Are Participants Really Cutting Back on Savings?

The new study, by the Vanguard Center for Retirement Research, acknowledges that last year 3.1% of 1.3 million participants in defined contribution plans administered by Vanguard stopped making contributions, and that that was up by 0.7% from the prior year. Additionally, the study acknowledges that, in any given year a small percentage of active participants ceases contributing, and that the economic and market downturn in 2008 appears to have raised that rate slightly.

However, the study says that while “recent surveys have suggested that a large group of households have ended retirement savings programs during the market turmoil of 2008,” the Vanguard data suggest a different conclusion. Those recent surveys included “Retirement Security or Insecurity? The Experience of Workers Aged 45 and Older” by AARP, which was cited in last fall’s hearings by the House Education and Labor Committee by Congressman George Miller (D-California) (see “Congress Considers Market Impact on Retirement Security“), and “Economic Downturn Puts Strain on Already Insufficient Retirement Savings” by TD Ameritrade Holding Corp. (see “Some Too Strained to Save“).

Vanguard notes that their analysis of administrative data using Vanguard DC plans suggests that “at least among DC plan participants, inertia remains quite powerful, so few active participants have ended their savings program.” The study notes that in 2006 and 2007, both “relatively benign periods for the economy and for financial markets,” about 2.5% of participants who were active contributors at the beginning of the year ceased contributing to their DC plans by the end of the year, even though they remained active employees and eligible to continue contributing to the plan.

Those figures weren’t much different than the 3.1% who took the same action in 2008.

Vanguard’s researchers—Cynthia A. Pagliaro and Stephen P. Utkus—dug deeper into the trend, noting that a month-by-month analysis of participants who stopped contributions uncovered a “seasonal pattern, with the number of those ending contributions increasing toward the end of the calendar year.”

The study notes that more research is needed to fully understand that dynamic, but noted that “it may be just a coincidence that the rate of discontinued contributions rises near the holidays, typically a time of competing spending priorities.” However, the researchers do acknowledge that perhaps the most prominent feature in the month-by-month analysis is the jump in those ending contributions in November. “One likely explanation is that participants were reacting to exceptional market volatility during that month,” they note.

Demographic Dynamics

Compared with participants who maintain their contribution levels, active participants who ceased making contributions tend to be slightly younger and less tenured, with lower household incomes. “At the same time,” Vanguard notes, “more than three-quarters of those ending contributions had household incomes of $50,000 or more, suggesting that the decision to stop saving is not isolated to lower-earnings households.”

Over three years, participants who continued to save in these plans (excluding new hires during the 2006 to 2008 period) saw contributions grow by 5% to 10%, though that increase reflects both wage increases and any increase in contribution rates.

As for what might account for the differences between this study and the others noted above, Vanguard says that might be due to a couple of factors:

• Other survey results include any type of retirement saving, “and so it reports on individuals ending discretionary retirement savings programs, like an occasional contribution to a savings or investment account;” or
• The other survey data are “partially capturing changes in intentions—households that were thinking of saving but chose not to.”

The Vanguard researchers also acknowledge the gap between self-reported behavior in surveys versus the reality in action evidenced on the recordkeeping systems where those intentions may—or may not—be acted on.