As investors search for income-generating assets in a low-rate environment, demand for ultra-high-yielding investments has spiked. As I’ve discussed in recent articles, this has been seen in preferred equities, derivative-based strategies, and highly levered investments such as mortgage REITs. While this has been an excellent strategy during the recent recovery period, it seems that many investors may not fully grasp the "big picture" economic risks which exist in these assets.
In my view, mortgage REITs are a fascinating segment of the stock market today. These companies purchase mortgages, often large baskets of mortgage-backed securities, and use high leverage to magnify returns. Truly, they operate much more similar to a traditional bank than a REIT, but they share the distributional tax benefits of REITs. It is still common to find mortgage REITs with dividend yields near or above 10%.
I have covered this segment in multiple articles over the past year, but a recent trend shift may soon cause a material market move. As I’ve discussed, it appears that most mortgage REITs are highly dependent on the Federal Reserve’s ongoing Q.E. program. Many mREITs are highly concentrated in low credit risk mortgage-backed securities guaranteed by Fannie Mae and Freddie Mac (OTCQB:FMCC). These same assets have been one of the top targets of Federal Reserve purchase as a means of stopping mortgage rates from spiking and encouraging housing demand.
This situation has led to a stellar recovery; however, as the Federal Reserve has flirted with tapering, there has been a decline in most mortgage REIT’s value. It is now believed that the Fed may begin tapering by November, which means that they may soon slow mortgage purchases. While the Federal Reserve’s decision is likely reasonable considering inflation concerns, I believe there is a significant chance this will soon result in a double-dip drawdown for many mortgage REITs.
The mREIT ARMOUR (NYSE:ARR) may be particularly at risk due to its high exposure to assets, which this change may impact. ARMOUR is a popular company with a dividend yield of 11.3% and a forward "P/E" of 11x. However, the company's significant exposure to agency residential mortgage-backed securities may be a considerable risk.
The Federal Reserve’s mandate is to moderate inflation and ideally maintain high employment. Through the second mandate, the Fed has found itself now owning over a third of all U.S. mortgage-backed securities. This is because the events of 2008 proved that what happens in the mortgage market eventually impacts employment.
That said, the Fed’s mortgage market interventions in 2008-2012 pale in comparison to their intervention last year, purchasing over a trillion in mortgages. Indeed, it is abundantly clear that the mortgage rates would be higher if not for these purchases, as seen in the difference in yields between mortgages and Treasury rates. See below:
Over the past two years, the Fed’s ownership of mortgage-backed securities has skyrocketed. Private investors such as banks have bought as well, but as that figure has slowed, the mortgage spread stopped declining.
In my view, mortgage-backed securities have likely never been as poor investments as they are today. Most currently have yields around 1.2%, which is quite negative in real terms (after accounting for rising inflation). Technically, those guaranteed have low credit risk since Fannie and Freddie are supposed to make up the difference in default or bankruptcy, but this guarantee is not necessarily guaranteed. While mortgage delinquencies have declined tremendously since last year, it is not surprising that banks have shown reduced demand for these assets as they’ve been kept up almost entirely by the Federal Reserve.
The result of the Fed’s demand has been a major compression in the spread between mortgage rates and Treasury rates. This spread is significant for mREITs since they can hedge against moves in the Treasury market through swaps but cannot hedge directly against moves in the mortgage markets.
As you can see below, there is an inverse relationship between the mortgage spread and ARR’s book value:
Of course, other factors have led to a decline in ARR's book value. However, it used to own non-agency credit risk securities, which carried greater credit risk and had higher yields. Today, around 85% of ARR's assets are in agency mortgage-backed securities. However, many of these securities drastically declined last year and had to be sold for a considerable loss - permanently impairing ARR's book value.
The truth is that we cannot know exactly how the mortgage market will react to the Federal Reserve’s tapering. Based on my research, it seems clear that private demand for these assets will only rise if their yield compared to Treasuries rises as well. If mortgage spreads were to return to normal levels of around 1.4%, then mortgage rates would rise an expected 50-60bps from today's levels.
While this would make mortgage-backed securities more attractive, a 50-60 bps rise in the spread would likely directly cause an ~20-30% decline in ARR's book value based on its mortgage spread sensitivity risk table (2020 Annual report pg. 70). This would likely be concurrent with a rise in rates overall. ARR is not fully hedged against rates, and according to its latest annual report, a 1% rise in rates would cause its book value to decline by around 12.2%. Thus, I believe ARR carries around 30-40% in downside risk from tapering and more if the mortgage market becomes volatile again.
Other economic factors could impact ARMOUR. There has been a slight but material drop in housing starts and home sales which implies the supply of new mortgages may decline. There has also been some interesting activity in the yield curve, which helps drive mREIT and bank net interest margins and, therefore, dividends. The spread was declining last time I covered the market, which was not great news for ARMOUR. However, the decline has been stagnant since, which may mean dividends are safe for now. Still, I do not believe this is enough reason to offset issues from the end of the Federal Reserve’s support.
Compared to most mREITs, ARR carries greater exposure to the potential impacts of an end to the Federal Reserve's support. Still, because of this, it has declined over 10% since its spring levels and has seen a small spike in short selling. This has caused a decline in ARR's valuation statistics:
ARR has a higher dividend yield of 11% and the lowest price to book ratio compared to peers. This is an indication that there could be a discount opportunity in ARR. That said, ARMOUR's book value has likely declined materially since last quarter due to the recent moves in mortgage rates. ARR's leverage is similar to peers but below the riskiest, like Orchid Island (ORC).
Overall, I believe ARR may be undervalued compared to its peers today. As such, ARR may not be a great short opportunity at the current moment since it may bounce. That said, I still believe its book value is likely to decline considerably by next year.
While many investors and analysts focus on the high yields and “low credit risk” of mortgage REITs, I believe many are failing to grasp the risk of a rise in mortgage spreads. It is the chief risk that most mortgage REITs are exposed to. It is even more dangerous when leverage is as high as today since the normalization of the mortgage spread and/or mortgage rates may cause ARR's book value to decline 30-40%. However, if the mortgage market experiences another wave of margin calls, as occurred last spring, this figure could be much larger.
We must ask whether or not the Federal Reserve aims to protect investor portfolios or citizens. Investors often assume that the Fed has their back since their mandates often have the effect of helping investors. However, it appears today that the Federal Reserve has saved the housing market from implosion through its mortgage purchase program. This gives them no material reason to support the mortgage market, particularly with home prices becoming extremely high and pricing many out of the market.
It is possible that a tapering-related crash does not occur. We live in a relatively uncertain period that is prone to surprising twists and turns. Still, I believe this crash is necessary for the U.S. real estate market to reach a more sustainable equilibrium. Mortgage REITs may be great investments at lower prices as they tend to decline well below their book value during drawdowns. Still, I would not necessarily assume that the Federal Reserve will always support mortgages or other assets. The Fed is subject to many uncertainties and obligations and is marching into uncharted waters with its current Q.E. program. Accordingly, I believe it is dangerous to assume the “Fed put” will exist perpetually.
At this point, I am bearish on ARR but would not short-sell it since it could be undervalued. Personally, I am not bullish on any mREITs today. However, investors who still wish to gain exposure to the mortgage market may want to take a look at Two Harbors (TWO) or PennyMac (PMT). They employ a similar strategy but use mortgage-serving rights to hedge against mortgage spread risks. This does not entirely protect them from tapering, but it leaves them in a safer position than ARMOUR.
This article was written by
Disclosure: I/we have no stock, option or similar derivative position in any of the companies mentioned, and no plans to initiate any such positions within the next 72 hours. 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.