In the 1995 and 1997 editions of The Intelligent Asset Allocator I poured a bit of cold water on the diversification value of Real Estate Investment Trusts ("REITs"). Noting a 0.88 correlation of annual returns for REITs and US small stocks I said:

The correlation coefficient for these 2 assets is quite high -- 0.88. REITs have been touted as effective "real estate" portfolio exposure. However, this high correlation indicates that the diversification value of this asset is overstated, as it behaves more like small company stocks (which REITs are) than real estate.

Since the REIT debacle of 1988-90 however, this sector has reinvented itself. Consider that the total value of US commercial real estate is currently about $3 trillion. A decade ago the total market cap of all REITs was well under $10 billion; according to NAREIT it is now $158 billion. Five years ago there were 7 REIT mutual funds, now there are 70.

What happened? Very simply, private real estate developers realized that the cost of capital in the equity markets had fallen to low levels as the public, hungry for equity of any sort, began to snap up REITs in the form of individual securities and mutual funds. In other words, owners and developers realized that they could acquire vast amounts of capital for the cost of a 5%-6% annual dividend. Why not? The ducks are quacking, feed 'em.

So, almost overnight REITs have become the proprietors of over 5% of commercial US real estate. This figure seems destined to mushroom in the coming years.

Astute investors should be made a bit uneasy by all this. For starters, when corporate treasurers decide en masse that it is a good idea to sell their company's equity, beware. In every poker game there's a patsy. If you don't know who it is, then it's you. The average long term return of initial public offerings ("IPOs") is considerably less than the market's. Secondly, there is a palpable gold rush mentality surrounding this phenomenon. Just about every popular magazine, right down to Time, fairly gushes about the opportunities offered to small investors by REITs. (NAREIT offers a nifty compilation of the most obvious of these. Take a look -- it'll raise the hair on the back of your neck.) Lastly, most REIT IPOs are now selling at a considerable premium to their underlying assets, always a danger sign in the real estate area.

For all the negatives, however, there is one upside to all of this. The nature of the business has changed. A decade ago the industry was dominated by staid property management companies typified by New Plan and Washington REIT -- conservative outfits with strong balance sheets who took risks only when a payoff seemed certain. Now the industry is dominated by companies like Vornado, run by highly aggressive property managers bent on rapidly acquiring vast tracts of income producing assets.

It turns out that this change in the nature of the industry has resulted in dramatically falling correlations with both small and large US stocks, as the below graphs demonstrate:

(Monthly returns of the NAREIT index before 1995, and the DFA REIT fund after were used. Rolling 3 year correlations are calculated.)

I've also graphed, for comparison, the correlations with the EAFE and Lehman Brothers long term government/corporate indexes. These correlations were fairly low to begin with, and do not seem to have changed much.

The bottom line? The nature of the REIT industry has changed, and not necessarily for the better. However, this change has at least improved the portfolio characteristics of the asset. The REIT industry, like real estate itself, is prone to boom/bust cycles, and often presents investors with striking buying opportunities, as occurred in 1990, and to a lesser extent, in 1994. One easy way of following REIT valuations is to keep abreast of the aggregate dividend yield of the NAREIT index. Unfortunately, this is not publicly available. A good proxy is the dividend yield of the Vanguard REIT Index fund. Currently, it's about 5%, fairly low by historical standards. In 1997 it was in excess of 7%, and in 1990 in excess of 9%.

This area deserves the attention of the alert asset allocator.

copyright (c) 1998, William J. Bernstein