pure amREITs get, because when investors hate non-Agency paper they really hate it.
During the first quarter of 2010, they sold off nearly $1 billion in Agency MBS for an average price of just over 105. They deployed that cash into just under $200 million of non-Agency MBS (private label), and committed to another $121 million of forward purchases in the non-Agency space over the period of Q2 and Q3.
That gives us a rough idea of their relative leverage, and a way to guess at how much prepayment and principal amortization they expect to have over the next few months on that Agency portfolio.
It looks like they are buying $1 in non-Agency MBS (purchase price, so maybe $1.20 to $1.50 face amount, averaging $1.39) for each $5 to $6 in Agency principal they sold or expect to be repaid.
With that shift away from Agencies, which probably averaged 7x or 8x leverage (they held more than $300 million in Agency MBS that were not pledged to repo), their all-in leverage ratio was under 3 times equity at the end of Q2, 2010. That will probably go up, since it sounded in their call like they expected the completion of the Fed buying program to bring wider spreads. That gave them half the leverage ratio held by the “pure” amREITs.
I was a bit surprised to hear in their last quarterly call that 91% of MFA’s Agency paper is Fannie Mae. In general, Fannie and Freddie paper trades very nearly the same (though the two of them used to wage the Public Relations version of all-out war to try to convince the market otherwise).
As a consumer of MBS over the years, the Agency that I found intriguing and different was Ginnie Mae (GNMA), the one agency that always had an explicit government guarantee. Ginnie also had a different class of borrower, being exclusively FHA and VA loans. Those were the small down payment (as little at 3%) loans given to young families and recent vets on what were usually “starter” homes. Subprime before the term got popular, in other words.
FHA/VA loans, besides being smaller on average, were also assumable. That created the potential for them to be VERY slow to prepay, because people selling houses could transfer the loans to the new owners, assuming the new owner qualified and was willing to pay the seller for all the increase in home value that might have occurred.
During the peak in mortgage rates in the early 80’s, there were tons of creative ways that people figured out to pass along these attractive 6% loans when new loans cost as much as 15%. Having the seller take back a second mortgage, for example.
But, I digress.
MFA is busy shifting its capital to the non-Agency paper that IVR and CIM buy (along with way too many special funds launched in 2008 and 2009 to buy distressed MBS).
The non-Agency stuff was yielding in the high single digits if safety was your primary concern. They said they’re be able to get some amount of financing at leverage of three or four to one, because not financing this paper would result in a lower ROE (Return on Equity) otherwise. Financing spreads are pretty steamy, though, running between L+100 and L+175 for 3 or 6 month financing.
The real benefit of the non-Agency paper right now is the deep discount prices, which mitigate the effect of historically high premium prices for the Agency paper, assuming that prepayments come equally in both kinds of underlying mortgages.
That said, MFA is still one of the more boring stocks in a boring group. I don’t expect any upside surprises in the dividend or the book value, and I don’t expect them to shine as active portfolio managers, either.
In a stroll down memory lane, I just printed out a mortgage-related stock group summary from September 1, 2006. Needless to say, of the 26 names on the list, you can count those that survive and own mortgages or mortgage securities today on the fingers of one hand.
MFA was trading within a few cents of it’s book value, which was listed at $6.94 per share. Today, it trades within a few cents of its book value, which was listed at the end of last quarter at $7.67 per share.