Sunk Costs and Exercising

How to cite

Epperson, Raphael; Habla, Wolfgang; Wibral, Matthias. 2023. "Sunk Costs and Exercising." AEA RCT Registry. AEARCTR-0011331. https://doi.org/10.1257/rct.11331-1.1

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

Raphael Epperson* , Wolfgang Habla , Matthias Wibral

* Primary investigator

Registry ID
AEARCTR-0011331
DOI
https://doi.org/10.1257/rct.11331-1.1
First registered
Last updated
Status
Completed
Trial dates
2023-03-01 – 2024-04-30
Countries
Germany ()
JEL codes
D91, I12
Registry record
http://www.socialscienceregistry.org/trials/11331

Abstract

This project investigates potential sunk-cost effects on exercising behavior by providing a discount on membership fees to randomly selected individuals who recently signed a 12-month or 24-month contract at a fitness chain. Individuals in treatment Early Discount receive a 50% discount on their membership fee for two consecutive months at an early stage of their membership (the 2nd and 3rd, or 3rd and 4th, month of the current membership contract, respectively). To gain a deeper understanding of how individuals mentally account for sunk costs, individuals in treatment Late Discount receive the same 50% discount on their membership fee for two consecutive months, but at a later stage of their membership (in the 8th and 9th, or 9th and 10th, month of the current contract, respectively). At the beginning of the intervention, individuals in both treatment groups receive an email in which they are informed about their discounts. Individuals in the control group also receive an email to control for reminder effects.

Intervention

The intervention consists of a discount on the membership fee of randomly selected members, along with an email informing individuals about their discount. The discount is either applied to two early months or two later months of the contract period.

Experimental design

Each individual is randomly assigned to one of three conditions. Individuals in treatment Early Discount receive a 50% discount on their membership fee for two consecutive months at an early stage of their membership. Individuals in treatment Late Discount receive a 50% discount on their membership fee for two consecutive months at a later stage of their membership. Individuals in the control group do not receive a discount.

At the beginning of the intervention, all individuals receive an email that informs them about some general aspects related to the fitness chain. Members in treatments Early Discount and Late Discount are additionally informed about their discount and its timing.

Design details

Randomization method
The randomization is stratified by two variables: the monthly membership fee of the purchased contract (five categories) and the sex of the individual (male or no male, the second category includes one individual without information). Hence, there are 10 strata. In each stratum, 60% of individuals are assigned to the control group, 20% are assigned to the treatment Early Discount, and 20% are assigned to the treatment Late Discount. If a stratum has a size that cannot be perfectly divided according to these shares (i.e., is not divisible by 5), the number of individuals assigned to a treatment for the remainder is determined by randomly drawing from a vector of five treatment assignments (3 control group, 1 Early Discount, 1 Late Discount) without replacement. The randomization is implemented with R.
Randomization unit
Individual
Clusters
3,071 individuals
Observations
3,071 individuals
Arms
Control group: ~60% Treatment Early Discount: ~20% Treatment Late Discount: ~20%
Primary outcomes
Average number of weekly gym visits
Secondary outcomes
Membership renewal at the end of the 12-month contracts

Cited by

No citation data available for this registration yet.

Keywords