Choi, J., Laibson, D., & Madrian, B. (2009). Mental accounting in portfolio choice: Evidence from a flypaper effect. American Economic Review, 99(5), 2085-2095.
- The study examined the impact of requiring employees at a large, U.S.-based company to actively choose the asset allocation for an employer’s matching 401(k) contributions, rather than automatically allocating such funds to the employer’s stock.
- The analysis was based on administrative data on employees’ demographic characteristics and 401(k) contributions. The study analyzed changes in outcomes that occurred when a firm began requiring employees to choose the funds to which the employer’s 401(k) matching contributions would be allocated.
- The study found that allowing employees to choose the asset allocation of their matched contributions reduced contributions to employer stock by 64 percentage points in the first year of the new policy. The study also found that employees’ own contribution allocations did not change significantly after the policy was enacted, suggesting that employees made their personal allocation decisions without considering the employer’s match allocation.
- The quality of causal evidence presented in this report is low. This means we are not confident that the estimated effects are attributable to the active-decision requirement for employer matched contributions. Other factors are likely to have contributed.
Active Portfolio Choice for Employer Matching Contributions
- The study found that employees who enrolled after March 2003 allocated, on average, 64 percentage points less of the employer match to employer stock in 2003 and 59 percentage points less in 2004 than those who enrolled before March 2003.
- Employees across both groups allocated nearly the same percentage of their own contributions to employer stock, evidence that employees made their personal allocation decisions without considering the employer match allocation.
- The year-end balance of matched contributions held in employer stock was 66 to 68 percentage points lower for employees hired after March 2003. The fraction of individual contributions held in employer stock was 2 to 4 percentage points lower for the intervention group.
Considerations for Interpreting the Findings
The study showed that individual 401(k) contributions to employer stock declined from about 60 percent in 1998 to about 20 percent in early 2003 (before the policy change) and that year-end balances of individual allocations in employer stock had declined by about the same magnitude over the period. Thus, there is some reason to believe that the change in the company’s 401(k) plan was a result of declining employee investment in employer stock. In addition, the study examined the policy change at only one point in time. This makes it difficult to conclude that the observed changes in outcomes were driven by the intervention itself, instead of other concurrent changes (for example, changes in market conditions).
Causal Evidence Rating
The quality of causal evidence presented in this report is low because the study examined only a single implementation of an intervention and did not demonstrate that the timing of the intervention was not influenced by existing trends. This low causal evidence rating means we are not confident that the estimated effects are attributable to the active-decision requirement for employer matched contributions. Other factors are likely to have contributed. To provide more convincing evidence of the intervention’s effects that satisfies Clearinghouse for Labor Evaluation and Research (CLEAR) criteria, the authors should consider examining how similar policy changes at other firms affected employees’ 401(k) allocations. If other implementations demonstrated similar effects, we could be more confident that changes in outcomes were driven by the intervention, and not some other concurrent change.