Priming or learning? The influence of pension policy information on individual preferences in Germany, Spain and the United States

Abstract

A promising approach to pension policy preferences focuses on the influence of policy related information. We advance this research programme by examining the impact of information about future pension benefits, including whether information effects occur through priming, learning or both. Drawing on a novel, split-sample survey experiment in the US, Germany and Spain, we examine the impact of information on forecasted pension replacement rates for 2040 on pension policy attitudes. Findings indicate that the information treatment increases support for the two outcomes considered: (i) increases in the pensionable age and (ii) greater spending on pensions relative to other social programmes. Analyses of heterogeneous treatment effects accounting for prior beliefs of participants show that information effects occur both through priming and learning. The study concludes that hard, non-partisan information increases support for reforms that foster the financial sustainability of pension systems, although the scope of information effects depends on contextual conditions.