ICI conclusions rest on shaky assumptions and exaggerate levels of Social Security benefits.
The thrust of a recent Investment Company Institute (ICI) study – “Who Participates in Retirement Plans, 2016” – seems to be “Don’t worry if young people and others aren’t saving for retirement. Retirement saving isn’t high on their agenda. They will start saving when they feel they need to.” This sanguine assessment of participation in retirement plans is buttressed by selective information on participation rates in retirement plans and amazing replacement rates from Social Security.
The ICI report starts with the basic question of how many workers are participating in a retirement plan. The report turns to Statistics of Income (SOI) tax data, which show 57 percent of tax filers ages 55-64 were actively participating in a retirement plan in 2016. (This number is compared to participation data from the Current Population Survey, which all analysts agree have gone off the rails since 2013.) The implication is that we should be happy with 57 percent, because this number includes older and richer people for whom retirement saving is important and Social Security produces very high benefits for the rest.
A couple of points on the 57 percent. First, it includes public employees – virtually all of whom are active participants. If public employees are taken out of the equation, participation would drop to about 52 percent for private sector workers. In addition, “active participants” include workers who have a plan but do not make a contribution in a given year. Taking these people out further reduces the participation rate. Finally, participation rates are not improving: according to the SOI data the rate was 56 percent in 2008 and 57 percent in 2016. So, participation rates are simply not as high as ICI reports, and they are not getting better over time.
In terms of Social Security benefits, ICI shows a chart with a 63-percent replacement rate – benefits relative to pre-retirement earnings – for those in the middle of the earnings distribution. How is such a high replacement rate possible? Two steps. First, select a really low measure of pre-retirement earnings. Among the options are average earnings in the last five years, average wage-indexed lifetime earnings as used by Social Security, and average inflation-indexed lifetime earnings. Average inflation-indexed lifetime earnings produces the lowest number and is always the measure of choice for those saying everything is fine. Second, assume everyone claims benefits at the Full Retirement Age (FRA), which will be 67 for the 1960s birth cohort. In fact, as the folks at ICI well know, most workers claim benefits before the FRA, and low earners are particularly likely to claim as soon as benefits become available at 62. Benefits claimed at 62 are only 70 percent of those claimed at 67. So realistic Social Security replacement rates tell us that all but the very lowest earners need to save for retirement.
The bottom line is that, contrary to the thrust of the ICI study, continuous saving is important across the earnings distribution. Just because middle-income households don’t cite retirement saving as a major motivation for saving does not mean that they are correct. It is not enough for people to start saving when they feel the need to. At current 401(k) contribution rates, intermittent saving simply will not produce adequate resources.