Inflation expectations and Policy preferences

How to cite

Bottan, Nicolas; Vera-Cossio, Diego; Hoffmann, Bridget. 2020. "Inflation expectations and Policy preferences." AEA RCT Registry. AEARCTR-0006743. https://doi.org/10.1257/rct.6743-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Nicolas Bottan* , Diego Vera-Cossio , Bridget Hoffmann

* Primary investigator

Registry ID
AEARCTR-0006743
DOI
https://doi.org/10.1257/rct.6743-1.0
First registered
Last updated
Status
In development
Trial dates
2020-12-22 – 2021-01-30
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/6743

Abstract

This online survey experiment has two objectives. First, to study how individuals update their inflation expectations. Second, to examine how these inflation expectations affect policy preferences and preferences for redistribution. Bayesian learning models are typically used when studying expectations, however the endogenous nature of individual's updating has not yet been addressed. We will launch a large-scale online survey experiment across 17 countries of Latin America and the Caribbean.

Experimental design

This information experiment is embedded in the BID/Cornell coronavirus follow-up study, that has the objective of studying how the coronavirus is impacting households in Latin America and the Caribbean.

Subjects are first be asked about their prior beliefs about inflation expectations over the next 12 months and the degree of confidence in their answer. Next subjects will be randomly assigned into four groups: control, information on past inflation, information on precision, information on past inflation and precision. After answering a section unrelated to this study, posterior beliefs are elicited (ie, they are re-asked the original question). After this, subjects are asked about perceptions about recovery of the economy and about policy and redistribution preferences.

Our main regression will exploit not only information provision, but heterogeneity in the precision of subjects in the guess they made during the baseline. This feedback on precision also has a random component as explained above. We will also exploit heterogeneity between countries, exploiting not only variation in information but also heterogeneity in responses by country and cohort characteristics (like inflation history).

Design details

Randomization method
Randomization done by the online survey platform
Randomization unit
Randomization is at the individual level
Clusters
Treatment is not clustered
Observations
Approximately 10,000
Arms
Approximately 2,000
Primary outcomes
Survey measures of inflation expectations
Secondary outcomes
Survey measures of policy preferences

Cited by

No citation data available for this registration yet.

Keywords