Taken together, the competition effect, which leads to reduced daily earnings, overwhelms the higher pay from cherry-picking.
The analysis found that after surge pricing went into effect, the number of drivers operating during peak hours in Guangzhou was approximately 20% higher than during the same periods in Shenzhen. Yet drivers in Guangzhou drove shorter distances, completed fewer trips and earned lower daily revenues compared with those in Shenzhen.
Not all drivers could simply add more workdays. Full-time drivers, arguably the most loyal and experienced, were already maxed out on the number of workdays, and as a result their weekly earnings fell. Part-time drivers, who could easily add more hours and days, did so and specifically targeted times when surge pricing would be in effect.
Drivers and ride-hailing companies could make surge pricing work better for themselves in a few ways. Full-time drivers could simply avoid the intensified competition by sitting out surges when they occur or by relocating to less competitive regions.
Ride-hailing companies also could help subsidise their more loyal drivers by taking lower commissions or providing bonuses. They could also do more to coordinate drivers’ schedules to minimise the effects of too much competition. In traditional taxi service, this is done by a central dispatch system. In decentralised ride-hailing, a company could provide a constantly updated map that displays all the drivers in an area and allow them to pick areas to work, depending on whether there is more or less competition nearby.