Central Bank Communication and House Price Expectations

How to cite

Kuang, Pei; Binder, Carola; Tang, Li. 2022. "Central Bank Communication and House Price Expectations." AEA RCT Registry. AEARCTR-0010024. https://doi.org/10.1257/rct.10024-1.1

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

Pei Kuang* , Carola Binder , Li Tang

* Primary investigator

Registry ID
AEARCTR-0010024
DOI
https://doi.org/10.1257/rct.10024-1.1
First registered
Last updated
Status
Completed
Trial dates
2022-02-21 – 2022-02-22
Countries
United States of America ()
JEL codes
D83, E31, E52, E71, R21
Registry record
http://www.socialscienceregistry.org/trials/10024

Abstract

This study analyses how US consumers form the expectation of house prices and respond to verbal and non-verbal communication of the central bank (Fed). Three surveys are implemented.

The first survey is an RCT. Prior to receiving the information treatment, all respondents provide their forecasts of national house price changes over the next 12 months, their inflation perceptions and expectations, their expectations and perceptions of the federal funds rate, and their response to a hypothetical investment question. The first treatment group is only told the level of the current federal funds rate. The second treatment group receives the same information, and also information about the projected federal funds rate over the next three years and in the longer run, from the Summary of Economic Projections. The third treatment group receives the same information as the second group, as well as a brief explanation of the mortgage rate channel of monetary policy. Respondents then provide their posteriors for the expectations of the variables of interest. We follow up with the respondents to our first survey about nine months later.

The second survey includes hypothetical vignettes in which we ask respondents to predict future house prices under different scenarios for future monetary policy. In the baseline scenario, participants are asked to imagine that the FOMC announces that it will keep the federal funds rate unchanged in the next meeting. In the shock scenario, respondents imagine that the FOMC unexpectedly raises the federal funds rate by 1 percentage point (p.p.). We use both open-ended and multiple choice questions to elicit the mechanisms that are on respondents’ minds when coming up with the prediction in the shock scenario, and we also ask about their personal experiences with different mechanisms.

Our third survey is an RCT which studies the response of consumer house price ex- pectations to part of Chair Powell’s speech at a recent press conference. The treatments are designed to allow us to distinguish between the effects of his wording, tone of voice, and visual appearance or body language. Prior to receiving the information treatment, all respondents provide their forecasts of house price changes over the next 12 months, expec- tations of 1-year ahead inflation, perceptions of current and recent changes in the policy interest rate, and 1-year ahead expectations of the policy interest rate. Respondents are randomly assigned to one of three treatment groups or the control group. The first treat- ment group receives a part of the transcript of the speech by Chair Powell which contains the announcement of the Fed’s decision on an interest rate hike and comments on recent economic conditions in the US. The second treatment group listens to an audio recording of the same speech. The third treatment group views and listens to a video recording of the same speech. Three questions test respondents’ understanding of the contents of the speech. Respondents then provide their posterior expectations and perceptions of the variables of interest. They also evaluate whether the wording, tone of voice, and body language of the Fed Chair send a positive or negative signal about future US economy.

Experimental design

Communication of current policy rates, interest rate hikes and explanation of the mortgage rate channel.

Design details

Randomization method
Randomisation by a computer
Randomization unit
individual
Clusters
0
Observations
3000 adults
Arms
700
Primary outcomes
House price forecasts

Cited by

No citation data available for this registration yet.

Keywords