Present Bias in Choices over Food and Money

How to cite

Danzer, Alexander; Zeidler, Helen. 2024. "Present Bias in Choices over Food and Money." AEA RCT Registry. AEARCTR-0014667. https://doi.org/10.1257/rct.14667-1.0

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

Alexander Danzer* , Helen Zeidler

* Primary investigator

Registry ID
AEARCTR-0014667
DOI
https://doi.org/10.1257/rct.14667-1.0
First registered
Last updated
Status
Completed
Trial dates
2019-11-11 – 2019-12-20
Countries
Germany (Bavaria)
JEL codes
D12, D01, C93, D91, I12
Registry record
http://www.socialscienceregistry.org/trials/14667

Abstract

This paper investigates time inconsistencies in food consumption based on a framed field experiment at a college canteen where participants repeatedly select and consume lunch menus. To test dynamically inconsistent behavior, the design features a continuous convex non-monetary budget in an entirely natural environment and satisfies the consume-on-receipt assumption. Leveraging 3,666 choices of different food healthiness, we find no time inconsistency at the aggregate meal level. Utility weight estimates at the dish level reveal that consumers‘ time inconsistencies are hidden by balancing behavior between food categories, like main dishes and desserts. Individuals who exert self-control in food consumption take up a commitment device as soon as it becomes available, while non-committers tend to be present-biased. Our results suggest that internal and external control strategies are substitutes. When measuring present bias with standard monetary tasks, we find no correlation between dynamic inconsistency in food and money choices.

Experimental design

We examine dynamic inconsistencies in real food consumption using a longitudinal
framed field experiment at a college canteen in Bavaria, Germany. In three separate
sessions participants choose food items from the canteen’s menu for immediate
and prospective consumption, and subsequently consume their lunch.

Design details

Randomization method
Out of the selected meals, one meal is randomly selected. This ensures incentive compatible choices, but is no intervention..
Randomization unit
meal level randomization
Clusters
73 individuals
Observations
3666 food items for 73 individuals in 3 sessions, each with several choices.
Arms
n.a.
Primary outcomes
Consumption choices and their healthiness; selected food items out of 2925 possible food item combinations

Cited by

No citation data available for this registration yet.

Keywords