UNRAVELLING THE BETA ANOMALY: AN ANALYSIS USING LIMITS TO ARBITRAGE, LOTTERY, AND DISTRESS CHARACTERISTICS

Abstract

Recent literature has demonstrated the risk-adjusted outperformance of low-beta securities over high-beta securities in the US and internationally. This phenomenon, which strikes at the core of the well-established Capital Asset Pricing Model, stands as an empirical puzzle as it is largely robust to a host of control variables, the time period studied, and the asset class under consideration. This paper is an attempt to unravel the low-beta anomaly along three distinct lines of well-documented research: (i) limits to arbitrage, (ii) distress risk, and (iii) lottery attributes. Amongst the three, we find lottery effects to be the dominant driver of the low-beta anomaly. Furthermore, the strength with which each lottery variable (MAX, volatility, skewness) explains the beta anomaly is positively correlated to the ease of computing the measure. Based on our finding that the beta anomaly is concentrated within stocks with high maximum daily returns (MAX) in the past month, we construct a MAX factor to quantify the role of lottery effects on the underperformance of high-beta stocks. Overall, the MAX factor is able to go further than size, value, and momentum in explaining the anomaly.