Risk Parity Resources: six pieces about REITs

What do the experts say about REITs as a distinct asset class? I’ve expressed my own doubts in the first two posts on real estate - but what’s the take from the world of professional investors? In this post, I gather and summarize six relevant sources on the topic.

💡

This is the third in a series on Real Estate investing in an RP Portfolio. The first previewed the discussion about REITs with my look at VNQ. The second applied Dalio’s three criteria to REITs to see if they count as a distinct asset class. More posts to come on backtesting, alternatives to REITs, and more.

Up until now, the Risk Parity Resources posts have all been focused on particular sources. This one takes a different approach, with mini-summaries of six on the same topic:

“Are REITs a Distinct Asset Class?”

Kizer and Gordon (2017)

This paper from Jared Kizer, CIO at Buckingham Strategic Wealth, and Sean Grover comes up first in every search I do on the topic, both because of the perfect title and the comprehensiveness of its approach. Kizer and Gordon tackle the question head on, and use four separate statistical methods to assess REITs.

The four tests are:
1. Low correlation with established asset classes such as broad market equities and government.

2. Statistically significant positive alpha with respect to generally accepted factor models.

3. Inability to be replicated, on a comovement basis, by a long-only portfolio holding established asset classes.

4. Improved mean-variance frontier when added to a portfolio holding established asset classes.

The bulk of the paper covers the data from the tests, and is well worth the read if you are interested. To cut to the chase; 1) REITs have low correlation but not especially so; 2) There is “no statistically significant alpha generation” with REITs as their returns are “well explained by traditional risk factors”; 3) REIT returns can be replicated by a portfolio of 66% small-cap value and 34% corporate bonds; and 4) REITs do not improve the mean-variance frontier when assed to stocks and bonds. It all adds up to a giant “no” in answering the question in the title of the paper.

💡

"These results, and the associated failure to satisfy our asset class criteria, lead us to conclude that REITs are not a distinct asset class."

“Real Estate Betas and the Implications for Asset Allocation”

This one by Peter Mladina, Director of Portfolio Research for Northern Trust Wealth Management, is the second most frequent paper that comes up, though I offer it here with a plot twist…I can’t access it. It’s behind a paywall for me, and may be for you too, though here is the link in case you want to try. Still, it’s well regarded enough that it has been summarized and relayed at least twice. The first of these is:

“Real Estate Isn’t Special”

Swedroe (2018)

This article by Larry Swedroe, author & principal and the director of research for Buckingham Strategic Wealth, reviews the findings by both Mladina and Kizer & Gordon. As for Mladina, Swedroe relays the approach of the paper, which was to use “a modified version of the Fama-French five-factor model to evaluate how well the returns and risks of publicly traded equity REITs and private real estate investments are explained by common stock and bond factors.”

Essentially, Mladina is tackling the third statistical test of Kizer & Gordon (to see if REIT returns can be explained by already-known risk factors), and adding to it by looking at multiple types of real estate returns, including privately held real estate. Swedroe summarizes the data, before relaying Mladina’s overall conclusion that “real estate would not be considered a separate source of return.”

Swedroe also reviews the Kizer & Gordon paper, as well, and then concludes the review by writing:

💡

"The findings from the two papers I reviewed demonstrate that returns to real estate are well-explained by exposure to common, systematic factors. These factors—market beta, size, value, term and default—comprise an uncorrelated set with compensated return premiums. In contrast, idiosyncratic real estate sector risk and misappraisals contribute to risk but not return. In other words, investors using factor exposures to determine their portfolio allocation do not need to consider adding real estate as a separate asset class."

“Reconsidering REITs in Your Investment Portfolio”

Rational Reminder/Felix? (2019)

This piece does not name the author, but I’m certain it's the work of Ben Felix, podcast host, YouTube star, and Portfolio Manager at PWL Capital. This post covers the same ground as Swedroe’s, with the same general conclusion.

Felix has also made a YouTube video treating the same topic that’s great if you prefer that format:

“REITs Aren’t a True Alternative”

This Morningstar article by Daniel Sotiroff, senior manager research analyst for Morningstar, is less math-heavy and reviews whether the common assumptions about REITs hold up to scrutiny. This is a pretty quick read, with another conclusion in support of the previous pieces, plus my own look into the topic. Sotiroff concludes:

💡

“Real estate investment trusts are included in most broad stock index funds, like Vanguard Total Stock Market ETF (VTI), where they represent 4% of the portfolio. They may provide a modest diversification benefit, but it probably won’t be much better than what other sectors of the market provide. And they may not hold up better than the market during downturns. Publicly traded REITs are best seen as publicly traded companies, and therefore subject to the same economic and market risks as other publicly traded firms.”

“REITs and Your Portfolio”

Finally, a piece saying that REITs are indeed distinct! I looked and looked, but every search brought me to the naysayers above, so I was glad to find this piece for balance. This one is by Rick Ferri, podcast host and author. Ferri writes that REITs are:

💡

“fundamentally different than other asset classes. REITs have unique characteristics from common stocks. They’re the collateralization of several individual properties, have a unique tax structure where the trust pays no federal tax, and the correlation of REITs indices with common stocks varies - this confirms a unique risk.”

The problem with Ferri’s argument here is that he provides no evidence, so these are just his assertions. Then, if we anticipate what sorts of evidence he would need to support his claim (an investigation into their historical correlations, whether their returns could be explained by already-established risk factors, etc.), it turns out that those questions were already answered in the negative by studies way more rigorous and systematic than his. It really doesn’t look like the evidence is on his side, yet nevertheless this article may be wroth reading to get a sense of the conventional “wisdom” surrounding REITs.