Financial Behavior: Work, Save, Retire.

January 15, 2013

401(k) Mutual Funds Mediocre

A spate of research in recent months shows that the mutual funds offered in employer 401(k)s have fairly unremarkable – though not disastrous – investment performance.

As with any academic study worth its salt, the various authors’ findings are complex and loaded with critical twists, turns, and footnotes.  Descriptions of three research papers on 401(k) plan returns can be found below. But here’s the gist:

• So-called Target Date Funds – investments are based on each employee’s planned retirement age – perform better than investments guided by financial advisers hired by one Oregon employer to advise its workers.  TDFs also outdo employees who go it alone.
• When the 401(k) plan’s trustee is also a mutual fund management company, it’s more reluctant to remove its own, poorly performing funds from the plans’ smorgasbord of funds.
• Employers select mutual funds that outperform a portfolio of randomly selected funds but underperform passive indexes.

There’s a common thread in many of these studies: the extra fees that  investors pay for advice or the stock pickers who manage their mutual fund often don’t translate to better returns.

The first of the following studies received funding from the Social Security Administration, which supports this blog:

Financial advice vs Target Date Funds

The investment returns for two approaches – using brokers for advice and buying target date funds (TDFs) – were compared by business professors John Chalmers at the University of Oregon and Jonathan Reuter at Boston College.

Target date funds came out on top.

“People who used brokers would’ve been, on average, significantly better off in a TDF fund,” Reuter said in an interview last month.  “That makes me think TDFs are a good default or substitute for financial advisers” in a 401(k) plan.

Using a unique data set – investment returns for members of the Oregon University System’s defined contribution plan – the researchers first compared members’ actual investment returns with how they would’ve fared if they had instead invested in a simulated portfolio of TDFs managed by Fidelity Investments from 1999 through 2009.  The TDFs out-performed the employees’ broker-directed portfolios by 3 percentage points per year.  TDFs also out-performed the portfolios of investors who picked their own mutual funds – though by about half as much (1.65 percentage points).

The authors also simply compared the performance of the two groups of actual investors.  They found that employees who used the financial advisers offered by the university’s plan earned 1.54 percentage points less per year on their investments than those who decided to go it alone.

The results were similar when the researchers adjusted their analyses to control for portfolio risk.  And a large chunk of the under-performance for advisers came in the form of fees, which totaled nearly 1 percent of their client’s assets.

Investors who used the brokers tended to have less education, lower incomes and ranked lower in financial literacy, indicating they may need help.  But this help “comes at a cost,” Reuter said.

Investors who use TDFs, rather than brokers, he added, “are likely to have larger 401(k) balances when they retire.”

Mutual fund managers as 401(k) trustees

Mutual fund management companies that also act as trustees for employers’ 401(k)s display “favoritism toward their own funds.”

Specifically, the trustees are more reluctant to remove poor performers from the plan’s selections of funds and are more likely to add them, according to a December paper by Veronika K. Pool and Irina Stefanescu at Indiana University, Bloomington, and Clemens Sialm at the University of Texas, Austin.

Using Securities and Exchange Commission filings on large 401(k) plans from 1998 through 2009, they compared decisions to drop mutual funds in plans in which some funds were affiliated with the trustee against drops by other plans in which they were not.  It turns out that the fund companies acting as trustees were less likely to delete poorly performing funds – and the worse a fund performed, the greater this tendency.

The worst-performing funds, the paper said, “have a probability of deletion of 27 percent for non-trustee funds and a probability of deletion of only 11 percent for trustee funds.”

While this favoritism may sound like a quirky side issue, it actually affects a huge share of all defined-contribution plans: 77 percent in their sample hired mutual fund companies to be trustees.

But wait a minute.  Just because a trustee holds on to a mutual fund doesn’t mean the fund’s performance will continue to lag.  Indeed, the fund company is in a unique position to possess information that the fund’s performance is about to improve.  It’s a wishful theory that didn’t hold up in the research, which found that the performance of trustee-run funds that should’ve been dropped deteriorates significantly the following year.

The researchers conclude that mutual fund families “display leniency” when faced with poorly performing funds within their own ranks.  “The decision to delete [or add] trustee funds is significantly less sensitive to poor fund performance than that of non-trustee funds.”

How employers’ mutual fund choices fare

This final paper analyzes how the returns to mutual funds in 43 different 401(k) plans compare with similar, randomly selected sets of funds.  This is a critical question, because employees are not very good at making their own investment decisions.

Returns to the employer-chosen mutual funds were a mixed bag.

Mutual funds in 401(k)s “outperform randomly selected funds of the same type,”  the research concludes.  But the 401(k) participants usually could’ve done better by keeping it simple and investing in indexes.

To read a research brief summarizing the findings, click on this link at the Center for Retirement Research at Boston College.  For another paper assessing the performance for stock mutual funds in 401(k)s, click here .

Full disclosure:  Some research cited in this post was funded by a grant from the U.S. Social Security Administration (SSA) through the Retirement Research Consortium, which also funds this blog.  The opinions and conclusions expressed are solely those of the blog’s author and do not represent the opinions or policy of SSA or any agency of the federal government.