The Deification of DB-ification
I recently stumbled across another of those “DC plans are becoming like DB plans” articles — you know, the so-called “DB-ification” of 401(k)s? This is all supposed to be a good thing, of course, but is it?
We are, of course, routinely told that defined benefit plans do (or did) a better job of providing adequate income in retirement than defined contribution plans — though we aren’t generally reminded that that assumes that workers have actually managed to accumulate service credits sufficient to vest in those benefits, and that those programs are properly funded.
However, this interest in emulation of DB plans by DC plans is a relatively recent focus, fueled in no small part by the success of Pension Protection Act-engendered trends, primarily auto-enrollment (after all, nobody asks people to fill out a form to be covered by their DB plan) and asset allocation fund defaults (ditto on asking participants to choose the investments in the DB plan). But while it’s said that imitation is the sincerest form of flattery, it’s not like those auto-designs were actually copied from DB plans.1
Don’t get me wrong — anything that turns employees into participants (and automatic enrollment surely does that) and helps them make better investment decisions (and, generally speaking, asset allocation solutions fulfill that need) has to be a good thing. But to suggest — as many of these reports do — that these trends are essentially helping DC programs mature into their more “responsible” DB counterparts seems a gross misinterpretation of what is going on.
If we’re really looking to bring the best of the DB approach to DC plans, we need look no further than the definition of a defined benefit. Defined benefit plans are funded — at least, they are supposed to be — with an eye toward the benefit that will be paid out. As the name suggests, the benefit is defined — and the decisions that are made about how much to contribute to the plan and how those contributions will be invested are also done with that in mind.
Defined contribution plans, on the other hand — even the “automatic,” DB-ified ones — have an entirely different focus. They are (still) mostly focused on how much participants can afford to put into them (or can be forced to put into them without them opting out), not how much you need to get out of them. Oh, sure, the PPA’s safe harbor automatic enrollment — and a growing number of DC plans — includes a provision to increase those contributions on an annual basis. But is it any replacement for the kind of true funding discipline that a defined benefit focus represents? More importantly, will it be enough to provide the same kind of retirement security that the DB system promised?
Let’s not kid ourselves — when they worked (and they didn’t always), DB designs were “better” not because they made decisions for individuals (though that helped), but because somebody else was generally doing the funding,2
but more importantly doing so based on specifically targeted outcomes.
We’re not likely to shift the funding paradigm — but there’s nothing to keep us from emphasizing the focus on the ultimate benefit, the outcome that we hope to achieve — and the discipline to fund those DC accounts so that they can provide it.
Another DB “innovation” — the annuitization of the benefit — is certainly talked about (though not yet widely adopted) in the context of DC plans. Ironically, the trend in DB plans seems to be to replace that with the lump sum option so prevalent in DC plans.
2Another significant DB/DC difference — and one that tends to get glossed over — is that DB plans not only don’t ask employees to sign up or make investment decisions — they — at least the ones in the private sector — generally don’t ask participants to fund them. Oh, sure, that DC plan frequently comes with a match, but the primary source of funding for most of these programs lies with the participant.