US-listed REITs account for nearly $1 trillion market cap out of the approximate $2 trillion global real estate estate sector. Not surprisingly, US-listed real estate funds, including active mutuals and exchange traded funds (ETFs), invest heavily in large cap domestic REITs. Listed funds manage roughly $170 billion of assets under management (AUM) across 190 products, however the vast majority of funds under performed the Morningstar Global Real Estate Index. Indeed, the Vanguard REIT ETF itself under performed not only the Morningstar Global Real Estate benchmark but also the S&P500 and displays higher volatility year-to-date, which suggests interest rate concerns are the over riding factor at play in the sector.
US-listed real estate funds are an interesting case study in both chronic under performance across the space and over concentration of AUM in popular products. The Vanguard REIT ETF holds 20% of all AUM and is nearly 7x larger than the next largest product. For the year-to-date period as of 10/14/2016, 72% of all listed products failed to match the returns of the Vanguard REIT ETF and a full 90% under performed the broad Morningstar Global Real Estate Index. We ran the same analysis over a five year period and the ratio of under performers mirrored the year-to-date results.
Our research suggests there a several factors at play here. First, US-listed funds tend to own the same basket of large cap REITs so other than cost, the typical real estate fund tracks the largest, most cost effective funds. Second, few US managers venture outside of large cap developed markets where non-correlated returns are available. Third, the concentration of AUM in a single product likely skews performance in the widely-held REITs, and fourth, US REITs are fully-valued and have been a drag on returns this year, as measured by higher volatility, beta and correlation versus the S&P500; that what worked well over the last few years, simply isn’t the case now. For example, the Vanguard REIT ETF alone shed $4 billion in market cap since the end of July (-11%) which likely impacted performance of other funds as well. And this shows up as well when we look at beta across the US-listed real estate fund space.
Few realize that US real estate funds are no longer non-correlated sources of alpha. In fact, over 80% of funds exhibit beta >1 versus the S&P500. The Vanguard REIT ETF showed a 1.1x beta versus the S&P500 year-to-date, realized volatility of 16.5% versus the S&P500’s volatility of 14%, and it under performed the S&P500 by approximately 200bps. It turns out that a 100bps in additional yield actually cost the investor 1000bps in alpha this year alone.
So what is going on here? Is this temporary or a harbinger of worse things to come? Real estate traditionally was a diversifying factor in a portfolio and that clearly is not the case for now. We believe, like fixed income, US REITs are already in a bear market but not the traditional bear market where losses occur over a relatively short period of time, asset prices reset and new capital re-enters. This bear is going to take a while, maybe two to three years, or longer. For the US REIT investor, it may end up feeling like a slow boil, marked by chronic under performance for the large cap, widely-held companies.
On the interest rate front, we think US REITs are trading in lockstep with fixed income. Large cap US REITs offer very limited appreciation potential as a result of oversupply and slow growth, trade at historically high valuations and have become quite crowded. In analyzing the real estate fund segment, the problem with US REITs becomes even clearer – extracting alpha from a crowded, highly correlated basket with above average volatility almost guarantees under performance. So, the question is: why aren’t managers diversifying into non-correlated, growth markets like Latin America?
By simply adding an allocation to properly diversified LatAm real estate this year, the average US manager could have improved performance substantially and reduced its risk profile at the same time. However, if the manager wants to move the needle, the allocation must be material. Some managers appear to have approached Latin America with a check-the-box mentality, wherein they add one or two names at a de minimis weight. Token allocations in the handful of widely-held Latin America names is not going to have an impact. To do so, the US manager will have to allocate at least five percent and ideally 10 percent but do so over a broad regional basket of real estate equities which presents a conundrum: many names are too small to be added individually. The solution? We believe a regionally diversified vehicle that offers exposure across the market cap structure without having to engage in single stock selection but offers compelling risk-adjusted exposure is a pretty good solution.