This research report was produced by The REIT Forum with assistance from REIT Forum Support.

Following reader feedback, we’ve enhanced the layout for this series. Your continued feedback is greatly appreciated, so please leave a comment with suggestions.

This article will be heavy on charts because we like to communicate with images, rather than words, whenever possible. Likewise, we will use several tables to more efficiently structure the data. Enjoy!

Mortgage REITs and Preferred Shares

We’ve consistently incorporated a significant allocation to preferred shares in our portfolio. We could simply hold the positions for income, but we take advantage of trading opportunities as well. Our goal is to maximize total returns, and this technique has worked wonderfully.

We also trade positions in the mortgage REIT common shares. We find this sector is particularly attractive because it can be so inefficient. Long term, share prices revolve around book value. In the short term, the price-to-book ratios can deviate materially. Simply by understanding that, you can put yourself in a better position.

Our other major source of allocations is equity REITs. While an investor might occasionally choose to trade an equity REIT position, the sector is a great fit for buy-and-hold investors.

We compare our performance against 4 ETFs that investors might use for exposure to our sectors:

(Source: The REIT Forum – Chart runs through 11/19/2020)

The 4 ETFs we use for comparison are:

Ticker Exposure
MORT One of the largest mortgage REIT ETFs
PFF One of the largest preferred share ETFs
VNQ Largest equity REIT ETF
KBWY The high-yield equity REIT ETF. Yes, it has been dreadful.

When investors think it isn’t possible to earn solid returns in preferred shares or mortgage REITs, we politely disagree. The sector has plenty of opportunities, but investors still need to be wary of the risks. We can’t simply reach for yield and hope for the best. When it comes to common shares, we need to be even more vigilant to protect our principal by regularly watching prices and updating estimates for book value and price targets.

Mortgage REITs

Today’s Lesson

Over the weekend, we published “AGNC Preferred Share Arbitrage“.

The article is built around the premise that AGNCM offers a materially worse value than AGNCP or AGNCO.

The market clearly agreed with us, as the relative values began shifting:

(Source: StreetSmart Edge)

This is a trade we had been suggesting to subscribers for the last two weeks and was highlighted when we wrote “Preferred Shares Week 228” on November 12th, 2020.

Same Dollar Amount or Same Share Count

Every investor knows that when you’re comparing two different investments, you need to use the same dollar amount for both positions. This is one of the most basic investing lessons we learn. However, it is also incomplete. Few investors go back to update that basic lesson to add a critical caveat. You’ll see it listed occasionally in some sources like the textbooks someone studies when they want to become a CFA Charterholder. However, that’s a pretty tiny niche of the total investing population.

What is this caveat?

When you’re comparing two roughly similar investments, you can create a third investment option. You could take either position and simply add cash to it. For instance, you might consider 1,000 shares of AGNCP plus cash against 1,000 shares of AGNCM.

In that manner, you would be able to contrast two portfolios that have the exact same balance, and they would have the same share count. Let’s look at how the picture changes.

This is the picture that a rookie investor will see:

Note: This table and the other tables will use the prices from the prior article because it simplifies the process.

Based on that simple table, they conclude AGNCM is better. It is important to emphasize that this mistake is so common that it was influencing the market price.

The problem here is that investors won’t even look out to 2024. That’s when AGNCM begins having a floating rate, specifically on 4/15/2024. AGNCP will join AGNCM in having a floating rate starting 4/15/2025.

We have checked the numbers and are confident that the difference between those two dates is equal to one year. Excellent math, I know. During that one year, AGNCM’s dividend rate will be lower than it is today, unless the short-term rates increase to at least 2.543%.

How can we know with such precision? We did some super-difficult math by subtracting one number from another:

Unless the short-term rate can contribute at least 2.543%, AGNCM’s dividend rate will be lower after 4/15/2024 than it is today.

When we compare two shares this way, things start to get a little messy because investors have to figure out how much of a change they expect in short-term rates. Here are a few different scenarios:

Now, some investors missed the change in the column titles or might still not realize that the fixed-rate period will end. So, let’s try it again:

Now we should all be on the same page.

Note: The short-term rate used here will be a benchmark interest rate. Since LIBOR will end, it probably won’t be LIBOR. That wouldn’t make sense. I expect another short-term benchmark will be chosen, and it will be similar enough that it won’t be an issue. Consequently, we won’t waste time on it. It should be obvious that AGNCM and AGNCP would use the SAME benchmark. Same isn’t an acronym, I just felt like yelling it at any person who would suggest otherwise.

We can clearly see that the fixed-rate dividend ends and the investor who owns 943 shares of AGNCM get less money than the investor who owns 1,000 shares of AGNCP.

READ ALSO  SE: Gold Hasn't Lost Its Allure, WGC India Director Says

Further, we can see that the gap widens if interest rates increase. Why would that happen? Because the investor with 1,000 shares of AGNCP is earning dividends on 1,000 shares rather than on 943 shares. We are quite certain that 1,000 is a larger number than 943, so this also makes sense.

This is kind of ugly because investors can’t predict where short-term rates will be in a few years. If anyone could predict this with incredible accuracy, they would be a billionaire, because there are options available for betting on the future of interest rates. To be clear, predicting with incredible accuracy is not the same as announcing that you were right years ago.

So, how do we simplify this concept?

We set the share counts to be equal, then we determine how much cash we would need to add to one side for the math to work:

In this simple example, we’ve found that 1,000 shares of AGNCM cost $1,340 more than 1,000 shares of AGNCP. Consequently, the investor who buys 1,000 shares of AGNCP simply keeps $1,340 in their account. This is a much simpler technique. Shares are callable at $25.00 each, which is another factor to consider. Let’s demonstrate how simple this math will be using call values.

For instance, what’s the call value on 1,000 shares of AGNCP? No calculators.


What’s the call value on 1,000 shares of AGNCM?


What’s the difference between those numbers?


What’s $0 plus $1,340?


I’m confident that every single reader knew those answers without even a slight pause.

Now, if I asked you to find the call value on 943 shares without a calculator, most of you will just decline to do the math (it’s $23,575). How about the difference in the total call value between $25,000 and $23,575? You can do it, but it still takes time. We want to make investing simple. That’s how we reduce mistakes. Other people get paid to make simple things complex. I get paid to make complex things simple.

Now, let’s assume that the investor earns 0.00% on their cash balance, no matter what happens in the market. Why? Well, it’s simple, and most investors don’t earn a high rate on cash anyway.

Example with 0% on Cash

We’re making two changes to the table:

  • Setting each position to 1,000 shares
  • Showing the difference in cash

This is what we get:

Those bright yellow cells reflect that we are assigning the investor a 0% rate of return on cash, despite a scenario where cash is expected to earn 1-2% by buying short-term securities.

You may notice that the difference in annual income has been normalized in this scenario. Now, regardless of the future interest rate, the difference in annual income is equal to $91.25.

That is inherently simpler! The investor following this technique no longer needs to make a guess about future interest rates. They simply accept that future interest rates will happen, and they design the math so that it doesn’t matter.

Maximizing Your Income

Are you trying to maximize your income over the next 4 years or over your lifetime? It matters.

Let’s say you only want to maximize your income over the next 4 years and then you plan to sell the position. Will the buyer be ignorant of the change in rates? That would be great for you, but it might not happen. So, you’ve got to plan for it. This is how you do that.

The investor who took 1,000 shares of AGNCP plus $1,340 in cash will start with more cash upfront, but investors think AGNCM will eventually catch up because of the higher yield. They could not possibly be more wrong:

We ran the hypothetical cash balances over 15 years assuming 1,000 shares in each portfolio and using 1% as the short-term rates. We’ve also assigned $0 in interest on cash for the entire period. If we paid interest on cash, then AGNCP wins by even more. Consequently, the investor in AGNCP flat out wins. The only scenario where the investor in AGNCM wins is one where the short-term rates are incredibly high in 2024. That’s it. No other scenario would enable AGNCM to even be competitive based on this gap in share prices. How could an investor win with AGNCM? Sell it to a sucker at a higher price. Too honest? Too transparent? Yeah, we hear that.

What About a Call?

A call could happen. It isn’t highly likely in these shares, but it is within the realm of possibility. What if a call happens? Remember that we designed the portfolio the easy way. We simply had an extra $1,340 in cash sitting in the AGNCP portfolio. The call value on 1,000 shares is $25,000. Even though AGNCM paid out a higher dividend rate for those first 3.5 years, it wouldn’t be nearly enough to make up for the head start. The AGNCP portfolio started with an extra $1,340 in cash. The extra dividends from the fixed rate over the first 3.5 years would only be $656.25. Again, we must stress that we are certain that $1,340 of cash is more valuable than $656.25 of cash.

READ ALSO  SE: What's 'Social Commerce'? What Does China's Pinduodo Do?

Want to bring taxes into the equation? What is the tax rate on having $1,340 of unspent cash? 0%. Can you get a tax rate lower than 0%? No, probably not.

Learn Something?

We have a very simple premise here. More money is better than less money. Our choice of AGNCP results in more money than the choice of AGNCM. If we look at AGNCO, we end up with similar math. However, we’re not going to do that. Why? Because that takes a substantial amount of time! For investors who want to commit this concept to memory, try running the same math on AGNCO.

We will close out the rest of the article with the charts we provide for readers to help them track the sector for both common shares and preferred shares.

Let the images begin!

Price-to-Book Value – Using Q3 2020 Book Value

All the mortgage REITs within our batch have reported their Q3 2020 earnings, so we have trailing book values for all of them.


(Source: The REIT Forum)

Remember that these are price-to-trailing book ratios. They are not using estimates of current book value. Book values have changed during Q3 2020 and, to a lesser degree, during Q4 2020.

Repeated note – there are two mortgage REITs we need to highlight here:

  • AG Mortgage Investment Trust – We are using the Q3 2020 book value reported by management, which does not deduct the value of accrued dividends for preferred shares. If the preferred dividends were paid, it would reduce common book value under these calculations. This method is accepted under GAAP.
  • MFA Financial reports “GAAP book value” and “economic book value”. We’ve chosen to use the GAAP book value to remain consistent.

Unfortunately, we have to repeat those bullet points every time we publish, because it regularly comes up if we don’t mention it.

Book values will have changed some already during Q4 2020. We aren’t including that in our public articles (except for index cards). Scott Kennedy provides frequent updates on estimated book value, ratings, and price targets through The REIT Forum.

Dividend Yields

Dividend yield often comes up in the comments, but picking based on dividend yield is stupid and regularly results in terrible performance. Don’t do it.


(Source: The REIT Forum)

This chart is still in the same order as the prior charts. Consequently, you know the highest price-to-book ratios (using trailing GAAP book value) for each segment will be at the top. If you see a mistake, please feel free to say something. Occasionally, the data for dividend rates requires a manual update.

Earning Yields

One of the next things investors may ask about is the yield using core earnings. This chart puts together the core earnings based on the consensus analyst estimate. Beware that the consensus estimate may not always be the best estimate.


(Source: The REIT Forum)

Consensus estimates aren’t always the best, and there are ways to increase “Core Earnings” through accounting decisions or modifying hedges. Consequently, investors should still take these values cautiously. We do not depend on the consensus estimate to make decisions.

Preferred Shares

After testing out a series on preferred shares, we decided to try merging it into the series on common shares. After all, we are still talking about positions in mortgage REITs. We don’t have any desire to cover preferred shares without cumulative dividends, so any preferred shares you see in our column will have cumulative dividends. You can verify that by using Quantum Online. We’ve included the links in the table below.

To better organize the table, we needed to abbreviate column names as follows:

  • Price = Recent Share Price
  • BoF = Bond or FTF (Fixed-to-Floating)
  • S-Yield = Stripped Yield
  • Coupon = Initial Fixed-Rate Coupon
  • FYoP = Floating Yield on Price
  • NCD = Next Call Date (the soonest shares could be called)
  • Note: For all FTF issues, the floating rate would start on NCD.
  • WCC = Worst Cash to Call (lowest net cash return possible from a call)
  • QO Link = Link to Quantum Online Page

Ticker Price BoF S-Yield Coupon FYoP NCD
AGNCM $23.43 FTF 7.43% 6.88% 4.92% 4/15/2024
AGNCN $23.85 FTF 7.43% 7.00% 5.66% 10/15/2022
AGNCO $23.11 FTF 7.12% 6.50% 5.71% 10/15/2024
AGNCP $22.30 FTF 6.95% 6.13% 5.58% 4/15/2025
NLY-D $25.41 7.38% 7.50% 7.38% 12/23/2020
NLY-F $23.63 FTF 7.35% 6.95% 5.52% 9/30/2022
NLY-G $22.20 FTF 7.32% 6.50% 4.95% 3/31/2023
NLY-I $23.75 FTF 7.11% 6.75% 5.49% 6/30/2024
MFO $25.40 Bond 7.96% 8.00% 7.96% 12/2/2020
ARR-C $24.27 7.24% 7.00% 7.24% 1/28/2025
DX-B $25.28 7.64% 7.63% 7.64% 12/27/2020
DX-C $23.95 FTF 7.29% 6.90% 6.01% 4/15/2025
CMO-E $24.35 7.80% 7.50% 7.80% 12/27/2020
EFC-A $22.45 FTF 7.59% 6.75% 6.09% 10/30/2024
NRZ-A $23.60 FTF 8.03% 7.50% 6.45% 8/15/2024
NRZ-B $22.67 FTF 7.94% 7.13% 6.53% 8/15/2024
NRZ-C $19.86 FTF 8.11% 6.38% 6.61% 2/15/2025
PMT-A $24.69 FTF 8.23% 8.13% 6.13% 3/15/2024
PMT-B $24.91 FTF 8.03% 8.00% 6.24% 6/15/2024
AIC $24.10 Bond 7.01% 6.75% 7.01% 12/27/2020
AIW $24.70 Bond 6.76% 6.63% 6.76% 12/27/2020
ANH-A $25.20 8.68% 8.63% 8.68% 12/27/2020
ANH-C $24.13 8.01% 7.63% 8.01% 12/27/2020
CIM-A $24.10 8.44% 8.00% 8.44% 10/30/2021
CIM-B $23.45 FTF 8.68% 8.00% 6.53% 3/30/2024
CIM-C $22.94 FTF 8.59% 7.75% 5.51% 9/30/2025
CIM-D $23.51 FTF 8.66% 8.00% 6.06% 3/30/2024
TWO-A $24.22 FTF 8.49% 8.13% 6.15% 4/27/2027
TWO-B $23.30 FTF 8.28% 7.63% 6.06% 7/27/2027
TWO-C $21.70 FTF 8.46% 7.25% 6.11% 1/27/2025
TWO-D $23.96 8.20% 7.75% 8.20% 12/27/2020
TWO-E $23.61 8.05% 7.50% 8.05% 12/27/2020
CHMI-A $25.49 8.15% 8.20% 8.15% 8/17/2022
CHMI-B $23.87 FTF 8.77% 8.25% 6.23% 4/15/2024
IVR-A $23.83 8.24% 7.75% 8.24% 12/27/2020
IVR-B $22.96 FTF 8.61% 7.75% 6.01% 12/27/2024
IVR-C $22.50 FTF 8.50% 7.50% 6.25% 9/27/2027
NYMTM $22.05 FTF 9.06% 7.88% 7.66% 1/15/2025
NYMTN $22.38 FTF 9.07% 8.00% 6.71% 10/15/2027
NYMTO $21.81 9.17% 7.88% 9.17% 12/27/2020
NYMTP $21.88 8.99% 7.75% 8.99% 12/27/2020
AAIC-B $19.82 9.00% 7.00% 9.00% 5/12/2022
AAIC-C $21.26 FTF 9.90% 8.25% 7.07% 3/30/2024
MFA-B $23.90 8.00% 7.50% 8.00% 12/27/2020
MFA-C $22.63 FTF 7.31% 6.50% 6.26% 3/31/2025
MITT-A $19.80 10.43% 8.25% 10.43% 12/27/2020
MITT-B $20.01 10.00% 8.00% 10.00% 12/27/2020
MITT-C $19.85 FTF 10.08% 8.00% 8.45% 9/17/2024
READ ALSO  Trump wants US government to restrict more goods & services from China

There are a few things you should know at the start:

  • When a share can be called on short notice, the annualized yield-to-call reaches absurd levels. Investors shouldn’t put too much weight on it. On the other hand, a negative number can be a significant concern. Consequently, we decided to include it in the chart.
  • We sort our spreadsheet for subscribers by risk ratings within each sector. We decided to use the same technique for this series, since it communicates more information to readers. You’ll notice a general correlation where lower risk correlates with a higher price and lower yield, though this link isn’t absolute.

For each metric, we have two charts. Why use two charts? Because it is much more convenient for readers who want to enlarge the charts. We simply can’t fit 40+ shares into a single chart and still have it show up well on a mobile device.

Share Prices

We will start with the prices:


(Source: The REIT Forum)


(Source: The REIT Forum)

That chart gives you a pretty quick feel for which shares are trading at a discount to call value. Each of these preferred shares has a call value of $25.00, but that doesn’t mean a share will be called. The company decides if they want to issue a call or not.

Dividend Yield

Let’s move onto the stripped yield. This is the way dividend yields should be handled for preferred shares:


(Source: The REIT Forum)


(Source: The REIT Forum)

Stripped yields are vastly more useful than “current” yields for preferred shares. The stripped yield uses the stripped price. That’s different from using the current price, because it means we already adjusted for dividend accrual. This makes the process easier for investors.

We can talk about shares using “regular prices”. Those are the prices an investor would actually use when entering an order.

However, we will provide the stripped yield to adjust for the dividend accrual. In the spreadsheets we host for subscribers, we include the actual ex-dividend date, or the projected ex-dividend date if the actual date isn’t yet known. If you’re planning to buy a share, it’s always wise to check if the shares just went ex-dividend so you can adjust your targets accordingly.

Floating Rate Dividend Yields

Since many of these shares switch over to floating rates, we also want to consider what the yield would be if the floating rate was in effect and shares were still at the current price. To demonstrate that, we use the “Floating Yield On Price”. If the share remains at a fixed rate indefinitely, then the value doesn’t change:


(Source: The REIT Forum)


(Source: The REIT Forum)

One point we need to emphasize here is that we are dealing with yields. A yield must involve the share price. We aren’t simply showing the new “rate” if the share began floating, we are adjusting the new rate for the stripped price.


You don’t have to follow us. You don’t have to care about our research. If you’re happy with a high dividend yield and a dwindling account value, there are plenty of authors who can cater to you. On the other hand, if you’re focused on generating total returns and want to use REITs to do it, you may love our research. Hit the “Follow” button beside my name to start seeing more of our research.

Make sure to leave a comment and let us know what you think of the layout.


  • None. Go read the part about relative values in preferred shares again.

Our method works. We know because we buy the same shares we recommend. We track our results on a real portfolio and we compare our returns with the major ETFs for our sector:

Those four ETFs are:

  • MORT – Major mortgage REIT ETF
  • PFF – The largest preferred share ETF
  • VNQ – The largest equity REIT ETF
  • KBWY – The high-yield equity REIT ETF

Sign up now. Take advantage of our October sale!

Disclosure: I am/we are long NLY-I, AGNCO, TWO-E, NRZ-C, NYMTM, CIM-A, AGNCP, NRZ, AGNC, NLY, SLRC, NYMT, GPMT, PMT. 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: As a reminder, Scott Kennedy also is an author for the REIT Forum. You may see his commentary featured in our articles and may notice an extremely high amount of overlap in our ratings, so subscribers reading this article should see Scott’s latest REIT Forum sector update for more detail.


  • bitcoinBitcoin (BTC) $ 36,555.00
  • ethereumEthereum (ETH) $ 1,260.15
  • tetherTether (USDT) $ 0.998805
  • litecoinLitecoin (LTC) $ 145.64
  • bitcoin-cashBitcoin Cash (BCH) $ 496.75