Publicly available information has given rise to financial analysis and research by academics, investment research organizations and Wall Street firms that evaluates and quantifies the economic benefits of using real estate in multi-asset portfolios.

"[On Wall Street], there's a tremendous amount of analytic coverage taking place on these companies," McAllister says. "Today, the ability of institutional investors to understand real estate as an asset class is better than it's ever been."

Some 45 analysts cover the REIT industry, according to Martin Cohen, co-chairman and co-CEO of Cohen & Steers Capital Management, Inc.

On the academic side, the research taken as a whole debunks the myth that private real estate returns are significantly less volatile than REIT returns. According to some financial experts, the myth persists because direct real estate is valued on an appraisal basis in the NCREIF (National Council of Real Estate Investment Fiduciaries) Property Index while REITs are priced daily in the market.

Appraisals smooth out the price changes of real estate and present an inaccurate picture of stable asset prices when the economic realities underneath direct real estate, like occupancy rates, rent levels and capitalization rates, are the same as they are for REITs, many industry observers say.

In research sponsored by NAREIT in 2003, Barry E. Feldman, a senior research consultant at Ibbotson Associates (a leading authority on asset allocation), used a transaction-based index for direct real estate (i.e., actual sales figures) instead of appraisals to study the effect of including both public and private real estate in multi-asset portfolios.

Feldman wrote in the Journal of Portfolio Management (September 2003) that "the use of the NCREIF Index results in unrealistically high direct real estate allocations." He found that direct real estate and REITs provide similar returns and have similar correlations with stocks. His work brought him to the conclusion that the best investment results are obtained by including both types of real estate investments in a multi-asset portfolio. The two asset classes are complementary and the optimal balance between them is 50/50, Feldman says.

After making its own adjustments for leverage, serial correlation and appraisal smoothing, Goldman Sachs Asset Management (GSAM) found in 2004 "that NAREIT and NCREIF returns and volatilities are not statistically different." Moreover, "better liquidity, transparency and implementation options lead us to prefer REITs over private real estate for strategic asset allocation," according to a 2004 GSAM report titled "‘Adjusting' NCREIF (Private) and NAREIT (Public) Returns: Are the average returns statistically different?"

UMWA's Adams agrees that appraisal smoothing produces unrealistically low standard deviations in raw NCREIF returns. UMWA therefore adjusts the data in its asset allocation modeling to obtain a truer picture of asset volatility.

What is the bottom line of all this research for pension plan managers?

"It is appropriate to consider REITs as part of a defined benefit plan's real estate allocation," GMAM's Behar says. "Over the long term, REIT investments produce rates of return that are comparable to the returns achieved on private market real estate investments."

Direct real estate accounted for approximately 85 percent of all real estate investing among the top 100 corporate defined benefit plans in 2004, according to Pensions & Investments survey data, leaving about 15 percent invested in REITs. In contrast, GMAM's current REIT allocation is running close to 30 percent of its total real estate holdings, Behar says.