Risk Parity 2.0?

(September 12, 2012) – Risk parity strategies have led many investors to strong returns over the last several years, but according to JP Morgan strategist Peter Rappoport, it’s not due to balanced risk.

Instead, when investors trade in the traditional 60/40 asset allocation approach for risk parity, they also tend to stop relying so much on fallible return forecasts, Rappoport argues.

He is the head of strategy at JP Morgan Asset Management, and recently published a report delving into the mechanisms behind risk parity. He and co-author Nicholas Nottebohm were continually hearing the same question from institutional clients, and the study was their attempt to come up with a satisfying answer. “Investors’ quandary now is whether risk parity has found the Holy Grail of asset allocation, or has just levered bonds during a long-lasting fixed income rally that may soon end,” the authors state in their paper.  

Their conclusion is a bold one: “Appealing as it may sound,” the authors argue, equal risk sharing “is meaningless, because every portfolio, even the 60/40, exhibits equal risk shares if we define the asset classes appropriately…Nothing needs to be bought or sold to go from one portfolio to the other, there is just a just reclassification. We should be wary of acting on any feature of a portfolio that changes depending on how we look at it. This simple but apparently unrecognized fact means that we have to look elsewhere for risk parity’s success.”

Many prominent risk parity experts might disagree with this verdict. For instance, Bob Prince, the co-chief investment officer of Bridgewater Associates, the world’s largest hedge fund and oft-cited as the originator of risk parity through its All-Weather fund, gave aiCIO his take on Rappoport’s conclusion: “The key to successful strategic asset allocation is reliable balance. Reliable balance is the goal, and reliable balance is unlikely to be achieved by holding equal risk shares across a series of asset classes. Rather, reliable balance is better achieved by holding equal exposures to all possible economic environments–inflations, deflations, strong growth and weak growth.”

In a practical sense, however, Rappoport and Nottebohm are largely in agreement with Prince. Despite their emphatic take-down of risk parity’s core principle, the authors see a slightly modified version of risk parity as institutional investors’ best technique for allocation assets.

After running numerous simulations, Rappoport found risk parity excels in the current economic climate in part due to its diminished reliance on return projections: “Forecasts plainly make errors; our simulation exercise suggests that these errors are big enough even under favorable circumstances to outweigh anything constructive the forecasts have to offer about asset returns. This reveals a dimension of practical asset allocation that rarely gets explicit consideration. It involves the extra layer of uncertainty created by unavoidable errors in return forecasts. In our simulation example, where asset Sharpe ratios were very close, risk parity turned out to be extremely successful uncertainty management.”

Uncertainty management has always been central to risk parity-as the name of Bridgewater’s hallmark fund suggests, it’s an approach for “All Weather.” But Rappoport and his colleagues at JP Morgan Asset Management argue that risk parity can be even better at what it already does well: offer strong returns in an uncertain and volatile environment. A risk parity allocation model that takes into account the surety-or lack thereof-behind every forecasted return gives investors a more nuanced and powerful tool, the report concludes.  

“If I were a CIO, I would want to be using this,” Rappoport told aiCIO. “You get a sense of how sensitive the outcome is going to be to changing those bounds of forecast certainty. In some cases, the outcome doesn’t change that much whether you’re absolutely or not-at-all sure about a projected return. It’s more a matter of playing around with this tool, and finding out what are the portfolios that are playing close to the edge. When someone provides or uses forecasts of returns, I think they imagine it as a ‘shrug your shoulders’ type thing-a best guess. Whereas in contrast, traditional asset allocation considers a forecast to be a laser-guided target. This is just a way of better matching the way you use your tools with what people actually intend them to be.”