In surprise, research shows REITs don't ape stock markets

Correlation drops dramatically over five years; investors holding too much stock, too little real estate, says Nareit economist

Nov 16, 2011 @ 1:31 pm

By Jeff Benjamin

Evaluating investments over a long time horizon could lead to dramatically altered portfolio weightings, according to some new research from the National Association of Real Estate Investment Trusts.

An analysis of volatility and correlation of real estate compared with the broad equity markets shows that over periods of five years or more, REITs are less correlated to stocks than any single sector of the broad equity markets.

“People are radically underestimating how much they should be investing in real estate, and they are overestimating how much they should be investing in stocks,” said Brad Case, an economist with Nareit.

Mr. Case makes the point by illustrating that performance correlation between stocks and real estate tends to shrink over longer holding periods. Between January 1990 and October 2011, for example, the average one-month performance correlation between REITS and stocks was 56.3%.

But by measuring the correlation over the same period in five-year increments, he found that stocks and REITS moved in tandem only 7.1% of the time.

Meanwhile, the various sectors of the broad market indexes, which are often used by advisers as portfolio diversifiers, actually become more correlated to the broad market over longer holding periods.

The consumer staples sector, for example, had a correlation to the S&P 500 of 65% over the one-month intervals dating back to 1990. But when studied on a five-year basis, the correlation jumped to nearly 88%.

Health care sector stocks over the same period saw correlation on a monthly basis of 61%, then jumped to nearly 90% when measured on a five-year basis.

“Long-term investors are making investment decisions based on data that day traders should be using, because they're using short-horizon data,” Mr. Case said. “If you have a five-year time horizon, use five-year data.”

RELATED ITEM The largest nontraded REITs »

The research, which Mr. Case is still developing with additional sets of criteria, illustrates the beneficial noncorrelated nature of stocks and real estate, despite the fact that most investors view real estate and stocks as being highly correlated. Volatility, somewhat predictably, also responds to longer time horizons.

During three-month cycles between the end of 1988 and June of this year, the annualized volatility of REITS was 18%. But when measured over three-and-a-half-year periods, the volatility dropped to below 4%.

According to Mr. Case, the volatility of REITS is essentially in line with that of equities, as is long-term performance, but the key is the lower correlation over longer time periods.

“You can handle high volatility if you have low correlation,” he said.


What do you think?

View comments

Recommended next


Hi! Glad you're here and we hope you like all the great work we do here at InvestmentNews. But what we do is expensive and is funded in part by our sponsors. So won't you show our sponsors a little love by whitelisting It'll help us continue to serve you.

Yes, show me how to whitelist

Ad blocker detected. Please whitelist us or give premium a try.


Subscribe and Save 60%

Premium Access
Print + Digital

Learn more
Subscribe to Print