Mutual Fund Performance: An Empirical Decomposition into Stock-Picking Talent, Style, Transactions Costs, and Expenses

Source: Wermers, R. (2000). Journal of Finance 55(4), 1655–1703.

TL;DR

Reconciles two conflicting literatures — holdings-based studies that find stock-picking skill and

net-return studies that find underperformance — by decomposing fund performance using **both quarterly

holdings and monthly net returns. Funds hold stocks that beat the market by 1.3% per year**, yet

their net returns underperform by ~1% per year. Of the 2.3% gap: 0.7% is the drag from

non-stock (cash) holdings and 1.6% is expenses plus transaction costs. Conclusion: funds pick

stocks well enough to cover their costs, supporting the value of active management. Sample: 1975–1994.

The idea

Prior work disagreed because it looked at different things: studies of fund holdings found managers

buy outperforming stocks, while studies of net returns to investors found underperformance versus

passive indexes (e.g., Gruber 1996: ~65 bps/yr shortfall, 1985–1994). Wermers merges a stock-holdings

database with a net-returns database for the same funds, so the full chain — from the stocks chosen,

through style tilts, trading costs, expenses, and cash drag, to the investor's net return — can be

decomposed within one sample using characteristic-based (size/BM/momentum) benchmarks.

Evidence

Gross stock holdings outperform the market by ~1.3%/yr. This splits into:

- ~71 bps/yr characteristic selectivity (stock-picking skill in excess of the manager's style).

- ~55–60 bps/yr from the characteristics of the stocks held (the style tilt itself).

Net returns underperform by ~1%/yr. The 2.3-percentage-point gap between the +1.3% gross

holdings result and the −1% net result decomposes as:

- 0.7 pp — underperformance of non-stock (cash/bond) holdings, i.e., cash drag.

- 1.6 pp — expenses and transaction costs combined.

Industry-average expense ratio ≈ 100 bps/yr vs ≈ 20 bps/yr for the Vanguard Index 500.

High-turnover funds hold better stocks and, notably, beat the Vanguard Index 500 on a net-return

basis — the trading appears to pay for itself.

Why it matters

Resolves the apparent contradiction between holdings and net-return studies: both are correct, and the

difference is costs and cash drag. It established that the average fund possesses genuine, measurable

stock-selection talent at the holdings level — a key distinction between gross skill (in the

portfolio) and net delivery (to the investor), and a foundation for characteristic-based

performance attribution.

Caveats

Holdings are observed only quarterly/semiannually, so trading costs are estimated, not observed.

Results are gross-of-tax and rely on characteristic benchmarks that may not span all priced risk.

"Skill covers costs" is an average statement; it does not imply investors capture positive net alpha.

Key references

Wermers, R. (2000) — Mutual Fund Performance: An Empirical Decomposition — Journal of Finance

Carhart, M. (1997) — On Persistence in Mutual Fund Performance — Journal of Finance

Daniel, K., Grinblatt, M., Titman, S. & Wermers, R. (1997) — Measuring Mutual Fund Performance with Characteristic-Based Benchmarks — Journal of Finance

Provenance: verified/generated from the paper's full text.