Comments and Discussion

William G. Gale:

This paper by James Choi, David Laibson, and Brigitte Madrian documents the well-known fact that many workers invest a significant share of their 401(k) contributions in the equity of their own employer, and briefly reviews the literature indicating that that share is too large. The basic argument here is correct. Asset returns on a well-diversified equity and bond portfolio will have a smaller variance than those on a concentrated equity portfolio. And, of course, holding employer stock increases the correlation between a worker's labor market and financial market outcomes, which is not likely to be a good thing.

The paper also documents that workers do not diversify away from employer stock when given the chance to, and even when bombarded with news stories illustrating the problems of holding employer stock. The evidence is compelling. Although the authors spend a lot of effort "proving" this result with a variety of regressions, readers should not come away thinking that the result is somehow sensitive to the specification. Figure 6 in the paper shows the stability of employer stock holdings over the relevant sample period. Although some variation is evident, the authors point out that the more relevant of the two series, employer stock as a share of equities in 401(k)s, varies relatively little. Moreover, the variation is especially small given that the optimal level for this series should be very close to zero.

Having bought into the basic findings, I will turn to two related points that may provide some perspective on the issues the authors address.¹ The first is that the current set of rules governing how workers may invest their 401(k) balances is essentially an accident of history. The 401(k) has come to play a far more central and critical role in the pension system than was envisioned when it was created in the late 1970s, yet the rules have changed little since then.

Twenty-five years ago, defined-benefit plans (together with certain types of traditional defined-contribution pension plans, such as employer-funded profit-sharing plans and money purchase plans) were workers' primary source of private pension coverage. These plans require workers to make almost no important financial choices before retirement. The firm enrolls all eligible workers, makes contributions on their behalf, and makes all the investment decisions or retains professional investment managers to do so. The worker's only real choices are when and in what form to collect benefits.

When 401(k) plans began their rapid spread in the early 1980s, they were viewed mainly as supplements to these traditional employer-funded plans. Since 401(k) participants were presumed to have their basic retirement income security needs covered by a traditional employer-funded plan and Social Security, they were given substantial discretion over their 401(k) choices, including whether to participate, how much to contribute, how to invest, and when and in what form to withdraw the funds.

Over the past twenty-five years, however, the pension landscape has changed dramatically. Many workers covered by an employer plan now have a 401(k) as their primary or only plan. Yet 401(k)s still operate in much the same way as in the early 1980s. Workers must still, for the most part, decide for themselves whether and how much to contribute, how to invest, and how and when to withdraw the funds. Imposing on workers the responsibility to make these choices may have been relatively harmless when 401(k)s were smaller, supplemental plans with limited coverage. The risk of workers making poor investment choices looms much larger now that 401(k)s have become the primary pension vehicle.

The second point is that government policies have encouraged this situation in several ways. As 401(k)s have expanded over the last twenty-five years, Congress has been enacting rules that implicitly or explicitly encourage overinvestment in company stock. First, the Employee Retirement Income Security Act of 1974 (ERISA) relieved employers of most fiduciary responsibility for investment losses if they allowed employees to direct their own investments—this was likely one factor...