Residential Mortgage Backed Security Funds (RMBS) are funds whose business it is to provide financing in the residential mortgage market, and a lot of these funds have been interesting lately because of their exceptionally high dividend yield. They are typically considered part of the REIT universe because of their favorable tax treatment.
These funds can and usually do have some percentage of their portfolio invested in "Agency-backed", that is "Government-backed", securities from various government-sponsored enterprises, such as the Government National Mortgage Association (Ginnie Mae) the Federal National Mortgage Association (or Fannie Mae) or the Federal Home Loan Mortgage Corporation, (or Freddie Mac).
Today's project was proposed by one of my followers, who asked whether the percentages of agency-backed mortgages in the various RMBS funds are stated somewhere. The obvious follow-on is: are the funds that are the more heavily invested in Agency-backed securities better in some way?
I've selected a group of REITS with the following criteria: Market capitalization of more than $750M and Operating Margin of greater than 75%. Since most of the RMBS in the current environment are borrowing money for essentially zero interest and lending it out at the current mortgage rate, these businesses are extremely profitable, and as REITs are required to distribute most of this income to shareholders, hence the high dividend payout.
I ended up with an even dozen funds as follows:
Note that the list resulted in three different classes of funds: Those that invested strictly in residential mortgages, those with a portion of residential and a portion of commercial, and one outlier, National Health Investors (NYSE:NHI) which actually owns and leases a number of health care facilities and has a small mortgage portfolio. For the purposes of this study that fund is in a different class but I will have more to say about that company later.
|% Agency Backed||Operating Margin||Div Yld||1-Yr Perf|
|CREXUS INVESTMENT CORP||CXS||20||87.47%||12.23%||-24.71%|
|CHIMERA INVESTMENT CORP||CIM||26||93.33%||19.48%||-36.73%|
|INVESCO MORTGAGE CAPITAL INC||IVR||64||93.38%||23.60%||-37.14%|
|MFA FINANCIAL INC||MFA||66||93.33%||14.77%||-18.43%|
|TWO HARBORS INVESTMENT CORP||TWO||81||91.78%||17.24%||-9.46%|
|ANNALY CAPITAL MANAGEMENT INC||NLY||90||94.00%||14.85%||-11.31%|
|ANWORTH MORTGAGE ASSET CORP||ANH||100||93.94%||14.09%||-10.30%|
|CYS INVESTMENTS INC||CYS||100||91.01%||15.12%||2.72%|
|CAPSTEAD MORTGAGE CORP||CMO||100||93.07%||14.08%||0.48%|
|HATTERAS FINANCIAL CORP||HTS||100||95.96%||15.00%||-15.31%|
|AMERICAN CAPITAL AGENCY CORP||AGNC||100||95.43%||19.49%||-3.20%|
The percentage of each fund's agency-backed portfolio is in the Management Discussion section of the most recent 10Q for each company. I have also included the current operating margin, dividend yield, and 52-week price performance.
A simple examination of the data suggests that the funds with the lowest agency-backed percentage did worse in market performance for the last year, and here is the scatter plot:
The statisticians will tell us that an R-squared correlation coefficient value of 0.62 is not especially strong, and says nothing about causality. Since dividend yield is a function of market price, one would also expect there to be some correlation between those two values, but in fact that is not the case:
The percentage of Agency backing had no correlation with Dividend Yield, nor, for that matter, with Operating Margin:
The sum of all of this is that there is no real advantage from an operating standpoint to having a portfolio that is 100% Agency-backed securities, but there is a marketplace value, or at least there was over the past 52 weeks, which was not directly related to the ability of the beaten-down funds to make money.
The two data points at the bottom of the first chart above were Chimera and Invesco. We know what happened to Chimera. The market punished the stock for management's revaluation of their portfolio to more closely conform to GAAP, the revaluation had nothing to do with their operating margins. The case of Invesco is less clear, they did have a 20-million share stock issue in July which amounts to about a 20% dilution of the float, but they announced recently a 7-million share buyback, which offsets a third of it.
Crexus Investment Group (CRX) is the third outlier, it is engaged mainly in commercial mortgages, has a much lower dividend payout, and does not really fit into the same class as the RMBS's above.
So what are we to make of all of this?
First of all, the percentage of agency-backed securities for each of these funds is listed in the 10Q.
Secondly, all of the funds in this little group are extremely profitable, and most pay exceptionally high dividends, and whether the security is 100% Agency-backed or not does not appear to have any bearing on the actual operations of the business, nor what kind of dividends they choose to pay out.
Thirdly, there was a market-price effect on some of the funds with lower percentages of Agency-backed securities, but they were due to factors other than their operating margins. I am not ready to argue with the conclusions in this article which suggests that these two funds may be due for a recovery, This article of a few days ago voices some concern about IVR's book value, and they recently cut their dividend. There are conflicting opinions.
Fourthly, overriding a lot of these issues, is the fact that most of these funds have their basis in the residential real estate market, and the value of the underlying properties are in question in most of the real estate markets in the country. No one is talking about this important issue, and naturally it interjects an element of risk into the equation which (hopefully) is compensated for by the nice, high dividends. These businesses are also vulnerable to changes in government policy, the solvency of the Government Sponsored Enterprises, and other issues beyond our control.
Now, a word about National Health Investors: Here is a business with almost an 80% operating margin engaged in the business of owning hospitals, and leasing them back to the companies that operate them, mainly to National Health Care (NHC). They also have a portion of their portfolio in assisted living places and other long term care facilities, and that business is not going to get any smaller in the next few years. The compelling reason for owing NHI is completely different from the rest of this group, because of the lower dividend, but you have to emotionally like where they are positioned in the marketplace to take advantage of demographic trends.
The world is chaotic, and there are no guarantees on anything, including investing in an RMBS with a 100% Agency-backed portfolio.
Disclosure: I am long CIM, CYS, TWO, AGNC, IVR.
Additional disclosure: I will be willing to hear arguments for and against reinvesting my nice dividend yield in one of the above RMBS funds.