Automatic Enrollment (Default Option)

Enrollment Decisions: Nudging People to Join

The first step in participating in a defined-contribution plan is enrollment. Most workers should find joining the plan very attractive. In the tax-deferred plans discussed here, contributions receive tax benefits, investment earnings accumulate before tax, and many employers match at least part of the employee’s contributions. The details depend on the plan and jurisdiction. For example, a common plan feature is that the employer will match 50 percent of the employee’s contributions up to some threshold, such as 6 percent of salary.

This match is virtually free money! It is offering an immediate 50 percent return on contributions. Taking full advantage of the match should be a no-brainer for all but the most impatient or cash-strapped households. Nevertheless, enrollment rates in such plans are far from 100 percent. Typically, younger, less-educated, and lower-income employees are less likely to join, but even high-paid workers sometimes fail to sign up.

To be sure, there are situations, say, for young workers with other pressing financial needs, in which it could be sensible not to join the retirement plan immediately even with an employer match. But in most cases, especially for anyone over thirty, the failure to join is usually a blunder, and a directional nudge (with an easy opt-out, of course) seems justified. How can we nudge these people to join sooner?

The solution to solving the enrollment problem is obvious, at least in hindsight. Make it easy! Make joining the default. The way plans were originally designed, when workers were first eligible to join (sometimes immediately upon employment) they would receive some forms to fill out. Employees who wanted to join would have to decide how much to put aside and how to allocate their investments among the funds offered in the plan. Forms can be a headache, and many employees just put them aside, treating them as sludge.

The alternative is to adopt what has become known as automatic enrollment. Here’s how it works. When an employee first becomes eligible, she receives a form indicating that she will be enrolled in the plan (at a specified savings rate and asset allocation), unless she actively fills out a form asking to opt out. It didn’t take a genius to realize this would work. In fact, one of us, a certifiable non-genius, suggested the idea in an article published in 1994. No one seemed to notice. A few firms, including the fast-food giant McDonald’s, did try the idea, but the policy was given a less-than-beautiful name: “negative election.” Even a good idea can be killed by a bad label.

In addition to the bad name, some firms worried that automatic enrollment might not be legal, since the firm would be taking an action without the explicit (active) permission of the employee. To help assuage this fear, the federal government issued a series of rulings and pronouncements that defined, approved, and promoted the use of automatic enrollment. That helped reassure cautious employers, though it inadvertently created a problem, to which we will return shortly.

An important milestone was the publication of an academic paper on automatic enrollment by Brigitte Madrian and Dennis Shea. Shea had worked at a firm that tried the idea, and he recruited Madrian, at that time an economics professor at the University of Chicago, to help him evaluate what happened. The firm had tried automatic enrollment because it was having trouble getting employees to join the plan using the usual opt-in design, even with a generous 50 percent match of contributions up to 6 percent of salary after one year of employment. To evaluate the effect of automatic enrollment, Madrian and Shea compared the enrollment rates of workers newly eligible for the plan in the year before the policy was changed with the enrollment rate in the plan with the improved default. The results were striking. In the opt-in system, only 49 percent of employees joined within a year of joining the firm, but once automatic enrollment was introduced, enrollment jumped to 86 percent. Only 14 percent opted out!

With the help of supportive legislation in 2006, automatic enrollment became more common. The reports discussed here show substantial increases in participation. A report by Vanguard in 2018, based on the 473 plans for which it served as record keeper, found that 59 percent of the employers in its sample used automatic enrollment, and those firms had an average participation rate of 93 percent. In contrast, firms requiring employees to opt in had a 47 percent participation rate. Hurray!

However, it is premature to declare victory. To stress what may seem like an obvious point, the benefits from making a policy the default option depend on the merits of that policy. If people are defaulted into a terrible pension plan, they can end up worse off. Even if they can easily opt out, inertia and procrastination might mean that they will not do so. A bad default can be sticky, especially if its bad aspects (such as the fees charged by the funds) are not very salient. The point is that high take-up of the default option cannot be viewed as a success in and of itself. We will reiterate this general point occasionally throughout the book, because it can be tempting to observe high take-up of a default option and declare victory. Please don’t!

The experience at the firm originally studied by Madrian and Shea illustrates this point. Note that when a firm uses automatic enrollment, it has to select a specific default investment rate and an investment strategy. In their case, the default was a 3 percent savings rate that was invested in a money market account (the least risky option). Unsurprisingly, if workers are saying “yeah, whatever” to joining the retirement plan, they are likely to accept these default details, which is unfortunate. That savings rate is too low, and the investment strategy is too conservative. Certainly, young workers should be investing some of their portfolio in stocks.

Of course, making a particular savings rate and investment option the default does not mean that everyone will accept those options. Some will actively choose something else, at least eventually. Nevertheless, many of the employees passively accepted the specific default options, even some who would have actively chosen to join in the previous regime. This group was nudged to their detriment, in terms of both how much they saved and how they invested.

We know this by looking at the behavior of people who joined under the previous regime. Before the introduction of automatic enrollment, the savings rates of those who actively chose to join were strongly influenced by the match formula. Recall that contributions were matched at a 50 percent rate up to a limit of 6 percent of salary. Unsurprisingly, about two-thirds of those enrolling chose a savings rate of exactly 6 percent. In contrast, when employees were enrolled automatically, the most common savings rate was the default: 3 percent. In other words, some of those who were automatically enrolled would have chosen a higher savings rate if they had been left to their own devices. Making matters worse, the default investment choice of the low-risk money market account meant that low contributions would earn tiny returns. Let’s underline what happened: more people were participating, which is good, but their savings rates were far too low and their investments were far too conservative. Automatically enrolling people in a bad or mediocre option has serious pitfalls. But both of these problems can be addressed by better choice architecture.