Behavioral Consumer Reactions to Product Size Changes

How to cite

Pires, Pedro; Wasserman, Lucio. 2026. "Behavioral Consumer Reactions to Product Size Changes." AEA RCT Registry. AEARCTR-0015352. https://doi.org/10.1257/rct.15352-1.2

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

Pedro Pires* , Lucio Wasserman

* Primary investigator

Registry ID
AEARCTR-0015352
DOI
https://doi.org/10.1257/rct.15352-1.2
First registered
Last updated
Status
In development
Trial dates
2026-03-28 – 2028-06-30
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/15352

Abstract

This study examines package size changes. It has two empirical parts: first, we exploit supermarket scanner data by developing an algorithm to identify size changes, which allows us to provide novel descriptive statistics and study how consumer respond to size and price changes. Second, we conduct an online experiment to investigate whether consumers respond differently to these two types of changes and to test whether different ways of displaying information causally affect consumer decisions. We also examine how package size changes affect the welfare of different groups of consumers and discuss how our results relate to different public policies. The empirical analyses are guided and complemented by theoretical models.

Intervention

Our study evaluates how consumers respond to changes in product attributes through an incentivized online experiment. Participants complete multiple shopping scenarios, with some (randomly chosen) participants receiving their chosen products through home delivery to ensure incentive-compatible decision-making. We reproduce key features of real-world shopping environments. The experiment includes different treatment conditions that modify how information is conveyed to consumers.

Experimental design

Participants complete a series of shopping scenarios with incentivized purchase decisions, where some participants will receive their chosen products through home delivery. After the first block of scenarios, participants are randomly assigned to different treatment conditions that modify how information is presented in the latter half of shopping scenarios. Each participant completes a total of six shopping scenarios.

Design details

Randomization method
Randomization is implemented through an online survey platform.
Randomization unit
Shopping environments (for the last three scenarios), the probability of size changes, the frequency of belief elicitation and incentives for belief elicitation are randomized at the individual level. These randomizations are independent of each other. Displayed product prices and sizes are randomized at the product by scenario by individual level. The shopping decisions we will implement through home delivery are selected first by randomly choosing individuals, then by randomly choosing a scenario for each of those individuals.
Clusters
N/A
Observations
We expect about 2,000 participants. The exact number will depend on the realized cost per participant.
Arms
Around 500 participants per shopping environment.
Primary outcomes
Consumer quantity demanded for different products in each shopping scenario. Beliefs about product attributes and their accuracy.

Cited by

No citation data available for this registration yet.

Keywords