Complexity in Central Bank Communications

How to cite

Godinho, Frederico. 2022. "Complexity in Central Bank Communications." AEA RCT Registry. AEARCTR-0010080. https://doi.org/10.1257/rct.10080-1.0

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

Frederico Godinho*

* Primary investigator

Registry ID
AEARCTR-0010080
DOI
https://doi.org/10.1257/rct.10080-1.0
First registered
Last updated
Status
Completed
Trial dates
2022-02-25 – 2022-03-11
Countries
United States of America ()
Registry record
http://www.socialscienceregistry.org/trials/10080

Abstract

Using data from a new small-scale survey ran under a Randomized Control Trial (RCT) framework, I measure the differences in treatment effects that come from different FOMC statements that express the same message, only changing the level of complexity in the choice of words used. The three treatments are based on the actual FOMC statement from December 15, 2021, and express the same message; however, the first treatment (T1) expresses the message with simplified language, and the third treatment (T3) with complex language. The second treatment (T2) is the actual FOMC statement.

I am consequently able to look at how the general public, a non-specialized audience, reacts to distinct policy statements that only differ in the complexity of the words used. For this audience, a simpler message is more effective in shaping households’ expectations and enhancing understanding, but less effective in stimulating trust in the central bank.

Experimental design

The survey was ran in March 2022, and it is divided in three parts: pre-treatment questions, presentation of treatment, and post-treatment questions. The three treatments are based on the actual FOMC statement from December 15, 2021, and express the same message; however, the first treatment (T1) expresses the message with simplified language, and the third treatment (T3) with complex language. The second treatment (T2) is the actual FOMC statement. The respondents were randomly assigned to one of four groups: T1, T2, T3 or control group, to whom no statement was shown.

Design details

Randomization method
Randomization done by a computer.
Randomization unit
Individual.
Clusters
1
Observations
2000
Arms
500 per group (Treatment 1, Treatment 2, Treatment 3, Control)
Primary outcomes
Understanding of central bank communications; engagement with the central bank; inflation expectations; GDP growth expectations

Cited by

No citation data available for this registration yet.

Keywords