As I mentioned in this earlier post I think there is an good investment opportunity in mortgage REITs (mREITs from now on). Its one of the the few remaining relatively undervalued sectors in the market and offers a compelling dividend yield of about 16%. In this post I’ll provide a basic economic framework to evaluate REITs and provide valuations for the big cap (>$1B) MLP names I’ll be covering.

In my first post on mREITs I described how the companies in this sector make money. Basically, the companies leverage up their equity and invest the funds in mortgage backed securities. The difference between what the mortgage investments yield and what they pay to borrow money determines their returns. Pretty simple. Of course, the devil is in the details. In the table below, I present a simple model for mortgage REIT economics using data for the large cap mREITs. Data is as of the quarter ending March 31, 2011. The companies in this table all have differences in their business model but the basic economics are the same. For example. NLY focuses exclusively in Agency mortgages, CIM primarily non-agency mortgages, and IVR does a little of both.

Lets walk through NLY to illustrate the model. The first line shows that NLY owns a portfolio of mortgages that yields 3.79%. The amount it costs for NLY to borrow money is shown on the second line, a cost of funds of 1.62%. This cost of funds line is net of any hedges the company may have. The difference between the portfolio’s yield and what it cost them to borrow is how much money they make before expenses, their net interest margin. 2.17% in NLY’s case. This is like the gross margin line for a typical corporation. The net interest margin is then multiplied by how much leverage they have (6.3 for NLY), i.e. how much money they’ve borrowed relative to their equity, and added to the yield on their equity to arrive at a gross ROE (return on equity), 17.46% for NLY. Then overall corporate expenses are subtracted to arrive at a net core ROE of 16.11% for NLY. The core ROE number is critical because it basically determined the dividend paying capacity for the mREIT. In general, you want core ROE percentages to be the same or higher than the dividend yield. Notice that this is the case for all the mREITs except CYS.

Most of the REITs in the table report core ROE which is different from GAAP ROE. Core ROE or core earnings are meant to represent the dividend paying capacity of the company. Items such as one time gains from the sale of mortgage securities go into GAAP ROE which may not be representative of the on-going dividends the firm can pay. Basically, pay attention to core ROE and look for it to be at or above the dividend yield of the stock. By sticking to this rule of thumb its hard to go wrong. If there is a quarter where the core ROE falls below the dividend yield then its time to look into the details. For example, for CYS in the table above the core ROE is 10.4% but their dividend yield is 19%. Not a good sign in the face of it. But diving into the details you find that CYS carries a bunch of MBS forwards (a type of hedge) on their books that impacts their core ROE but not their dividend paying capacity.

In today’s environment of low short term interest rates mREITs are making good money. The net interest margins in the table above are pretty high relative to history. Also, because of these large margins the mREITs have been able to keep their leverage ratios low. For example, in the past, NLY has run leverage ratios of between 8 and 12. It looks like the good money making conditions will persist until the end of 2012 at least according to most analysts. When net interest margins begin to decline mREITs will be able to increase leverage ratios to maintain good returns.

What is happening now is that there is some fear in the market that mREIT margins will decline as interest rates increase and thus their dividend yields will decrease. For this reason mREITs continue to trade at reasonable valuations. If and when interest rates begin to rise mREITs have a few tools at their disposal to mitigate the impact. One, the low leverage rations give them room to increase leverage as interest margins decline. Second, most REITs hedge a portion of any potential interest rate impact, especially on their cost of funds, through interest rate swaps. And when this does start to happen an investor will be able to see it by monitoring the parameters I listed in the economic model. History has shown that an investor can see these big impacts coming way before they impact the stock price in a big way.

Finally, in the table below I show the big cap mREITs, their dividends, dividend yields, stock price as of last Friday and their valuations. Price to book is my favorite valuation metric for mREITs. Normally, price to book levels of 1.1 have proven to be good entry points but each stock is slightly different and you should do your own research.

In summary, I think mREITs represent decent value in today’s market. A lid is being kept on valuations due to the fear of imminent large rises in interest rates. These fears are overblown and underestimate the tools the mREITS have to mitigate the impact of rising rates. An average dividend yield of 16% for this group of mREITs is an attractive risk reward tradeoff.

Disclosure: Long CIM, NLY

Tagged , , , , , , , ,

14 thoughts on “ Mortgage REIT economics remain compelling

  1. Paul:

    I’m long on NLY, CIM, HTS & RSO. 25% in each. Looking for a 5th candidate. AGNC, ARR, CYS?


    1. Hey Rick. Why do you need a 5th? I’d just put more money in NLY. Its’ 10 times bigger than its nearest competitor and the lowest risk mREIT out there. Size and history will matter if things start heading south. If you still want a 5th, either AGNC or CYS are pretty good. CYS is cheaper on a P/B basis but you should understand the MBS forwards that they use to hedge interest rate risk. ARR is too small for my tastes and its leverage is too high for this environment. RSO is not an mREIT per say – I consider them more of a commercial real estate player/specialty finance company, i.e. a more risky space. Why own them when you can get higher yield in a government guaranteed mortgage REIT? Food for thought anyway…


  2. Why a 5th? I am concerned about having mose than 25 or 30K in a single stock. The problem is my level of education and capacity to read financial statements and discern. To that effect your recommendation of sticking with the big NLY is well taken. BTW, the best of these positions has been HTS.

    Do you understand the recent receeding in this whole group? Are there any indicators one must understand and observe to warrant a sell? Do you have stop loss oders in the long REIT positions?

    Thank you Paul


    1. Rick, the best way to look at diversification is by percentages of your portfolio. If you stick with hard dollar numbers, as your portfolio grows, you will end up with way too many stocks. A max position size of 2%-5% is usually a good target to start with. As you get more comfortable analyzing stocks you can go higher if it fits your style.

      Here is another issue. Placing equal bets on stocks is not a way to make superior returns. Some stocks are better than others. To really increase your chances of making good returns you need to size your positions relative to the opportunity. You can start out with equal positions but as you get familiar with the companies and figure out which one is doing better, or cheaper, etc… you should increase your positions in those good ones vs adding another one to your portfolio. Position sizing is a very important thing to lear. I recommend two books; Fortune’s Formula and The Dhando Investor.

      Lastly, the price moves for this group have been very small recently, nothing to write home about. Look back in history at how much this group moves. It moves a lot – very volatile. And yes, there are indicators to watch for this group. Look at my spreadsheet in my mortgage reit economics post. The indicator that drives returns for this group is the net interest spread. I’m working on a post on risks in mortgage reits that will help in this regard. As for understanding the financials of this group more I recommend two things; pick a mREIt, read their Q1 earnings release and listen to the conference call. Write down your questons and then email them to and I’ll walk you through them. Its not that hard to understand this group.

      Personally, I do not use stop losses on this group. Sometimes when they go down I buy more. It depends on the fundamentals. I do consider stocks in this group a trade, for me meaning a holding from 1 to 3 years. So, I have specific buy points and sell points for each holding before I ever take a position.

      Hope that helps.


  3. Paul:

    I sort of instictively started with 100k divided into 4 stocks purchased over the course of a year. Presently 15% of my folio is in these products in almost equaly %ages. They have been the “bell” part of the bar-bell. A short adventure in and out SLV and PSLV was productive. Just an exeriment to see how stop loss orders work.

    By end of year I will oblide you to review my folio to see how this new student has faired and see where a little tweaking might be necessary.

    As far as Net Interest Spread… I am guessing one would look for a high interest spread and Low Leverage?

    Looking forward to your next post on this subject.


    1. TWO is decent. Its probably got the best hedge book of the mREITs. I like them. But for about the same risk I prefer AGNC or CYS which sport higher yields. For the about the same yield as TWO you can pick up the best in class behemoth in the space NLY.


  4. Hi,

    I would like to know why don’t you take into account NLY’s Other Income (which in the income statement is around 243 million USD) to come up with the Net Core Roe? In addition, could you shed some light on how did you calculate the OPEX? As a percentage of? Thank you

    1. Core ROE is trying to capture what the mREITs make on their core business of making money off the interest spread. It helps to compare apple to apples across the different mREITs. For NLY, that $241.7 million of other income as of March 31, 2011 (not $243million) is a mainly ($169.3 million) unrealized gains on interest rate swaps which is a one time event, easily reversible in subsequent quarters, not representative of their on-going earnings power.

      Opex is calculated as a percentage of equity. Unlike other mREITs, NLY reports it as a percentage of assets (0.23% of assets for the March 2011 quarter). Since most other mREITs report it as a percentage of equity I do it that way. Just need to pick a consistent way to compare across companies.

      Hope that helps.


    1. Hi Keith. It depends. For this particular post, most of the data came from the individual earnings reports of the mREITs, or their 10-Qs, or their investor presentations. All the data to calculate these metrics is reported by the companies.

      In general, if I did not crunch the data myself, I post the source of my data in the table or chart I use. For example, I use credit suisse charts quite often.


      1. Thanks Paul, the logic and analysis definitely makes sense and I’ve been tracking REIT’s and MREIT’s. But I’m not great at compiling the data. I’m much better at analyzing the information. Any suggestion on how to put together these charts or an eaiser way to obtain the necessary research data? For instance, I could not find Core ROE on any financial statements or the 10Q’s.

        1. Kieth, unfortunately there is no easy way. I read every quarterly report and listen to ever conference call. You could try and see if your broker offers some free research from one of the big banks. My broker, TD Ameritrade, offers Credit Suisse reports. I sometimes use analyst reports for reference but I never solely rely on their numbers, particularly their forecasts.


Comments are closed.