Over the last 20 years, institutional investors have raced to employ the endowment model with its large allocation to alternatives to diversify away from the equity risk of in a traditional 60-40 portfolio. But new research from quantitative asset manager PanAgora finds investors would have been better off taking a risk-parity approach over the last 10 years if they wanted diversification.
But both models, designed to provide growth and minimize volatility over a market cycle, underperformed a simpler portfolio with 60 percent in global equities and 40 percent in global bonds over the last 10 years. Risk parity invests in risk premiums associated with equity, interest rates, and inflation, but it uses leverage to meet return targets.
The endowment model’s weakness is that returns of alternative investments, while appearing to be very different from equities, are actually closely correlated to stocks, the paper argued.
“If you talk to endowments, they say they have less of a liquidity need so they have opportunities to seek returns in a different way. In this paper, we asked the question of how diversifying are these alternatives really?” said Bryan Belton — director of multi-asset at PanAgora, which manages equities and multi-asset portfolios — in a phone interview. Risk parity represents about $15 billion of the firm’s $45 billion under management. For the study, PanAgora used endowment data from the National Association of College and University Business Officers (NACUBO) and HFR’s risk parity index.
According to PanAgora, private equity, venture capital, and real estate investments all have exposure to equity risk. “Alternatives act like risky assets with significant positive correlations to stocks and commodities and a negative correlation to bonds. There is illiquidity premium and possibly alpha in alternative investments, but they are offset by the embedded equity risk,” wrote the authors of the report, including Belton.
Another way to measure diversification is tracking error — in this case the return difference between the endowment portfolio and one that is 60-40. The higher the tracking error, the bigger the difference. PanAgora found that the endowment model’s tracking error to be small, 3.2 percent, which would be considered a benchmark hugger when compared to other active managers. Risk parity had a tracking error of 6 percent.
The endowment model is also highly correlated to a 60-40 portfolio, with a 0.98 correlation (1 is perfectly correlated. The HFR Risk Parity Index had a correlation of 0.79).
“Why? A lot of endowments have private equity exposure and that’s really correlated to public equities,” Belton said as an example. “It’s alternative in name, but it tends to do well when equities do well and vice versa,” he added.
“Risk parity is more alternative or less like a 60-40 portfolio than an endowment,” said Belton. Belton added that risk parity is also cheaper and more transparent, investing in asset classes that have no capacity constraints.