Read Achilles markRead Achilles

Finance

Where Did the Small-Cap Premium Go?

For a generation, the small-cap premium was the closest thing empirical finance had to a free lunch on the syllabus: small companies, over long samples, appeared to return more than their risk seemed to warrant. I…

♥ 0 Sign in to like 4 comments

This post’s retention & record terms are stated in full in the colophon below.

For a generation, the small-cap premium was the closest thing empirical finance had to a free lunch on the syllabus: small companies, over long samples, appeared to return more than their risk seemed to warrant. I taught it for years. Then, roughly when everyone had learned to order it, it stopped arriving — and the argument about why has become more instructive than the premium ever was. [author assertion]

The first interpretation is that the premium was real, and being known killed it. Publication is a tax on anomalies. Once an effect is in the textbooks, in the products, in the mandates of every fund with "small-cap tilt" in its name, capital crowds into the trade until the excess return is competed down to the cost of harvesting it. On this reading, the premium did not fail. It succeeded, publicly, and success was fatal. [author assertion]

The second interpretation is less flattering to my profession: the premium was substantially mismeasured. In the samples I have examined, the effect concentrates in the smallest and least liquid names — stocks that are cheap to count in a database and expensive to actually own — and in particular corners of the calendar. The tradable version of the premium was always thinner than the paper version, and a good deal of what my generation taught as a return was, in part, a measurement of the gap between database prices and executable ones. [author assertion]

The third interpretation is that size was never the point. Small firms are exposed — to credit conditions, to local economies, to the refinancing calendar — in ways that the standard risk models capture imperfectly. What looked like a premium for smallness may have been compensation for risks the models omit, which would explain both why it appeared and why it does not sit still to be harvested: the compensation varies with the very conditions that make it necessary. [author assertion]

A word, too, about the calendar, since students always ask. Effects in equity markets have a habit of living in particular months and dying in others, and the small-cap story was never innocent of this. When a premium can only be found by those willing to rebalance into illiquidity at the most congested moment of the year, one should ask whether the premium is a return to size — or a return to showing up, with cash, when showing up is hardest. The two hypotheses have identical charts and very different morals. [author assertion]

My retired verdict is that all three are true, in proportions we will never estimate cleanly, because the data-generating process refuses to hold still while we estimate it. Markets learn, rules change, the composition of "small" changes. We are fitting lines through the wake of a moving ship and arguing about the ship. [author assertion]

Which is the general lesson I would leave with students, and it extends well beyond this one anomaly. Every empirical regularity in finance is a photograph of a market at a particular moment in its institutional development. Expect decay as the pattern becomes known. Expect measurement to flatter the pattern. And do not build a retirement — or a syllabus — on the assumption that either force will politely refrain. [author assertion]

Anderson Smith

All posts by Anderson Smith → · Follow Anderson Smith by RSS

The Board 20 essays 0 corrections Agent citations: —— — citation attached ≠ claim validated As of 16 November 2026 Full board →

Provenance

About this version

How this post was verified, labelled and shared — in plain language first. The complete machine-readable record is kept below, word for word, for anyone who wants to check it.

Version
1 · in good standing
Author
Anderson Smith · verified at publication
Published
6 Nov 2026
Claims
7 labelled — 7 author assertion
Citations
No external citations
Distribution
Shared with the author’s consent — including to agents acting for readers, under the author’s per-post opt-in
Content fingerprint
9d895813…2736d3
Full verification record

The verbatim graded record of this version: identifiers, claim classes, citation targets, hashes and standing, exactly as issued.

This record is also published in structured text for agents and tools — the same fields, word for word. View structured text

House editorial — Read AchillesAuthor verification: verifiedStanding: governingHash: 9d89581319d03ede0ae88e70ed9dfee7349b5ffa51fd3667e65cf9e4592736d3

Version & standing

Post post-where-did-the-small-cap-premium-go-58ce86 · version 1 (post-where-did-the-small-cap-premium-go-58ce86-v1)

@anderson-smith · verified state at publish: verified

Published: 2026-11-06T23:46:06.530Z · Standing: governing

Claims & citations

  • c-1: author_assertion
  • c-2: author_assertion
  • c-3: author_assertion
  • c-4: author_assertion
  • c-5: author_assertion
  • c-6: author_assertion
  • c-7: author_assertion

    Disclosure & trust

    When your words reach the end of their retention period they are destroyed, on the schedule above. What the platform keeps after that — indefinitely, with no expiry — is a skeleton of acts: that an act occurred, its class and time, the bundle and section it touched, its change class, a hash that identifies the record without containing it, and the keys that link a correction to what it corrects. The skeleton is not your words and cannot reproduce them. Your elections to shorten a period, to ask ephemerally, or to skeletonise now govern your words; they do not shorten, and cannot remove, the skeleton — it is the record that lets the platform prove what it did with your work after the words are gone. The platform will never produce your skeleton by choice against you; like any record it can be reached by a court order the platform cannot refuse, and if that happens you will be told wherever the law allows and the production will state, on its face, that it is a skeleton and that the words expired on this schedule.

    When your words reach the end of their retention period they are destroyed, on the schedule above. What the platform keeps after that — indefinitely, with no expiry — is a skeleton of acts: that an act occurred, its class and time, the bundle and section it touched, its change class, a hash that identifies the record without containing it, and the keys that link a correction to what it corrects. The skeleton is not your words and cannot reproduce them. Your elections to shorten a period, to ask ephemerally, or to skeletonise now govern your words; they do not shorten, and cannot remove, the skeleton — it is the record that lets the platform prove what it did with your work after the words are gone. The platform will never produce your skeleton by choice against you; like any record it can be reached by a court order the platform cannot refuse, and if that happens you will be told wherever the law allows and the production will state, on its face, that it is a skeleton and that the words expired on this schedule.

    Integrity & distribution

    Content hash: 9d89581319d03ede0ae88e70ed9dfee7349b5ffa51fd3667e65cf9e4592736d3

    Distribution: opted_in

    Canonical presentation: https://blog.readachilles.com/post-where-did-the-small-cap-premium-go-58ce86/

    What we retain, and why — the retention disclosure

    The retention disclosure for this version is the disclosure line of record, stated in full in the colophon fields above.