By Harold Heuschmidt
Risk parity concepts are increasingly becoming a key component of many investors’ portfolios. The strategy of diversifying assets across various classes based on risk rather than capital has convinced many investors and achieved convincing results over the years.
Risk parity strategies work on scientifically- substantiated basic assumptions. Allocation decisions based on assumptions about the covariance of asset classes deliver more reliable results than those made on future return forecasts. A direct connection exists between risk and return in the medium to long term. Long-term Sharpe ratios are roughly equal across all asset classes. Finally, low average correlation between selected asset classes leads to considerable diversification effects for the overall portfolio.
The risk-parity concept seems intuitive – with asset distribution based on risk parity, capital is not simply invested in various asset classes such as equities, bonds, short-term interest rates and commodities; the risks of holding these assets are also balanced. With the risk parity approach, all asset classes contribute the same risk to the portfolio. More capital flows into asset classes with lower risk and less capital into asset classes with higher volatility, i.e. higher risk. This makes it possible to avoid overweighting strongly fluctuating asset classes. The idea behind risk parity is that in efficient markets, future performance cannot be forecast accurately and exploiting the benefits of diversification offers the best return per unit of risk.
In June of this year, however, all risk parity funds suffered setbacks. Statements by the Chairman of the US Federal Reserve, Ben Bernanke, on a possible reduction of quantitative easing sent shock waves through markets. Consequently, almost all asset classes delivered negative performance contributions to risk parity.
Tail events such as those in June can negatively impact risk parity performance in the short term. Even if the correlations of individual asset classes are negative in the long term, the occurrence of brief phases of simultaneously falling equity and bond prices cannot be ruled out. Based on US data, we found only seven months since 1945 in which both markets dropped by more than one standard deviation at the same time.
Such tail events are not only rare, their occurrence cannot be forecast. Gratifyingly, tail events of this kind occur more frequently at the right end of the distribution than the left. In the same period, equities and bonds simultaneously recorded gains in 19 months which exceeded one standard deviation.
The risk parity approach is underpinned by the knowledge that the underperformance of an asset class versus its long-term risk premium is very likely to be offset by the outperformance of another. So it comes as no surprise that various studies show that the risk parity approach also pays off in times of rising interest rates. We back-tested the performance of a simple risk parity portfolio consisting of US equities and bonds from 1956 to 1980. During this period of sharply rising interest rates, diversification across two asset classes would have generated an attractive average return of 6% per annum.
Aquila’s approach aims to maximise diversification benefits through the careful selection of asset classes based on the availability of risk premia, for their low correlations to each other and with specific reference to liquidity. In non-systematic approaches, investment decisions are frequently driven by emotion, particularly at inflexion points. Our risk parity funds use a time-tested, systematic approach which automatically triggers a reduction of exposure in bouts of weakness.
Harold Heuschmidt is head of quant fund management at Aquila Capital
Comments