Analyzing the Differences Between

A critical component of REIT investing is understanding the relationship between public and private real estate investments. Joseph L. Pagliari, Jr. is the co-author (along with Kevin Scherer and Rich Monopoli) of "Public v. Private Real Estate Equities: A More Refined Comparison" (Journal of Portfolio Management). He is also a clinical assistant professor and the associate director of the real estate program at the Kellogg School of Management, Northwestern University. Portfolio spoke with Pagliari to gain his insights on weighing the two investments.

Pagliari and his fellow authors found that over a 24-year period publicly traded REITs had an average return that was 500 basis points higher than private real estate returns as measured by the NCREIF Property Index. The authors set out to determine if the platform truly mattered. After adjusting for a number of features of the two indices they found no statistically significant difference.

Portfolio: What advice would you give to an investor who is exploring investing in either public or private real estate?
Pagliari: For the lion's share of investors, public real estate is certainly the preferable investment vehicle. While it's not exactly on point, it's nevertheless interesting to note that, since the inception of the NAREIT Equity Index (in 1972) through the year ended 2003, the NAREIT Equity Index has outperformed the S&P 500 in 19 of those 32 years, or about 59 percent of the time. Over this time period, the S&P 500 has averaged an annual return of 13 percent while the NAREIT index averaged 14.2 percent. Moreover, the S&P 500 has generated these returns with greater volatility: the standard deviation of the annual S&P 500 returns is approximately 18 percent but is 12 percent for the NAREIT index. For most of us, the exposure to the idiosyncratic risks (i.e., non-market risks which otherwise can be diversified away) of holding a concentrated (private) real estate portfolio is imprudent.

There are two other investment possibilities, besides public real estate, that can also avoid this idiosyncratic risk. First, a number of the large advisory firms (i.e., those institutionally oriented investment managers who contribute to the NCREIF Property Index) are starting to open up their commingled investment products to individual investors. As they rectify the problems of heretofore-infrequent valuations, these products may become more of a competitive threat to the REITs in the future.

Second, private (non-institutional) investment vehicles oriented toward individual investors are gaining significant momentum. In part because of the heavy upfront fees and costs charged by many of these funds, their structure is reminiscent of the old syndication days prior to the Tax Reform Act of 1986.

Then, as now, the fear is that some of the sponsors/promoters of these vehicles are more adept at money-raising activities (than actually identifying and managing prudent real estate investments). If so, the concern is that their lack of investment discipline will adversely influence the real estate market's overall risk/return characteristics. Of course, it remains to be seen whether or not this comes to pass.

Portfolio: How does the platform itself factor into investor preferences?
Pagliari: While our research suggests that the platform did not matter in terms of the statistical properties of the REIT and real estate return distributions, the platform may matter with regard to liquidity, governance, transparency, control, executive compensation, etc. Clearly, the apparent clientele effect hints at these characteristics being valued differently by large and small investors.

Though our research did not address this point empirically, it seems sensible that larger firms have more negotiating leverage than do smaller firms and, accordingly, large firms are able to extract concessions (e.g., fee/compensation levels, discretion, governance, etc.) from the advisory firms that smaller firms are not. Moreover, REITs are legally obligated to treat all shareholders equally—clearly the advisory firms differentiate their products and pricing based upon investment size.

Total Returns by Asset Class In Percent, 1985: Q2–2001:Q2

This is for illustrative purposes only and not indicative of any investment. Source: Ibbotson Associates, Inc.

Another factor might also be the differing levels of liquidity for large and small funds. In other words, a large pension fund with passive REIT weights (i.e., each firm is held in proportion to its market capitalization) will experience more adverse price pressures (that is buying will push prices up, while selling will push prices down) in moving a given percentage of its investment than will a small firm moving the same percentage of its investments. Of course, this is particularly true with the more thinly capitalized REITs. These adverse pricing pressures may also play into the observed clientele effect.

Lastly, it has been suggested to me that there is simply not enough REIT product available to satisfy the investment demands of the larger pension plans. I believe this wrongly ignores the talents and guile of the investment bankers and other parties involved in converting private real estate equity to public—should the demand for such product exist.

Portfolio: In comparing the performance of public and private real estate markets, you and your fellow authors eliminated non-"core" REIT firms (e.g., hotels, health care, golf courses, race tracks, etc.) from the NAREIT index in order to make the returns directly comparable to those reported in the NCREIF Property Index. As a result, the risk/return performance of the NAREIT index improved–in the sense that the risk remained essentially unchanged while returns increased by approximately 100 basis points. What does this say about those REITs that you removed from consideration?
Pagliari: In making our comparisons, my co-authors and I excluded the non-core firms from the NAREIT Equity Index (as described below) and we reconstituted the NCREIF Property Index such that they mirrored, on a year-by-year basis, the property-type orientations of the core REIT firms–all done on the basis of asset values. In essence, this meant reducing NCREIF's allocation to the office sector (the worst-performing property type in NCREIF) and increasing the allocation to the apartment and retail sectors (the best-performing property types in NCREIF).

From this result, some people have mistakenly concluded that these non-core sectors themselves offer poor risk-adjusted return opportunities. They may be mistaken because they confuse the risk/return performance of the operating business with the risk/return performance of the financial structure (e.g., master lease) used to comply with the IRS regulations concerning active management. In other words, the NAREIT returns reported in the non-core sectors are most often financing transactions where the operating risks and rewards are left to, for example, the master lessee.

Since we "rolled up" the NAREIT and NCREIF indices (after having made our adjustments), we did not explicitly look more narrowly within a given property type to examine whether or not public and private returns differ. I could be convinced that the advantages of REIT ownership (branding, centralized management, etc.) are most important in property types with a good deal of segmentation and heterogeneity like the mall business and are the least important in property types which are homogeneous like the apartment sector. And, accordingly, you will see the largest return differentials in the non-commodity sectors. Again, the empirical work has not been prepared however.

Portfolio: For many years, plan sponsors were limited to direct real estate investment programs when implementing their real estate allocations. Now that they can freely choose between direct investment and publicly traded real estate stocks, how do you think sponsors compare these two alternatives today?
Pagliari: We found that it appears that the large pension (endowment and union) funds have a preference for private real estate equities, whereas the smaller pension funds have a preference for public real estate equities. We specifically examined pension funds because they are sophisticated investors (often supported by sophisticated consultants), and they clearly have the opportunity to invest in either (the public or private) real estate platform.

Yet, as of the end of 2001 (the end of our study period), almost 90 percent of pension fund capital for the top 25 pension funds (which control more than half of the assets of the top 1,000 funds) allocated to real estate equities was invested in private-market vehicles: $80 billion in private versus $12 billion in public. More specifically, each of the top 25 pension funds with an allocation to private real estate equities has, on average, $54 billion of assets committed to various investment vehicles—most of which is allocated to stocks and bonds. However, each of the top 25 pension funds with an allocation to REITs has, on average, only $39 billion of assets committed to various investment vehicles.

In other words, the funds with a commitment to private real estate are generally 40 percent larger in terms of total asset size than those funds investing in public real estate. Perhaps as telling are the 10 funds that are among the 25 largest commitments to private real estate as well as among the 25 largest commitments to REITs. When examining those pension funds, the private-market allocation dominates the public-market allocation by a ratio of approximately 6:1. This substantive discrepancy suggests an argument in favor of a clientele effect by which large institutional investors tend to favor private real estate equities, while individual and small institutional investors favor public real estate equities.

Portfolio: Over the past few years, REITs have increasingly used joint ventures as a way to expand their operations while limiting some of their exposure. This is just one of the many trends that has come to the forefront since your research was completed. In what way does this impact the comparison between public and private returns?
Pagliari: Since 2001, the pace of joint ventures has stepped up substantially. And, by and large, I view this as an effective way for REITs to grow assets under management (which, arguably, adds to the firm's ability to improve upon its economies of scale) as well as creating additional sources of fee income and the potential to earn a "promoted" interest (i.e., many of these JVs provide for the REIT to receive an inordinate share of the JV's profits upon the realization of some threshold).

Should the level of JV activity sustain itself or expand in the future, then the REIT returns will reflect some combination of underlying real estate returns combined with the profitability of the fee income and "promotes" (as discussed above), along with other factors (e.g., leverage, franchise-building, etc.) also baked into the REIT returns. This is, by and large, a good thing for REIT shareholders. However, the nature of the NCREIF reporting (for private-market assets) generally excludes the issues of fees, promotes and the like (as well as leverage). So, while NCREIF is reporting portfolio-level returns at the asset level, NAREIT is reporting entity-level returns at the equity level. These are fundamentally different perspectives—and the nature of our paper was an attempt to disentangle these various factors. My point about increasing JV activity is simply to point out that the process of disentanglement will be more difficult going forward.