Information and Social Interaction in Savings Decisions in the United States
Elderly poverty in the U.S. decreased dramatically during the twentieth century; from 1960 to 1995, the official poverty rate of those aged 65 and above fell from 35% to 10%. Social Security is often credited as a partial contributor to the decline.1 Despite this, for most U.S. families, employers’ pensions and not Social Security are the main source of cash income during retirement, and many of these pension plans are now reliant on voluntary employee enrollment. Individual choice about how much and when to save provides flexibility and autonomy, but low levels of savings can lead to crises in elderly households, and strains on the economic system which must support them.
Over the last 25 years, traditional pension plans’ mandatory employee participation have been partly replaced with Tax Deferred Account (TDA) retirement plans such as 401(k)s, where employees choose whether to participate and how much to save. As a result, most U.S. workers now have to make a decision about how much to save for their retirement, instead of being passive participants in their employer’s pension plan. This makes it very important for individuals to understand how retirement savings decisions are made. Deciding how to save for retirement requires complicated calculations and knowledge about financial options to reach optimal outcomes. Thus, information about these calculations and rules, as well as about others’ decisions, should have a large impact on savings behavior. While 71% of Fortune 500 Companies systematically hold financial information sessions, the meetings are often poorly attended.
This evaluation studies the influence of information and social networks on university employees' decisions to enroll in a voluntary TDA retirement plan. In an effort to increase TDA enrollment, this university set up a benefits fair to disseminate information about different benefits plans offered by the university.
Researchers offered a $20 reward for attending the fair to a random group of employees within a subset of randomly selected university departments. There were two distinct groups studied in the experiment: 2,039 employees who received the letter promising the monetary reward for attending the fair, and people in the same department as someone who received a letter. The study traced the reward's effects on recipients' fair attendance and TDA enrollment as well, as the social network's effects on fair attendance and TDA enrollment of non-recipients in the same department as reward recipients.
Small financial incentives successfully induced treated employees, as well as members of their peer groups, to attend a benefits fair. Twenty-one percent of individuals affected by the experiment, whether directly or indirectly through their social network, attended the fair, compared to only 5% in comparison departments. People in treated departments were also more likely to enroll in the TDA after the fair. The program increased participation after one year with 5% of the 4,000 non-enrolled employees in treated departments compared to 4% of the comparison departments.
The direct effect is no larger than the indirect effect: in treated departments, those who received the letter and those who did not are about as likely to subsequently enroll in the TDA. This suggests that an individual’s decision to participate in the TDA is affected by small changes in the environment, and not only by the information content of the fair. However, these effects remain very small compared to the changes in enrollment which can be effected by altering default enrollment rules.
1 National Bureau of Economic Research (NBER), “Social Security and Elderly Poverty,” http://www.nber.org/aginghealth/summer04/w10466.html.
