The rise in alternative investments among public pension plans has become a major concern for those wary of risk taking in the public sector. Several reports have previously covered this trend. However, few have systematically investigated the kinds of plans that have made this shift, as well as the impact it had on the overall investment returns and volatility. A recent study by the Center for Retirement Research at Boston College (CRR) attempts to do just that.
Covering nearly 95% of public pension assets in the US, the study shows what has been shown before: the move towards alternative investments over the last decade was significant. In aggregate, the share of alternative investments increased from 9% to 24% between 2005 and 2015. Back in 2005, the maximum share of alternatives was under 30%, and half the plans held less than 10% in alternatives. In 2015, the maximum share was 50%, and only 9% of the plans held less than 10% in alternatives.
What plan characteristics are associated with a higher alternatives allocation? The authors look at a number of variables, including plan sizes, funded ratios, payments of required contributions, return assumptions, and how early the plans started investing in alternatives. They find that only the last two variables are statistically significant in predicting the share of alternatives. Plans that have above-average return assumptions tend to invest more in alternatives in hope of higher expected returns. And plans that invested early in alternatives (before the Great Recession) seem to be more interested in these investments, and thus tend to hold more of them. These results are not surprising.
The results for the effects of alternatives on returns and volatility, however, are more interesting. The study examines four major alternative categories: private equity, hedge funds, real estate, and commodities. It finds that overall, plans that allocate more to alternatives earn lower returns.
What explains this? A deeper look into the specific alternative categories reveals that while private equity and real estate are associated with higher returns, hedge funds and commodities are correlated with lower investment performance. Among these relationships, however, only the effect of hedge funds is statistically significant, which explains the overall negative correlation between alternatives and returns. The fact that other relationships are not statistically significant also implies that public pension plans could have fared just as well investing in traditional equities (which indeed has been demonstrated by the experiences of several plans).
What about the effects on return volatility? The study finds that alternatives, as a group, do not have a statistically significant effect on volatility. That sounds puzzling to many. Numerous financial experts believe that the shift to alternatives to boost returns in a low yield environment entails higher risk due to the standard risk-return tradeoff in finance. Others, especially public plan managers, believe that alternative investments help enhance returns while reducing risk allegedly thanks to their low correlations with other asset classes.
The reality, according to the study, is a bit more complicated when looking at the individual alternative categories. Depending on the time frames, real estate and commodities are found to be positively correlated with volatility, while hedge funds are associated with lower volatility (with statistical significance). Private equity, interestingly, does not have a statistically significant effect on volatility. This is quite odd, as private equity itself is expected to have highly volatile returns, and their returns are strongly correlated with those of traditional equities.
One important lesson from the study is that not all alternative investments are the same. Different alternative categories may have different effects on risk and returns. Investing in alternatives does not necessarily yield higher returns, and the additional risk involved is not clear either.
However, it does not mean that the current investments made by public plans are not highly risky. There is a critical difference between marginal risk and total risk. The total risk assumed by public plans may have well increased over time since interest rates have greatly declined over the past several decades and since the share of fixed income and cash held by those plans has dropped considerably over the same period. That is not necessarily inconsistent with the fact that the marginal risk from replacing one unit of traditional stocks with one unit of alternatives is insignificant.
It should also be noted that the time frame studied is relatively short (only 10 years), and thus the results may not be robust enough to make a precise judgement on these asset classes. Despite this limitation, the CRR study provides useful insights for policymakers and serves a good model for future research.