ICI Viewpoints

Economic data make it clear: younger and lower-paid workers don’t place the same priority on retirement savings as their older and better-paid counterparts. The question for retirement policy is: are they behaving irrationally?

The author of an August 14 opinion piece apparently believes they are irrational. Criticizing a recent paper by me and my colleague, she falsely charges that the Investment Company Institute (ICI), the leading trade association for regulated funds globally, believes “steady saving for retirement is not important.” In addition to misrepresenting the study’s conclusion, the author accused us of using “selective information on participation rates” and “amazing” Social Security replacement rates.

Let’s set the record straight. ICI doesn’t play games with data, nor does it minimize the importance of retirement saving.

Our study starts with the participation rate in retirement plans—the fraction of workers either covered by a defined benefit (DB) plan or who contribute to a defined contribution (DC) plan. That statistic is often used to gauge the success of the US retirement system—but, as we show, this simple statistic understates the importance of these plans to retirees.

The key takeaway from our paper, “Who Participates in Retirement Plans, 2016,” is that more people benefit from employer plans than is commonly understood. This is for two reasons. First, the most widely cited statistics undercount participation. Second, many younger and lower-income participants who are not participating in retirement plans today will do so later in their careers or have a spouse who participates.

Looking beyond the headline statistic, we find that participation goes up with both income and age. Overall, 64 percent of workers aged 26 to 64 participate in a retirement plan, either directly or through a spouse. If we exclude workers with income less than $20,000 per person, the rate increases to 72 percent. These workers may rationally choose not to save for retirement because Social Security benefits will replace a large share of their earnings. If we also exclude younger workers (aged 26 to 45) with slightly higher income (up to $40,000 per person), the rate increases to 78 percent. These workers may rationally choose to delay saving for retirement until their earnings are higher and other priorities—education, starting a family, buying a house—have been taken care of.

So let’s look at the three criticisms raised about our study.

First, the suggestion that we were selective in reporting participation rates is unfounded. The retirement plan participation rates reported in the ICI study are derived from publicly available tabulations of Form W-2 data done by the IRS Statistics of Income Division. The IRS looks at all workers who received a Form W-2, which—as we explain clearly in our report—includes both private-sector and government workers. Using tax data, the IRS identifies active participants in retirement plans, which—contrary to what is stated in the opinion piece—does not include workers who were eligible to participate in a plan but chose not to. Active participants include workers who were covered by a DB plan, who had contributions (by either the employer, the employee, or both) made to a DC plan, or who had both. The only “selection” we did was on age, limiting the analysis to workers aged 26 to 64.

Second, the claim that our Social Security replacement rates were “amazing.” The replacement rates in our paper are derived from Congressional Budget Office projections. We chose to present replacement rates that reflect the design of the system—that is, benefits claimed at the full benefit age. Workers born in the 1960s, for example, will be able to claim full Social Security benefits at age 67, although they will have the option to claim as early as age 62 (with benefits reduced 30 percent) or as late as age 70 (with benefits increased 24 percent).

But what drives savers’ behavior is the pattern of Social Security benefits: replacement rates are much higher for workers with lower lifetime earnings. This holds true whether we present replacement rates for benefits claimed at age 62, 65, 67, or 70. Because of this pattern, when workers need to start saving for retirement differs substantially based on lifetime earnings—a point I have illustrated in academic articles and a book.

Finally, our conclusion. Contrary to our critic’s claim, ICI doesn’t believe that people shouldn’t save for retirement (they should). Nor do we state that the retirement system is perfect (it isn’t).  

Here is the final sentence from the study’s conclusion: “The incentives faced by both employees and employers should be taken into account when crafting pension reforms, and realistic goals should be set for increasing employer-based retirement plan coverage.”

In other words: let’s think this through before we act.

Peter Brady
Senior Economic Adviser
Investment Company Institute