Evidence on the relationship between employment shocks and preferences for redistribution is mixed - on stated outcomes - and sparse - on revealed ones. In an incentivized survey of US workers, we measure the relationship between repeated labour shocks and both stated and revealed preferences. We measure the former by support on seven different policies and the latter through donations. We examine experiences of both mild (having to reduce working hours) and hard shocks (unemployment), as well as past unemployment during formative years. We find evidence of adaptation to unemployment on policy preferences and compounding for milder shocks on donations, suggesting that repeated shocks are not independent in relation to preferences for redistribution. Our results show that unemployment may impact preferences in a self-interested way, while milder shocks may lead to broader support for redistribution.

---

1 October 2025

---

Journal of Economic Behavior and Organization 2382025

---

DOI: 10.1016/j.jebo.2025.107239

https://www.sciencedirect.com/science/article/pii/S0167268125003580