Mortgage REITs: Does Doubling the Leverage Make Them a Good Investment?
Having recently passed the four-year anniversary of the Lehman Brothers collapse, it’s tempting to believe that our economy and capital markets have learned from their mistakes. After all, big banks are heavily scrutinized (see J.P. Morgan’s “London Whale” debacle), fancy new regulations are in place or being proposed (see Dodd–Frank and Basel III), and skepticism seems to be a permanent part of investor psyche. You could be forgiven for thinking that, big, bad Wall Street might have caught us off guard before, but it won’t again.
Nevertheless, every now and then Wall Street reminds us that it has “fallen off the wagon,” so to speak, and reverted back to scary old ways of the bad old days. One of those old tricks is to add unnecessary leverage to assets that may not be safe to begin with. And one of those reminders came recently when UBS announced a new 2× leveraged mortgage REIT exchange-traded note (ETN).
I couldn’t blame you for asking, “What is a mortgage REIT, and why is this so bad?” Let’s examine what’s happening in the mortgage REIT market before looking at this offering in particular.
Brief Background on Mortgage REITs
REITs typically purchase 30-year agency mortgage pools (issued by Fannie Mae and Freddie Mac) and lever them up by six or eight times. Mechanically, the REIT takes a new, purchased agency mortgage-backed securities (MBS) pool and enters into a repurchase agreement (repo) with a dealer, where the dealer gives the REIT cash. Then the REIT purchases another agency MBS pool.
REITs repeat this process until they have achieved six to eight times leverage.
The appeal of this model is easy to understand once you consider the shape of the yield curve. Repo rates are based on the short-term end of the curve (let’s assume a cost of 25–50 bps, for example; a basis point is 1/100th of a percent); agency MBS pools, in contrast, price off of the belly (5- to 10-year) part of the yield curve, where rates are higher. In short, the model allows the funds to capture some of this difference in interest rates.
For the better part of the last few years, 30-year current coupon MBS pools yielded 2.5% to 3.5%. In addition to incurring the repo cost, REITs typically also enter into derivatives to hedge the interest rate risk of the fixed-rate mortgage pools. The resulting net spread — after the costs of hedging — has been around 2% for some of the larger agency-only REITs.
American Capital Agency (Ticker: AGNC), for instance, had a net interest spread of 2.14% in 3Q2011 and a spread of 1.60% in 2Q2012. Remember that these interest spreads are being levered six or eight times, which means that the leveraged net interest spread ran between 12% and 18% on average.
Ah, so that’s how these REITs were able to pay out such large dividends!
How Did We Get Here, and Why Is This Scary?
The short answer is that the Fed’s quantitative easing (QE) programs have resulted in extremely distorted pricing of both duration and credit risks. Although QE is one of my favorite topics to describe (mostly because it hits so close to home), I will not go through an in-depth explanation of how QE has distorted fixed-income risk assets. Interested readers can go back to an earlier post on Inside Investing, titled “The QE Aftermath: What it Means and How it’s (not) Different”
In short, there’s an insatiable appetite for yield in the fixed-income markets. Think of it as being like a “yield piñata”: Investors far and wide are scrambling to pick up as much yield as possible before it’s all gone. Investment-grade corporate bonds, high-yield, leveraged loans, non-agency MBS, and commercial mortgage-backed securities (CMBS) are just a few asset classes that have seen prices charge higher.
Both institutional and retail investors have been smitten with mortgage REITs and the 10%+ yields available from AGNC, Two Harbors Investment (Ticker: TWO), Armour Residential REIT (Ticker: ARR), Western Asset Mortgage Capital (Ticker: WMC), and others. Yes, these have had a remarkable run in terms of share appreciation and total return from dividends, but it’s important to remember how this has occurred. The Fed has purchased well over $1 trillion of agency MBS thus far, and this figure is only growing with the QE3 plan of an additional $40 billion per month.
As stocks continued to deliver strong returns, mortgage REITs took advantage of this opportunity by selling additional stock. Because some of these companies are compensated by assets under management (AUM), they had every incentive to keep selling shares to grow the REIT as large as possible. The chart below from J.P. Morgan shows mortgage REIT MBS holdings over the past 10 years in a hockey stick–shaped chart. Does this look healthy?
I’d be scared to own mortgage REITs even before you double the leverage, for the following reasons:
Record high MBS prices. Shortly after QE3 was announced, 30-year 3% Fannie Mae mortgage pools traded at 106 cents on the dollar, for a projected yield to maturity between 1.8% and 2%. As the REITs purchase new bonds, the yield at purchase is significantly lower than what’s already on their books.
Extremely tight spreads. Spreads over a comparable maturity U.S. Treasury or swap are also at historically low levels; thus, these bonds don’t have the ability to tighten in spread to offset a decline in price.
Expected increase in prepayments. Given the recent drop in mortgage rates, there has been a dramatic pickup in refinancing activity as measured by the refinancing index. The problem for the mortgage REITs is that an increasing amount of their earlier purchased bonds yielding 3% or higher will be returned at par (100), instead of where they may be currently trading.
Margin compression. The implication of reasons one through three is that mortgage REITs are facing and will face strong margin compression in the coming quarters. High-yielding bonds are running off and being replaced with lower-yielding securities, while the cost of their funding hasn’t changed. This is a perfect combination for future dividend cuts, which have already begun in some cases.
Dividend popularity. With dividend paying stocks now in vogue, mortgage REITs are trading under a halo; they are one of the few places to find a significant yield. As quickly as these stocks have come into favor, they can also fall out of favor. Stocks that were trading at a premium to book value could quickly find themselves trading at a discount.
So, Should We Buy These on Leverage?
Wall Street has created a product to magnify the return of an already extremely leveraged product. The 2× levered exchange-traded fund (ETF) will essentially create a vehicle that will be 12–18× levered to agency MBS.
The cynic in me wonders whether UBS has created this product solely so that institutional clients can bet against agency MBS by shorting this ETF at a time when MBS prices have never been higher.
The average investor likely has no idea how mortgage REITs operate and what circumstances could cause a material fall in their share prices. As an investor who deals in the MBS markets regularly, I am downright frightened that such a product is being created at a time when so many red flags are apparent.
I am not saying that MBS prices cannot continue to rise or that this product will be a failure, but the dark side of QE, the yield-chasing investors, and the capitalizing nature of Wall Street have created a product ripe for trouble down the road. Caveat emptor!
If you liked this post, don’t forget to subscribe to the Enterprising Investor.
All posts are the opinion of the author. As such, they should not be construed as investment advice, nor do the opinions expressed necessarily reflect the views of CFA Institute or the author’s employer.
“After all, big banks are heavily scrutinized (see J.P. Morgan’s “London Whale” debacle)”
The London whale debacle is evidence for precisely the opposite propostion. If the banks were remotely adequately supervised the “whale” would never have been able to take these kind of risks.
Hi David, is it possible to subscribe for Inside Investing and economicmusings.com articles by email ? I don’t usually use twitter or tumblr.
Wimal, you can subscribe via an RSS feed like Google Reader (Outlook has a built in RSS feed as well).
Dave –
Which interest rate path do you think is worse for RMBS REITS: continued rate declines on the order of 50-100 bps over the next 2 years; gradual rate increases on the order of 50-100 bps over the next 2 years (assume begins in 6 months)?
Casey- I actually think rates down 50bps would be worse for the mREITs. When I say mREITs, I mean the GSE focused mREITs such as NLY, AGNC etc. This is because their funding costs (repo) would not materially fall while prepayments would continue to speed up and reinvestment opportunities would be even worse. We have seen this play out since this article, but you’ll see more and more dividend cuts and margin pressure.
Very nice article. Thanks for sharing your thoughts about what is mortgage that will have a good investment.