McCombs Professor Shows that Affiliated Funds Bias Affects Participant Accounts
Clemens Sialm, a professor at the McCombs School of Business at the University of Texas at Austin is an author on a Pension Research Council working paper that suggests that mutual fund companies that act as 401(k) plan service providers display significant favoritism toward affiliated funds. Sialm coauthored the paper with Veronika K. Pool from Indiana University, Bloomington, and Irina Stefanescu from the Board of Governors of the Federal Reserve System. The researchers collected information about mutual fund options menus from a large sample of 401(k) plans from 1998 to 2009, using information from firms’ annual Form 11-K filings with the Securities and Exchange Commission.
Sialm, Pool, and Stefanescu found that affiliated plans are more likely to be added to 401(k) plan fund menus and less likely to be removed than unaffiliated plans. The researchers found the biggest relative difference in the treatment of affiliated and unaffiliated funds in the worst performing funds. Mutual funds in the lowest decile of performance (based on their performance over the previous three years), are deleted at a rate of 25.5% per year if they are not affiliated with the plan’s trustee, but they are only deleted at a rate of 13.7% if the fund is affiliated with the trustee.
The researchers write: “Protecting poorly-performing funds by keeping them on the menu helps mutual fund families to dampen the outflow of capital triggered by poor performance and, as a result, mitigates fund distress.”
The researchers tried to determine the impact of menu favoritism on overall allocation of plan assets by looking at the sensitivity of plan participants to the performance of affiliated and unaffiliated funds. Sialm, Pool, and Stefanescu found that participants in their study were usually not sensitive to poor-performing mutual funds, and their selections did not mitigate the impact of the menu bias toward affiliated mutual funds. This suggests that retirement plan participants are affected by affiliation bias.
