Risk parity outperforms 60/40 portfolios, says Aquila

A risk parity approach to investing has outperformed the traditional 60%/40% equity/bonds portfolio since 2004, delivering positive performance in each calendar year

Risk parity strategy seeks balance

Ibn Khaldun, the 14th-century historian and one of the leading jurists of his day, once observed that in the world of existence or in science, “it is from the relations existing among data that one finds out the unknown from the known”. However, “things of the future”, he cautioned, “belong to the supernatural and cannot be known”.

Things may have moved on since the days of Khaldun – advances in quantitative analysis have transformed the way we see and understand the world – but he was fundamentally right about the power of prediction. We make certain observations and inferences based on what we know but consistently reliable forecasting techniques continue to be elusive.

Occasional genius aside, this certainly applies to predicting the returns of broad asset classes or even sectors, investment themes or geographies, let alone individual securities. We all seek an informational or methodological edge but ultimately all we can do is guide our capital towards where the essential positive propensities reside.

Risk parity is rooted in the observation that certain asset classes display a ‘long random walk upward’ over time and the conviction that effective diversification in the face of uncertainty is the cornerstone of successful investing. Underpinned by its market agnosticism, interest in the concept is growing.

Investors are becoming aware that a well-constructed risk parity approach, which blends uncorrelated assets on a risk-equalised basis, can deliver superior risk-adjusted returns over time.

To understand risk parity, it is worth revisiting the genesis of the ideas that aim to define and optimise the relationship between investment risk and reward. Nobel Prize winner Harry Markowitz introduced the efficient frontier concept more than 60 years ago. This went on to become an important part of modern portfolio theory, which for all its shortcomings is still a useful reference point.

Markowitz suggested that by using three metrics – mean return, the anticipated dispersion of returns and correlation with other assets – optimal portfolios could be created, running from low to high risk and low to high return. The opportunity set illustrating the highest level of return for a given level of risk could be represented along a notional efficient frontier.

The tidy world that Markowitz envisaged was made up of markets with continuous pricing where participants had equal access to finance, perfect information and behaved rationally. Markowitz himself accepted these stretching assumptions were like “studying the motion of objects on earth under the assumption that the earth has no air” .

Markowitz highlighted the benefits of diversification but in the real world, with air and cross-currents, his process contributed to an environment where investment practitioners continue to underestimate risk. This can be seen, for example, in balanced portfolios that carry unintended skews and other risks.

They will have higher tail risk than more effectively diversified portfolios, are unlikely to perform consistently across the economic cycle and tend to deliver lower risk-adjusted returns. These shortcomings have become abundantly clear in the last decade.

Risk parity seeks to re-introduce equilibrium into a portfolio’s risk profile by combining assets that will perform differently across the economic cycle. To do so, blend several different asset classes, ideally ones that display sustainably extractable risk premiums and show a low degree of correlation to one another. Risk equalise the capital allocations to each (chart 1).

As the volatility of individual asset classes changes over time, regular portfolio rebalancing can ensure that the overall risk-equalised structure is maintained. This is very different in nature to rebalancing based on capital allocations. In its purest form the blended portfolio should deliver the same returns as the combined market aggregates with much less risk.

Applied effectively the approach can contribute to a truly diversified portfolio: a portfolio that can deliver ‘all-weather’ performance with less dramatic drawdowns; generate superior risk-adjusted returns over the long term; and exhibit low correlations to markets and other investment strategies, including traditional hedge funds.

In practice investors can apply risk parity in a number of ways, including as a diversified growth engine, an alternative to balanced multi-asset allocations or as a diversifier to hedge fund holdings.

The approaches themselves are diverse. Some, like Aquila Capital, are entirely systematic while others are discretionary. Some place less emphasis on the liquidity of the asset classes used and others do not necessarily optimise the choice of asset classes to be mixed or even the instruments through which to capture the risk premia involved.

For Aquila the starting point in a three-step investment process is an allocation to equities, government bonds, commodities and short-term interest rates, each accounting for 25% of risk. Asset classes are chosen with specific reference to liquidity and the availability of risk premiums. Together, these factors can influence the generation of smart systematic beta.

A tactical overlay is used to fine-tune the asset allocations. The overlay, driven by a series of behavioural factors, can result in allocations shifting up to 10% in either direction, altering weightings in a way that can be a potential source of alpha.

Risk is managed through FundCreator, one of the world’s most advanced risk management systems developed by Harry Kat and Helder Palaro. Its objective is to give the portfolio a consistent risk profile characterised by a normal distribution of returns with zero excess skewness and kurtosis. FundCreator ensures portfolio volatility remains within pre-determined limits and establishes well-defined downside floors to limit monthly losses.

The strategy is currently available via customised mandates and three Ucits-compliant funds, offering high levels of transparency and daily liquidity. The AC Risk Parity 7 Fund has an annualised volatility target of 7%, the AC Risk Parity 12 Fund targets 12% and the AC Risk Parity 17 Fund targets 17%. These targets are calculated from five-year rolling volatilities, which we believe is the most effective reference point: long enough to prohibit the fund being ‘whipped out’ of positions too early.

Maximum monthly loss limits are set at 4% for AC Risk Parity 7 Fund, 7% for AC Risk Parity 12 and 10% for AC Risk Parity 17 Fund. Asset liquidity is particularly important in this context as part of the risk management framework involves the ability to liquidate holdings and reduce market exposure quickly if volatility increases. If this action is triggered, the risk-equalised asset profile will remain unchanged. However, overall exposure is reduced only to be reset at the start of the next calendar month.

The low correlation between the asset classes held within the strategy has delivered strong all-weather performance. It has helped to shelter investors both from bond and equity market drawdowns.

Bond bear markets, for example, are associated with rising interest rates and a steepening yield curve. With a well-constructed risk parity portfolio, the expectation is that as rates rise, gains from the other asset classes will help offset any deterioration from the fixed income segment of the portfolio.

The direction of yield movements is not the only feature to influence the success or otherwise of bond investments: the speed and size of any rate rises and the element of surprise are important too. In 2005-7 when the benchmark lending rate rose by 5%, the contribution of bonds to the performance of Aquila’s risk parity strategy was only mildly negative. In 2009, the worst year for long-term government bonds since 1926, the strategy’s returns were positive.

We believe risk parity embodies some quite profound observations about the nature of markets and the limits to knowledge. In the context of complex systems with multiple feedback loops, where the existence of investment models can change the nature of behaviour itself, today’s forecasting techniques are inadequate.

Using risk parity is a measured response to our inability to anticipate what might happen and a more effective route to portfolio diversification.

The proof of its effectiveness can be found in low levels of cross-correlation and steady Sharpe ratios. Risk parity’s role as a portfolio diversifier is invaluable.

Stuart MacDonald, a managing director at Aquila Capital, wrote this article.