Federal Reserve Data Offers Sweeping View of DC Plans

“The release of the Federal Reserve’s 2016 Survey of Consumer Finances (SCF) is a great opportunity to see how a strengthening economy, the continued maturation of the 401(k) system, and steady stock market returns have affected workers’ retirement wealth,” say researchers with the Center for Retirement Research (CRR) at Boston College.

According to CRR white paper “401(k)/IRA Holdings in 2016: An update from the SCF,” the big advantage of the survey is it provides information not only on 401(k) balances—much of which is available from financial services firms—but also on household holdings in individual retirement accounts (IRAs), which are largely rollovers from 401(k)s.

“Essentially, 401(k)s serve as the collection mechanism for retirement saving, and IRAs serve as the resting place,” the report explains.

The good news in the updated Federal Reserve data is a slight increase in participation rates and greater use of target-date funds; the bad news is flat total contribution rates, persistently high fees and significant leakage.

“The SCF shows, for households approaching retirement, an increase in 401(k) plan balances from $111,000 in 2013 to $135,000 in 2016,” CRR researchers . “But only about half of households have 401(k)/IRA balances, and, as defined benefit [DB] plans phase out in the private sector, the rest will have no source of retirement income other than Social Security.”

The analysis concludes that 401(k) plans “could work much better and balances would be higher if all plans were fully automatic.” Particularly important are mechanisms such as automatic enrollment—for both existing and new employees—and automatic escalation in the default contribution rate. Additionally, it would be helpful if “default contribution rates were set at realistic levels,” i.e., much higher.

Crucial to note, CRR says, is that participation rates in plans without auto-enrollment actually declined between 2013 and 2016. “To the extent that plans without auto-enrollment constitute a larger share of total participants than is often reported, the decline in their participation rate would noticeably slow the pace of improvement,” the CRR concludes.

Also troubling, average employee contribution rates declined between 2015 and 2016.

“The decline can be attributed mainly to auto-enrollment, which increases participation rates but has a depressing effect on contributions,” CRR notes. “The reason is that default contribution levels are often set at 3% or lower, and [as] less than 40% of plans with auto-enrollment have auto-escalation in the default contribution, many of those who are enrolled at low contribution rates remain at those rates. Employer contributions bring the total average deferral rate to around 11%.”

The full brief, including dataset, is available for download here.