Savings Boost from Auto-Enrollment Wanes Over Time

Forty percent of U.S. private-sector workers in a 401(k) retirement plan are in plans with automatic enrollment, and the widely agreed-upon story is that these plans work well.

Now comes a more nuanced assessment, which finds they aren’t working quite as well as everyone had hoped.

The study, conducted by some of the pioneers in auto-enrollment research, shows that numerous dynamics significantly reduce how much is being saved in 401(k)s. Workers often leave the firms before their employer matching contributions have fully vested, withdraw money from savings, or opt out of the automatic increases in contributions designed to accelerate their savings incrementally.

Auto-enrollment still results in more saving than when workers are left to their own devices. But their often-overlooked decisions “meaningfully reduce the impact of automatic policies on accumulation in the U.S. retirement savings system,” the researchers concluded from their analysis of nine 401k plans.  

Four of the companies they studied had recently adopted auto-enrollment. The other five added a second feature: automatic increases in how much employees contribute to their savings plans. The goal here is not only to encourage more people to save – but to save more over time. Two of these firms already had auto-enrollment in place and just introduced the automatic contribution increases, and three firms introduced both features simultaneously.

To test the plans’ effectiveness, the analysis compared the rate of saving for thousands of employees hired by the companies within a year of the new auto-enrollment policies with thousands who had joined the previous year and were unaffected by policies put in place after they were hired.  

Initially, the affected workers saved substantially more than the workers who lacked auto-enrollment plans. But the saving rate diminished as the researchers incorporated workers’ real-world decisions about how much or whether to save and whether they would stick with the automatic contributions increases embedded in the plan design.

Among the four firms that adopted auto-enrollment only, the average saving rate initially was 2.2 percent more of workers’ incomes than the rate among employees hired prior to the policy’s adoption. But this gap shrinks over time to 0.6 percent when the rosy assumptions – that employees stick with their initial saving rate for all five years of the analysis, never withdraw money from their accounts, and fully vest – are dropped, and the data used in the analysis reflect workers’ real-world behavior.

The saving rate also eroded at the firms that automatically increased workers’ contribution rates. One factor was that less than half of them accepted the first scheduled increase, a number the researchers called “surprisingly high.” The workers also withdrew money from their accounts or missed out on vesting of their employers’ contributions.

At the firms with auto-enrollment that later added auto-escalation, the gaps in the saving rate between the employees hired before and after the change shrank from 1.8 percent of incomes initially to 0.3 percent using actual behavior. At the firms that simultaneously adopted both features, the gap fell from 3.5 percent to 0.8 percent after the rosy assumptions were dropped.

“Medium- and long-run dynamics,” the researchers concluded, “undermine the effect of automatic enrollment and default savings-rate auto-escalation on retirement savings.” 

To read this study by James Choi, David Laibson, Jordan Cammarota, Richard Lombardo, and John Beshears, see “Smaller Than We Thought? The Effect of Automatic Savings Policies.”

The research reported herein was performed pursuant to a grant from the U.S. Social Security Administration (SSA) funded as part of the Retirement and Disability Research Consortium. The opinions and conclusions expressed are solely those of the authors and do not represent the opinions or policy of SSA or any agency of the Federal Government. Neither the United States Government nor any agency thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of the contents of this report. Reference herein to any specific commercial product, process or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply endorsement, recommendation or favoring by the United States Government or any agency thereof.