By LARRY SWEDROE
The “conventional wisdom” — ideas that become so commonly accepted that they go unquestioned — on investing is often wrong. An example of that is that rising interest rates are bad for real estate returns. Let’s examine the evidence for REIT returns in particular.
REITs and interest rates
Randy Anderson, Eli Beracha and Spencer Propper, authors of the study The Effect of Interest Rates on REITs Valuation and Future Returns, published in the August 2022 issue of The Journal of Investing, analyzed the relation between interest rates, REITs (real estate investment trusts) valuation and future returns using a dataset on equity REITs that spanned the period 1993-2019. They investigated whether, and to what extent, nominal and real interest rates as well as interest rate quality and term spreads were related to the valuation of REITs (measured by different metrics) and REITs returns in the medium run. Their data sample covered the 27-year period 1993-2019 and 400 REITs. They collected data on both the level of the nominal and real 10-year rates as well as the quality interest spread (Moody’s seasoned Baa-rated corporate bond yield minus the yield on the 10-year Treasury with constant maturity) and the term interest spread (the 10-year Treasury constant yield minus the 3-month Treasury yield).
To measure the level of equity REITs valuation, they used the following metrics:
NAV premium: The average percent premium (or discount) of equity REITs price per share relative to their net asset value (NAV) estimation. Higher values suggest that equity REITs are trading at higher prices relative to their intrinsic asset value and therefore imply higher valuation and vice versa.
P/FFO: The average ratio of equity REITs price to funds from operation (FFO) per share. Higher values are evidence of higher equity REITs trading prices relative to the funds they generate from operation, which imply higher valuation and vice versa.
Implied CAP: The average implied capitalisation rate (CAP) of equity REITs. Because CAP rate is the ratio of net operating income (NOI) to price, lower values of implied CAP mean that equity REITs trade for high prices relative to their NOI and point to higher valuation and vice versa.
(P/FFO)/PE SP500: The value of P/FFO, as defined previously, divided by the average price-to-earnings ratio (PE) of the S&P 500 (PE SP500). Higher values of this measure suggest that REITs are trading at a high price relative to the S&P 500 when funds from operation and earnings are accounted for, which implies higher REITs valuation and vice versa.
Dividend yield/Baa yield: The average dividend yield of REITs divided by the yield of Baa-rated corporate bonds. Lower values show that REITs yields are lower relative to the yield of corporate bonds and imply higher REITs valuation and vice versa.
Dividend yield/dividend yield SP500: The average dividend yield of REITs divided by the dividend yield of the S&P 500. Lower values show that REITs yields are lower relative to the dividend yield of S&P 500 firms and imply higher REITs valuation and vice versa.
They referred to the first three valuation metrics (NAV premium, P/FFO and implied CAP) as fundamental metrics because they value REITs with respect to their own fundamentals. They referred to the remaining three valuation metrics as relative metrics because they value REITs relative to other investable assets. Following is a summary of their findings:
On average, equity REITs traded at a slight premium of 3.7 percent to their asset value, an implied CAP of 8.4 percent, and a P/FFO ratio of 12.2 (an FFO yield of 8.2 percent). On a relative basis, equity REITs P/FFO was 0.6 times the PE of S&P 500 firms on average. The dividend yield of equity REITs was slightly lower (0.82 times) than the yield of Baa-rated corporate bonds, but more than three times (3.29 times) the dividend yield of S&P 500 firms on average.
Equity REITs valuations were rising during periods of improving economic conditions and falling during periods of worsening economic conditions—consistent with general economic intuition. However, the R-squared values (correlations) were relatively low—the explanatory power was not high.
When the initial equity REIT valuation level was low (risk perception high), rising nominal interest rates (indicative of a stronger economy) were associated with higher equity REIT valuation (higher NAV premium, higher P/FFO and lower implied CAP). However, when the initial equity REIT valuation was high (risk perception low), the relation between changes in the nominal rate and absolute equity REIT valuation was statistically insignificant. In other words, the absolute level of interest rates mattered. Similar results were found when examining REIT valuations relative to the S&P 500.
A widening quality (default) spread was associated with a lower equity REITs valuation when the initial REITs valuation was low. However, when the initial REITs valuation level was high, the coefficients of quality were mostly statistically insignificant.
REITs’ future three-year returns were negatively related to the level of the nominal rate—a higher level of the nominal rate was associated with lower future returns and vice versa.
Once the nominal interest rate was accounted for, lower equity REITs valuations were associated with higher future returns.
A higher level of quality spread was associated with higher future equity REITs returns.
Once the quality interest rate spread was accounted for, lower equity REITs valuation level was associated with higher future returns.
Their findings led Anderson, Beracha and Propper to conclude: “REITs valuations increase in both absolute terms and relative to other investment alternatives when nominal or real interest are rising, as well as when interest quality spread narrows—all signal strengthening economic conditions. The results are mostly driven by periods when initial REITs valuations are low. … Additionally, our analysis provides evidence that lower REITs valuations are followed by periods of higher REITs performance, even when we control for the interest rate environment. Overall, our findings are consistent with our economic intuition and suggest that REITs valuations are overcorrected by markets during periods of strengthening or weakening economic conditions.” They added: “Higher levels of nominal rates or lower levels of quality interest spread — both indicators of healthy economic conditions — are associated with lower future returns, perhaps reflecting an overcapitalisation of these economic conditions into REITs prices. Simultaneously, higher levels of REITs valuation, measured by different valuation metrics, yield lower future returns.”
Another example of conventional wisdom being wrong is that REITs are a unique asset class because, in aggregate, they historically have had relatively low correlation with both stocks and bonds, and their returns have not been well explained by the single-factor CAPM. For example, during the period January 1978-July 2022, the correlation of the Dow Jones Select REIT Index to the S&P 500 Index was 0.60 and just 0.06 to five-year Treasuries.
Are REITs a distinct asset class?
Research, including the 2017 study Are REITs a Distinct Asset Class? by Jared Kizer and Sean Grover, and the 2018 study Real Estate Betas and the Implications for Asset Allocation by Peter Mladina, has found that the returns of REITS are well explained by common factors from asset pricing models: market beta, size, value, term and default. For example, Mladina found that all five factors had economically large and statistically significant (at the 1 percent confidence level) factor betas. The betas were: market beta, 0.58; size, 0.38; value, 0.72; term, 1.04; default, 0.45. He also found that two-thirds of excess REIT risk was explained by compensated factor risk. In other words, REITs could be considered a hybrid asset class, showing return and risk behaviours common to both stocks and bonds, with some diversification benefits.
Given the findings that REITs are a hybrid asset that loads heavily on term risk, it should not be a surprise that REIT valuations and returns can be impacted by changes in interest rates, though the relationship is ambiguous. As we discussed, while rising rates would negatively impact REITs because of their term premium exposure, rising rates can also reflect a strong economy, and REITs have significant exposure to the economic cycle risk of stocks through their exposures to the market beta and default factors. Thus, rising rates do not doom REITs to having poor returns. To demonstrate the point, we will examine the returns to the Dow Jones U.S. Select REIT Index during a period of rising interest rates.
The federal funds rate bottomed out on June 25, 2003, at 1 percent. Over the next several years, the Fed kept raising the rate until it peaked at 5.25 percent on June 29, 2006. On June 25, 2003, the five-year Treasury note yielded 2.3 percent. On June 29, 2006, the yield had risen 2.9 percentage points, to 5.2 percent.
How did REITs perform during that period of sharply rising rates? From July 2003 through June 2006, the Dow Jones U.S. Select REIT Index returned 27.7 percent per annum, providing a total return of 112.7 percent. During the same period, the S&P 500 Index returned 11.4 percent per annum, providing a total return of 39.3 percent.
If rising rates are supposedly bad for REITs (and for stocks in general), why did they produce such great returns? The reason is that the impact of rising rates on REIT returns depends on the sources of those rising rates. If rising rates reflect strong economic growth, the expected returns to REIT investments might also be good. This could reflect stronger demand as well as the likelihood of a falling risk premium, which causes valuations—for example, price-to-earnings ratios—to rise.
On the other hand, if interest rates are rising because inflation is growing faster than expected, the markets could become concerned that the Fed will begin tightening monetary policy. That would likely put a damper on economic growth and probably cause a rise in the risk premium, which causes valuations to fall. Thus, there are some periods when rising interest rates are likely good for REITs, and some periods when they are likely to have a negative impact. That explains why the correlations have been close to zero over the long term.
Investor takeaways
The fact that, from January 1978 through July 2022, the monthly correlation of the Dow Jones U.S. Select REIT Index with five-year Treasuries was just 0.07 indicates that there has been very little correlation between the level of interest rates on REIT returns. Thus, it should be no surprise that changes in interest rates have had an ambiguous relationship with REIT valuations and returns. As Anderson, Beracha and Propper demonstrated, the absolute starting level of rates matters, the strength of economic conditions matters (rising rates signal a stronger economy), and credit spreads matter as well.
© The Evidence-Based Investor MMXXIV. All rights reserved. Unauthorised use and/ or duplication of this material without express and written permission is strictly prohibited.