I Had to Laugh

In an investment forum I enjoy, I saw a comment that made me laugh.

One investor, responding to a lengthy discussion of asset mix, hedging, funding and risk in Agency mortgage REITs (amREITs) said that they would stick with easy-to-understand banks, and let the experts play in mortgage REIT land.

Why did I laugh?

Because every bank I know of has a mortgage REIT inside it.  Not only that, banks run at far higher leverage ratios on their mortgage portfolios than any of the publicly traded mREITs or amREITs.

Let’s start with mortgages that aren’t guaranteed by government.  Those loans are the old-fashioned portfolio loans like Bailey Brothers (Jimmy Stewart’s mythical S&L) made within a community.  They are subject to “100% risk weight,” or 8% reserve capital requirements.

We should note that new BIS standards will raise this amount, but that’s the level it’s been at since the Basel II Accord was adopted in 2004.

Put another way, if the bank holds 8 cents of capital for each dollar of non-guaranteed mortgage loan, they are levered 11.5 to 1.  That is obviously higher than the 3:1 or so that mREITs that buy non-Agency paper use.  In Agency MBS, the numbers are even more “scary” at banks.  Agency MBS are “20% risk weight” assets.  That means 8% times 20%, or 1.6% capital reserves are required.  Theoretically, a bank can be an amREIT running 61.5 to 1 leverage if they only hold that minimal 1.6%.

In practice, most banks have higher measures of risk for their Agency MBS, and typically hold 3% to 6% in capital to deal with the price and spread risk of that product.  They report the excess above the required amount as part of their capital.  We do, after all, see a lot of banks carrying capital north of 10%, which tends to make investors pretty comfortable.

Now let’s compare that with the publicly traded amREITs.  They are running leverage of 7:1 up to around 8:1 at the end of their reporting periods (and I suspect a bit higher during the quarters).  That’s the same as a bank carrying as much as 15% capital against a portfolio that requires only 1.6%.

Now a few words on the other side of the story.  First of all, banks, especially S&L’s and community banks, fund themselves with “stickier” funding sources like deposits.  Those can take up to several years to reset, and don’t carry with them the risk of nightly mark-to-markets.  That’s not to say the banks skate free on market value.  They have to report their capital based on market value for any asset held for sale or available for sale.  They also don’t fund themselves exclusively by taking in deposits.  The S&L’s are huge users of FHLB (Federal Home Loan Bank) advances and loans, especially for their mortgages.  There is literally trillions of dollars of MBS and mortgage loans funded this way.  The FHLB can and does make margin calls, so that funding has a lot of the risk profile that the amREITs have with their repo funding.

For both amREITs and banks, loss of value in their mortgage portfolios is a hit to capital.  How it gets reported may vary, but the bottom line is the same.

So, when I see someone say that amREITs are too complicated, and banks are “simple,” I shake my head and wonder whether ignorance is truly bliss, since banks have a whole lot more going on than simply running a portfolio of liquid bonds (MBS).

When you think about it, the main reason banks are tight with credit for entrepeneurs and consumers is the fact that they can run a profitable “amREIT” with virtually no credit risk, and truly astounding ROE.  We like our 12% to 18% dividends, and worry about the risk being taken to get that kind of return.  If we broke out the Agency MBS and its funding from most banks, we’d be amazed to see returns well north of 30%, and the risk that attends that level of return.



8 Responses to I Had to Laugh

  1. Bruce B says:


    If the stand alone amREITs are conservative, why does the efficient market allow their yields to be so high? These yields for “safe” investments have always been a source of puzzlement.


    • hhill51 says:

      Several arguments support the idea that banks can afford to have higher leveage than amREITs. First is multiple sources of financing, some provided by government. Second is multiple business activities, especially all the things they do that nick us for fees.
      Having said that, it is a mystery to me why amREITs aren’t trading closer to 10% yield. With higher leverage and far more rate-hike storm clouds on the horizon, they traded to single digit yields and higher book multiples in 2004/2005.

  2. Bruce B says:


    That makes sense. I am glad you agree with me, in a sense. The yields seem as if they should be lower, especially given that we won’t see rate hikes here until 2015 at the earliest. Maybe sooner if savers demonstrate in front of the Fed with pitchforks that they show a willingness to use.


  3. Jim D says:


    I just discovered your site and want to thank you for it. It is extremely difficult to find good commentary about the amreits. I bought earlier this year during the time everyone was worried about the effects of the FED not purchasing anymore and just before the GSE buyouts, so pricing wasn’t bad. But, I thought that things looked reasonably well on the macro front and really couldn’t believe the level of yield. But, I also, thought that the general uncertainty about all manner of things is keeping the stock prices “nervous” in the amreits. Your site will help me going forward as I attempt to keep abreast of macro changes and determine just what it means for each amreit.

    My thesis has been that as we move thru the more difficult part of the cycle, if it happens in a reasonable way, a decline in current dividend should eventually rationalize to a more normalized yield situation because stability in the fearful part of the cycle has been reached and the share price, though initially faltering, would eventually recover. I mean, if ANH dividends .70/year in the future, at 10% yield that’s still close to current share price. So, a buy and hold income investor might experience a temporary share price dip, but eventually recover close to initial purchase price or better (mine is lower with ANH), all the while receiving dividends. Does that make any sense?

    • hhill51 says:

      That’s certainly the theory behind my core investments in the less risky amREITs. (By less risky, I am using shorthand for their duration mismatch and funding cost hedging programs.) Historically that gave less downside when spreads widened, but the game has really been screwed up by the Fed being there. At times over the past two years, fixed-rate MBS spreads have tightened “too much” given their extension risk without the mitigating rate resets the ARMs enjoy.
      I did a Roth conversion early in 2010, and leave a healthy chunk (about 35%) of that money in amREITs bought in 2009.

      • Jim D says:

        Do you view ANH and CMO as the less volatile or more safe of the amreits? I own a basket of them, including AGNC and CIM, but would probably make additional purchases in the more stable ones. I really do like Lloyd McAdams (hope I got the last name correct) and benefit greatly from listening to him at conference replays on the internet. Funny, he has the shortest slide presentations couple with the most information. But, I don’t know CMO although I have discovered their preferred lately.



      • hhill51 says:

        CMO used to be my “favorite” because they kept their duration gap very near zero. They blew it somewhat in the recent weird prepay pattern, where Agency buyouts caused a surge in prepays for new-issue ARMs.
        The hit to capital from paying a premium for those ARMs combined with the mismatch the prepays caused because the liability stream (locked in by swaps) was based on larger balances knocked them off their previously very well balanced portfolio strategy. That said, I still own my core (2009 purchase) CMO common, and recently added CMO-B, though the latter was at an embarrassingly high price vs the call price.
        ANH is still my largest amREIT in spite of their lower yield and less aggressive posture, and then I have some pieces of others. Clearly the home run hitters that have both high dividend yield and high portfolio duration have been far better total return investments over the past 18 months. So be it.

  4. Jim D says:


    Thanks. It’s nice we have the conference calls coming up. These amreits are fascinating companies. When, you plug in the uncertainties in the macro environment and the particular strategy of each management team, if does present a lot to think about. It’s interesting.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: