In a previous blog we examined direct investment in real estate as one option for investors wanting real estate exposure. Real Estate Investment Trusts (REITs) are an alternative to directly holding real estate, but is one a substitute for the other?
REITs allow a partial ownership of a number of individual real estate properties and many REITs are publicly traded. A REIT is a mutual fund trust, so that investors are unitholders in the trust, and the trust holds a passive interest in a portfolio of real properties. Three advantages REITs have over direct ownership of properties are diversification, liquidity and regular distributions. For companies the primary reason for structuring as a REIT is that distributions are taxed only in the hands of the unit holders. In Canada to qualify as a REIT, the trust must receive 90% of its income from passive revenue, primarily rental income. The absence of tax at the trust level makes REIT investments particularly attractive where tax is deferred (RRSPs) or tax free (TFSAs). From the perspective of the REIT owners, the structure accesses pools of capital from private investors that would otherwise be unavailable.
REIT returns are highly correlated with equities1 (correlation 0.63), particularly small company equities (correlation 0.74) when the correlation is measured over quarterly data. This contrasts with the short-term behaviour of direct real estate investments which have a correlation with equities of close to zero. Given that both are means of investing in property it would be reasonable to assume that the performance of both is driven by the same underlying economic forces, so the discrepancy between direct real state and REITs might be viewed as a puzzle. The puzzle is partially resolved if we recall from Part 1 that direct real estate returns are artificially smooth. Over any 3 year period the performance of direct real estate and REITs is increasing correlated (0.43) but falls short of unity, suggesting other factors at work other than the artificially smooth nature of direct real estate returns.
We examined the performance of REITs based on U.S. data, as illustrated in the table below.
| Annualized Return % | Annualized Standard Deviation % | |
|---|---|---|
| Dow Jones U.S. Select REIT Index | 12.72 | 17.29 |
| Russell 2000 Index | 11.68 | 17.68 |
| Russell 3000 Index | 11.95 | 13.82 |
| Bloomberg Barclays U.S. Government Bond Index | 7.43 | 8.16 |
Source: PWL Capital, Dimensional Fund Advisors
The return and volatility of the REIT index is closest to the Russell 2000 Index which is an index of smaller U.S. companies. REITs are definitely not “as safe as houses”. One reason for the, perhaps surprising, volatility of REIT returns is that U.S. REITs have debt ratios of 30-40%. A more detailed analysis2 reveals that only 55% of REIT returns can be explained by assuming they behave like the stock market and other factors are at play that justify treating REITs as a separate asset class from equities. The high leverage of REITs leads to a concern that REITs might be vulnerable to rising interest rates. The same study found no relationship between the performance of REITs and contemporaneous interest rates.
1 Unless indicated otherwise, much of the information in this blog is drawn from two sources: Asset Management: a systematic approach to factor investing, Andrew Ang (2014) and Expected Returns: An Investor’s Guide to Harvesting Market Rewards, Antti Ilmanen (2011).
2 DFA Quarterly Institutional Review, Q2 2015, Real Estate Investment Trusts (available on request).
