There was a period, back in the late 1990s, when 401(k) plans were hailed as the key to every household's financial dreams. Put money in your 401(k), the thinking went, and you will retire a millionaire, thanks to the magic of the stock market, which would reliably deliver 15% annual returns.*
A 401(k) plan is a defined contribution (DC) plan (other types include 403(b), 457, and SEP-IRA plans, as well as IRAs), meaning that participants put some of their income into the plan (along with, in some cases, additional money contributed by their employer); when they retire, they get that money back, along with any investment returns it has earned along the way. The lure of high investment returns is what made DC plans so attractive. By contrast, people with traditional defined benefit (DB) pensions--which pay a predetermined amount each year in retirement--were seen as suckers.
After two stock market collapses, defined contribution plans are no longer seen as a solution to all of life's financial problems, and for many well-known reasons. First, people don't save enough to begin with. Second, DC plans aren't reliable retirement savings vehicles, since it's too easy to take money out of a 401(k) plan. Third, the stock market doesn't always return 15% (on an inflation-adjusted basis, it's still well below 2000 levels). Fourth, and most importantly, DC plans place all that investment risk on the individual participant.
But wait, there's more! In addition to these very real problems, 401(k) plans are generally run by the asset management industry, which (surprise!) does not always have your interests at heart.
Most defined contribution plans allow participants to select from a menu of investment options, largely mutual funds. The question is, who decides what's on the menu? About three-quarters of the time, these decisions are made by a mutual fund company acting as the plan's trustee.** And (surprise again!) mutual fund companies are more likely to push their own funds onto employees than other companies' funds--despite the fact that they are legally obligated to act in the best interests of plan participants.
Although many people have suspected this all along, we now have convincing evidence from a paper by Veronika Pool, Clemens Sialm, and Irina Stefanescu aptly titled "It Pays To Set the Menu" (which was sent to me by two different readers). The paper uses a clever empirical approach. In any given year, a mutual fund may be included in 401(k) plans overseen by that fund's sponsoring company (e.g., the Fidelity Magellan Fund may be included in a plan whose trustee is Fidelity) and also in other plans not overseen by that company. It turns out that there are many such funds.
Let's say such a fund has a bad year. If plan trustees are acting solely in the interests of their participants, we would expect the identity of the trustee not to affect the chances that the fund is dropped from the investment menu. But that's not the case: a poorly-performing fund is much less likely to be dropped from a menu controlled by its sponsoring fund company than from a menu controlled by a third party. Fund companies are also much more likely to add their poorly performing funds to plan menus that they control.
But maybe the fund companies are keeping their poorly-performing funds because they know that they are likely to do better in the future. After all, past performance is no guarantee of future results, and maybe they are anticipating mean reversion. Not so much. Those poorly-performing funds that fund companies keep on plan menus continue to do worse than other funds--by 3.6% per year, after adjusting for risk. Finally, we can't count on individual rational actors to bail us out here: plan participants do not correct for fund companies' bias by proactively shifting their money out of these poorly-performing funds.
This is a special and particularly maddening example of the problem that the vast bulk of our retirement savings, other than Social Security, are funneled through the asset management industry, which charges often-hefty fees to manage our money, despite its proven inability to beat the market. (Theoretically speaking, since the asset management industry is most of the market, in aggregate it should match the market--before fees.) This would be a necessary inconvenience if competition drove fees down to reasonable levels, like the 6 to 20 basis points that Vanguard charges to invest in just about any major segment of the global stock market. But instead, structural factors, like mutual fund companies' control over investment options in DC plans, take the edge off of competition, allowing these rents to persist. (As of 2009, domestic stock mutual funds in 401(k) plans, in aggregate, charged 74 basis points in expenses--and cut into returns further with transaction costs.)
Even if we solve the more-publicized problems with DC plans--under-saving, pre-retirement leakage, etc.--they will still suffer from this constant drain of asset management fees. The UPenn Journal of Business Law recently published my (first!) law review article, which proposes a relatively far-reaching solution: Reinterpreting existing law to impose a strong presumption in favor of low-cost index funds for tax-advantaged, employer-sponsored DC plans, including 401(k)s. Even if you don't agree with that solution, something needs to be done.
Because with each passing year, Americans are becoming less prepared for retirement, not more.
* In 1998 and 1999, investors' expectations of returns over the next twelve months ranged roughly from 12% to 15%. See a working paper by Robin Greenwood and Andrei Shleifer.
** Under the Employee Retirement Income Security Act of 1974 (ERISA), employer-sponsored plans must be held in trust for the exclusive benefit of participants and beneficiaries.>