By now we assume that most plan sponsors are well-acquainted with the automatic features of retirement plans. For those who are not, the thought of familiarizing themselves can be daunting. While auto-features can be extremely helpful, they typically work only if properly designed. So there are a number of factors that plan committees should consider.
To take a step back, it’s worth considering the impetus for them. Behavioral finance has received attention in recent years, as the economics profession migrated away from a perception of people as rational self-interested profit-maximizers and toward the view that we’re all actually riddled with biases that mess with our financial judgment. Daniel Kahneman is the founding evangelist of this discipline, and he was awarded a Nobel Prize in Economics for his work.
Behavioral finance theory has been applied directly to retirement plan design. Research detailed in “The Power of Suggestion: Inertia in 401(k) Participation and Savings Behavior,” a white paper by Bridgette C. Madrian and Dennis F. Shea, found that noneconomic factors can have a strong influence on straightforward economic issues—specifically with respect to the decisions participants need to make to maximize the likelihood of retirement success. For example, participants viewed default decisions as informed suggestions and deferred to them—which demonstrates the power and responsibility that sponsors possess.
The behavioral term that’s received the most attention lately is “nudge.” This refers to hints that are built into the choices we are required to make in every aspect of our lives. In terms of retirement plans, we could think of “auto” and default decisions as nudges—they’re not mandatory, but they represent the plan architect’s preferred choices.
Does the practice match the theory? Absolutely. Data show us that auto-features are the most effective plan design options that sponsors have. Automatic enrollment can add 15% to 30% participation when implemented properly.
The most common objections
Despite the proven results, whenever I discuss any auto-features with a committee, I know I’m going to get at least three objections:
1) It’s too paternalistic. The most common concern I hear is that employees will feel like they’re being told or forced to do something. My response is to remind the committee that employees are not being forced to do anything. Auto-enrollment requires a 30-day notification period. If they don’t want to join the plan, all they have to do is opt out.
Said differently, and in concert with the behavioral finance approach, a sponsor that offers an auto-enrollment plan is simply saying to its participants: “We assume you want to save for retirement, so that’s our default. If you don’t, just let us know.” Auto-enrollment is so common that most employees have already experienced it anyway. It simply isn’t viewed as paternalistic by employees.
2) Many of our employees can’t afford to contribute. Employers know their employees. They hear the stories and requests for raises, no question. If they assume from this that their employees can’t save, it creates a dangerous situation.
Again, if they can’t afford to save, they can opt out. I typically tell committees with this concern, about one of my clients—a manufacturing company. The vast majority of its workforce is lower-paid factory workers. The company had concerns about auto-enrollment but took the leap six years ago, defaulting at 4%. Today it has 93% participation. Only 7% of the workers have indicated they can’t save for retirement.
3) I will have a line of angry participants at my door. This concern typically comes from the person on the committee who interfaces with the employees and payroll. Visions of angry mobs of people wanting their money back dance in his head. In almost every situation, this never materializes. First, implementing auto-enrollment doesn’t affect the existing participant base—just new hires. Second, for those employers concerned about this, there’s an easy solution. Besides meeting the normal notification requirements, they can include a one-page notice in the new-hire kit. Print in big letters: “We have an automatic-enrollment retirement plan. If you do not wish to participate, you must opt out, or you will be enrolled at 4%.”
If you’re still reading, then maybe I have convinced you—or at least have your interest. Let’s next look at each auto-feature and the items a committee should consider when implementing it:
In terms of using participants’ inertia to their benefit, auto-enrollment is where the ball starts rolling. Sponsors need to ensure participants are auto-enrolled at the right level. Simply put, participants should, theoretically, be auto-enrolled at a level that ensures they’ll maximize any employer match. Unfortunately, there is a reality that must be addressed—the higher the auto-enrollment percentage, the higher the opt-out rate.
So if you auto-enroll at 3%, you can likely expect to see about 4% to 6% of your employees decline participation in the plan. If you auto-enroll at 6%, the number who decline will dramatically increase, on average, to over 10%.
Another consideration is that a successful auto-enrollment initiative will directly result in higher match costs. In theory, when sponsors offer a match, they do so with the assumption that everyone will use it. In reality, sponsors often use the total match cost from the year before as a guide for budgeting the following year. Implementing auto-enrollment can result in a pretty ugly surprise if the match cost spikes.
Lastly, auto-enrollment works best when paired with a shorter eligibility period. Trying to auto-enroll an employee after one year of service will dramatically decrease acceptance. Eligibility and entry periods should be considered as part of the auto-enrollment discussion.
Sponsors should also consider that auto-enrollment is not right for all situations. High-turnover organizations and companies with large numbers of lower-paid employees who work inconsistent hours will likely find auto-enrollment to be more of an administrative burden than a benefit.
New employees get the benefit of auto-enrollment, but automatic re-enrollment can sweep existing employees into the participant pool. In a re-enrollment, anyone who defers below the auto-enrollment rate is put through the process. As an example, assume we have a plan that auto-enrolls at 4% and the committee elects to do a one-time re-enrollment. All employees who currently defer below 4%—including those who defer nothing—are auto-enrolled at that rate. If they do not decline, they remain in the plan.
To be clear, this is a progressive approach. Auto-enrollment of new hires works partly because the employees are new. New job, new employer, new paycheck, new start. Re-enrolling existing employees does not benefit from this “newness” and will see significantly higher opt-out rates. This doesn’t mean we don’t recommend it, but we do tell our clients who are considering it that they will likely catch some feedback from employees.
Automatic escalation is the lesser-known little cousin of auto-enrollment. While auto-enrollment addresses the problem of getting participants to defer, it does not address getting them to defer enough to be able to retire comfortably. This is where auto-escalation comes in. Plans with auto-escalation automatically increase the deferral rate of their participants by 1% annually until they hit a maximum. The increase can happen on a single day each year or on the employee’s anniversary date.
Typically we recommend aligning auto-increase with annual compensation adjustments or the beginning of the year. Doing it at these times tends to be less noticeable to participants because their net paycheck is changing anyway.
While auto-escalation can be effective, it has not been as widely adopted as its more popular cousin. Generally, we recommend adding the two simultaneously.
There are a few other considerations for committees when discussing auto-escalation. The most important one is whether the auto-escalation will apply only to those employees you have auto-enrolled or to all employees. For example, if a plan has 4% auto-enrollment, but a participant manually enrolled at 6%, would that rate auto-increase? The answer may seem to be a simple yes, but that’s often not how the feature is set up. Sponsors should work with their recordkeeper and adviser to determine how best to set up auto-escalation.
Auto-features are some of the most effective plan design options we have—when designed and implemented properly. Sponsors should seek to build simple nudges into their plan’s choice architecture to steer participant decisions in a beneficial direction.
Andrew Zito, AIF, is executive vice president, retirement plan services, at LAMCO Advisory Services, an independent retirement plan consulting and advisory firm.
This feature is to provide general information only, does not constitute legal or tax advice, and cannot be used or substituted for legal or tax advice. Any opinions of the author do not necessarily reflect the stance of Institutional Shareholder Services or its affiliates.