Team incentives at work: Evidence from a retail bank

How to cite

Thon, Max; Alfitian, Jakob. 2025. "Team incentives at work: Evidence from a retail bank." AEA RCT Registry. AEARCTR-0016118. https://doi.org/10.1257/rct.16118-1.0

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

Max Thon* , Jakob Alfitian

* Primary investigator

Registry ID
AEARCTR-0016118
DOI
https://doi.org/10.1257/rct.16118-1.0
First registered
Last updated
Status
In development
Trial dates
2025-06-02 – 2025-08-30
Countries
Germany ()
Registry record
http://www.socialscienceregistry.org/trials/16118

Abstract

Team-based incentives are a widely used management practice aimed at increasing employee performance and motivation. In collaboration with a large German bank, we conduct a randomized field experiment introducing a team incentive scheme for sales employees. First, we aim to provide causal evidence on the effect of team incentives on performance. Second, by combining firm data on individual performance with employee survey responses, we examine heterogeneous effects of the incentive program on performance outcomes.

Intervention

The retail bank specializes in selling foreign currencies. Our intervention consists of paying a variable compensation for each sale. Specifically, we pay 0.3% on the portion of revenue exceeding €400 up to €600, and 0.5% on the portion exceeding €600. The bonus is capped at a sale value of €1,000, which means the maximum achievable bonus amounts to €2.60 per transaction.

Experimental design

We collaborate with a German retail bank and introduce a team incentive in randomly selected branches. The team incentive works as described in the field "Intervention". We assign the treatment on clusters defined by 2-digit postal code regions, such that all branches in the same region are either allocated to the treatment or control condition.

Design details

Randomization method
Clustered randomization was implemented using a computer program, assigning 51 clusters, defined by 2-digit postal code regions, to either the treatment or control group. All branches and their employees within a cluster were assigned to the respective treatment arm.
Randomization unit
2-digit postal code regions
Clusters
51 clusters, defined by 2-digit postal code regions
Observations
Depends on the number of transactions.
Arms
Treatment: 163 employees, 38 branches, 26 cluster; Control: 157 employees, 39 branches, 25 cluster
Primary outcomes
Transaction amount.
Secondary outcomes
Total sales per branch, total sales per employee, heterogeneity analysis with respect to workers pre treatment performance and survey outcomes (worker perception, characteristics, and preferences), absenteeism (subject to data availability).

Cited by

No citation data available for this registration yet.

Keywords