Two decades ago, researchers studied how automatically enrolling people in retirement plans affects savings in the U.S. Instead of making employees opt into a 401(k) plan, they were enrolled by default. The results were dramatic: at one company, automatic enrollment boosted 401(k) participation by nearly 70 percentage points five months after hire and by 28 percentage points after four years, compared to an opt-in system.
“At the time, people thought such a small nudge wouldn’t have a big impact on a decision as important as retirement savings,” the researchers explained. “Seeing these huge effects was a real surprise.”
These findings helped spur laws like the 2006 Pension Protection Act, which pushed companies to use automatic enrollment and automatic escalation (where employees’ contribution rates rise gradually, often by 1% per year). But with these practices now common, the researchers wanted to know: how do they hold up in the real world?
“Automatic enrollment is everywhere now, and we wanted to test it under real-world conditions,” they said. “What are its limits? How can we poke holes in our own findings? Researchers should be their own toughest critics, not cheerleaders.”
Limited impact
In their new study, they found that the effects of automatic enrollment and escalation were modest. Once they accounted for job-switching and savings habits, they found that automatic enrollment raised annual savings by just 0.6% of income, and automatic escalation by only 0.3%. These numbers were about 72% smaller than earlier estimates that left out important factors.
The researchers reached this conclusion by analyzing data from over 118,000 employees at nine companies that introduced these nudges between 2003 and 2011. They compared people hired before and after the policies were introduced to see how workers’ savings behavior changed.
Their new approach differed from earlier studies in a few key ways. First, they included employees who quit their jobs. About 4% of U.S. workers leave their jobs each month, and in this study, 40% of those who quit cashed out their 401(k) balances, often losing their employer’s matching contributions before they were fully vested. Previous studies ignored these workers, focusing only on those who stayed at the same firm.
Following through
Second, they tracked whether people actually followed through with automatic escalation. Surprisingly, only 40% of workers stuck to the automatic contribution increases—far below the 85% reported in a major 2013 study.
Taken together, these factors resulted in much smaller savings than expected.
Some people, the researchers note, are natural savers who don’t need nudges. Automatic enrollment has the biggest effect on those who are less motivated to save. But the new study shows that these reluctant savers often reverse the nudge by cashing out when they switch jobs.
“The job-switching moment is a big weak spot in the U.S. retirement system,” the researchers said. “A lot of savings flow out at that point. You could stop people from withdrawing their money early, but that might discourage them from saving in the first place. People might think, ‘If I can’t touch my 401(k) money until I’m 60, why contribute at all?’”
A better approach
If nudges are only somewhat effective and limiting withdrawals could backfire, what’s the solution? The researchers have become more open to forced savings schemes, like those in Singapore and Australia, where people are required to save a large chunk of their income. But they note that forcing everyone to save 15% of their pay could cause financial stress for some.
They also wonder if people cash out their 401(k)s when they leave a job simply because they need the money. Surveys show that many Americans can’t come up with $400 in an emergency. This has led the researchers to study savings plans in the UK designed to help people build emergency funds alongside retirement savings.
All of this points to the importance of looking at a person’s total financial health, not just their retirement savings.
