In Defense of Auto Enrollment

An from yesterday’s Wall Street Journal was disconcerting — and misleading. It said the

re may be unintended consequences to the law that allows firms to automatically enroll their employees into 401(k) plans. In one of many scenarios, an analysis done by the Employee Benefits Research Institute for the WSJ found that up to 40 percent of workers might have elected to save more than the typical 3 percent default contribution rate used by most of today’s plans. My first thought was that auto-enrollment is making a positive difference for workers who would otherwise not save, but it appears that it can cause complacency among those would normally save more. So what should we do? Throw up our arms in despair and do away with auto enrollment?

As I was digging deeper into what could be the problem, I received an email from Jack Towarnicky, an employee benefits attorney. He thought that the WSJ article focused only on a minority of participants and did not present the whole story. He referenced a recently issued Vanguard study, “How America Saves, 2011,” and highlighted Figure 23 on page 25, which shows the difference in voluntary versus automatic enrollment by age, pay and service. Vanguard data shows only 29 percent of workers with a year or less of service voluntarily enroll, while 75 percent of similarly-situated workers enroll in plans with automatic enrollment.

So, looking at only one year, the tradeoff is, as Towarnicky states:

  • 29 percent of new hires would have participated anyway, and maybe 40 percent of them would have elected a contribution of greater than 3 percent so, 40 percent of 29 percent — that is, 12 percent of new hires (at best) would have contributed at a higher level. The other 17 percent apparently would have contributed 3 percent or less anyway.
  • But, with 75 percent enrollment, it means 46 percent more are participating whereas otherwise they might not have enrolled, and;
  • All 75 percent are “teed-up” for automatic escalation in subsequent years.

Indeed, the lead in the article suggests that automatic enrollment and the use of 3 percent as the default was a direct and proximate result of the Pension Protection Act (PPA) of 2006. Automatic enrollment has been around for well over a decade before PPA 2006 and 3 percent had become the “standard” default contribution percentage before PPA, having migrated from 2 percent in the earliest automatic plans. What PPA did change, and what is not mentioned in the article, is the addition of Qualified Default Investment Alternatives (QDIA) requirements, which freed up employers to use something other than a money market as the default investment.

The article focuses on workers who say they would have elected a higher contribution rate under a voluntary system. However, each one was specifically notified of their opportunity to enroll at a higher rate – but all failed to take action. That is typical. There are studies which confirm many state their intention to start saving or to increase contributions, but fail to follow through. As a result, more and more plans incorporate automatic enrollment and automatic escalation features.

Similarly, the use of average savings rates of participants is probably misleading. The article cites a decline from 7.9 percent to 7.3 percent in the average contribution percentage among participants in AonHewitt administered plans from 2006 to 2009.

However, the populations are very different, as the group in 2009 includes many who joined at a 3 percent contribution rate due to automatic enrollment. So, at a minimum, the article should have also shown the increase in average deferral percentages (ADPs) for those same years — a more accurate measure of how automatic enrollment changes savings behavior.

Despite what the article failed to mention, however, there will be those who insist that 401(k) plans are a failure and will trot out government-guaranteed retirement accounts as the answer. (See my blog post on GRAs.) But this remedy would worsen, not cure, the problem of low retirement savings rates.

Comments (1)

Trackback URL | Comments RSS Feed

  1. As Ms. Villarreal and Mr. Towarnicky correctly note, the WSJ story cherry-picked EBRI’s data and wholly misrepresents our research on auto-enrollment. A detailed response rebutting the WSJ article, by EBRI Research Director Jack VanDerhei, is posted on EBRI’s blog at

    — Stephen Blakely

Leave a Reply

If you want a picture to show with your comment, go get a Gravatar.