Are REITs Still Good Portfolio Diversifiers?

2012 February 27
by Kyle Bumpus
from → Investing And Investments

Portfolio diversification is really simple. Step one, find at least three different mutually non-correlated  asset classes. Step two, buy those asset classes, preferably in equal amounts. Step three, re-balance. It doesn’t really matter if you re-balance annually, quarterly, in accordance with some pre-determined allocation shifts, or anything like that so long as you re-balance. Done!

Unfortunately, those three or four magical mutually uncorrelated  asset classes simply don’t exist. You see, asset class correlations shift over time. Sometimes (usually), short-term bonds are an excellent diversifier of domestic equities. Other times, they’re not. Ditto for TIPS. Ditto for commodities. Ditto for real estate. Ditto even for gold. Unfortunately, we are in an age of increasing correlations. According to Morningstar, 9 out of 11 broad asset class indexes they track have become increasingly correlated with the S&P 500 over the past decade, including foreign stocks, gold, REITs, and commodities. The two indexes that actually became less correlated with the S&P 500 are those tracking the aggregate US bond market and the 7-10 year treasury bond market.

REITs Have Become Almost 100% Correlated With The S&P 500

According to Morningstar, the Trailing 12 Month (TTM) average daily correlation of the Dow Jones US Real Estate Total Return USD index stood at 0.59 at the beginning of 2002. A 0.59 correlation coefficient represents a moderate but not spectacular diversification opportunity. Still, even a moderate diversification benefit is well worth chasing after. Unfortunately, at the beginning of 2012 the TTM average daily correlation of that same index stood at 0.91, yielding pretty much no diversification benefit over the prior year.

Should we be concerned that REITs no longer seem to be adequate diversifiers? In a word, no. At least not yet. Remember that “correlations drift over time” statement I made above? It’s not just a theory. It really happens. A lot. I fully intend to remain invested in Real Estate Investment Trusts going forward and recommend you do as well. There are a few obvious reasons why the broader stock market and REITs have become so correlated over the past few years and a few compelling reasons to believe they won’t stay that way, for the most part.

Why Have Correlations Increased?

The S&P 500 now includes several large REITs - The S&P decided to finally allow REITs into the index back in 2001. According to REIT.com, there was only a single REIT in the S&P 500 at the beginning of 2002, Equity Residential (EQR), and that had only been a part of the index since November of 2001. As of now, there are 15 REITs in the S&P 500 with almost 2% of the index being made up of real estate related companies. To be sure, 2% isn’t a large number, but it does make sense that the S&P 500 would become slightly more correlated with REITs over the past decade as they have increased from 0% to 2% of the index. It may be that REITs will go on to comprise even more of the S&P 500 in the future as more private real estate holdings (the vast majority of domestic real estate is still privately owned) are “REITized.” Before 2001, flows off assets into REITs and REIT funds were determined more by fundamentals than anything else. Since their inclusion in the S&P 500, however, flows become much more tied to the broader market. Unfortunately, this isn’t going to change.

“Correlation moves to 1 in a crisis” - There is a lot of truth to this statement. When the fit hits the shan, risky asset classes tend to all drop together as investors flock to the safety of short-term treasuries and gold. Then, as the economy begins to recover, that money tends to pour back into risky assets and they all more or less rise together in a “rising tide lifts all ships” kinda way. Therefore, I don’t think it should really come as a surprise to anybody that correlations are up across the board over the last 3 years. The economy has been through a lot!

REITS have become more accepted by conventional wisdom - Ten or fifteen years ago, not many experts would have recommended a large REIT allocation for most investors. These days, they are recommended by almost everybody. I don’t have any numbers to back this up, but just based on the “prevailing wisdom” I would be very surprised if the percentage of the average portfolio dedicated to REITs hasn’t increased significantly over the last decade. More and more people are buying REITs as a matter of course; they are no longer exotic investments of just a few sophisticated investors. I think this trend is likely to continue.

Why Will Correlations Probably Decrease Again?

But all is not lost! While in-depth analysis seems to suggest it is likely (but not guaranteed) that REITs will be at least slightly more correlated to the broader market in the future than in the past, there are reasons to believe the recent run of very high levels of correlation is temporary.

Separation tends to occur following a crisis - As mentioned above, correlation moves to one in a crisis. The upshot of this is that correlation has usually, in the past, moved away from one after the recovery. Indeed, since the middle of January daily correlations of REITs and the broader market have dropped to below 0.60. Of course, you can’t really conclude anything based on just 30 days of data, but hopefully this is a step in the right direction.

Owning real estate is a distinct business - Owning and operating commercial real estate is just a different business than companies in most other industries are engaged in. Rents tend to be relatively stable, at least compared to the earnings of companies in most other industries. It’s true that lower corporate profits could lead to decreased demand for commercial real estate, but the long-term nature of most lease contracts gives REITs at least somewhat of a buffer. Additionally, real estate can often be re-purposed to fit with shifting demand. Old industrial properties are being turned into residential lofts in urban areas across the country, for example. None of this is to say that the commercial real estate market is immune from economic troubles; obviously it’s not, it’s just that those troubles tend to come in slightly different cycles than the broader economy. I don’t foresee this changing anytime soon.

REITs are extremely exposed to hiccups in the financial market – This is an understatement. Since REITs are required by law to pay out at least 90% of their net income as dividends to shareholders, they depend heavily on the capital markets for expansion. The financial panic really threw most real estate companies for a loop as infusions of outside capital became nearly impossible to come by. Fortunately, the financial crisis was a once-in-a-lifetime occurrence. Will it happen again? Absolutely! But it won’t happen regularly. It could be decades before a comparable crisis hits again. Or at least, I hope so.

We’ve been here before - This is not the first time we’ve seen correlations over 0.90. It’s happened several times in the past, in fact. Correlations moved downward then and there’s no reason to think this time is any different.


Did you enjoy this article?


Please subscribe to our blog via RSS Feed and get great new content delivered straight to your desktop every day!

Or if you prefer, you can have daily updates delivered to you via Email.


3 Responses leave one →
  1. 2012 February 27

    Negative or low correlations between different asset classes with expected positive real returns effectively create a free lunch thanks to rebalancing. Sooner or later portfolio managers ought to catch on to this effect and arbitrage it out just as any other free lunch in the markets. It’s entirely likely that this is what we’ve been seeing over the past decade or so and I would expect more of the same going forward.

  2. 2012 February 27

    Mike, the only flaw I see with your reasoning is that the free lunch of diversification is not a new discovery. We’ve known about it academically since the 50′s. Why would portfolio managers just now be getting around to arbitraging this inefficiency? I believe there is something more at play. Rather, different investors buy different assets for different reasons. So long as that remains true, I don’t see how the diversification bonus will ever completely disappear. I do agree that it will probably narrow going forward, at least a little.

  3. 2012 February 27

    Hey, Benjamin Graham wrote the book on value investing but it took 40+ years till Fama/French/DFA made it “official” :) But, yes, I certainly agree that diversification is still important and some rebalancing benefit will always be there.

Leave a Reply

Note: You can use basic XHTML in your comments. Your email address will never be published.

Subscribe to this comment feed via RSS