Deposit Contracts with Performance-Based Refund as a Commitment Device: A Field Experiment

How to cite

Hiller, Maximilian; Mertins, Vanessa. 2022. "Deposit Contracts with Performance-Based Refund as a Commitment Device: A Field Experiment." AEA RCT Registry. AEARCTR-0009431. https://doi.org/10.1257/rct.9431-1.0

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

Maximilian Hiller* , Vanessa Mertins

* Primary investigator

Registry ID
AEARCTR-0009431
DOI
https://doi.org/10.1257/rct.9431-1.0
First registered
Last updated
Status
In development
Trial dates
2022-05-15 – 2022-09-30
Countries
Germany ()
JEL codes
C93, I120, D91
Registry record
http://www.socialscienceregistry.org/trials/9431

Abstract

The pure attempt to achieve health related goals is already positive behavior, which, however, is rarely rewarded economically. Commitment devices and motivational tools such as deposit contracts, whose effectiveness has been empirically proven, often only reward successes.
The performance evaluation and pay are dichotomous: Success across the board is rewarded with a bonus, while every failure, no matter how small, carries a penalty. Therefore, acceptance rates tend to be low. But only individuals who accept them benefit from those contracts. To tackle this issue, we designed a deposit contract with a performance-based refund. To tackle these issues, we implement an additional feature in deposit contracts. Individuals can decide every day whether they would like to opt-in in a deposit contract but in addition to a standard deposit contract, where people lose money if they fail and get an additional reward if they reach the predefined goal, we implement a piece rate for performance-based deposit refunds. For every percent of goal achievement, the participants take forward to full (100%) goal achievement, they receive one percent deposit refund. For example, if a participant with a daily goal of reaching 10.000 steps performs 9.000 steps, he receives a refund of 90% of the previous made deposits. Still, a reward is only paid when the goal is reached (100%). With this procedure participants are not penalized disproportionally but also not rewarded, if they do not manage to achieve 100% goal achievement. We empirically test this device in a field experiment (randomized controlled trial) with individuals who want to achieve a health goal.

Experimental design

We focus on a setting where the participants want to increase physical activity by reaching an individual calculated step goal. The goal is calculated based on baseline phase. At time of running the baseline e, participants are not aware that we assign them a goal and how it will be determined. After the baseline, the participants learn about the goal. Afterwards, all participants are randomized into the following treatments:

a) T0 Control Group
b) T1 Reward
c) T3 Standard Deposit Contract
e) T3 Deposit Contract with Performance-Based Refund

Participants in T0 Control are simply told that they will get a participation fee (20 Euro) at the end of the study, regardless of how often and to what extent they actively tried and actually have achieved their goal. In T1 Bonus, we offer participants a reward: participants will receive the same information as in the T0 Control but gain an additional bonus (0.25 Euro) for each day on which they have reached or exceeded their individual goal (max. 20 Euro additional) after an opt-in. In T3 Standard Deposit Contract participants are offered to opt-in in a deposit contract where they could earn additional money (0.25 Euro; max. 20 Euro additional) for each day they achieve or exceed their goal and lose money (2.00 Euro) if they fail. T3 Deposit Contract with Performance-Based Refund needs a 2 Euro front-loaded deposit if accepted. Participants receive a performance-based deposit refund (1:1) and an additional bonus for each day on which they have reached or exceeded their individual goal (0.25 Euro; max. 20 Euro additional).

Design details

Randomization method
Randomization into the four groups done by a computerized random draw.
Randomization unit
Individual
Clusters
100 individuals
Observations
100 individuals
Arms
a) T0 Control group: 25 Individuals b) T1 Treatment 1: 25 Individuals c) T2 Treatment 2: 25 Individuals d) T3 Treatment 3: 25 Individuals
Primary outcomes
take-up rate, successful goal achievements and (average) degree of target achievement
Secondary outcomes
Average costs per treatment, average cost per individual (especially important in comparison to the control and bonus treatments).

Cited by

No citation data available for this registration yet.

Keywords