Back to Portfolio for the Future™

Risk Parity: Riding an Unpleasant Arc

September 27, 2015

10-30-09 © gunnar3000The risk parity strategy has been around at least since the mid-1990s. According to the origin story available from the Bridgewater website, its roots go further back than that. The story begins when Ray Dalio was a clerk at the New York Stock Exchange back in August 1971, which was also when Richard Nixon suspended the convertibility of the dollar into gold, thus creating a sharp break between the old Bretton Woods regime and the float of every-currency-against-every-other that has been the status quo since.

When he heard the news, Dalio expected that stock prices would plummet the following day. Instead, they rose. This catalyzed an epiphany: not just about his own fallibility but about everyone’s . Unique, supposedly once-in-a-hundred year occurrences actually happen fairly often.

Four years later, Dalio founded Bridgewater Associates, an investment management firm that has been very successful, and that now manages more than $160 billion. A quarter century after Dalio’s Nixon-inspired epiphany, in 1996, Bridgewater in turn launched All Weather, the flagship fund of the “risk parity” movement.

Controversy

Now, in the middle of the second decade of the new millennium, there is a good deal of to-do about risk parity, some of it involving the contention that this strategy made a significant contribution to the August 2015 market turmoil.

Specifically, analysts at Bank of America Merrill Lynch contend that the heavy borrowing that risk-parity funds have to do in order to produce a return creates a risk that compounds volatility. Further, there is a positive feedback loop here. The B of A team contended that funds following this strategy tie their own leverage to the amount of volatility they expect in the near future. Thus, increased volatility means deleveraging, which in turn can increase volatility.

Let’s back up a bit: risk parity is a portfolio strategy that targets risk levels across the portfolio’s asset classes, distributing risk equally at the expense of any fixed proportion among the classes. Take the simplest example: a fund manager has a traditional bifurcation of its portfolio into equity and debt: 60% to the former, 40% to the latter. The manager then discovers that 85% of its risk is on the equity side. The simple solution is to sell stocks and use the proceeds to buy bonds, and to keep doing this until the risk is equally distributed, whatever the percentages might be at that point.

The problem, though, is that equities also provide most of the upside for such a portfolio, and that once equity and debt are at risk parity, the portfolio’s return will be, for most investors, unacceptably low. The natural response to this: leverage. RP achieved with borrowed money can do what unleveraged RP cannot.

The Contrast with MVO

Risk Parity is often contrasted with mean variance optimization as systems for risk management. In MVO, which is in turn tied to modern portfolio theory. Under MVO, portfolio allocation is designed to maximize return given a specific level of risk. In other words, the goal is to stay on the risk/return frontier. The expected return from an asset class represents the probability-weighted average of the possible return for each asset, with standard deviation employed as a proxy for uncertainty.

The usual critique of RP is precisely that it doesn’t optimize return against risk, and thus it ends up inside the frontier. But the case against its systemic effect depends upon the alleged tie between the borrowing the strategy requires and the vol.

Dali0 and his team deny such a tie in the practice of Bridgewater’s All Weather Fund.

It appears that All Weather is not doing well in the present weather. Bloomberg, citing an anonymous source, says it is down 4.8 year to date, and that most of that drop came about in August.

Well … nobody does well all the time, and in any big picture – even in any medium sized picture, August 2015 is not going to seem an epochal moment. It wasn’t, say, September 2008.

Nor does the charge that RP funds in general are responsible for that month’s turbulence likely to persuade anyone not a special pleader. RP funds represent a very small part of the over-all market, and it is unlikely they have such systemic effects at present.

A more serious charge though, is that RP is large enough to have become a crowded trade, and to suffer from the usual difficulties of such plays. Andrew Lo says that RP is just one of a number of recent examples of strategies that performed well before they became the common chatter of people who watch cable television channels about finance. Once they did become the chatter, though, they underperformed, and once that became obvious, they underperform still worse as many tried to exit all at once. RP may be somewhere in that unpleasant arc.