Betting Against Beta (BAB) is a strategy that exploits the empirically flat (or inverted) security market line by going long leveraged low-beta assets and short high-beta assets. The concept and factor were formalized by Frazzini and Pedersen (2014).
The anomaly
The capm predicts that expected returns increase linearly with beta. Empirically, the security market line is too flat: high-beta assets earn lower risk-adjusted returns than the CAPM predicts, and low-beta assets earn higher risk-adjusted returns. This has been documented since Black, Jensen, and Scholes (1972).
Theoretical explanation: leverage constraints
Many investors (mutual funds, pension funds, individuals) cannot use leverage freely. To reach higher expected returns, they tilt toward high-beta assets instead of leveraging the tangency portfolio. This excess demand for high-beta assets bids up their prices and compresses their expected returns. Low-beta assets, which require leverage to achieve competitive returns, are neglected and earn a premium.
The model generates five testable predictions, all confirmed empirically:
- High beta is associated with low alpha (flat SML)
- A BAB factor produces significant positive risk-adjusted returns
- BAB returns are low when funding constraints tighten
- Increased funding liquidity risk compresses betas toward one
- More constrained investors hold riskier assets
BAB factor construction
- Rank all stocks by estimated beta
- Go long the low-beta half (leveraged to beta 1) and short the high-beta half (de-leveraged to beta 1)
- The portfolio is dollar-neutral and beta-neutral by construction
The resulting factor earns significant positive returns across: US equities, 20 international equity markets, Treasury bonds, corporate bonds, and futures.
Relationship to other factors
BAB is related to but distinct from residual-volatility. The low-volatility anomaly (Ang et al. 2006) documents that high idiosyncratic volatility stocks earn low returns. BAB focuses specifically on systematic risk (beta) rather than total or idiosyncratic volatility, and provides a leverage-constraint theory rather than appealing to behavioral biases.
The FF5 model partially absorbs the low-beta anomaly through RMW and CMA loadings, as low-beta stocks tend to be profitable conservative investors. However, BAB retains significant alpha after controlling for FF5 factors.
Practical significance
BAB was among the first factors documented across multiple asset classes (not just equities), strengthening the case that leverage constraints are a fundamental friction in financial markets rather than a statistical artifact. AQR incorporates BAB insights into its multi-strategy approach.