There’s a Better Way to Harness Risk Parity Strategies, Says NISA

Risk parity is an allocation strategy that seeks to deliver superior returns and go one better than the asset allocations based on modern portfolio theory. But too often, risk parity has fallen short, disappointing investors. So NISA Investment Advisors has put together an alternative that relies more on real-world returns rather than benchmarks.

Risk parity returns, at least outwardly, typically look better than other asset mixes because they rely on leverage to amp up outcomes, according to a new NISA research paper. The risk parity results, beating benchmarks, often have little to do with managers’ investing skill, the paper argues—the debt is the special sauce.

“Risk parity, as a strategy, has helped many investors achieve higher risk-adjusted returns, in no small part because of the adoption of leverage,” comments David Eichhorn, NISA’s CEO and head of investment strategies. “But it is important not to confuse this outcome with manager skill/alpha.”

The central tenet of modern portfolio theory is that the best diversification can be reached by matching different assets against each other, with riskier ones offsetting less risky ones. Risk is matched up on a graph with expected returns, with each investment choice represented by a plot point. One popular outgrowth of MPT is the now-classic 60-40 equities-bonds mix.

Want the latest institutional investment industry news and insights? Sign up for CIO newsletters. 

Risk parity tinkers with MPT and divides assets into equal shares, such as stocks, bonds, commodities and Treasury inflation-protected securities. Then managers ladle on dollops of borrowed money to get a bigger bang for this buck.

Nonetheless, this is hardly a magic formula, and idiosyncratic application from managers produces losers as well as winners. As the NISA paper notes, “dispersion in performance over risk parity managers has been substantial.” Every manager uses different assumptions about how to treat the ingredients of the allocations: expected returns, volatility and correlations.

Complicating the picture, as NISA points out, is that risk parity is hard to benchmark. The upshot is that a lot of the proclaimed alpha (benchmark-beating) returns are not that great and owe more to leverage. And when the cost of borrowing and manager fees get factored in, those returns don’t look so special.

A better idea, the paper contends, is to use manager performances, as measured by Hedge Fund Research’s indexes, and then adopt a passive approach, letting the assets ride along with the indexes.

To Eichhorn, “Because these weights can be readily ascertained from manager time series either statically or dynamically, we believe that a passive approach exists that investors can utilize to seek better risk-adjusted performance and undoubtedly lower manager fees.”