Loss aversion in the riskless choice of two goods

Last registered on May 26, 2022

View Trial History

Pre-Trial

Trial Information

General Information

Title

Loss aversion in the riskless choice of two goods

RCT ID

AEARCTR-0009494

Initial registration date

May 25, 2022

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

May 26, 2022, 11:45 AM EDT

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

Last updated

May 26, 2022, 4:05 PM EDT

Last updated is the most recent time when changes to the trial's registration were published.

Locations

Country

Japan

Region

Osaka

Primary Investigator

Name

Tomoharu Mori

Affiliation

Ritsumeikan University

Contact Primary Investigator

Other Primary Investigator(s)

PI Name

Yasuhiro Nakamto Nakamoto

PI Affiliation

Kansai University

Contact Investigator

PI Name

Naoko Okuyama

PI Affiliation

Yokohama National University

Contact Investigator

Additional Trial Information

Status

In development

Start date

2022-06-07

End date

2022-09-23

Keywords

Behavior, Lab

Additional Keywords

Loss aversion; Laboratory experiment; Riskless choice; Substitute goods; Reference-dependent utility model

JEL code(s)

C91, D91

Secondary IDs

Prior work

This trial does not extend or rely on any prior RCTs.

Abstract

Making the experimental comparison of risky and riskless two-goods choice, Chung, Glimcher, and Tymula (2019, American Economic Journal: Microeconomics) find that loss averse behaiviors are not observed under riskless choice, which is not consistent to the prediction of the prospect theory in Tversky and Kahneman (1991, Quarterly Journal of Economics). Specifically, the loss range of their task questions, which measures the degrees of loss aversion, is that the quantities of two goods both decrease, compared with the initially given endowment (See Table 1 in Chung et al. 2019). Alternatively, turning to the original paper of loss-averse preferences by Tversky and Kahneman (1991), the loss range is that the quantity of a good decreases but that of the other good increases. Furthermore, they show that the differences of the initial quantities of two goods affect the curvature of the indifference curve in the loss range (See Figure III in Tversky and Kahneman); however, Chung et al (2019) give the fixed quantities as the initial endowment.

Following the experimental design in Chung et al (2019), this study re-examines the loss averse propensity of riskless choice by differing the initial quantities of two goods as in Figure III in Tversky and Kahneman (1991) where the relationship between two goods is not substitute as in Chung et al. (2019). Moreover, this study newly observes the indifference curve of the loss range in the case of two goods that are substitute (orange juices of two companies).

External Link(s)

Registration Citation

Citation

Mori, Tomoharu, Yasuhiro Nakamto Nakamoto and Naoko Okuyama. 2022. "Loss aversion in the riskless choice of two goods." AEA RCT Registry. May 26. https://doi.org/10.1257/rct.9494-1.1

Sponsors & Partners

Experimental Details

Interventions

Intervention(s)

In the experiment, participants are given an endowment of two goods and then face options regarding the combination of changes in the two goods. The number of goods in the endowment differs within participant. The type of goods used in the experiment and the order of endowment differ between participants.

Intervention Start Date

2022-06-07

Intervention End Date

2022-09-23

Primary Outcomes

Primary Outcomes (end points)

Participants’ choices of the two goods in the experiment

Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)

Secondary Outcomes (explanation)

Experimental Design

Experimental Design Details

In the experiment, participants are given an endowment of two goods and then face options regarding the combination of changes in the two goods. For example, given an endowment of eight juices and eight chocolates, participants choose one of two options: X) four more juices and four more chocolates, or Y) two more juices and seven more chocolates. To estimate the utility function and the indifference curve, participants make many choices: a choice block consists of 38 choices.

Each participant responds to all four selection blocks. The endowments and options differ in the choice blocks. The number of goods in the endowment is as follows.

B1) endowment of good A is 0, and endowment of good B is 4,
B2) endowment of good A is 4, and endowment of good B is 4,
B3) endowment of good A is 8, and endowment of good B is 4,
B4) endowment of good A is 8, and endowment of good B is 8.

To investigate the loss averse propensity in the riskless choice of two goods, our choice task is based on the gain and loss ranges. As for the gain range, the experimental design in this study is fundamentally the same as Chung et al. (2019), except the initially given quantities. In the loss range, Chung et al (2019) focus on the case that both quantities of two goods decrease, while our study considers the case that the quantity of a good decreases but that of the other increases, following the original paper of the loss averse propensity by Tversky and Kahneman (1991).

Blocks B1 and B2 are choices regarding the gain range, and block B3 is the choice when only one of the goods is in the loss range. Block B4 is the choice when two goods are in the loss range.

The main experimental treatment is a 2x2 design. One is the goods to be used and the other is the order of the blocks. Two goods used in Chung et al (2019) are a beverage and a snack whose relationship is not substitute. In treatment Beverage-Snack of our experiment, participants will be asked to make decisions for a beverage and a snack (chocolate and orange juice), and in treatment Beverage-Beverage, participants will be asked to make choices for two kinds of beverages (orange juices of two popular companies) which are considered substitutable.

As for the order of the blocks, we cannot try every combination, so we will implement some order. In treatment Order1, the order is B1, B2, B3, and B4. In treatment Order2, the order is B3, B2, B1, and B4. If feasible, we implement a different order: in additional treatment Order3, the order is B2, B3, B1, and B4.

The following questions will be asked in the post-experiment questionnaire.

Q1) Did you understand the content of today's experiment (what it is about and how to perform the procedure)? (4-point scale)
Q2) Please tell us your gender. (Male/Female/Other)
Q3) Please choose the option that best describes how much you like or dislike each of the following foods used in this experiment. (7-point scale)
Q4) How much do you think each of the foods used in the experiment today cost?

Randomization Method

Randomization done in office by a computer

Randomization Unit

Experimental Sessions

Was the treatment clustered?

Experiment Characteristics

Sample size: planned number of clusters

12 experimental sessions

Sample size: planned number of observations

32 participants for 6 experimental treatments (192 participants in total)

Sample size (or number of clusters) by treatment arms

32 participants for each of the following treatments
1) Beverage-Snack, Order1
2) Beverage-Snack, Order2
3) Beverage-Beverage, Order1
4) Beverage-Beverage, Order2
5) Beverage-Snack, Order3
6) Beverage-Beverage, Order3

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

Supporting Documents and Materials

IRB