Low Volatility Research: Academic Papers Summarised

This page gathers short, paraphrased summaries of research papers that bear on the low volatility effect. They were listed on the original site for readers who wanted the academic background. Readers should consult the original papers, which can be found through SSRN and the journals named below, before relying on any finding.

Explanations for the volatility effect

Authors: David Blitz, Eric Falkenstein and Pim Van Vliet (2013).

The Capital Asset Pricing Model (CAPM) says that higher risk should bring higher return. Empirical data, however, show a relationship that is flat or even reversed. The authors survey the many explanations scholars have offered for this and sort each one by the CAPM assumption it challenges. A notable theme is that several explanations rest on behaviour that is sensible given the rules and incentives investors face, which could explain why the effect has lasted so long. The paper concludes that fixing the reasons CAPM fails might itself be an opportunity.

Attention and news sentiment

Authors: Peter Hafez and Junqiang Xie, RavenPack Quantitative Research (2013).

The researchers treat media coverage, sentiment trends, price trends and return volatility as different measures of investor attention. In their backtest, a portfolio that bought stocks with positive news sentiment and low volatility while shorting those with negative sentiment and high volatility produced an annualised spread of 16.3 per cent. That figure applies only to the paper’s own test period.

Volatility and expected returns

Authors: Andrew Ang, Robert Hodrick, Yuhang Xing and Xiaoyan Zhang, Journal of Finance.

This study asks how volatility risk is rewarded in the cross-section of stock returns. Shares that were highly sensitive to shifts in overall market volatility earned low average returns. So did shares with high idiosyncratic volatility, meaning company-specific swings, and exposure to market-wide volatility did not account for it. Size, book-to-market, momentum, liquidity and trading volume did not explain the results either.

International evidence on idiosyncratic volatility

Authors: the same four, Journal of Financial Economics, 2009 edition of a paper first printed in 2008.

Extending the work abroad, the team found that stocks with recently high idiosyncratic volatility tend to deliver low returns in later periods around the world. Across 23 developed markets, the gap between the most and least volatile quintile portfolios averaged about 1.31 percentage points per month, after allowing for world market, size and value factors. The effect appeared in each G7 country. For the United States they ruled out trading frictions, information flow and higher moments as explanations, and the co-movement of these returns across countries hints at broad factors that are hard to diversify away.

Alternatives to cap weighting

Authors: Robert Arnott, Vitali Kalesnik, Paul Moghtader and Craig Scholl, Journal of Indexes, 2010.

For more than forty years, CAPM beta and market-capitalisation-weighted indexes have been the default for core equity holdings. The authors describe challengers: fundamental weighting by the economic size of firms, minimum variance portfolios aimed at matching objectives and construction, and equal weighting. Historically these approaches have offered higher returns and lower beta. As investors lower their risk budgets, the authors ask whether a broader mix of indexes and betas could serve them better. They compare four families of quasi-passive index construction from a global angle and discuss combining several of them in one diversified portfolio.

Reading these studies carefully

Academic results describe averages over particular samples, and the samples differ in country, era and method. A strategy that looked strong in a backtest may behave differently once trading costs, taxes and changing market conditions are included. Results of this kind do not promise outcomes for any individual investor, and anyone making decisions should speak to a qualified adviser. See the glossary for definitions of terms used above.