Mortgage-based Real Estate Investment Trusts (mREITs) are usually characterized as volatile and risky. There may be a question as to whether volatility and risk are the same thing, or whether volatility and risk are distinct and separate concepts, but I think there is little doubt that the more volatile a stock is, the greater the risk involved in owning that stock. I would also go out on a limb and say that, indeed, mREITs are (or, at least, can be) risky investments - and are so independent of their volatility (perceived or actual). The possibility of rising interest rates (especially for the short-term loans mREITs tend to use for leverage) and prepayment of mortgages held by an mREIT (or a general increase in the overall prepayment rate) are perhaps the most commonly acknowledged risks involved in owning stock in mREITs. At the very least, these two factors can adversely affect the attractive dividends characteristic of mREITs, and the potential for changes in dividend yields might be seen as another risk factor.
Given the nature of mREITs, the interest paid for short-term loans and the interest received from mortgage holdings - and particularly the difference between the two - constitutes an inherent risk, the cost of doing business, as it were. The question I would like to address here is whether volatility is also inherent to mREITs, and the answer to that matter would help to determine if, indeed, volatility and risk are two words for the same thing. The response to both issues is, I believe, "no," and I hope to show why I've come to this conclusion.
To begin with, there are two ways in which to discuss volatility: the volatility of a stock itself, which is a measure of the distance between the highest prices and the lowest prices a stock reaches over a given period of time; and the volatility of a stock in relation to the volatility of the stock market in general (usually computed using the S&P 500) - which relation is called the beta of a stock. The volatility of a stock by itself can be represented as a percentage, where the higher the percentage, the more volatile the stock's behavior for the period being measured. A stock's beta would be represented by a number reflecting the relationship between the stock's volatility and the volatility of the market, where a beta of 1 indicates that the volatility of a given stock is the same as the volatility of the market, a beta greater than 1 indicates the stock is more volatile than the market, and a beta of less than 1 means the stock is less volatile than the market.
A good example of market volatility and beta can be found if we examine the August betas for SPDR S&P 500 Trust ETF (SPY), iShares S&P 500 Index ETF (IVV), SPDR Dow Jones Industrial Average ETF (DIA), iShares Dow Jones US Total Market Index ETF (IYY), PowerShares QQQ Trust ETF (NASDAQ Index) (QQQ), Technology Select Sector SPDR ETF (XLK), and Vanguard Total Stock Market ETF (VTI). (Note: IVV, IYY and XLK are included since some analysts consider SPY, DIA and QQQ to be inaccurate representations of their respective indexes. VTI is included because it is largely considered one of the more accurate total-market indexed funds. Note, also, that the data that follows is drawn from FINVIZ.com and BAML, reflects market activity for August, and was collected on September 1, 2012. Blank spaces or readings of 0.00 indicate data that was not available. Year-To-Date Performance data is provided primarily for readers' interest, although some interesting observations can be made about a company's performance with respect to its volatility and/or riskiness.)
On the basis of the data presented in table 1 it would seem that the market in general was rather subdued in terms of volatility both for the last week in August and for the whole month of August, as indicated by the low volatility readings for all seven ETFs (all are less than 1%). The most volatile sector of the market was technology (as indicated by QQQ and XLK), which tends to be fairly typical of that sector. Note also that, as one would expect, the betas for these funds are in keeping with general market volatility, with both SPY and IVV being almost dead on, and with DIA and QQQ being the furthest off the mark (as one might expect, given that these ETFs follow the Dow and NASDAQ respectively), although only marginally.
For further illustration, let's take a look at the data for a few companies. The companies I have selected are chosen for no particular reason other than that they illustrate varying levels of volatility and beta, and are recognizable companies: Apple (AAPL), AIG Insurance (AIG), Brunswick Recreation (BC), Avis Budget Rental (CAR), Goodyear Tire & Rubber (GT), Las Vegas Sands (LVS), Altria Group (MO), Pfizer (PFE), Pier 1 Imports (PIR), Schwab US Small-Cap ETF (SCHA), AT&T (T), and Exxon Mobil (XOM). (Note: the following table adds the BAML Risk rating as well as beta figures calculated by BAML, reflecting performance over the past 60 months. BAML does not provide this information for ETFs, apparently.)
As we saw in table 1, the ETFs all exhibited volatility less than 1%, and recorded a beta in the immediate vicinity of 1. This happens in table 2, as well, where we note that Pfizer, AT&T and Exxon-Mobil all have volatilities close to 1% and also have betas below 1. We also note that Apple and Altria Group have volatilities just above 1%, and have betas near 1 (in fact, Altria's beta of 0.39 is the lowest beta in the group). We can also see that not only are the betas for these five companies low for the month of August, but their betas are also less than 1 in the longer-term, 60-month calculation provided by BAML; indeed, of the five companies, Pfizer, AT&T and Exxon-Mobil are rated as "low risk" by BAML, while Altria is rated as a "medium risk." Only Apple is given a "high risk" rating, which can perhaps be explained by the tendency to consider the technology sector to involve higher risk than other sectors; however, Apple's stockholders are probably laughing all the way to the bank, inasmuch as Apple's year-to-date performance is over 64% - the largest gain of all of the listed stocks.
Of the companies having the lowest volatilities, the only anomalous findings concern Schwab Small-Cap ETF. While its volatility for the last week of August - as well as for the whole month of August - are in line with those of Apple, Altria, Pfizer, AT&T and Exxon-Mobil, its beta for the month is an astounding 11.91 (the highest beta listed on FINVIZ for the month)! A beta of such a magnitude would indicate a substantial divergence between the performance of Schwab and the S&P in general, and, indeed, during the month of August Schwab went from over-performing the S&P to under-performing it for a substantial part of the month, then returning to over-perform during the last week of the month. No doubt, this flip-flop involved days where the performance of Schwab exceeded the movement (both downward and upward) of the S&P as a whole (and vice versa).
As for the rest of the companies represented in table 2, they illustrate fairly well how one might expect that a company having high volatility would also tend to have a high beta. Even here, though, we must be careful to note that the relationship between volatility and beta is not a directly proportional one. Certainly Avis-Budget Rental, given the high volatility it exhibits for August, would merit a high beta, but Pier 1 Imports earned a very similar beta with substantially lower volatility. Likewise, Goodyear Tire & Rubber's beta of 2.63 is based on volatility somewhat higher than that of AIG Insurance, which earned a beta of 3.43. All things considered, then, while there may be some correlation between a given stock's volatility and that stock's beta, there is no direct, proportional connection.
Just as there may be some correlation between volatility and beta over a short period, there seems to be a correlation between a stock's short-term beta and its longer-term, 60-month beta, but here there is something of a curiosity: while short-term betas for our selected companies range from 0.39 to 11.91, the longer-term betas for these companies are all below 2. Now, while there is no requirement that betas range as high as 11 or higher, there is no theoretical limit holding betas below a given level. Intuitively, the most reasonable interpretation of this data is that, for the past 60 months, these stocks have not been that grossly divergent from the stock market in general. Indeed, examining the Chicago Options' ^VIX index reveals that from approximately 2007 until earlier this year, the S&P 500 has been rather volatile, and that volatility may have served as a mask to the volatility of the companies within the market. The gradual decrease in market volatility this year may, then, account for the appearance of greater volatility in individual companies. This would be something interesting to follow-up on, if one were so inclined.
One last point: there may be some relationship between a company's beta and its performance. We note that of those companies having lower betas only Apple has an impressive year-to-date performance of 64.69% through August, while most of the more volatile companies post better year-to-date figures than the less-volatile Altria, Pfizer, AT&T and Exxon-Mobil. Only Goodyear Tire & Rubber was decidedly deflated at -13.90%, while Las Vegas Sands did barely better than break even at 0.19%.
Let's turn now to the subject at hand: mREITs. It would be rather difficult to collect data for every mREIT on the market, let alone for every REIT in general. I have opted to select those mREITs that are either most popular or which have received some measure of attention recently: American Capital Agency (AGNC), Armour Residential (ARR), Chimera (CIM), Capstead (CMO), CreXus Investment (CXS), CYS Investments (CYS), Hatteras Financial (HTS), Invesco (IVR), MFA Financial (MFA), American Capital Mortgage (MTGE), Annaly (NLY), New York Mortgage Trust (NYMT), and Two Harbors (TWO). There are several points of interest when comparing the volatility - and risk - found with these companies to that of the other holdings we have considered so far, some of which have struck me as intriguing (hopefully you, as well), and some which have given me reason to consider why certain notable characteristics of the 60-month betas should look the way they do. Let's take a look at how the mREITs stack up:
At first glance, we are presented with a set of companies all of which exhibited only moderate volatility for the last week of August and the month of August as a whole, with Chimera, CreXus, Invesco and New York Mortgage having both readings above 1 (with CreXus and New York Mortgage barely over the 1% mark), and with American Capital and Two Harbors posting monthly volatilities above 1. Things only get better when we examine the monthly betas for these companies: only Chimera was more volatile than the market (among those companies for which data was provided), and Chimera has had something of a rough ride recently. Several of the mREITs were significantly below market volatility, with Armour Residential clocking in at an impressive 0.16. More on the short-term betas in a moment.
With respect to the BAML data, of those companies for which risk ratings are given, only Hatteras and Annaly earn "medium" ratings, while American Capital, Armour, CreXus, CYS, and Invesco are considered "high" risks. Of our group of mREITs, only Chimera (perhaps no surprise) and MFA earn longer-term, 60-month betas above 1, although Capstead and Annaly are very close to market volatility at 0.99 each. The remainder of the group has been, to a greater or lesser degree, more stable than the stock market in general, with Two Harbors and New York Mortgage reporting significantly more stable numbers with 0.48 and 0.55, respectively. It would seem odd to have a group of supposedly "volatile" and "risky" investments such as these mREITs being significantly less volatile than the market as a whole, unless we are able to account for the lower betas by appealing to a volatile market overall. This was suggested as a means of accounting for the lower 60-month betas for some fairly volatile companies in table 2; perhaps the same phenomenon is at work here, but this approach gives rise to something of a problem.
When we looked at table 2, we considered the possibility that the gradual lessening of market volatility exposed some companies whose high level of volatility was masked by a volatile marketplace; thus, companies such as Avis-Budget and Pier 1 Imports, which had 60-month betas of 1.51 and 1.86, respectively, ended up in August to have substantially larger short-term betas of 4.82 and 4.30, again respectively. This should not come as a surprise since, in a volatile market, one would expect to find a large number of companies that were volatile, while in a relatively calmer marketplace the "normal" volatilities of some companies will show up with substantially higher betas. What, then, of our group of mREITs which have both low long-term betas as well as low short-term betas?
One could reply that the similarity between longer-term and shorter-term betas indicates that the characterizing of mREITs as "volatile" is incorrect, at least in any meaningful way. I am not inclined to agree with this, however; it is unlikely that the investors and the analysts who have considered mREITs to be volatile are all wrong. But if mREITs are, indeed, volatile, what can account for the current lack of volatility? As luck would have it, a mixture of factors having direct impact on mREITs have suppressed the conditions that would normally cause much of the risk usually associated with mREITs.
"Operation Twist" was initiated by the Fed as a means of lowering the interest rate on long-term borrowing by the government's purchase of long-term treasuries and mortgage-based securities (thereby raising the price of these instruments, which would lower their yield). Unfortunately, this has only increased long-term borrowing by 15% - not enough to have significant impact on the economy overall. No doubt part of this failure is due to the impact of the state of the economy on individuals' credit rating, particularly an unstable employment environment. As an indirect result, there has been a suppression of the refinancing of mortgages which would diminish the interest income mREITs depend on. It may be of some interest that Freddie Mac has been given permission to increase interest rates on mortgages.
Add to Operation Twist the establishment of a "zero-interest-rate policy" (ZIRP), which has dramatically cut the cost most mREITs encounter in using short-term loans for leverage in acquiring mortgages and mortgage-based securities, and you have an environment where an mREIT is able to secure high-interest-rate mortgages with minimal cost. This enables mREITs to offer shareholders dividend yields ranging from 10% - 16%, which has been a driving force in helping these stocks achieve the year-to-date performances they've been seeing. Inasmuch as the Fed has announced that ZIRP will continue until sometime after 2014, this gives mREITs some breathing room in which to prepare for the time when short-term interest rates begin increasing. That some mREIT managers take this breathing room seriously may account for the fact that many mREITs are releasing stock periodically to build up cash reserves, which can then be used to replace the short-term loans currently being used for leverage.
To conclude, while mREITs may be risky investments, that risk is primarily due to the fact that the long-term interest that serves as income, minus the short-term interest that serves as a liability, are factors by and large out of the mREIT's control. Volatility is a concern only when and if the interest rates should be subject to fluctuations. As long as there are market factors placing limits on the interest rates, however, volatility diminishes, giving investors an opportunity to enjoy much-welcomed income.