A pension fund's "value" manager beat its benchmark by 3% last year. Skill or factor tilt? If a basic value index returned 4% above the benchmark over the same period, the manager actually subtracted 1% of value-add. The exam tests whether you can split returns into factor exposures and the residual that is genuine alpha.
Capital asset pricing model (CAPM) says the market is the only priced risk. Empirically that fails. Small caps earn more than their beta predicts. Value stocks earn more than growth. Low-volatility stocks earn more than high-volatility ones, a direct contradiction of CAPM. Each anomaly is a factor that CAPM does not see.
A factor is a pervasive driver of returns. It needs three properties: many assets load on it, the loading is measurable, and the loading earns a premium that does not diversify away. If a tilt is exam-relevant only for one asset class, it is not a factor; it is a sector bet.
KEY: The CAPM is a one-factor model where the single factor is the market. Multifactor models nest CAPM as a special case and add factors that empirically explain returns CAPM...
Common mistakes
- Treating factor premiums as alpha. A value manager earning 4% extra return when the value premium itself was 4% has zero alpha. Trap: the question states "Alpha is 4%" and the candidate accepts it without checking the factor decomposition. The right answer requires regression against the appropriate multi-factor benchmark.
- Using the wrong benchmark. Benchmarking a small-cap value manager to the S&P 500 inflates alpha. The S&P 500 has neither the size nor value tilt the manager carries. Trap: a "10% alpha" against an all-cap blend benchmark collapses to 1% against a small-cap value benchmark. Choose the benchmark first, then measure.
- Ignoring the no-trade region. A manager whose target weight is 5.05% versus a current weight of 5.00% does not need to trade if transaction costs exceed the alpha pickup. Trap: candidates assume rebalancing-to-target is always optimal. Exam questions give a small drift and a high cost. The right answer is "do not trade."
Bottom line
- Factors are pervasive sources of return that price every asset; the market is one factor, alongside priced value, size, momentum, low-volatility, and quality.
- Multifactor model: , where each is the premium on a factor-mimicking portfolio.
- Factor premiums are not alpha: alpha is the residual after subtracting every factor contribution, so a 4% return matched by a 4% factor premium leaves zero alpha.
- Benchmark choice is decisive; alpha against the wrong benchmark is mostly a style premium, judged by seven effective-benchmark properties.
Exam shortcut
When the question gives a manager's return AND a list of factor returns and loadings, decompose: subtract each factor contribution from the return; what remains is alpha. If the question says "the manager has skill" and shows a 3% gross alpha that fully decomposes into size and value premiums, the right answer is "no skill: the return is factor exposure." Memory aid: "Alpha is what is left after the factors...
The full lesson (about 3,411 words, 23 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
- 12
- 13
- 14
- 15
- 16
- 17
Browse all free FRM Part II lessons or jump into free FRM Part II practice questions.