Pricing, Salience, and Waste: A Field Experiment on About-to-Expire Goods in Supermarkets

How to cite

Yan, Andong. 2026. "Pricing, Salience, and Waste: A Field Experiment on About-to-Expire Goods in Supermarkets." AEA RCT Registry. AEARCTR-0017591. https://doi.org/10.1257/rct.17591-1.0

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

Andong Yan*

* Primary investigator

Registry ID
AEARCTR-0017591
DOI
https://doi.org/10.1257/rct.17591-1.0
First registered
Last updated
Status
On going
Trial dates
2025-12-22 – 2026-02-15
Countries
Hong Kong ()
Registry record
http://www.socialscienceregistry.org/trials/17591

Abstract

Food waste in grocery retail is a major sustainability challenge, particularly for about-to-expire perishable products where pricing decisions involve a trade-off between waste reduction and profitability. This research project presents evidence from a large-scale field experiment conducted in partnership with a leading supermarket chain in Hong Kong to evaluate alternative clearance strategies for about-to-expire items. We implement a randomized controlled trial across 56 stores, testing three operational interventions: (i) deeper price markdowns on about-to-expire perishable products, (ii) bundle-based incentives that provide additional discounts when consumers purchase multiple units of the same about-to-expire item, and (iii) changes in in-store display that increase the visibility of about-to-expire fresh produce by consolidating these items in designated locations. The experiment targets high-waste categories including fresh meat, seafood, fruits, and vegetables, and is executed under real-world retail constraints with detailed measurement of sales, inventory, and waste. The analysis aims to quantify the causal effects of pricing depth, purchase incentives, and product salience on demand, waste reduction, and profits, with the goal of identifying scalable operational rules that reduce food waste without compromising retailer margins.

Experimental design

Participating stores are assigned to one of several experimental groups. The interventions include: (i) adjustments to markdown depth for about-to-expire products, (ii) promotional incentives that encourage customers to purchase multiple units of the same about-to-expire item, and (iii) changes in in-store display that increase the visibility of about-to-expire fresh produce. A control group of stores continues to follow standard clearance procedures during the study period.

Design details

Randomization method
Randomization is conducted at the store level using a computer-generated random assignment. Stores are assigned in advance to the control group or one of the treatment groups by the research team using programs. Randomization is performed conditional on basic store characteristics (such as region and store format) to ensure balance across experimental groups.
Randomization unit
The unit of randomization is the store. Participating supermarkets are randomly assigned at the store level to either the control group or one of the treatment groups. All interventions are implemented uniformly within each store.
Clusters
56 supermarket stores in total.
Observations
Approximately tens of millions of SKU–store–day observations, generated from daily transaction, inventory, and waste records across approximately 200 supermarket stores, each carrying around 1,000 SKUs, over the experimental period.
Arms
Control (business-as-usual clearance practices): ~160 stores Treatment Arm 1 (markdown depth adjustment for about-to-expire items): 18 stores Treatment Arm 2 (multi-unit purchase incentive for selected about-to-expire items): 18 stores Treatment Arm 3 (centralized display for about-to-expire fresh produce): 20 stores
Primary outcomes
Waste quantity of about-to-expire products Store-level profitability / gross margin from targeted categories Revenue (sales value) and volume sold of about-to-expire products

Cited by

No citation data available for this registration yet.

Keywords