As institutional investors increase allocations to alternative investments, it becomes increasingly challenging to understand risk.
A new paper, co-authored by Spencer Couts, assistant professor of real estate at the University of Southern California; Andrei S. Goncalves, an assistant professor of finance at the University of North Carolina; and Andrea Rossi, an assistant professor of finance at the University of Arizona, introduced an unsmoothing technique for real estate and hedge funds to improve the measurement of risk exposure and risk-adjusted performance.
The researchers reviewed the existing techniques to unsmooth illiquid returns, says Rossi. “And we know that, in many cases, these returns that are recorded are based on net asset values. They are not fully mark-to-market, just because of the illiquid nature of the underlying asset investment.”
They found the current techniques are good for predicting volatility, but do a poor job at uncovering the true correlation between the illiquid investment and the proxies for risk in the public market. “Basically, what we do in this article is we propose a new methodology that is specifically trying to address this problem,” he says. “And as we started presenting, the early versions, like in internal seminars, some of the most enthusiastic responses we got were actually from the [chief investment officers] of the endowments of our universities.”
Other common unsmoothing methods take the history of past returns for funds and use statistical models to infer the probable true return after adjusting for the fact that the assets aren’t properly mark-to-market, notes Rossi. “Basically, we do something similar in terms of statistical methods that we use, but essentially if I wanted to visualize what our method does: we put each fund that we know ex-ante, given the strategies that they have — say fixed income-based hedge fund strategies or core diversified U.S. real estate funds; we put them together in a panel and we basically add an additional restriction that makes it so that the index — so the overall average return across this fund — is also unsmoothed.”
This method eliminates a lot of noise in the models, he adds. “By pulling together all the data for hundreds of funds that have similar exposure . . . , we’re able to further eliminate some of that noise, some of that estimation error that makes it difficult to recover the true correlation between the returns of these funds and the returns of indexes that proxy for those risks.”
At the moment, the technique for unsmoothing illiquid fund returns is meant for open-ended funds without a finite life, says Rossi, noting it wouldn’t work for some other alternative asset classes like private equity. But the researchers are working on adapting the methodology so it can be applied to private equity as well.
The paper demonstrated that, on average, actual exposures to risk are larger than previously thought. “What we show is that it’s not only the true volatility that is underreported, but it’s also the amount of correlation of the returns of these funds that is underestimated,” he says.
Pointing to real estate investment trusts as an example, Rossi notes that, when looking at the true correlation of private commercial real estate funds with REITs, the premium earned on the private investments is sufficient to compensate for illiquidity. “Pension funds definitely invest in private real estate. They might think that their investments are less volatile than investing in REITS and, because of that, they’re worth investing in, even though the returns are fairly similar. What we show is that, actually, the true correlation is high, meaning that the risk return trade-off is pretty similar.”
Overall, institutional investors can use unsmoothing mechanisms to better understand the risk-return trade-off. “What we’re showing is, it’s not just the volatility that is higher than you expect,” he says. “It’s also that they correlate a lot more with public markets than you’d expect. And once you account for that, essentially the risk-return trade-off doesn’t look as good as you would think.”
The paper was presented at the Northern Finance Association’s 2020 conference. The Canadian Investment Review is a proud partner of the NFA conference.