The Real ‘Coverage’ Gap

Teresa Ghilarducci doesn’t give American employers—or advisors—much credit. 

She recently penned an article in Forbes titled “Employers Can’t Provide Retirement Plans. Let’s Stop Pretending They Can.” The article’s premise is false on its face, of course. Employers[1] have long provided retirement plans, and by the hundreds of thousands. 
 
Her point appears to be that all employers won’t provide retirement plans, or at least to date haven’t. She attempts to prove her case by pointing to the coverage gap—a real issue, and one that we’ve written about previously. Unfortunately, she tries to make the case—and exaggerates it—by citing as a reference a dataset that has been drawn into question by none other than the non-partisan Employee Benefit Research Institute (EBRI), and more recently Forbes contributor Andrew Biggs (who in a recent Forbes article also cites the EBRI research). Relying on this dataset, Ghilarducci claims that retirement plan participation actually declined between 2015 and 2019. 
  

She has a solution, of course, though she doesn’t call it by name. In the article she claims that, “if every worker had an employer that contributed 3% of their salary in a retirement plan at work it would cost employers—well, her figures suggest it would be about half what they currently do. That’s not completely illogical—employer contributions averaged 5.2% of worker pay in 2018, according to the Plan Sponsor Council of America’s 62nd Annual Survey of Profit Sharing and 401(k) Plans. It’s no accident that Ghilarducci invokes a 3% rate—it’s the basic assumption underlying her Guaranteed Retirement Account (GRA) proposal, which she has advocated—with some “fluidity” in the underlying assumptions—for some time now. 

Indeed, not specifically mentioned in this particular article, but surely implicit in Ghilarducci’s premise that employers can’t/won’t offer retirement programs, is the mandatory contribution aspect of her GRA—oh, and the elimination of the current tax preferences of the 401(k) to “pay” for it.

More than that, one might well wonder if the employer “spend” Ghilarducci advocates is only half of the current rate, could it be “enough?” Well, a couple of years back I asked Employee Benefit Research Institute (EBRI) Research Director Jack VanDerhei to run the GRA program assumptions—for younger workers only (ages 26-30)—and asked him to compare that to what those same workers might get if they simply continued in their 401(k)s. To do so he took actual balances, contribution rates and investment choices across multiple recordkeepers from more than 600,000 401(k) participants, looking at those currently ages 26-30, including those with zero contributions, with 1,000 alternative simulated outcomes for stochastic rate of returns based on Ibbotson time series (with fees between 43 and 54 bps), including the impact of job change (an assumption was made that 401(k) participants would continue to work for employers who sponsored 401(k) plans), cashouts, hardship distributions, loan defaults, and with contributions based on observed participant data as a function of age and income and asset allocation based on observed participant data as a function of age. For the GRA, EBRI assumed no cashouts, hardship distributions or loan defaults (they aren’t allowed), assumed a deterministic 7% nominal return with no fees, and took their assumptions about the 3% mandatory contributions. And then compared the two outcomes at age 65.

The result? Well, let’s just say that if current rates of saving aren’t sufficient, 3%—even mandatory, and even with no leakage—won’t match the performance, the “contribution” of the 401(k). 

She is correct in that most of the current coverage “gap” is among employers with fewer than 100 workers, and that many of those employ workers that are lower-income, part-time, part-year workers. But she’s off base in her characterization of the participants in that system. According to EBRI, in 2016, more than 80% of 401(k) participants (for whom this information was available) made less than $100,000 per year, and more than half of 401(k) participants made less than $50,000. Even more importantly, moderate income workers participate when they have the option: In 2017, Vanguard reports that more than two-thirds of workers earning between $30,000 and $50,000 save in their 401(k). 

There is a coverage gap—though not as wide, nor as deep, as Ghilarducci assumes. In fact, the system she chooses to characterize as a “failure” works remarkably well for those who have access to it—thanks to the hundreds of thousands of employers that have… voluntarily, supported and encouraged by tax considerations, chosen to provide them. 

The real “coverage” gap is that they often don’t get the credit they deserve for doing so.

Footnote

[1] She not only has issues with employers. She goes on to claim that “Now we are left with self-interested advisors, tax breaks worth $250 billion (of which 70% go to benefit the top 20% of earners), and less than half the workforce with a retirement plan at work.”