TL;DR
Low-volatility stocks have delivered roughly market-level returns with materially lower drawdowns since the 1970s. On a Sharpe-ratio basis, that's a massive anomaly the CAPM can't explain.
The anomaly that breaks textbook finance
The Capital Asset Pricing Model says risk and return scale linearly. The riskiest stocks (highest beta) should earn the highest returns; the safest stocks (lowest beta) should earn the lowest. This is the foundational claim of modern finance.
Then the data showed up. Andrea Frazzini and Lasse Pedersen at AQR published "Betting Against Beta" in 2014, documenting that across 23 countries and 50+ years of data, the lowest-beta stocks have consistently delivered higher Sharpe ratios than the highest-beta stocks. In some studies, they've delivered higher absolute returns too. Robert Haugen had been showing similar results for decades before, but Frazzini-Pedersen made it impossible to ignore.
This isn't a small effect. Their data shows the long-low-beta / short-high-beta portfolio earning roughly 8% annualised excess return with very high statistical significance. If markets were efficient in the CAPM sense, that couldn't exist.
The numbers
What the academic data shows for US stocks 1972–2024:
- Bottom-quintile beta stocks: ~10% annualised return, ~13% volatility, Sharpe ~0.55
- Top-quintile beta stocks: ~8% annualised return, ~28% volatility, Sharpe ~0.20
- Broad market: ~10% annualised return, ~16% volatility, Sharpe ~0.45
The low-vol portfolio matches or beats the market's return with materially less volatility. The high-vol portfolio earns less and bounces around far more. This is the opposite of what risk-return theory predicts.
The pattern holds globally. Asness, Frazzini and Pedersen's later work documented the same anomaly in European, Asian and emerging markets — usually with smaller magnitudes but consistent direction.
Why it exists
Two competing explanations:
- Behavioural overpaying for lottery tickets. Investors systematically overpay for high-beta, high-volatility stocks because they offer the chance of huge gains. The expected return on these "lottery" stocks is low because they're bid up to unsustainable levels. Low-vol stocks are correspondingly underpriced.
- Leverage constraints. Many institutional investors can't use leverage (insurance companies, pension funds with regulatory caps). Their only way to hit return targets is to buy high-beta stocks — which pushes their prices up and depresses their expected returns. Investors who can use modest leverage on low-beta stocks earn the excess.
The two explanations aren't mutually exclusive. Both probably contribute. The implication for either story is that the anomaly should persist as long as those constraints persist — which is structurally, forever.
How it fits with other factors
Low volatility overlaps with quality but isn't the same thing. Quality is defined by accounting fundamentals (profitability, earnings stability); low-vol is defined purely by price behaviour. In practice the two correlate at roughly 0.6 — high overlap but not identical.
It also overlaps with the dividend-paying segment of the market. Utilities, consumer staples and healthcare dominate low-vol portfolios. Some critics argue low-vol is really just a sector bet rather than a true factor. The defence: even within sectors, lower-vol names tend to outperform higher-vol ones, so the effect isn't pure sector exposure.
The relationship between low-vol and Buffett's lifelong outperformance is striking. The 2018 paper "Buffett's Alpha" (Frazzini, Kabiller, Pedersen) showed that Berkshire's long-run alpha is fully explained by exposure to four factors: value, profitability, low-volatility (BAB), and modest leverage. See our Buffett factor article for the full decomposition.
Practical use for FIRE planners
Low-vol is one of the most useful factor tilts for retirees specifically. Two reasons:
- Sequence-of-returns protection. Lower drawdowns in the years just before and just after FI directly reduce sequence risk. A 25% drop in year 2 of retirement at a 4% withdrawal rate is much more recoverable than a 40% drop.
- Behavioural tolerability. Low-vol portfolios are the easiest to actually hold through bear markets because they don't fall as far.
Real-world implementations:
- iShares MSCI Min Vol USA (USMV): $30bn AUM, 0.15% expense ratio. The default US low-vol ETF.
- iShares Edge MSCI World Minimum Volatility (MVOL/MINV): UCITS, 0.30%, ISA-eligible.
- Invesco S&P 500 Low Volatility (SPLV): alternative US implementation, 0.25%.
Most FIRE planners shouldn't go 100% low-vol — that gives up some equity upside in bull markets. A reasonable approach is to shift 20–30% of equity exposure into low-vol in the 5 years before retirement, tapering back to normal once you're 10 years past the FI date. Test the strategy in our factor comparison tool and see how it changes your survival curve.
Frequently asked questions
- Is low-volatility the same as defensive?
- Closely related but not identical. Low-vol is a quantitative screen on historical price volatility; defensive sector tilts (utilities, staples) often overlap but aren't the same construction.
- Should pre-retirees overweight low-vol?
- There's a case for it. Lower drawdown in the years just before and just after FI dramatically improves sequence-of-returns risk. Our [bond tents article](/blog/bond-tents-early-retirement) covers similar ground.
- Does low-vol underperform in bull markets?
- Yes, modestly. In strong bull runs like 2017 and 2021, low-vol typically lagged by 3–5% as high-beta names led. Over full cycles the gap closes because of the lower drawdowns in bear markets.
Stress-test your own FIRE plan
FIRE Wealth OS runs your savings rate and expenses against every historical market starting point since 1871. Free to use, no card required.