CAPM compresses every priced risk into a single number: beta against the market. That works as a textbook benchmark; it fails as a real-world risk model. A bank's loan book has interest-rate, credit-spread, and term-structure exposures that the market beta doesn't see. The exam tests when a single factor is enough and when you need three, five, or ten.
A single-factor model assumes the only systematic risk is the market. Empirically, that is wrong. Small-cap stocks earn higher returns than CAPM predicts. Value stocks earn higher returns than growth stocks even after adjusting for beta. Bonds with similar durations but different credit ratings have different return distributions. Each pattern is a factor that CAPM ignores.
For a risk manager, the practical case is even sharper. A long-short hedge fund has near-zero market beta but huge exposures to size, momentum, and sector factors. A single-beta hedge is useless. The hedge fund needs a model with enough factors to actually capture its risk.
Common mistakes
- Treating APT as identifying specific factors. APT is silent on which factors matter: that is empirical work. A choice that says "APT requires the market portfolio" or "APT identifies size and value" is wrong. APT only requires that some pervasive factor structure exists. Trap: choices that conflate APT's logic with Fama-French's empirics.
- Confusing factor risk premium with factor return. is the expected return on a factor-mimicking portfolio in equilibrium: it is the COMPENSATION for bearing the factor risk, not the realized factor return. A negative premium (like the inflation example) means investors PAY to hold inflation-hedging assets.
- Adding CAPM and a multifactor model's predictions. A given asset has ONE expected return; pricing models give competing forecasts of it. You don't sum CAPM's prediction with Fama-French's prediction. Trap: a choice that adds CAPM expected return to a Fama-French additional premium is wrong because the Fama-French model already includes the market term.
Bottom line
- APT assumes returns are driven by multiple factors and rests on a factor model, diversification, and no arbitrage. It needs no homogeneous expectations, no mean-variance preferences, and no market portfolio.
- APT vs. CAPM: APT is multi-factor and free of CAPM's restrictive assumptions. CAPM is a special case of APT with one factor (the market).
- Multifactor model: . Expected excess return is the sum of factor betas times factor risk premiums.
- Fama-French 3-factor: market, size (SMB, small-cap minus large-cap), value (HML, high minus low book-to-market). An empirical add-on with positive US premiums that explains cross-sectional returns better than CAPM.
Exam shortcut
When a question gives factor betas and factor premiums, the answer is almost always plain arithmetic. The trap is reading the wrong sign on a premium or confusing realized factor returns with equilibrium premiums. Watch for negative factor premiums on hedging factors. Memory aid: "APT is general; CAPM is the one-factor special case." And: SMB = small wins, HML = value wins.
The full lesson (about 2,846 words, 19 min read) adds 2 worked examples, all 6 common mistakes, a self-check, free in the app.
Learning objectives
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10
- 11
Browse all free FRM Part I lessons or jump into free FRM Part I practice questions.