That is the finding of Lauren H. Cohen, a professor at the Harvard Business School, and his co-author, Breno Schmidt at the University of California Marshall School of Business, who just released their study, Attracting Flows by Attracting Big Clients: Conflicts of Interest and Mutual Fund Portfolio Choice.   They look at instances where corporations (plan sponsors) use a mutual fund company as the trustee of their 401(k) plan, while at the same time using that fund company’s mutual funds as investment options in the 401(k) plan.  Cohen and Schmidt’s study is the first to look at the economic impact of this obvious conflict of interest.

Fund companies can potentially gain large inflows by having their funds included in 401(k) plans.  As the trustee shares fiduciary responsibility over the plan investments and ends up having a sizable number of its funds included in the plan, mutual fund families may find it valuable to secure trustee relationships.

Collectively, fund assets are estimated to be overweighted by approximately $24 billion dollars across the mutual fund industry.

Methodology

Cohen and Schmidt began by identifying situations where public corporations have mutual fund companies as their 401(k) plan trustee.  Although this data is publicly available, it is not provided through any of the commercial data services.  The authors used Form 11-K documents (filed with the SEC) and Form 5500 documents (filed with the Department of Labor) to obtain 401(k) plan data, including the identity of the trustee and the plan assets.  They identified 899 companies with fund companies as their plan trustee, with an average plan size of $533 million.  The data used was from 1993-2003.

The average revenue (based on expense ratios) to fund companies from these plans is $4.2 million.  By contrast, the fund companies earn an average of $150,000 from their position as a trustee.  The large difference between these two sources of income suggests the primary motivation to become a trustee may be to earn revenue from asset management, not from the trustee relationship.

In their sample, a total of 251 fund companies served as trustees of at least one plan, representing approximately 80% of the assets in the mutual fund industry.  There were 1,929 individual funds holding plan assets.

The critical step was to identify whether the fund company overweighted the holdings of the plan sponsor’s stock.  The authors looked at the individual holdings of each fund, using the fund’s quarterly reports, and calculated the percentage of shares outstanding.  They determined the percentage of shares outstanding the fund family (or individual mutual fund) was holding of every stock. 

Using a regression framework, they controlled for stock and fund specific characteristics (e.g., company size, book-to-market ratio, Total Net Assets, momentum, and industry), providing an estimate of how much larger of a percentage of shares outstanding the trustee holds, as compared to the average fund family. 

Their estimate is a 47% overweighting.  Thus, if the average fund holds 10% of shares outstanding, the trustee holds 14.7% of shares outstanding of the sponsor company's assets.

Significant Findings

The authors investigated the causes and implications of fund companies overweighting the plan sponsor’s stock.  Among these findings are the following:

• Overweighting is pervasive across the fund industry, and it is more pronounced for smaller fund complexes and larger plans.

• If the overweighting is driven by superior information afforded to the trustee by the trustee relationship, the implication would be that the fund companies could use this information to outperform on the sponsor's stock.  The trustee overweighting in the sponsor firm's stock, however, is not accompanied by any investment outperformance. 

• Larger plans correlate with larger overweighting.  This further supports the belief that fund companies are proactively overweighting in order to secure a trustee relationship that gathers more assets.

• Overweighting in trustee stock is six times greater when comparing funds in the plan to funds not in the plan, but in the same trustee fund family.  This is consistent with overweighting being more severe in those funds that have greater benefits.

• By looking at holdings at the time of a switchover, the authors more precisely measure the effect of a fund company becoming a trustee. About 3.4% of plans switch trustees each year, representing 58 instances of a change in the sample data.  In these instances, Cohen and Schmidt show the old trustee strongly decreases its position in the stock after it stops being the trustee, while the new one progressively increases its position on the stock when it becomes the trustee.

Implications for Advisors

To determine the net loss to investors, the authors calculate the Sharpe Ratio of each fund, and compare this to its value with optimal weighting (i.e., without any overweighting) and show investors lose up to  60 basis points per year in risk-adjusted returns for the average fund in their sample.

Assuming a plan participant begins investing at age 30 and retires at age 65, and makes annual contributions of $3,320 per year, overweighting in the average fund will cost this investor up to $42,313 (5% of a total roughly $730,000) over the 35 year period.

While plan members lose from this conflict of interest, sponsor companies stand to gain.  Fund companies support the price of the sponsor’s stock, by buying shares at times when other fund companies are selling the security.   In fact, Cohen and Schmidt show that fund companies increase their already overweighted positions by 11.45% “at exactly those times when the sponsor firm may find it most valuable.”  Fund companies also have a 19.8% smaller probability of selling the sponsor’s security when other funds are sellers.  Of course, all this may come at the expense of the fund investors.

The authors advise that “financial advisors should be cognizant of the fact that if they invest in a fund family that is the trustee of a company's 401(k) plan, that fund family has an empirically overweighted position in the given company in its mutual funds.  The overweighting is most severe in those mutual funds of the trustee family that also happen to be included in the 401(k) plan of the company.”  As we note above, commercial data services do not provide the necessary information to determine whether a fund is the trustee of any plans.

The authors do not provide data on specific funds or fund companies and their status as trustees.

Display article as PDF for printing.

Would you like to send this article to a friend?