Using Fund Overweight Holdings as a Factor in Index-Enhanced Portfolios

Summary

This report examines stocks that mutual funds hold at a larger portfolio weight than their benchmark index. It describes these holdings as tending toward larger, higher-valued companies and certain sectors, and reports that many outperformed during fund holding periods. The proposed factor is binary: a stock receives a positive indicator when it is a fund holding overweight relative to the benchmark. The reported cross-sectional premium was significant in 62.2% of months, but its direction varied; among significant months, it was positive in 55.4% and negative in 44.6%.

The report finds that stronger historical fund performance was associated with better long-short returns for the holdings, while factors built from weaker-performing funds could improve index-enhancement portfolios more because premium direction persisted more reliably. It suggests suppressing the factor when its predicted direction is negative. With this timing approach and the top-performing half of funds, the report says annualized excess returns for CSI 300 or CSI 500 enhancement strategies could rise by more than 0.6%. These findings are historical and subject to factor decay, model specification errors, and changes in market patterns; portfolio holdings are also reported with a delay.

Key ideas

• A fund-overweight stock is one whose weight in fund holdings exceeds its weight in the benchmark index.
• The proposed factor assigns an indicator to overweight holdings and tests its relationship with cross-sectional stock returns.
• The reported factor premium changes direction, so the signal is not consistently positive.
• Funds with weaker past performance produced a factor whose premium direction persisted more reliably in the enhancement model.
• The report proposes setting the factor weight to zero when its predicted stock-selection direction is negative.
• The reported portfolio improvements rely on historical patterns and may not persist.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.