- 1Adding momentum to the Fama-French 3-factor model creates the "Carhart 4-factor model"
- 2Most mutual fund "alpha" disappears when momentum exposure is included
- 3Fund performance persists—but only for the worst funds (losers keep losing)
- 4Expenses and transaction costs are the best predictors of future fund returns
- 5This paper made momentum a standard factor in academic finance
#The Paper at a Glance
Title: On persistence in mutual fund performance
Authors: Mark M. Carhart
Published: Journal of Finance, 1997
DOI: 10.1111/j.1540-6261.1997.tb03808.x
Carhart's paper served two purposes: it introduced the four-factor model (adding momentum to Fama-French) and used it to demolish the myth of persistent mutual fund outperformance.
#The Four-Factor Model
Before Carhart: The Fama-French 3-Factor Model
The standard model for evaluating performance included:
- 1Market (MKT): The overall stock market return
- 2Size (SMB): Small stocks minus big stocks
- 3Value (HML): High book-to-market minus low book-to-market
What Carhart Added
Carhart added a fourth factor:
- 1Momentum (UMD): Up minus down—past winners minus past losers
This addition was motivated by Jegadeesh and Titman's (1993) discovery. Many funds that appeared to outperform the 3-factor model were simply holding momentum stocks.
| Model | Factors | Fund "Alpha" |
|---|---|---|
| CAPM (1-factor) | Market only | Many funds show alpha |
| Fama-French (3-factor) | Market + Size + Value | Fewer funds show alpha |
| Carhart (4-factor) | Market + Size + Value + Momentum | Almost no funds show alpha |
#What Happened to Fund Performance
The Disappearance of Alpha
Carhart studied 1,892 U.S. equity mutual funds from 1962-1993. His findings were devastating for the active management industry:
- Before expenses: The average fund matches the market (zero alpha)
- After expenses: The average fund underperforms by roughly 1.0% per year
- Top performers: Most past winners owed their returns to momentum exposure, not skill
Performance Persistence
Previous researchers found evidence that top-performing funds continued to outperform. Carhart showed this was mostly a momentum effect:
- Funds that bought recent winners (momentum stocks) outperformed when momentum worked
- Once you account for momentum factor exposure, the persistence largely disappears
- Exception: The worst-performing funds DO persistently underperform. Bad funds stay bad.
What Actually Predicts Fund Returns
| Predictor | Predictive Power |
|---|---|
| Past returns | Weak (mostly factor exposure) |
| Expense ratio | Strong (negative) |
| Turnover | Moderate (negative) |
| Fund size | Moderate (negative) |
| Manager tenure | Weak |
The best predictor of future fund returns? Low fees. Not past performance, not manager pedigree, not Morningstar stars.
#Why This Paper Matters
1. It Made Momentum "Official"
Before Carhart, momentum was a curious anomaly. After Carhart, it became a standard factor used by academics and practitioners worldwide. No serious factor model today omits momentum.
2. It Advanced the Case for Passive Investing
If most fund "alpha" is really just factor exposure, investors can get the same exposures cheaper through factor-based strategies or index funds. This paper helped fuel the passive investing revolution.
3. It Changed How We Evaluate Performance
Today, any serious performance evaluation uses at least four factors. Claiming "alpha" without adjusting for momentum exposure is considered incomplete.
#The Active Management Paradox
Carhart's findings create a paradox for active managers:
- 1The average fund can't beat the market after fees (it's a zero-sum game minus costs)
- 2The best funds' outperformance is largely explained by systematic factor tilts
- 3Those factor tilts can be achieved at a fraction of the cost through quantitative strategies
This doesn't mean all active management is worthless—but it does mean investors should be skeptical of claimed "alpha" and check whether it's really just factor exposure.
#How This Applies to Our Rankings
Our ranking system embodies Carhart's key insight: factors explain most of the variation in stock returns. Rather than paying a fund manager 1% per year to (unknowingly) tilt toward momentum stocks, investors can use systematic factor scores to build their own factor-tilted portfolios.
Our six factors go beyond Carhart's four, adding profitability, low volatility, investment quality, and short interest. But the core principle is the same: systematic factor exposure, transparently applied, at low cost.
Explore our factor-based rankings →
#Academic Source
Carhart, M. M. (1997). "On persistence in mutual fund performance." Journal of Finance, 52(1), 57-82.
Last updated: February 1, 2026