Effective Policy Communication: Target vs. Instrument

How to cite

D'Acunto, Francesco; Weber, Michael. 2020. "Effective Policy Communication: Target vs. Instrument." AEA RCT Registry. AEARCTR-0006597. https://doi.org/10.1257/rct.6597-1.0

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

Francesco D'Acunto* , Michael Weber

* Primary investigator

Registry ID
AEARCTR-0006597
DOI
https://doi.org/10.1257/rct.6597-1.0
First registered
Last updated
Status
On going
Trial dates
2020-06-01 – 2021-01-31
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/6597

Abstract

Communication targeting households and firms has become a stand-alone policy tool of many central banks. But which forms of communication, if any, can reach ordinary people and manage their economic expectations effectively? In this large-scale randomized control trial, we provide randomly three groups of a representative population of men in Finland with different forms of communication about the same policies that the European Central Bank (ECB) has implemented in April 2020 to reduce the negative shocks due to the COVID-19 pandemic. The communication about the same policies comes from the same source and medium (the twitter account of the Governor of the Bank of Finland, Dr. Olli Rehn), but one group faces a form of target communication, which only discusses the final objective of the policy without discussing the legal instruments the ECB will use to reach such objective, and another group faces a form of instrument communication, which discusses the technical instruments the ECB uses and which form the policy. A control group of subjects faces a tweet discussing a period of economic crisis in Finnish history without any specific connection to the COVID-19 policies implemented by the ECB. Our trial aims to study the effects of different forms of policy communication on the average consumer as well as the heterogeneity of these effects across dimensions we observe in the data---cognitive abilities as a proxy for sophistication and prior income-change beliefs.

Intervention

We propose a within-subject experimental design with three experimental arms: a control group, in which subjects read a tweet by Dr. Rehn referring to the economic crisis of Finland after the First World War and unrelated to current events and policies; a target-communication group, in which subjects read a tweet that discusses the objectives of the ECB's policies implemented in April 2020 to overcome the negative economic shock due to the COVID-19 crisis; an instrument-communication group, in which subjects read a tweet that discusses the legal instrument and the size of the PEPP program the ECB implemented in April 2020. We elicit subjects' beliefs about their households' income both before and after the experimental treatments are administered. To reduce the concern of demand effects, we elicit point estimates in the first instance and full probability distributions ("Manski question") after the treatment.

Experimental design

The design is a within-subject experimental design with 3 experimental arms: a control group, a treatment-communication group, and an instrument-communication group. We do not randomize or stratify any other dimensions.

The manipulation consists of showing subjects a tweet from Dr. Olli Rehn's twitter account that discusses a situation of economic crisis in the history of Finland (control group), the objectives of the ECB policies implemented to overcome the COVID-19-induced economic crisis, and the instruments (PEPP) implemented by the ECB to overcome the COVID-19-induced economic crisis. All tweets are true tweets from the official account of Dr. Rehn.

Design details

Randomization method
Randomization is performed by a computer at Statistics Finland after subjects are recruited for a survey experiment.
Randomization unit
Individual
Clusters
5100
Observations
5100
Arms
1700 in the control group, 1700 in the target-communication group, and 1700 in the target-communication group
Primary outcomes
1. Change in numerical household income expectations 2. Change in trust towards the central bank and its policies 3. Change in qualitative household income expectations
Secondary outcomes
NA

Cited by

No citation data available for this registration yet.

Keywords