Commitment to Safety: A Demand Experiment from the Ugandan Transit Industry

Last registered on January 16, 2023


Trial Information

General Information

Commitment to Safety: A Demand Experiment from the Ugandan Transit Industry
Initial registration date
January 09, 2023

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published
January 09, 2023, 5:08 PM EST

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Last updated
January 16, 2023, 11:43 AM EST

Last updated is the most recent time when changes to the trial's registration were published.


Primary Investigator

University of Zurich Econ Dep

Other Primary Investigator(s)

Additional Trial Information

On going
Start date
End date
Secondary IDs
Prior work
This trial is based on or builds upon one or more prior RCTs.
This project investigates the demand for commitment and the role of peer pressure in shaping the uptake of commitment contracts for safe driving. I propose that such contracts can serve as a justification for evading the social norm of speeding, thereby mitigating the status signaling aspect of driving fast. To test this mechanism, I design a demand experiment using GPS technology to offer contracts providing monetary rewards for safe driving. By manipulating the conditions under which the contracts are offered, I aim to quantify the demand for incentives to drive safely and to test whether the possibility to use the contract as a justification to drive slow influences the uptake of these contracts.
External Link(s)

Registration Citation

Raisaro, Claude. 2023. "Commitment to Safety: A Demand Experiment from the Ugandan Transit Industry." AEA RCT Registry. January 16.
Experimental Details


Intervention Start Date
Intervention End Date

Primary Outcomes

Primary Outcomes (end points)
- Dummy variable indicating whether respondent picked any (weakly) dominated contract
- Discrete variable indicating willingness to pay for the contract
Primary Outcomes (explanation)
A contract is defined as weakly dominated if the best payment under the contract is less or equal to the outside option. The outside option is always a flat payment, i.e. a payment that is delivered regardless of the driving behavior.

Willingness to pay: the maximum amount of money the respondent is willing to forgo in order to get the contract (real-stakes).

Secondary Outcomes

Secondary Outcomes (end points)
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
Respondents choose between two daily contracts offered for a period of ten days: (i) a flat payment contract regardless of the driving behavior (UCT), or (ii) performance-based contract that pays more if no speeding violations occur. A speeding violation is defined as surpassing 50 kph in urban areas of Kampala, according to the GPS device installed in the respondent's motorbike.

Each choice is made under one of the following regimes:
(1) T1 private regime: all choices and the offer of the contract are kept private;
(2) T2 public excuse regime: choices are made privately; the respondent is informed that the contract offered will be made public to his peers;
(3) T3 public with no excuse regime: the respondent is informed that (a) the contract offered will be made public to his peers and (b) the forgone alternative payment (UCT if performance contract picked or vice versa) will also be made public to the peers.
Peers are defined as co-workers operating from the same taxi station.

Under a given regime, each respondent takes nine consecutive choices (henceforth, a set of choices) between the performance-based contract and a UCT. More details are provided below. An additional choice between two fixed payments is also provided.

Each respondent is informed that only one of the choices will be picked at random. Their preferred contract for the randomly picked binary choice (performance vs UCT) will be offered. To maintain incentive compatibility for all binary choices, the respondent is informed that a positive probability to be selected is assigned to all binary choices in the set.

Each respondent is asked to make three sets of binary choice under the three regimes, plus the flat payment vs flat payment choice, for a total of 30 binary choice. The order of the exposure to the regimes is randomized. I exploit both within and across-respondent variation. I intend to run two analysis: one considers the answers to the first regime each respondent is exposed to; the second analysis exploits within-subject variation and controls for the order of exposure to the three regimes. I intend to verify differential treatment by income (above/below median) and pressure by peers at the workplace to drive fast (measures collected in AEARCTR-0010625).
Experimental Design Details
Each respondent takes nine consecutive choices (henceforth, a set of choices) between the performance-based contract and a UCT. The former is the same across choices and pays UGX 2,000 regardless of the driving behavior and UGX 6,000 if no speeding violations are committed in that given day; the UCT ranges from UGX 1,000 to UGX 9,000.

The contracts offered are used in a follow-up experiment where I study the impact of these contracts on driving behavior and labor supply.

The choice between two flat payments has 2 purposes: to check the understanding of the task; and to allow random allocation of a sub-sample of respondents to UCT in a subsequent field experiment that will leverage the results of this experiment.
Randomization Method
Randomization done in office by a computer
Randomization Unit
I run two sets of analysis:
1) across-subject design for the choices made in the first-exposed regime. The randomization unit is the respondent
2) within-subject design across three different regimes shown in random order. The randomization unit is the respondent-by-regime, clustered at the respondent level
Was the treatment clustered?

Experiment Characteristics

Sample size: planned number of clusters
360 subjects
Sample size: planned number of observations
1080 observations in within subjects design.
Sample size (or number of clusters) by treatment arms
1/3 in private regime, 1/3 in public with excuse regime, 1/3 in public with no excuse regime
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Pilot work suggests that for detecting a 0.1 increase in the share of contracts taken up, with a control mean of .2 and a standard deviation of 0.35, a sample of 388 individuals is required.

Institutional Review Boards (IRBs)

IRB Name
MUREC Research Ethics Committee
IRB Approval Date
IRB Approval Number


Post Trial Information

Study Withdrawal

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information


Is the intervention completed?
Data Collection Complete
Data Publication

Data Publication

Is public data available?

Program Files

Program Files
Reports, Papers & Other Materials

Relevant Paper(s)

Reports & Other Materials