Should risk controlled equity be seen as a smart beta?
According to the Financial Times lexicon, “Smart beta strategies attempt to deliver a better risk and return trade off than traditional market cap weighted indices”1. So do risk controlled equities pass this test? We decided to find out using the evaluation criteria set by Ashley Lester and Fred Dopfel in their paper Improving investment outcomes with advanced beta: moving beyond elementary smart beta2 to find out.
Dopfel and Lester proposed that, to qualify as an “advanced beta”, a fund needs to meet:
- a “value add” condition, i.e. provide something better than is already available; and
- a “portfolio demand” condition, i.e. help the investor attain a desired outcome.
We investigated four candidate risk controlled equity betas – an equity collar, a volatility cap, a volatility target and a variable volatility cap3 – and compared them with a set of four broadly adopted smart betas from the MSCI family of indices – MSCI World (global equities), Small Cap, Value, Momentum and Minimum Volatility. We added the Merrill Lynch Global Broad Market Index (GBMI) for global bonds, as well as cash.
To test the value add condition, we looked at the correlations of the excess returns versus global equities for both the smart betas and the risk controlled equity portfolios. We wanted to see if the latter could be explained by the existing smart betas. We went on to investigate whether risk controlled betas could be built using our extended universe of assets. Our conclusion was that risk controlled equity betas cannot be replicated using a combination of standard and smart betas. And while risk controlled equity betas do not tend to add excess return relative to global equities, they do offer a better Sharpe ratio and a lower risk of capital loss.
These two findings indicate that risk controlled equity betas satisfy the value add criteria and thus can bring beneficial characteristics when considered in a total portfolio context. This suggests that risk controlled betas could be considered as a new set of portfolio building blocks that offer something different than existing equity, smart beta, bond and cash components of a portfolio.
To test the “portfolio demand” condition, we built an efficient frontier for the universe of assets with and without the risk controlled equity betas. We then compared the risk versus return trade-off using Sharpe ratios and tail risks using conditional value at risk (CVaR) to see whether the introduction of risk controlled equity betas can help improve expected outcomes for investors.
When we constructed these efficient risk and return portfolios, we found that risk controlled equities helped improve Sharpe ratios. This indicates that risk controlled equity betas satisfy the portfolio demand criterion by helping investors attain an outcome that they couldn’t otherwise have achieved. Thus the addition of risk controlled equities to a portfolio provides the opportunity either to recycle risk (i.e. increase returns for a given level of risk) or to minimise the reduction in expected returns as a portfolio is de-risked.
The implication of our analysis is far greater than extending the definition of smart betas. It shows that the inclusion of risk controlled equities in all but the most risk-seeking of portfolios can not only improve returns for a given level of risk, but also reduce drawdowns. Therefore these techniques should be considered an important long-term strategic element in many portfolios and be seen as much more than a short-term tool to protect against equity shocks.
3These were examined in greater detail in Managing investment outcomes with volatility control, Mike Hodgson and Andy Connell, Schroders July 2016.
Our full research paper can be found below.