Tests of the CAPM from the 1970s onward found patterns it could not explain.
Stephen Ross (1976) proposed that returns are driven by several systematic factors, not one. APT does not say which factors; it says that an asset's expected return is the risk-free rate plus the sum of each factor exposure times that factor's premium.
SMB (small minus big) is the return of small stocks minus large stocks. HML (high minus low) is the return of high book-to-market (value) stocks minus low book-to-market (growth) stocks. The Carhart model adds momentum; the Fama-French five-factor model adds profitability (RMW) and investment (CMA).
A fund has loadings of b = 1.1, s = 0.4 and h = minus 0.2. Expected premiums are market 5%, SMB 2% and HML 3%.
With a 3% risk-free rate, expected return is 8.7%. The loadings say the fund tilts toward small companies and slightly toward growth.
Two explanations compete. The risk view: small and value companies are riskier in ways beta misses (distress risk, sensitivity to recessions), so the premium is fair compensation. The behavioural view: investors make systematic errors, such as overpaying for exciting growth stories and underreacting to news, and limits to arbitrage stop professionals from correcting them.
Factor or smart beta ETFs package these tilts at low cost. Cautions:
A manager who beats the market by holding small value stocks has not shown skill, only factor exposure that an ETF could deliver cheaply. True alpha is what remains after controlling for all the factors.
A value ETF has loadings of market 0.95, SMB 0.1 and HML 0.6. With expected premiums of 5%, 2% and 3%:
With a 3% risk-free rate, the expected return is 9.75%. If the ETF charges 0.25% and a broad index fund 0.05%, the value tilt must earn at least 0.2 points a year more to pay for itself.
Momentum strategies buy recent winners and short recent losers. In sharp market rebounds after crashes, recent losers (often the most beaten-down, highest-beta stocks) surge, and momentum suffers severe losses. The rebound of spring 2009 produced one of the worst months in the factor's history. Factors carry their own tail risks.
Quality (profitable, stable, low-debt companies), low volatility (stocks with low past volatility or beta) and investment (CMA, conservative minus aggressive asset growth) are common in practice. Many are correlated with each other: low-volatility stocks tend to be high quality and value-tilted.
McLean and Pontiff (2016) studied nearly 100 published return patterns. Their returns were about a quarter lower out of the original sample and roughly 60% lower after publication, consistent with a mix of data mining and investors trading the effect away. Expect smaller premiums than backtests show.
Tilting toward factors that look cheap or have recently done well sounds appealing, but evidence that investors can time factors profitably is weak. A steady, diversified factor allocation held through long underperformance is more realistic, and harder to stick with.
A factor regression whose alpha has a t-statistic of 0.6 shows no evidence of skill after accounting for factor exposures. If the fund charges active fees, the investor is paying active prices for exposures available cheaply in factor ETFs.
Look up the fact sheet of one value ETF, one small-cap ETF and one momentum ETF. Record each one's fee, number of holdings and 10-year return against the broad market. Note in which years each one lagged.
Educational content only; not investment, tax or legal advice. Companies, tickers and figures are illustrative examples, not recommendations. Rules, rates and limits change, so check current details with the regulator, exchange or tax authority before relying on them. Worked examples are not forecasts. Written by Elizabeta Dimoska. RiskStock is not a registered dealer or adviser.