Narrow Bracketing and Rationality: Evidence from a Continuous Choice Experiment

How to cite

DONG, Wanxin; Zheng, Jiakun; Miao, Bin; Zhou, Ling. 2025. "Narrow Bracketing and Rationality: Evidence from a Continuous Choice Experiment." AEA RCT Registry. AEARCTR-0016827. https://doi.org/10.1257/rct.16827-1.0

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

Wanxin DONG* , Jiakun Zheng , Bin Miao , Ling Zhou

* Primary investigator

Registry ID
AEARCTR-0016827
DOI
https://doi.org/10.1257/rct.16827-1.0
First registered
Last updated
Status
In development
Trial dates
2025-09-25 – 2026-07-31
Countries
Private
JEL codes
D12, D81, D91
Registry record
http://www.socialscienceregistry.org/trials/16827

Abstract

This study aims to test narrow bracketing in continuous choices and measure individuals' economic rationality using budgetary experiments within and between different frames. We conduct a within-subject laboratory experiment in which participants make 22 budgetary choices in each treatment. In the narrow frame, participants receive some initial risky points in the X account and the Y account, and distribute extra points between the two accounts. The total points depend on the sum of the initial points and the points distributed to each account. In the broad frame, participants instead directly choose the total points they want to receive from the X account and the Y account. Two frames are presented in random order at the individual level. Using revealed preference analysis based on Afriat’s Theorem, we measure economic rationality by assessing the consistency with utility maximization with each frame and across both frames combined. We also attempt to employ structural econometrics to estimate individual preferences under different model specifications and to quantify individuals’ degree of narrow bracketing, following Zheng and Zhou (2025)'s work published in Experimental Economics. This, in turn, allows us to conduct welfare analysis.

Experimental design

The experiment is a within-subject design with two treatment groups-- the narrow frame and the broad frame. In each frame, participants make 22 independent continuous choices (Choi et al., 2014; Halevy, 2018). This experimental design was first introduced by Choi et al. (2007) and used in many previous studies in revealed preference analysis.
In the narrow frame, participants receive some initial risky points in the X account and the Y account. Their task is to distribute a certain number of points between the two accounts on a budget constraint line. The total points depend on the sum of the initial points and the points distributed to each account. In the broad frame, participants' task is to directly choose the total points they want to receive from the X account and the Y account on a budget constraint line. At the end of the experiment, one of the decision rounds in the two frames will be randomly selected for bonus payments based on participants' choices. Two frames are presented in random order at the individual level. The orders of choices for all participants are also random in each frame.

Design details

Randomization method
Randomization done by a computer. This is a within-subject experimental design. We have random orders of two frames, and in each frame, the budgetary choices are also random.
Randomization unit
Individual. The computer will generate a random order of experimental parts and a random order of decisions within each part for all subjects independently.
Clusters
300 individuals.
Observations
300 individuals.
Arms
300 individuals.
Primary outcomes
Critical cost efficiency index (CCEI; Afriat, 1972); Money pump index (MPI; Echenique et al., 2011); Cheaper asset fraction.

Cited by

No citation data available for this registration yet.

Keywords