Policy Uncertainty and Information Flows: Evidence from Pension Reform Expectations

Working Paper: CEPR ID: DP13988

Authors: Marco Francesconi; Emanuele Ciani; Adeline Delavande; Ben Etheridge

Abstract: Subjective expectations about future policy play an important role in individuals’ welfare. We examine how workers’ expectations about pension reform vary with proximity to reforms, information cost, and aggregate information acquisition. We construct a new pan-European dataset of reform implementations and government announcements, and combine it with individual-level representative survey data on expectations about future reforms and country-level data on online search. We find: (1) Expectations are revised upward by about 10 percentage points in the year leading up to a reform, from a median of 50%, regardless of whether the reform is announced; (2) Aggregate online search increases after announcements, when the cost of information is lower; (3) Reform announcements and online information gathering are substitutes in the formation of expectations; (4) Expectations do not converge as a result of announcements or implementations; (5) The effect of information on expectations varies substantially across workers and systematically with observed characteristics that proxy cognitive ability and information value. These findings, interpreted using a model of rational inattention, reveal substantial informational rigidities, with welfare costs that run into trillions of Euros.

Keywords: Expectations; Retirement; Pension Reform; Uncertainty; Reform Announcement; Online Search; Rational Inattention

JEL Codes: C8; D84; D91; J14


Causal Claims Network Graph

Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.


Causal Claims

CauseEffect
Reform announcements (E69)Online information gathering (L86)
Announcements (Y70)Convergence of expectations (D84)
Information (D83)Expectations (D84)
Proximity of a reform (P41)Increase in expectations (D84)
Announcement of reforms (E69)Increase in online search (L86)

Back to index