Behavioral Finance – the Next Frontier

All too often the innovations honored with a Nobel Prize fly under the radar of “regular” Americans. But that wasn’t the case last week when the work of University of Chicago’s Richard Thaler was acknowledged.

Thaler was, of course, recognized by the Royal Swedish Academy of Sciences, who said that his focus on limited rationality, social preferences and lack of self-control has “built a bridge between the economic and psychological analyses of individual decision-making.” More plainly, to my reading, Thaler (finally) managed to prove to economists that human beings don’t (always) act rationally and/or in their own self-interest.

Now, anybody who has ever actually interacted with human beings knows this. Indeed, in some ways the most amazing thing about Thaler’s insights of this reality is that it is seen as being innovative by economists.¹ I still remember reading the report that Thaler and Schlomo Benartzi authored way back in 2004, “Save for Tomorrow: Using Behavioral Economics to Increase Employee Saving.” That’s where (among other things) I first learned about the concept of what we today call contribution acceleration, based on the premise that people are more likely to act (and act more aggressively) on their good (but painful) intentions in the future than if they had to do so today.

There’s no denying that Thaler’s work has had a big impact on retirement savings (about $29.6 billion worth, according to one estimate). And if Thaler and Benartzi did not exactly create the notion of automatic enrollment, they at least freed it from the “dark” connotations of “negative election,” as it was called at the time.

Today we may wonder at – but no longer question – the notions that human beings rely on heuristics (mental shortcuts) when making complex decisions, that they fear loss more than they value gain, that they tend to diversify across the number of options provided, without regard to what lies within those choices, and that they tend to treat “old” money differently than “new” money. For this, Prof. Thaler and his collaborators over the years deserve our thanks.

That said, it may be worth remembering that while we tend to assume that plan fiduciaries are rational in all their decisions, they too are human beings making complex decisions. Consider that:

• Many remain hesitant to “impose” automatic enrollment for concerns about negative response from workers, though multiple surveys suggest workers would appreciate the move.
• Many continue to auto-enroll new hires, but not current workers.
• Many extend auto-enrollment to eligible workers – once.
• Many choose to implement auto-enrollment – and then wait 3 to 4 years to start contribution acceleration.
• Long-standing (and probably ill-considered) fund choices are routinely mapped during a recordkeeping conversion. Perhaps through multiple conversions.
• Plan committees often seem more worried about the negative reaction to removing a poor-performing fund than the possibility of being sued later on for keeping it on the menu.

That doesn’t mean that there aren’t any number of positive, rational reasons for those decisions (see “Why the ‘Ideal’ Plan Isn’t”). Indeed, most of us are rightly hesitant to superimpose our imperfect judgments on “other people’s” money – even on those on whose behalf fiduciaries are admonished to act.

But as we commemorate – and celebrate – those behavioral finance “nudges” that have done so much to buoy individual retirement security, perhaps some of those fiduciary decisions are worth (re) considering as well.

- Nevin E. Adams, JD

3 Things You Should Know About Automatic Enrollment

One of the most celebrated plan design features of the 401(k) era is automatic enrollment. Nearly as old as the 401(k) itself, once upon a time it was called a “negative election.” But regardless of the name, the concept has been extraordinarily effective at not only getting, but keeping, workers saving via their workplace retirement plans.

However, adoption of the design, after a surge in the wake of the passage of the Pension Protection Act of 2006, now seems to have plateaued. Moreover, current data suggests that, while automatic enrollment adoption has certainly had a positive impact on retirement outcomes, we’re not getting as much mileage from it as we might.

So, here are three things that plan sponsors — and others — should know about automatic enrollment.

1. You don’t have to default contributions at 3%.

Three percent was the standard default contribution rate for automatic enrollment plans long before the Pension Protection Act of 2006 incorporated it as part of its auto-enroll safe harbor. Originally cited in a now-obscure IRS regulation years before the advent of the PPA, in the years to follow, it was largely embraced because it was seen as little enough that it wouldn’t spur massive opt-outs by automatically enrolled participants.

With more than a couple of decades of experience under our belts (a third of that under the auspices of the PPA), we know a couple of things. First, that 3% is indeed too small an amount to spur most auto-enrollees to opt out. In fact, there have been any number of studies — and some real-world experience — suggesting that a defaulted contribution rate twice as high would produce very nearly the same result.

What many plan sponsors may not know is that while that the auto-enrollment safe harbor of the PPA calls for a minimum starting deferral of 3%, it is a floor, not a ceiling.

But another, and more important, thing that we’ve all known from the very beginning is that a 3% rate of deferral is not enough.

2. If you are going to default at 3%, make sure you accelerate the contribution rate.

Automatic enrollment and contribution acceleration have always been separate things: the former a decision made by the plan sponsor, with the participant having the ability to opt-out; the latter a voluntary decision by the participant, facilitated by the plan sponsor.

These two traditionally separate concepts were wedded in the PPA’s automatic enrollment safe harbor, and for a very sound reason: As noted above, a 3% deferral is not enough.

Current survey data suggests that plan sponsors continue to separate these two design choices, and that tells me two things: 

• most plan sponsors who adopt automatic enrollment designs aren’t doing so with an eye toward taking advantage of the PPA safe harbor; and
• while many plan sponsors are willing to make one monetary decision on behalf of their workers, they apparently aren’t nearly as willing to make two, however integral it might be to retirement security. 

3. Automatic enrollment isn’t just for new hires.

Despite our extended history with automatic enrollment, and the PPA’s safe harbor that contemplates its extension to all eligible workers, to date most plans — roughly two-thirds — that have adopted automatic enrollment have done so only for new hires. Over the years, I have heard a variety of explanations for this trend — anything from a hesitation to “suddenly” take contributions from long-time workers (who have ostensibly declined to take advantage of previous opportunities to join) to a general resistance to running the risk of stirring up trouble with existing workers.

However, to me, the most logical explanation is economic: Just take the likely increase in participation rate (generally from 70% to 95% or so) resulting from automatic enrollment, and then figure out the increase in matching dollars that would result, particularly for an employee population that is likely more tenured and highly compensated.

Little wonder that so many decide to “let sleeping dogs lie.” Though in all my years of experience working with 401(k) plans, I have never heard of even a single participant who objected to being automatically enrolled. On the other hand, I’ve heard dozens of stories of long-tenured workers who had, for a variety of reasons, put off signing up for their 401(k) — but who, when they finally were enrolled, were oh-so-very-grateful for that “start,” however delayed.

And that, perhaps as much as anything else, is something plan sponsors should know.

- Nevin E. Adams, JD