Peer Effects Among Teachers: A Study of Retirement Investments

Using a unique matched-panel dataset that combines detailed demographic information from the Florida Department of Education's annual survey of school districts with investment information from the Florida State Board of Administration for 2002-2009, this paper looks for the presence of peer effects among Florida Defined Contribution plan participants at the school level. Overall, the regression results suggest a small, but significant peer effect on asset allocations and activity level. A 1 standard deviation change in initial peer equity yielded a 0.04 standard deviation change in equity reallocation. This is in comparison to a 1 standard deviation change in the individual's own initial equity allocation, which would yield a 0.16 standard deviation change in equity allocation. For activity level, a 5.2% change in peer activity level increased the likelihood of a participant being active by 1.8%. The findings are reinforced by similar analysis using false peer groups. These findings suggest the presence of a social multiplier for coworker investment decisions.