Coronavirus Pandemic Shock, Economic Preferences and Beliefs: An Online Survey Experiment

How to cite

Jiao, Peiran; Duch, Raymond. 2020. "Coronavirus Pandemic Shock, Economic Preferences and Beliefs: An Online Survey Experiment." AEA RCT Registry. AEARCTR-0005731. https://doi.org/10.1257/rct.5731-1.0

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

Peiran Jiao* , Raymond Duch

* Primary investigator

Registry ID
AEARCTR-0005731
DOI
https://doi.org/10.1257/rct.5731-1.0
First registered
Last updated
Status
In development
Trial dates
2020-05-01 – 2021-05-01
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/5731

Abstract

Preferences and beliefs determine individual decision making. A growing body of research suggests that personal experience has a significant impact on these constructs. The diversity and breadth of the COVID-19 pandemic provide a unique, although grime, opportunity to identify the impact of catastrophic experience on economic reasoning and behaviour in both the short and long.

The COVID-19 pandemic has killed over a hundred thousand individuals around the world. For many this pandemic will have a particularly dramatic and gruesome effect because they live in or near one of the many epicentres. Many communities, though, have experienced comparatively minor – and in some cases essentially no – impact from this pandemic. We will explore this heterogeneity in online survey experiments in China, Italy, Chile, India, the UK and the US. We will use 1) geographical discontinuity and 2) “pre-treatment” measures of economic preferences to establish the causal influence of COVID-19 exposure on economic preferences and beliefs.

We will measure two groups of outcome variables. The first group represents the most widely studied economic preferences, namely risk aversion, time discounting and pro-social behaviour. The second group is belief updating, in both health and economic domains. We conjecture that individuals most intensely exposed to the CORVID-19 virus, and the resulting human suffering and deaths, will exhibit significantly different economic preferences and belief updating than those with much lower exposure. Our proposed field experiment is extremely timely and will provide novel insights into the economic aftermath of this tragic event, as well as necessary policy responses.

Experimental design

In our survey, we include an information treatment to test how people update belief given new information. We conduct this test in two domains of beliefs, health and economic respectively. In each domain, we provide two types of information, either neutral or positive. After reading the information, subjects are asked a belief question and an action intention question. Each subject gets one type of information in each domain. Each domain treatment is independent and within each domain subjects are randomly assigned to an information treatment.

In the health domain, subjects are randomly assigned to one of the following information vignettes:
Neutral: Some experienced medical scientists have started working on the vaccination for COVID-19.
Positive: Some experienced medical scientists have started working on the vaccination for COVID-19, and they have had a breakthrough finding.

Subjects are then asked about:
Their belief that there could be a COVID outbreak in the following year;
Their intention to get more private health insurance.

In the economic domain, subjects are randomly assigned to one of the following information vignettes:
Neutral: Some economists have studied the economic impact of COVID-19.
Positive: Some economists have studied the economic impact of COVID-19, and predicted that the economy will bounce back in the second half of 2020.

Subjects are then asked about:
Belief about stock market and unemployment in the following year
Intention to invest in financial markets

Design details

Randomization method
Randomization is done within the survey platform, Qualtrics. Before a respondent start the information treatment sections of the survey, the survey platform will randomly draw one treatment to appear.
Randomization unit
Randomization unit is done among survey respondents within each targeted district of the respective countries.
Clusters
We plan to distribute the survey in 12 provinces in Italy, 10 cities in China and 12 municipalities in Chile.
Observations
We plan to collect 150 valid survey responses within each of the targeted districts in each of three countries.
Arms
We will have two information treatments: in each treatment arm within each target district, we will have 75 observations per treatment arm.
Primary outcomes
We measure two groups of outcome variables: 1) economic preferences, namely risk aversion, time discounting and pro-social behaviour; 2) belief updating, in both health and economic domains.

Cited by

No citation data available for this registration yet.

Keywords