Lyxor – Improving traditional risk parity strategies by considering more appropriate risk measures than historical volatility

Risk parity is a generic term used by the asset management industry to designate portfolio construction methodologies based on risk budgeting. Whereas the standard approach to risk parity is used to design a well-diversified strategy, a new generation of risk parity strategies is emerging that considers more sophisticated risk measures.

WHAT IS THE ORIGINAL RISK PARITY STRATEGY?

Since 2008, risk parity is a popular approach to building well-diversified portfolios placing risk management at the heart of the portfolio construction process. This explains why an increasing number of pension funds and other institutional investors are now using this approach both within asset classes, and notably for the development of smart beta equity and bond benchmarks, and across asset classes, that is for the redefinition of their long-term investment policy portfolios.

This methodology ensures that the risk contributions will be identical for all constituent assets of the portfolio. The attractive performance of such strategies in recent years undoubtedly explains the success of risk parity diversified funds based on equities, bonds and commodities. For instance, Invesco manages about USD 22 billion using a risk parity strategy. Another commercial success is Bridgewater’s “All Weather Fund”, which is one of the largest hedge funds in the world.

TRADITIONAL RISK PARITY STRATEGIES SUFFER FROM A NUMBER OF SHORTCOMINGS

A risk parity portfolio is typically defined by considering volatility as the risk measure. However, these strategies also suffer from a number of shortcomings.

Standard approaches to risk parity are based on portfolio volatility as the risk measure, implying that upside risk is penalized as much as downside risk in obvious contradiction with investors’ preferences.

Moreover it is not obvious indeed that historical volatility of market returns has the same meaning for equities, sovereign bonds, corporate bonds, inflation-linked bonds, commodities, etc. In particular, market volatility does not fully reflect the risk taken by an investor when investing in bonds, because it only partially incorporates default risk. For example, the recent crisis in the Eurozone sovereign bond markets has not been accompanied by a massive increase in the historical market volatility.

Typical risk parity strategies inevitably involve a substantial overweighting of bonds with respect to equities, which might be a problem in a low bond yield environment, with mean-reversion implying that a drop in long-term bond prices might be more likely than a further increase in bond prices. More generally, risk parity strategies do not take into account changing economic environments and in particular time-varying risk premia.

Perhaps as a consequence of these shortcomings, the performance of risk parity funds has been disappointing overall in 2013. Indeed, most of them posted negative or flat performances in a context of strong equity returns. Moreover, dispersion among their performances was high, with as much as a 20% difference between the best and worst performers.

RISK PARITY NEW GENERATION: WHY DIRECTIONAL RISKS MATTER

A new generation of risk parity strategies – sometimes known as conditional risk parity strategies to emphasize their higher degree of reactivity to changes in market conditions – introduces a new dimension, which is the directional risk, as reflected in particular in the first moment (expected return) and third moment (skewness) of asset return distributions, in addition to volatility.
This new approach recognizes that risk cannot only be summarized as the daily fluctuations of prices. In Exhibit 1, the prices of three assets are reported with the same 20% volatility, For the standard risk parity approach, there is no difference between the three assets because they present the same volatility patterns. For the conditional risk parity approach, the three assets may not present the same risk patterns because they do not have, for instance, the same directional risks as can be seen by the very different trends.

For instance, if we consider a standard risk parity portfolio of equities and bonds, we notice that the allocation is the same in June 2008 and July 2013. This implies that these two periods present the same patterns in terms of relative levels of stock and bond volatility.. However, these two periods vastly differ in terms of directional risks as can be measured through valuation, trend, risk premium, economic condition, etc.

By reintroducing expected returns into the risk measure, the conditional risk parity strategy corrects one drawback of the traditional risk parity, which is its lack of sensitivity to yield levels. The use of conditional risk parity strategies avoid the sample dependency to the historical market volatility of bond returns and overweighting of bonds in a low interest rate environment. This new generation of risk parity strategies is therefore better equipped to deal with the rising interest rate challenge.

In this framework, risk parity becomes a robust active management model, which is less aggressive than the Markowitz model. This approach is therefore an interesting alternative to traditional active management approaches when portfolio managers want active management without concentrating their portfolios on a small number of aggressive bets. This explains why conditional risk parity strategies are also often called active risk parity strategies by the asset management industry.