Every day, 10 people die on the roads of Kampala, Uganda’s capital.
Road accidents cost Uganda US$1.2 billion annually, which is about 5% of its GDP. The cost typically arises from healthcare spending. Families face crippling medical bills and businesses lose workers.
Motorbike taxis, which are popular in Uganda, are a leading cause of accidents. They are responsible for 64% of all recorded accidents – mostly as a result of speeding.
Why do so many motorbike taxi drivers in Uganda speed? The common wisdom suggests that they do it for financial reasons. Higher speed translates to more trips, and more trips mean more income.
But a closer look reveals a more complex reality: speeding isn’t just a money decision – it’s about social pressure among motorbike taxi drivers and the need to adhere with behaviours that signal masculinity. Most drivers are male.
Uganda’s current approaches to counter speeding include fines and awareness campaigns. There is little evidence that these methods have been effective.
My recent study in Kampala challenges these traditional road safety approaches, which often fail to change behaviour. I am a behavioural economist, and my findings show offering financial incentives can work – but only if these incentives provide drivers with a socially acceptable reason to slow down.
Financial incentives need to be made public, and only work when they allow motorbike taxi drivers to justify safer behaviour to their peers. This is key, because getting road safety incentives right saves lives. It also reduces healthcare costs, lowers fuel consumption and emissions, and helps shift harmful social norms that encourage reckless driving.
Why drivers speed
My research finds that speeding among motorbike taxi drivers isn’t just a financial decision in Uganda, it’s a social one. Drivers work in tight-knit communities where reputation matters as much as income.
I collected data from a representative sample of 386 passengers and found commuters prefer safer drivers and are willing to pay up to 8% more for careful driving. Yet, speeding remains the norm.
The reason? Driving fast is a status symbol for motorbike taxi drivers.
I carried out an experiment to test whether drivers who speed are perceived more positively by their co-workers. Results are clear: fast drivers are perceived as more skilled and have a higher social status, measured as their ability to influence decisions at their taxi stations.
This presents a policy challenge: how can financial incentives encourage safer driving without making drivers feel like they are losing respect among their peers?
To test how financial incentives could encourage safer driving, I conducted an experiment in which a research team offered 360 drivers two options:
-
a contract that paid them a daily incentive of UGSh6,000 (US$1.64) – roughly a third of their daily income – for observing speed limits
-
or an equivalent lump sum cash payment with no conditions attached, including limiting a driver’s speed.
But the framing of these choices mattered.
-
Some drivers knew their decision would be private, meaning no one else would know if they took the safe-driving contract.
-
Others knew that only the safe-driving contract would be public, while the alternative lump sum cash option remained private – giving them a socially acceptable reason to slow down.
-
A third group knew their decision would be fully public, meaning their peers would see if they chose the safe-driving contract over the lump sum.
The results were clear. Twice as many drivers accepted the safe-driving contract when it was public and provided a justification for slower speeds.
Why? Because when the incentive was visible but also justified, drivers could explain their decision as a financial one:
I’m not driving slower because I’m scared, I’m doing it because I’m getting paid.
The design of this experiment allowed me to answer the question: what mechanism favours socially desirable behaviours when incentives are offered?
But would the drivers actually slow down?
Did it work?
To see whether these contracts actually changed driving behaviour, I conducted an impact experiment, offering incentives for two weeks and tracking drivers for six months.
Drivers were randomly offered one of the following contracts:
-
a private safe-driving contract – where only the driver knew about the financial reward
-
a public safe-driving contract – where their peers knew they were being paid to slow down
-
a control group – who received a contract consisting of a simple cash payment with no conditions.
The results were striking. While both safe-driving contracts reduced speeding, the public contract was nearly twice as effective as the private one. The most significant reductions were seen in extreme speeding (occurrences of 80km/h or more) – the kind most likely to cause severe accidents.
The key takeaway is that visibility makes incentives work, but only when it provides justification. If a driver had to publicly choose the safe-driving contract over another cash offer, it lost effectiveness. But when structured as a justifiable contract, it allowed drivers to slow down without social consequences.
Reframing safe driving as a smart decision, not just a rule, is important. Featuring respected drivers in safety programmes can potentially help shift perceptions of what makes a “good” driver.
Finally, drivers operate in tight social networks. Policies should be developed with their input rather than imposed externally. Programmes that actively engage drivers will be more widely accepted and successful.
Rethinking how incentives shape behaviour
Speeding is often framed as a problem of reckless individuals making bad choices. My research shows that’s rarely the case – rather it’s about social incentives and peer influence.
A poorly designed financial incentive may slow drivers down temporarily, but it won’t change long-term behaviour. Incentives that help drivers escape the social pressure of adopting risky behaviours may shift norms – creating lasting improvements in road safety, economic efficiency and environmental impact.
Claude Raisaro receives funding from the Swiss National Science Foundation (grant no. 195266), the Forschungskredit of the University of Zurich (grant no. FK-22-020), the Swiss Re Foundation for Research in Development Economics, and SurveyCTO. He is affiliated with Mistra Center for Sustainable Markets at Stockholm School of Economics.
