Alpha
Alpha is the part of an investment's return not explained by its market — what a manager added, or lost, by choosing differently from the index, after adjusting for the risk taken. A fund that returned 12% when its market rose 10% at a beta of 1.0 earned 2% of alpha; at a beta of 1.5 it earned negative alpha, because the market alone would have given 15%. It is what active management sells and mostly fails to deliver.
- Also known as
- Excess return, Manager alpha, Jensen's alpha
- Used at
- manager evaluation, fund marketing, hedge fund fee justification ("we charge for alpha"), and the argument between active and passive investing
- Binding
- no — a statistic from a regression of the fund's returns on the market's (Jensen's alpha, 1968)
- Typical evidence
- the average active fund's alpha after fees is negative; the S&P SPIVA scorecards find 85–90% of US large-cap active funds underperform over fifteen years; persistence of alpha among the winners is weak
- Typical fee logic
- index funds charge for beta at a few basis points; active managers and hedge funds charge 1–2% plus performance fees for alpha; when the alpha is not there the fee is the only certain outcome
- Parties
- fund managers, the investors who pay for it, the consultants who try to identify it in advance, and the academics who keep finding little of it
How alpha works in practice
A manager runs a fintech equity fund. Over five years it returned 11% a year; the sector index returned 9%. Did she add value? First, the beta: her fund was more volatile than the index — a beta of 1.3 — so the market's 9% should have given her 11.7% for the risk taken. Her alpha is −0.7% a year: she earned less than her risk deserved, and her investors paid a fee for the privilege. Had her beta been 0.8, the market would have explained only 7.2% of her return and her alpha would have been 3.8% — genuine skill, or luck, distinguishable only over a longer period than most investors wait.
The arithmetic explains the industry's shape. Beta is cheap — an index fund delivers the market's return for almost nothing — so active managers are paid only for what they add beyond it, and the evidence is that few add anything after fees and fewer do so persistently. Money moved to index funds; the managers who kept assets were those with a demonstrable alpha — some quantitative funds, some concentrated stock-pickers, private-markets managers whose alpha is harder to measure and easier to claim.
For fintech companies the relevant alpha is the wealth platforms': a robo-adviser sells beta at 0.25% and does not pretend otherwise; a wealth manager at 1% has to explain what the other 0.75% buys.
Example
A pension fund reviews two managers over ten years against the same global equity index, which returned 8% a year. Manager A: 9.5% a year, beta 1.2 — the market explained 9.6%; alpha −0.1% before a 0.8% fee: −0.9% net. Manager B: 8.7% a year, beta 0.85 — the market explained 6.8%; alpha 1.9% before a 0.9% fee: 1.0% net. The pension's board had preferred A for years on the headline; the review moves €200m from A to an index fund at 0.05% and keeps B, whose lower beta and positive alpha were invisible in a table of returns. Over the next decade A's 9.5% and B's 8.7% both turn out to have been the previous decade's numbers, which is the other lesson.
Common mistakes
- Judging a manager on return without beta — high beta in a rising market looks like skill. - Alpha measured against the wrong benchmark, which manufactures it. - Paying active fees for beta. - Reading five years of alpha as skill when the persistence data say it may be noise. - Confusing alpha with total return in a pitch, which is where most of the confusion is produced.