Source: Carhart, M. M. (1997) · The Journal of Finance 52(1), 57–82 · DOI: 10.1111/j.1540-6261.1997.tb03808.x
TL;DR
Introduces the four-factor model — Fama-French three factors plus a one-year momentum factor (PR1YR/UMD) — and uses it on a survivor-bias-free sample of equity funds (1962–1993) to show that apparent persistence in mutual-fund performance is almost entirely explained by momentum exposure and costs, not skill. Funds holding last year's winners look good until you adjust for the momentum factor; the only persistence the model cannot explain is the strong, persistent underperformance of the worst funds, driven by expenses.
What anomaly it documents
Predictor: a fund's prior one-year return predicts its near-term return — the "hot hands" effect of Hendricks, Patel & Zeckhauser (1993).
Direction / mechanism: the predictability is mechanical exposure to the Jegadeesh-Titman one-year stock-momentum factor, not stock-picking skill. Funds that did well were accidentally holding momentum winners; individual funds that actually chase momentum earn lower net returns because transaction costs consume the gains.
OSAP predictor: momentum (this paper defines the fourth, momentum, factor used throughout performance evaluation).
How to construct it
The four-factor model adds a momentum factor to market, SMB, HML:
PR1YR = equal-weight average of firms in the top 30% of eleven-month returns lagged one month minus the bottom 30%, over all NYSE/Amex/Nasdaq stocks, re-formed monthly. (SMB and HML are taken directly from Fama-French.)
For the fund test: sort funds each January 1 (1963–1993) into deciles on the prior calendar-year return, equal-weight monthly, then regress decile returns on the four factors; attribute the spread to factor loadings, expenses, and turnover rather than to alpha.
Evidence and replication
Result
Magnitude
Source
Decile 1−10 spread, raw monthly excess return
≈67 bp/month (annualized ≈8%); 24% in the ranking year
this paper
of which momentum (PR1YR) explains
≈31 bp/month — almost half
this paper
CAPM alpha, top vs. bottom decile
+22 bp/mo (2.6%/yr) vs. −45 bp/mo (−5.4%/yr)
this paper
4-factor alpha residual spread
≈28 bp/mo, of which 20 bp is just decile 9 vs. 10
this paper
Decile-1 best funds' 4-factor alpha
essentially zero / does not cover costs
this paper
Key points:
The CAPM cannot explain the decile spread (betas nearly identical across deciles), but the four-factor model absorbs most of it, driven by SMB and especially the pronounced PR1YR loading pattern (top deciles load positively, bottom deciles negatively).
Costs matter one-for-one: expenses reduce performance roughly dollar-for-dollar, turnover hurts (≈0.95% of trades), and load funds underperform no-loads by ≈80 bp/year holding expenses constant.
The only unexplained persistence is strong underperformance by the worst-return funds — the bottom thirtieth still underperforms T-bills by ≈25 bp/month the following year.
Out of sample, the broad conclusion — most fund "skill" is factor exposure and luck — was reinforced by Fama & French (2010).
Why it might work
Mechanical / factor exposure: "hot hands" is the momentum anomaly leaking through fund holdings, not manager skill.
Costs as a persistent drag: expense and transaction-cost differences are sticky, producing genuine persistent underperformance among high-cost funds.
Momentum itself remains theoretically unexplained (behavioral underreaction vs. risk).
Limitations and risks
Attribution depends on the factor model; momentum is not grounded in a risk theory.
Measuring whether a fund actually follows momentum is imperfect in the sample.
Fund data must be handled carefully for survivorship and look-ahead (this paper's sample is explicitly survivor-bias-free).
Key references
Carhart, M. (1997) — On Persistence in Mutual Fund Performance — Journal of Finance
Jegadeesh, N. & Titman, S. (1993) — Returns to Buying Winners and Selling Losers — Journal of Finance
Fama, E. & French, K. (1993) — Common Risk Factors in the Returns on Stocks and Bonds — Journal of Financial Economics
Fama, E. & French, K. (2010) — Luck versus Skill in the Cross-Section of Mutual Fund Returns — Journal of Finance
Provenance: verified/generated from the paper's full text.
