This paper asks whether expanding pension benefits increases political participation among retirees, testing policy feedback and habit-formation theories with a causal design. The authors leverage a large pension reform in Illinois that raised benefits and use a regression discontinuity approach on administrative records to estimate turnout effects among retired public employees across multiple elections.
Regression discontinuity design Administrative data covering 93,000 public employees No detectable turnout effect following the benefit increase
Data and methods
The study uses administrative data on roughly 93,000 public employees affected by an Illinois pension reform. The reform increased benefits by 36%, and the authors apply a regression discontinuity design to compare turnout for those just above and below the eligibility cutoff across twelve elections. The design is presented as a causal test of whether material benefit gains change retirees' participation.
Main findings
Across the analyzed elections the authors report no measurable effect of the pension increase on voter turnout. The null result holds in subgroup analyses noted in the abstract, including low-income groups, and is interpreted relative to expectations from policy feedback and habit-formation accounts.
- Treatment: a 36% benefit increase from an Illinois pension reform.
- Design: regression discontinuity on administrative turnout records across twelve elections.
- Outcome: no detectable turnout change, including among lower-income retirees.
