Three, two, or only one cheer for 401(k)s?

Anita Monk and her neighbour John Everett dance during a Christmas Dinner event for older people.
Editor's note:

This article originally appeared on Real Clear Markets on January 11, 2017.

When an economist from AEI and journalists associated with places as varied as National Review and Mother Jones all say there isn’t a retirement crisis, can they possibly be wrong?

Last week, The Wall Street Journal reported that some of the early supporters of 401(k)s now lament “the 401(k) revolution”, because 401(k)s have been replacing traditional pensions instead of supplementing them.[1]  The article notes, accurately, that 401(k)s were originally designed for an entirely different purpose – to enable corporate executives to defer taxes on their bonuses – but that instead they have become the vehicle of choice for corporate retirement plans generally.

It then notes some of the many ways in which our 401(k)-centric retirement system differs from the – idealized – traditional pension:

  • Insufficient Saving for Retirement That employees, not employers, decide how much to save for retirement.  As a result, many observers believe employees are not saving enough, even after optional employer matches are taken into account, to account for our increasingly longer lives and for the more costly medical care that is now available.
  • Individual Retirement Accounts Subject to Market Crashes & Higher Fees Unlike traditional pensions, individual employees rather than employers got the benefit of market surges, but they also bore the risk of market crashes.  Employers also are able to use their greater expertise and scale to negotiate lower fees in traditional pensions.
  • Loss of Lifetime Protection An issue not mentioned is that 401(k)s, unlike traditional pensions, generally do not provide lifetime income, as very few include annuities.  The Employee Benefit Research Institute (EBRI) currently estimates that more than 40% of retirees will run out of money in retirement by outliving their savings, so this is more than a theoretical concern.

The article did admit that most private workers never had a traditional pension (38% in 1979, vs 13% today), that they weren’t portable when people changed jobs, and that in the go-go years of the 1980’s and 90’s the booming stock market made 401(k)s boom, too.

“Celebrating” 401(k)s   Criticism by its creators of what has become our dominant private retirement system struck a nerve.  Beyond the predictable reactions from 401k providers, the article provoked immediate criticism from columnists associated with publications as varied as Forbes, National Review, and Mother Jones.  (I think the best of these is that of Andrew Biggs, an AEI economist, in Forbes2.)  They said we should “celebrate the 401(k) revolution,” and had plenty of arguments:

  • The golden age of pensions wasn’t so golden.” At peak, traditional pensions covered at most 40% of working Americans.  When workers changed jobs, their pension didn’t follow them.
  • Employers have given up on traditional pensions, and they’re not going back.

Up ‘til now, I wouldn’t disagree.  Unfortunately, they went much further.

  • There’s no retirement ‘crisis’:

    • Retirement plans cover many more people today than pensions ever did.
    • Most of today’s retirees are comfortable.
    • Retirement savings are sufficient. They’re much greater now than they were in the ‘golden age of pensions’”.

I think both sides in this debate are kidding themselves.  I empathize with those who support traditional pensions, and think we should do everything possible to preserve the plans we have.  The traditional pension is an extraordinarily useful retirement plan, and remains so for tens of millions.

Nonetheless, traditional pension use will continue to decline and its advocates themselves helped bring that decline about.  Ironically, the very protections required for traditional pensions under the Employee Retirement Income Security Act of 1974 led most employers to decide they didn’t want to offer pensions at all.  It was ERISA that started the death of DB plans in the mid-1970’s[3]; the availability of a tax-deferred 401(k) alternative merely accelerated it.

“Yes, Virginia, there is a Retirement Crisis.”   But just as traditional pension advocates overstate their case, so, too, do the advocates of 401(k)s.  Here’s why:

  • People are worried   Many more people today who have yet to retire are worried about retirement than was the case in the 1970’s and 80’s.  For the past 15 years, people have listed running out of money in retirement as their #1 financial fear – more than half are worried about running out of money in retirement.[4]  In the late 70’s, by comparison, about a quarter felt that way.
    Many economists, who assume both that people do numbers and that their preferences are rational, think such fears are irrational.  They take comfort that models of optimal lifetime consumption predict consumption declines in retirement about as much as, on average, it does.[5]  From this, they infer that the level of retirement savings is sufficient.  It’s worth noting, however, that other economists disagree.[6]  Personally, I’m skeptical that people who don’t even bother to calculate retirement income nonetheless, miraculously, save optimally.

Those who claim there’s no retirement crisis point to surveys of people who are already retired, most of whom say they are comfortable.  I think these claims of comfort derive far more from social psychology than economics – they’d rather say they’re OK than admit they didn’t save enough after there’s nothing they can do about it anyway.  Also, many of those already retired have traditional pensions, which their children won’t.
No, retirement plan coverage hasn’t increased very much, if at all.  In the 1970’s just under half of all private sector workers were in some sort of retirement plan, most of them in a traditional pension that paid lifetime income.  Today, after the 401(k) “revolution”, the portion of all full-time private sector workers in an employer-provided retirement plan is (drum roll, please) just under half.[7]
Yes, more money is going into retirement plans, but is that because we have 401(k)s or because – since we’re living longer and investment returns are lower – we need to save more?  Saying that the 401k revolution caused increased retirement savings is probably backwards.  Had we not provided both legal & financial incentives to move from DB to DC, wouldn’t retirement savings have gone up anyway, and by more?  In the public sector, which largely retained traditional pensions, retirement costs have increased, too.[8]
And, by the way, are retirement savings growing enough?  Although retirement savings have grown, many observers believe they’re still insufficient to offset (a) the lower future investment returns that most of us now expect from all investments, DB or DC, (b) increased longevity after retirement, (c) the changes in administrative, marketing, and investment costs, and (d) the actual costs incurred by seniors, given that health care costs have risen more rapidly than general inflation.  If we have 10,000 people a day reaching retirement age and, as estimated by the Employee Benefit Research Institute, more than 40% of retirees will run out of money in retirement, should anyone pretend that retirement problems aren’t urgent?
And what about people who save less, who live longer than average or have higher than average medical bills?  After all, they’re only tens of millions of them.

Personally, I’m trying to be realistic, but also optimistic.  I empathize with traditional pension advocates yearning for an idealized past, but don’t want to join them with their heads in the sand.  DC plans are now the dominant private retirement vehicle.  That’s a fact and it will not change.

What we need is neither “celebration” nor handwringing, but real reforms.  It’s now clear that 401(k)s can be improved.  ERISA could be interpreted so that small businesses could join 401(k)s without becoming fiduciaries.  Automatic enrollment and automatic escalation of savings could become universal (or, dare I say it, maybe even required).  Lifetime income could become a part of all retirement savings.

Unfortunately, even these relatively modest changes have yet to occur.  The congressional debate on retirement security has been stalled over whether a 401(k)-type program should be universal, without even reaching the questions of what kind of retirement security it should provide.

I’m still hopeful that, in about 10-20 years, we will achieve 401(k)s with adequate savings and lifetime income security.  Nonetheless, I can’t bring myself to applaud a 401(k) model that will probably require several decades more to be efficient and which pretty much completely diverts all other retirement policy debates while doing so.