Cover | Published in June 2017

Shaking Things Up

The case for re-enrollment to improve participant outcomes

By Judy Ward | June 2017
Art by Valeria Petrone

Re-enrollment is a plan design strategy that helps many employees better their retirement outcomes by increasing participation and deferral rates and/or improving investment allocations. And yet, 89.3% of sponsors surveyed for the 2016 PLANSPONSOR Defined Contribution (DC) Survey said they had re-enrolled no employees or participants in the past 12 through 18 months. Among those that had, 1.9% re-enrolled participants not invested in the default investment, 4% re-enrolled participants saving below the default deferral rate, and 8.6% re-enrolled nonparticipating employees.
For many plan sponsors, re-enrollment remains a relatively new concept, says Lynda Abend, chief data officer at John Hancock Retirement Plan Services in Boston. “Automatic enrollment really started with the PPA [Pension Protection Act], so it’s about 10 years old. It took four or five years before employers got comfortable with it, and then most automatically enrolled new employees at 3%,” she says. “In the past five years, we’ve made good progress in working with plans that had auto-enrolled at 3%, [prompting them] to move to a higher initial deferral, and getting sponsors comfortable with auto-increases. Now, the next step in this evolution is re-enrollment.”
Re-enrollment can improve participants’ investment diversification, get participants on track to defer enough for their retirement, and bring nonparticipating employees into a plan. “If a plan sponsor cares about the financial well-being of its work force, [it] should care about this. And if a plan sponsor cares about employees transitioning into retirement in an orderly way, [it] should care about this,” says Jerry Patterson, senior vice president of retirement and income solutions at Principal Financial Group in Des Moines, Iowa. “There is no tool, no calculator, no seminar that can do for your employees what we together can do through best-practice plan design.”
Re-enrollment can take various forms and be used to accomplish several purposes, the main of which we discuss on the following pages.

Improving Asset Allocations
Re-enrollment in its earliest form focused on reallocating the balances of a plan’s current participants to its qualified default investment alternative (QDIA), says Cynthia Pagliaro, senior research analyst at the Vanguard Center for Investor Research in Valley Forge, Pennsylvania. In that form, re-enrollment “is not aimed at getting participants to save more,” she says, but at moving them into a proper asset allocation.
While automatically enrolled new hires get access to the age-appropriate allocations and diversification of a QDIA, there has not been a big, voluntary migration of longer-tenured employees to their plan’s default investment, Patterson says. “It’s almost like a dart board for some people, when they select their investments,” he says. “And most human beings are highly disengaged on these issues to begin with, so it’s not surprising that most participants do not revisit their initial investment selection.”
In a paper Vanguard released in March, “Re-enrollment: One year later,” the firm presented a follow-up to a 2016 case study of a large-plan client’s investment-focused re-enrollment. Over that year, diversification improved considerably, with just 7% of participants still holding an “extreme” position of 0% or 100% equity. Eighty-one percent of plan assets remained in the default target-date funds (TDFs).
The original, February 2016 paper on the case study, “Reshaping Participant Outcomes Through Re-enrollment,” also showed how re-enrollment into a passive target-date fund family can lower participant fees substantially. The average expense ratio paid annually by the large plan’s participants decreased from 41 basis points (bps) to 10 basis points after re-enrollment.
The current paper shows that 80% of re-enrolled participants stuck with the default as their sole plan investment, and 20% did not. Of that 20%, 12% of participants did a partial opt-out—still holding the default but also moving some money to one or more other plan investments—and 8% did a full opt-out from the default. Pagliaro characterizes that opt-out pattern as typical for an investment-focused re-enrollment. “Over time, we see a small percentage of people leave the target-date funds altogether,” she says. “And sometimes participants take a smaller position in a target-date fund, where they are not 100% allocated to it.”
The idea of defaulting existing participants who already made an investment election into a plan’s QDIA gives some sponsors pause, however. They worry about employee reaction—more so than when they consider a re-enrollment focused on bringing participant savings up to their plan’s default deferral rate, says Rob Austin, director of retirement research at Aon Hewitt in Charlotte, North Carolina.
With so much published industry research showing the positive impact that increasing participants’ deferral rates has on savings outcomes, Austin says, more employers feel comfortable telling participants the reasoning for that re-enrollment decision. “It’s a little trickier for a plan sponsor to say to current participants, ‘We know what the best investment is for you,’” he says. “Especially with target-date funds, the only variable that comes into play is how old you are. To pigeonhole those people into the default investment, that takes a lot of intestinal fortitude for sponsors.”
When re-enrolling current participants into a QDIA, Pagliaro says, it is important to make clear that they can opt out of the change. The opt-outs mostly fall into the “do-it-yourself” group that feels more confident about investing, she says. “They tend to be people who are older and wealthier, and they have a preference for building their own portfolio,” she says. “Typically, they opt out of the target-date default and then rebuild the portfolio they had before the re-enrollment. Often, in their investment choices, these participants demonstrate rational portfolio-construction skills.”
But the vast majority of participants do not want to build their own portfolio, Pagliaro says. “They’d prefer that investment professionals make those decisions,” she says. “Through re-enrollment, you’re helping those folks to get a better-diversified portfolio. The vast majority of re-enrolled participants stick with it, to their benefit.”

Boosting Participants' Deferral Rate 
Another approach to re-enrollment focuses on improving deferrals. Sponsors sometimes re-enroll employees saving below the default deferral rate, raising them to that rate, or to the rate needed to collect the maximum match. To determine whether taking this step makes sense, an employer should work with an adviser to perform a participant retirement readiness study, recommends Tom Foster, assistant vice president, strategic relationships at MassMutual Financial Group in Enfield, Connecticut. “The key to all this is doing that study first,” he says. A study can look at the percentage of participants on track to retire at an appropriate age—say, 65 or 67— with a sufficient percentage of their pre-retirement income—say, 75% or 80%, including Social Security.
Some employers do that study and find that their work force saves sufficiently and does not need re-enrollment to the plan’s default savings rate. “But, in many other cases, many employees are woefully underprepared for retirement,” Foster says.
T. Rowe Price Retirement Plan Services often sees plans implement this type of re-enrollment when they do a recordkeeper conversion or have corporate mergers/acquisitions. “Re-enrollment can be a real accelerant to helping improve the retirement outcome for their employee population,” says Aimee DeCamillo, head of retirement plan services for the firm, in Baltimore.
Asked to cite a best practice for re-enrollment that includes the deferral rate, Jerry Patterson, of Principal Financial Group, says, a 6% rate, with 1% auto-escalation, is emerging as a good number. “The experience of our plan clients shows that if you auto-enroll employees at 8% or above, you begin to see a meaningful fall-off in the ‘stick rate,’” he says. “More employees say, ‘That seems like a lot.’”
The biggest argument for sponsors increasing the deferral rate in a re-enrollment is simply because it works to help participants save enough for retirement, says Nathan Voris, managing director of business strategy at Schwab Retirement Plan Services in Richfield, Ohio. “You build in these positive participant behaviors every time you do a re-enrollment,” he says.
On the challenging side, a re-enrollment that includes deferral rates could increase an employer’s match costs significantly, if a plan has many very low-saving participants. “It depends on the plan design,” says Aimee DeCamillo of T. Rowe Price Retirement Plan Services. “You have to do an analysis and look at, what are the incremental costs to the employer? And what are the ways to offset those?”
A T. Rowe Price white paper, “Getting Beyond Ordinary—Managing Plan Costs in Automatic Programs,” discusses some of the ways an employer can offset the potential cost increases accompanying automatic features. These can include making changes to the employer contribution’s structure, such as shifting to a stretch match or moving the timing of the employer contribution to the end of the year so employees who depart before that do not get that year’s match. An employer also can control its costs by changing the vesting design, such as requiring longer service for 100% vesting. Additionally, by altering a plan’s eligibility rules the employer can limit the financial impact: Potential moves include changing eligibility timing for new hires to let them enroll immediately but get no employer contribution for one year.
In terms of participant-level retirement readiness, it also helps to tell participants why an employer wants to increase their retirement-savings rate, Foster says. “An employer may feel concerned that employees will think, ‘Why does my employer think it should make these decisions for me?’” he says. “If we can show employees, ‘Here is where you are, and here is where you need to be,’ then I think that it’s going to be perceived positively.”

Doing the Backsweep
When Schwab Retirement Plan Services’ plan clients perform a re-enrollment, they typically include current participants and nonparticipating eligible employees. All are re-enrolled into the qualified default investment alternative (QDIA). Those saving below the default deferral rate are increased to that rate. “It gets more people closer to success,” says Nathan Voris, a managing director at Schwab “And we see very little, if any, participant backlash.”

Aon Hewitt’s biennial Trends and Experience in Defined Contribution Plans Survey shows that the percentage of employers doing a backsweep—re-enrolling eligible nonparticipants—has held steady in the past several years in the 14% to 16% range, says Rob Austin of Aon Hewitt. “About half of those companies do it as a one-time sweep, and the other half keep re-enrolling people every year,” he says.
Principal Financial Group strongly believes in plans doing a backsweep of eligible employees every year, because employees’ reasons for not saving often get eliminated over time, says Jerry Patterson of the firm. “We did some participant research, and it found that 80% of participants said they had a neutral or positive reaction to plan sponsors re-enrolling employees who had opted out of the plan,” he says. “They said it didn’t upset them.”
Schwab finds the “stick rate” of a backsweep usually runs in the 85% to 95% range of employees, Voris says. “The No. 1 piece of feedback we get from sponsors about a sweep is, ‘I don’t want to make my participants angry,’” he says. “But if someone opted out of the plan seven years ago, chances are that the reasons for that person have changed since then.”
According to Lynda Abend of of John Hancock Retirement Plan Services, opt-out rates over the long term for a backsweep follow an interesting pattern. “We find that, when our clients use a default deferral rate of 3% in a sweep, the long-term opt-out rate is 10%,” she says. “If a sponsor does a sweep at 6%, the long-term opt-out rate is 4%.” She attributes the lower opt-out rate accompanying a higher-deferral re-enrollment to the tendency of participants to get more enthusiastic about saving for retirement once they start accumulating a substantial account balance. A higher initial deferral builds the balance more quickly.
Among Principal’s plan clients, the average participation rate runs 7% higher for plans that implement a backsweep along with auto-enrollment of new employees, versus only auto-enrolling new hires, Patterson says. “And we see a 17% higher average deferral rate for plans that implement a sweep,” he says.
Re-enrollments that include both low-deferring participants and nonparticipating employees substantially boost the percentage on track to, when including Social Security, replace at least 70% of their pre-retirement income, John Hancock finds. Among its plan clients that do re-enrollment that way, “There is a 10-percentage-point improvement, on average, in the percentage of participants being ready for retirement,” Abend says.
On the downside, employers often worry about how a backsweep will impact their costs, Austin says. “You can bet you are going to see an increase in participation. So the company needs to put in more match money for those new participants, which is a hard-dollar expense for the employer,” he says. “This is one reason why some companies set a low initial default rate.” An employer may e.g., match up to 6% of pay but do a backsweep at a 3% default deferral ad couple it with auto-escalation.
“So the longer employees stay with the company and the plan, they will get auto-escalated and get up to the 6% match,” he adds. That way, he says, employers ensure that they concentrate their match money on longer-tenured employees.
Employers doing a backsweep and worried about employees feeling frustrated should clearly explain the rationale to anyone affected, Patterson suggests.
“They’re concerned that employees will think, ‘You’re being Big Brother, and redeciding an issue that I’ve already decided for myself,’” he says. “It’s important to position this as something the plan sponsor is doing because it cares about employees’ financial well-being,” he says. “We also emphasize with employees that we don’t assume one size fits all, so we give them tools that they can use to figure out whether this is right for them.”

“When our plan clients implement automatic increases, we see 19% of participants opt out,” Principal Financial Group’s Jerry Patterson says. “Which means that 81% of participants are saying, ‘You know, you’re right, I can afford to contribute another 1% to my 401(k).’”
Always include automatic escalation in a re-enrollment, recommends Lynda Abend of John Hancock Retirement Plan Services. “If re-enrolled participants have a default deferral rate of 3% or 4% and no auto-increases, that is never going to get somebody ready for retirement,” she says. “Oftentimes, if you enroll employees at 3%, you are giving them the impression that is enough, when it isn’t.” A common emerging best practice is 1% auto-escalation a year, up to a 10% ceiling, she adds.
Among sponsors responding to the 2016 PLANSPONSOR Defined Contribution (DC) Survey, 58.3% do not perform auto-escalations. Just 17.5% default participants into auto-escalation at the time of enrollment unless they opt out, while 17.8% require participants to opt in to auto-escalation.
Corporate culture plays a big part in companies’ comfort with defaulting participants into auto-escalation, says Nathan Voris of Schwab Retirement Plan Services.
In terms of the impact on participants’ outcomes, he says, “It is almost always the right thing to do.” But employers’ philosophies still sometimes prevent them from going ahead.
Then, “there is that ‘Field of Dreams’ philosophy of ‘If you build it, they will come,’” Voris says, meaning participants will actively decide to opt in to auto-escalation. “For others on a plan committee, they are more paternalistic and comfortable with defaulting into auto-escalation. [However it is done,] there is no debating at this point that auto-escalation works.”