Citation
Choi, J., Laibson, D., & Madrian, B. (2005). Are empowerment and education enough? Underdiversification in 401(k) plans. Brookings Papers on Economic Activity 2005, (2), 151-213.
Highlights
- The study examined factors that can affect diversification from employer stock in 401(k) plans. The study analyzed the impact of loosened diversification restrictions and media coverage of highly publicized events on 401(k) holdings in employer stock.
- The authors assessed factors affecting diversification in 401(k) plans through four studies. The first was an interrupted time series analysis of employer stock holdings at an unnamed company. The other three studies explored the impact of media attention to the Enron, WorldCom, and Global Crossing bankruptcies through regression analysis. All analyses were based on transaction-level and annual data from 1997 to 2003 from Hewitt Associates, a benefits administration company.
- The authors attributed only a 2.4 percentage point decline in employer stock holdings to news coverage of the Enron, WorldCom, and Global Crossing bankruptcies across the entire Hewitt 401(k) Index.
- The quality of causal evidence presented in this report is low. In the first study, the authors did not account for prior trends in the outcome variable and in the second through fourth studies they did not provide information a reader would need to assess changes in sample composition. This low rating means we are not confident that the estimated effects are attributable to any of the interventions studied, whether loosened diversification restrictions or media coverage of the Enron, Global Crossing, and WorldCom bankruptcies. Other factors are likely to have contributed.
Intervention Examined
Factors Promoting 401(k) Diversification
Findings
- Before the change in diversification policy, an average employee held at least 90 percent of his or her employer matching account in employer stock. By late 2003, that proportion had dropped to 84 percent (statistical significance not assessed).
- News of the Enron, WorldCom, and Global Crossing bankruptcies had a negative effect on investments in employer stock; however, the authors attributed at most a 2.4 percentage point decline in employer stock holdings to news coverage of those bankruptcies.
Considerations for Interpreting the Findings
Studies presented in this report relied on two quasi-experimental designs: an interrupted time series and a comparison group design with regression analysis. The first study, which examined the impact of relaxed restrictions governing employee 401(k) diversification, used an interrupted time series design but failed to provide evidence that the intervention was not caused by underlying factors or trends in employees’ investment behavior, introducing many possible confounds related to prior trends in the outcomes of interest.
The other three studies employed a longitudinal regression analysis framework but did not provide sample sizes at the beginning and end of the period of interest. Changes in the size and composition of the study sample could affect the outcome variable in ways that do not reflect the impact of the news items.
Causal Evidence Rating
The quality of causal evidence presented in this report is low. This low rating means we are not confident that the estimated effects are attributable to any of the interventions studied, whether loosened diversification restrictions or media coverage of the Enron, Global Crossing, and WorldCom bankruptcies. Other factors are likely to have contributed.
In the first study, the authors did not account for prior trends in the outcome variable. To provide more convincing evidence of the intervention’s effects that satisfies Clearinghouse for Labor Evaluation and Research (CLEAR) criteria, the authors must successfully demonstrate that the policy change was not caused by some underlying factor or trend that might also change employees’ choices to invest in employer stock.
In the second through fourth studies, the authors did not provide information on potential changes in sample composition. To provide more convincing evidence of the intervention’s effects that satisfies CLEAR criteria, the authors should examine changes in the composition of the sample over time and demonstrate that shifts in the sample cannot account for the observed changes in employer stock holdings.