Obesity and Income: Wealth Signaling with Body Mass

Last registered on July 02, 2019

View Trial History

Pre-Trial

Trial Information

General Information

Title

Obesity and Income: Wealth Signaling with Body Mass

RCT ID

AEARCTR-0004332

Initial registration date

June 21, 2019

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published

July 02, 2019, 3:20 PM EDT

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Locations

Country

Malawi

Region

Salima

Primary Investigator

Name

Elisa Macchi

Affiliation

MIT Department of Economics

Contact Primary Investigator

Other Primary Investigator(s)

Additional Trial Information

Status

Completed

Start date

2019-06-22

End date

2019-06-23

Keywords

Health

Additional Keywords

JEL code(s)

Secondary IDs

Abstract

The alarming obesity rates and consequent major-chronic-diseases incidence in developing
countries have lead to numerous calls for targeted obesity prevention policies; most
interventions’ focus is Pigovian taxation. In the presence of social-image concerns, however,
the standard incentive-based taxation is not optimal and instead, norm-based interventions
might prove effective. Given the positive correlation between body mass and wealth in developing
countries, high body mass is likely to be a positive signal for socio-economic status,
yet this has never been empirically tested. This project proposes an experimental design to test whether individuals perceive body mass as a signal of socio-economic status, in low-income contexts. In the experiment, I randomly assigning photo-morphed variation of the same portrait and elicit perceived
income. To verify robustness of the findings, I investigate how the comparative statics when I experimentally reduce asymmetric information about wealth. Lastly, I investigate individuals' beliefs on returns from high body mass in dating and credit markets, as well as whether individuals with high body mass have a beauty or respect premium.

External Link(s)

Registration Citation

Citation

Macchi, Elisa. 2019. "Obesity and Income: Wealth Signaling with Body Mass." AEA RCT Registry. July 02. https://doi.org/10.1257/rct.4332-1.0

Former Citation

Macchi, Elisa. 2019. "Obesity and Income: Wealth Signaling with Body Mass." AEA RCT Registry. July 02. https://www.socialscienceregistry.org/trials/4332/history/49149

Sponsors & Partners

Experimental Details

Interventions

Intervention(s)

Intervention Start Date

2019-06-22

Intervention End Date

2019-06-23

Primary Outcomes

Primary Outcomes (end points)

The primary outcomes are portrayed individuals’ perceived wealth; Perceived Dating Opportunities; Perceived Status; Perceived Creditworthiness and Perceived Attractiveness.

Primary Outcomes (explanation)

Perceived wealth: 0-1 continuous variable; standardised perceived number of individuals in the participant's village who would rate the portrayed individual as rich, as perceived by the participant.

Secondary Outcomes

Secondary Outcomes (end points)

Secondary outcomes investigate are perceived creditworthiness, dating opportunities and respect in the community. These outcomes are meant at capturing the returns from being overweight/obese.

Secondary Outcomes (explanation)

Perceived Status: 0-1 continuous variable; standardised perceived number of individuals in the village who would follow the portrayed individual monitions.
Perceived Creditworthiness: 0-1 continuous variable; standardised perceived number of individuals in the village who would lend money to the portrayed individual.
Perceived Attractiveness: 0-1 continuous variable; standardised perceived number of individuals in the village who would find the person attractive.
Perceived Status: 0-1 continuous variable; standardised perceived number of individuals in the village who would respect the portrayed individual authority.
Perceived Dating Opportunities: 0-1 continuous variable; standardised perceived number of opposite-gender individuals in the village who would go on a date with the portrayed individual.

Experimental Design

In this experiment, I investigate whether high body mass is perceived as a wealth signal in Malawi. The experiment tests 3 hypothesis: 1) the causal effect of body-mass on perceived wealth
(second order beliefs); 2) the causal effect of wealth information provision on the relationship
between body mass and perceived wealth; 3) the existence of beliefs on positive returns from body mass.

Experimental Design Details

Each participant is randomly assigned to one single portrait and is asked to answer to five questions, which elicit his/her second order beliefs on four different characteristics of the portrayed
individual: creditworthiness, dating opportunities, attractiveness, wealth and status. The four questions are as follows:
Q1) Wealth: Out of 10 people, how many would think that PIC* is rich?
Q2) Status: PIC states that is morally unacceptable to litter. Out of 10 people, how many
would follow this monition?
Q3) Attractiveness: Out of 10 people, how many would think that PIC is rich?
Q4) Creditworthiness: Imagine PIC asks for a small loan to a friend or a micro-financing
agency, which he/she will return on his/her way back. If he/she was to ask to 10 people,
how many would give him/her the money?
Q5) Dating: How many women/men would go on a date with PIC?
The experiment exploits a two-by-two design. The main treatment effect is the variation in portraits’ body mass. I construct a set of ten pictures pairs from ten original portraits of individuals living in Kampala, Uganda.The second variation relates to the level of wealth asymmetric information. Asymmetric
information on the actual socio-economic status of the portrayed individual comes from the
information provided with the portrait. Half of the participants are randomly assigned to Questionnaire
A, the remaining half to Questionnaire B. The paradigms are identical, except that in
Questionnaire B participants are provided information that the portrayed individual owns a car.
The variation is obtained as follows:
• Questionnaire A: PIC is a men/woman from Lilongwe. PIC is traveling for
work and stops at your village. We would like to ask you a set of questions on PIC. As
these questions are about perception, there is no right or wrong answer.
• Questionnaire B: PIC is a men/woman from Lilongwe. PIC is traveling for
work and stops at your village. PIC enters the village on his/her car. We would like to
ask you a set of questions about PIC. As these questions are about perception, there is no
right or wrong answer.

* PIC describes the person in the picture.The MDE for the main outcome is 0.45 sd. or a 1.12 change in the number of individuals who would rate the portrayed individual to be wealthy, which corresponds to a 20% change.

Randomization Method

randomisation by computer

Randomization Unit

individual

Was the treatment clustered?

Experiment Characteristics

Sample size: planned number of clusters

350 individuals

Sample size: planned number of observations

350 individuals

Sample size (or number of clusters) by treatment arms

350 individuals

Minimum detectable effect size for main outcomes (accounting for sample design and clustering)

The MDE for the main outcome is 0.45 sd. or a 1.12 change in the number of individuals who would rate the portrayed individual to be wealthy, which corresponds to a 20% change.

Supporting Documents and Materials

IRB