Where Buffett's Investment Experience Meets Multifactor Research

Multifactor studies link valuation and profitability to historical average returns. These findings overlap with Buffett’s discussions of business quality and purchase price, although the methods and samples differ.

Source dates and applicability

The main papers were published from 1992 to 2015. Carhart studies funds from 1962 to 1993; the momentum and profitability studies use separate historical samples. Shareholder letters are named for the reporting year and generally issued the following year. As of October 2026, these sources support a comparison of investment ideas, without a new test of recent returns or current implementation costs.

Many of their findings echo Buffett’s experience. Here are four of the most representative points of overlap.

1. Novy-Marx’s Finding: Both Cheap and Good

One of Novy-Marx’s central findings is that companies with high gross profitability (gross profits divided by total assets) deliver higher average returns even at relatively higher valuations. Controlling for profitability also substantially improves the performance of value strategies.

In his 1979 shareholder letter, Buffett noted that a poor business bought cheaply is inferior to a good business bought at a reasonable price. In his 1982 shareholder letter, discussing acquisitions paid for with the buyer’s undervalued shares, he explained that even a fair price for the target can cost too much in actual value, turning a marvelous business into a terrible buy.

2. The Risks of Indiscriminate Expansion

CMA, the investment factor in the Fama-French five-factor model, indicates that companies investing more aggressively have lower average returns after controlling for other factors. The paper also identifies a particularly poor-performing group: small companies with high investment and low profitability, which it calls lethal portfolios.

Berkshire is a diversified holding company, and Buffett has repeatedly emphasized that capital allocation should focus on capital efficiency rather than simply pursuing scale. In his 2014 shareholder letter, for example, he used the LTV conglomerate as a cautionary example.

3. Frequent Trading Costs Significantly Drag on Performance

Carhart’s results are straightforward: expense ratios, turnover, and sales charges, or loads, all have significant associations with poorer risk-adjusted performance. The regression coefficients suggest that investors receive essentially no corresponding benefit for the costs they bear.

Table V reports coefficients of about -1.54 for expenses and -0.95 for turnover. A one-percentage-point increase in the expense ratio corresponds to about 1.54 percentage points less annual abnormal return. Turnover requires attention to the table’s scaling; the author interprets its result as roughly 0.95% net cost per round trip. The two coefficients require their respective units.

In his 1983 shareholder letter, Buffett called excessive market activity a pickpocket of business. He also worked through the arithmetic: if shares turn over 100% in a year, with a 1% commission charged on both purchases and sales, shareholders pay 2% of the company’s net worth each year for that transfer. That consumes one-sixth of a company’s 12% ROE. It is an expensive game of musical chairs and a form of self-imposed taxation.

4. The Main Disagreement: Short-Term Investment Opportunities

One major difference between Buffett and these quantitative studies concerns short-term momentum. Jegadeesh and Titman’s research shows that buying past winners and selling past losers produces significantly positive returns over holding periods of 3 to 12 months. The most successful 12-month/3-month strategy earns approximately 1.31% per month, with results affected by whether the most recent week is skipped.

This comparison concerns Buffett’s public discussions of business analysis and the papers’ short-term momentum signals. They use different information and holding periods; these papers alone do not establish which approach performs better in the current market.

5. References

1. Eugene F. Fama and Kenneth R. French (1992), The Cross-Section of Expected Stock Returns, The Journal of Finance 47(2): 427–465. Background on size, book-to-market, and average returns.
2. Eugene F. Fama and Kenneth R. French (1993), Common Risk Factors in the Returns on Stocks and Bonds, Journal of Financial Economics 33(1): 3–56. Background on the three-factor stock model; this is a separate work from the 1992 paper above.
3. Narasimhan Jegadeesh and Sheridan Titman (1993), Returns to Buying Winners and Selling Losers: Implications for Stock Market Efficiency, The Journal of Finance 48(1): 65–91. Section II and Table I support section 4: the 12-month formation/3-month holding strategy earns 1.31% monthly without a gap and 1.49% with a one-week gap.
4. Mark M. Carhart (1997), On Persistence in Mutual Fund Performance, The Journal of Finance 52(1): 57–82. Table V supports section 3’s expense, turnover, and load regressions, including the -1.54 and -0.95 coefficients.
5. Robert Novy-Marx (2013), The Other Side of Value: The Gross Profitability Premium, Journal of Financial Economics 108(1): 1–28; see also NBER Working Paper 15940 (2010). Supports section 1; gross profitability is gross profits divided by total assets, rather than the conventional gross margin.
6. Eugene F. Fama and Kenneth R. French (2015), A Five-Factor Asset Pricing Model, Journal of Financial Economics 116(1): 1–22. The CMA definition and Table XI support section 2’s investment factor and low-profitability, high-investment small-stock portfolios.
7. Warren E. Buffett, Berkshire Hathaway shareholder letters for 1979, 1982, 1983, and 2014. These support, respectively, good versus cheap businesses, acquisitions paid for with undervalued shares, the turnover-cost example, and LTV’s capital allocation mistakes.