The Cross-Section of Volatility and Expected Returns
Source: Ang, Hodrick, Xing & Zhang (2006) · The Journal of Finance · DOI: 10.1111/j.1540-6261.2006.00836.x
TL;DR
Stocks with high idiosyncratic volatility earn abnormally low average returns — the "idiosyncratic volatility puzzle." Separately, stocks with high sensitivity to innovations in aggregate volatility also earn low returns. The IVOL result is a puzzle because standard theory predicts idiosyncratic risk should be either unpriced or positively priced.
What anomaly it documents
Two related findings:
Aggregate volatility risk: stocks that covary positively with innovations in market volatility (proxied by changes in VIX) earn lower returns — aggregate volatility risk carries a negative price (it is a hedge).
Idiosyncratic volatility puzzle (the famous one): stocks with high recent idiosyncratic volatility earn low subsequent returns, with large negative alphas relative to the Fama-French model.
Predictor: idiosyncratic volatility (IVOL) — the standard deviation of residuals from a Fama-French 3-factor regression on daily returns within the prior month.
Direction:negative — high IVOL predicts low returns.
OSAP predictor: RealizedVol (the realized/idiosyncratic volatility family).
How to construct it
Sorting variable: IVOL = std. dev. of daily residuals from a FF3 regression estimated over the prior month (the "1/1/1" convention: 1-month formation, daily data, 1-month hold).
Universe: US common stocks (1963–2000 in the paper); apply a price/liquidity filter to avoid microstructure noise.
Portfolio formation: monthly; sort into quintiles by prior-month IVOL.
Long / short to capture the anomaly: long low-IVOL, short high-IVOL (the profitable direction, since high IVOL underperforms).
Weighting: value-weighted in the headline result (the puzzle survives value-weighting, which distinguishes it from pure microcap effects).
Rebalancing: monthly.
Evidence and replication
Period
Sharpe (approx)
Ann. Alpha
T-stat
Source
IS (1963–2000), high−low IVOL
—
large negative (≈ −1%+/mo on high-IVOL quintile)
significant
this paper
OOS (post-2006)
persists
weaker but present
—
post-publication
OSAP replication
clear, negative-IVOL relation
—
—
Chen & Zimmermann 2022
High-IVOL stocks earned strikingly low returns — on the order of −1% per month or more in alpha for the highest quintile.
The puzzle is robust to value-weighting and to controls for size, value, momentum, and liquidity — which is why it generated a large follow-up literature rather than being dismissed as a microcap artifact.
It is closely related to the low-volatility / low-beta family (Frazzini-Pedersen BAB, low-vol investing).
Why it might work
The negative IVOL–return relation contradicts simple theory, so explanations focus on frictions and preferences:
Lottery preferences / MAX effect: Bali, Cakici & Whitelaw (2011) show investors overpay for stocks with lottery-like (high-skew, high-vol) payoffs, depressing their returns.
Arbitrage asymmetry + short-sale constraints: Stambaugh, Yu & Yuan (2015) argue overpricing dominates among high-IVOL stocks because it is harder to arbitrage away (shorting is constrained), so high IVOL coincides with overpricing and low future returns.
Aggregate volatility hedging: for the first result, stocks that pay off when volatility spikes are valuable hedges and thus command lower returns.
Limitations and risks
Definition sensitivity: results vary with the IVOL window (daily-in-month vs longer), the factor model used for residuals, and the holding period; some critics show the effect weakens under alternative conventions and is entangled with a one-month return reversal.
Short leg costs: capturing it requires shorting high-IVOL, often hard-to-borrow, lottery-type names.
Crowding: the low-vol family is now a large, popular strategy, compressing the premium.
No free full text: paywalled; see DOI.
Key references
Ang, A., Hodrick, R., Xing, Y. & Zhang, X. (2006) — The Cross-Section of Volatility and Expected Returns — Journal of Finance — DOI: 10.1111/j.1540-6261.2006.00836.x
Bali, T., Cakici, N. & Whitelaw, R. (2011) — Maxing Out: Stocks as Lotteries and the Cross-Section of Expected Returns — Journal of Financial Economics
Stambaugh, R., Yu, J. & Yuan, Y. (2015) — Arbitrage Asymmetry and the Idiosyncratic Volatility Puzzle — Journal of Finance
Frazzini, A. & Pedersen, L. (2014) — Betting Against Beta — Journal of Financial Economics
Chen, A. & Zimmermann, T. (2022) — Open Source Cross-Sectional Asset Pricing — Critical Finance Review
