Quant Evidence Atlas
Strategy dossier · Evidence B

Betting Against Beta

Leverage-constrained investors may overpay for high-beta assets, leaving lower risk-adjusted returns than in levered low-beta portfolios.

Evidence BRetail feasibility CMonthly

The research question

The hypothesis is economically coherent and documented across markets. Reproducing the academic portfolio, however, requires leverage, shorting and careful beta estimation.

The useful question is not whether a chart once went up. It is whether the hypothesis has a clear economic mechanism, appears outside one hand-picked sample, can be reconstructed with information that was available at the time, and remains plausible after realistic implementation frictions.

Minimum viable rule

Build beta-ranked low-risk and high-risk portfolios, lever the low-beta side and de-lever the high-beta side toward beta neutrality.

Universe: Equities and other liquid asset classes with financing and shorting access. A credible test fixes the universe, timestamps, missing-data policy and rebalance convention before model selection. Results should be shown both gross and net of an explicit cost model.

Implementation audit

Disclose the beta window, shrinkage, volatility adjustment, weighting, leverage cap, financing rate and borrow cost. Small-stock exposure and stale betas should be tested explicitly.

At minimum, an audit should report turnover, worst peak-to-trough loss, recovery time, exposure concentration and sensitivity to neighboring parameter choices. A strategy that works only at one exact lookback or threshold deserves a lower level of confidence.

How the idea can fail

Funding costs can rise exactly when leverage is least available. Short squeezes, beta instability, deleveraging and construction choices can dominate the theoretical premium.

Failure conditions belong in the strategy definition. They provide a disciplined reason to investigate or stop, instead of changing the story after a loss.

Primary source

Betting Against Beta
Frazzini & Pedersen (2014), Journal of Financial Economics 111(1), 1–25.
DOI record

This page is an independent educational synthesis. It does not reproduce the paper and does not claim that published historical results are currently achievable.