If you are a regular reader of Defined Benefit, then you already know that we don’t care for defined contribution plans like 401(k)s that have many failings when compared to traditional pensions. One of these failings is that they provide little in retirement security for low-income working families.

Let’s take a look back at the history of the 401(k) to get a better picture. The 401(k) was never intended to be the primary workplace retirement savings plan for most workers. It was initially created as a minor provision in the tax code to allow wealthy corporate executives to put a portion of their high salaries in tax-advantaged savings accounts. In the 80’s the Reagan administration changed their interpretation of the minor tax law provision regarding 401(k) accounts. Afterwards, more companies began offering them to their employees. This new interpretation, coupled with a series of laws that drove the decline of private sector pensions, has dramatically moved more and more workers away from defined benefit pensions to defined contribution 401(k) plans over the last three decades. The results of this shift have not been good.

A report from the Center for Retirement Research at Boston College examines the question of why low-income workers don’t have workplace retirement plans. They explain that there are four links in the chain of having a workplace retirement plan:

  1. A worker must have a job
  2. That job must offer a retirement plan
  3. The worker must be eligible to participate in that plan
  4. If eligible, the worker must choose to participate in the plan

The researchers found that only 22 percent of low-income workers participated in a workplace retirement plan. The weakest links in the chain were having a job and having a job that offers a retirement plan. Eligibility rates and take-up rates (choosing to participate in the plan) were high and were similar to higher-paid workers. However, when the researchers removed employers that offer pensions and focused only on employers that offer 401(k)s, the participation rate dropped even lower, down to a frightening 12 percent. This is largely driven by the fact that the take-up rate is lower with 401(k) plans.

A recent study looks at this same issue from a different angle. The study, from a professor at the University of Kansas and a researcher at the Social Security Administration, examines 401(k) participation rates based on education. What they found largely tracks with the study from CRR. The less education a worker has, the less likely that worker is to participate in a 401(k). In fact, a college graduate saves 26 percent more than a high school graduate. Again, part of the issue is access: 83 percent of college graduates have access to a workplace retirement plan compared with only 62 percent of high school graduates and 43 percent of high school dropouts. Even if workers with different levels of education both have access to a 401(k), the college graduate is more likely to participate and is likely to contribute more to her 401(k). Over time, these different savings levels are going to widen, increasing inequality in retirement. The data already tells us this is happening and it’s only going to get worse as future generations of retirees rely more heavily on 401(k)-style plans.

The take-away from both of these studies is this: the move away from defined benefit pensions has been particularly bad for low-income workers. The authors of both reports advocate for auto-enrollment of workers in 401(k) plans, something that defined benefit pensions already have. Auto-enrollment eliminates the need for workers to choose to participate. With traditional pensions, they also don’t have to decide how much to contribute and they don’t have to make any investment decisions; those decisions are made by professionals. The U.S. has been participating in a three decades long experiment with 401(k)s. The results of these studies prove, yet again, that it’s time for us to admit: 401(k)s have failed. And they’ve especially failed low-income workers.