The California Public Employees’ Retirement System (CalPERS) made headlines in February when it decided to focus at least 75 percent of its $16 billion real estate portfolio on core properties. But a new study suggests that a core-focused approach fails to provide risk-reducing diversification for pension funds.

“Pension funds have tended to say, ‘The losses in opportunistic funds and value-added funds were horrible. Let’s get out of them and put all our real estate money in core funds,’” says Brad Case, senior vice president, research and industry information at the National Association of Real Estate Investment Trusts (NAREIT). “That will not reduce risk. The only way you can do it is to combine private and public real estate holdings.”

In its February paper “Optimizing Risk and Return in Pension Fund Real Estate: REITs, Private Equity Real Estate and the Blended Portfolio Advantage,” NAREIT says two decades of actual performance data show that the optimal diversification for a pension fund came from a real estate portfolio with about two-thirds private real estate and one-third REITs. Pension funds currently put an average of only 9 percent of their real estate allocation into REITs, Case says.

“Part of it is that despite the fact that REITs have been around since the early 60s, it is only in the past 10 years that we have had the depth in the market that pension funds need, in terms of being tradable,” says Anatole Pevnev, a principal at real estate investment consultant The Townsend Group. “It is just becoming a more mature asset class.”

Also, private real estate traditionally has a very low correlation with public equities and debt, Pevnev says. “From a pension fund perspective, that is very attractive,” he says. “And the returns long term are in the 8 percent to 10 percent range, which is right on target with a lot of the actuarial returns assumed by these plans. The tradeoff is that you do not have access to the capital whenever you want it.” On the flip side, some pension fund executives cite as a negative that “liquidity does result in greater volatility for REITs,” he says.

Case agrees that some pension fund executives may see REITs as more volatile than private real estate, but he says that a closer look at the numbers disproves that idea. “The short-term correlation between REITs and the stock market is in the 52 percent range, and for longer holding periods of two years or more, it goes down to 40 percent” he says. “Private-property values have a very low correlation when you look at one quarter, but over holding periods of two years, it goes to 40 percent.”

The diversification benefit stems from REITs moving differently than private real estate in the real estate cycle for both downturns and recoveries, Case says. “In the last real estate market cycle, REIT returns peaked approximately one year ahead of those of private real estate funds,” the paper says. “REITs also began their bull market approximately three years ahead of private funds.”

“No matter how you invest in real estate on the private side of the market, the assets move in a certain cycle,” Case says. “They move together, pretty much nationwide, and pretty much for all property types. If you put a substantial percentage of the allocation on the public side, the returns will follow the real estate cycle, but with different timing, smoothing the returns of the overall portfolio.”