Price Caps Harm Riders and Drivers

A Case Study from Bangalore and Hyderabad

By Bharat Chandar, Ranjana Menon and Calum You

Dynamic pricing is a key algorithmic tool in balancing Uber’s marketplace. In Bangalore, price restrictions have been set at a level that binds, preventing the algorithm from setting prices at a healthy level. As a result, service quality degrades for riders and drivers miss out on significant earnings potential.

How does dynamic pricing improve efficiency?

The role of Uber’s marketplace is to enable riders to travel and drivers to earn to the greatest extent possible. To do this, we use dynamic pricing to maintain a healthy balance between supply and demand.

As shown above, the level of demand for rides varies greatly between different times and places. Whenever there are many more riders demanding rides than there are drivers to supply them, our dynamic pricing algorithm will raise prices. This encourages riders who do not want to pay a premium to either wait or seek alternative transport, and also gives drivers incentives to come out onto the road with higher earnings.

(You can read more about dynamic pricing here).

However, price controls can prevent the marketplace from balancing supply and demand and lead to unfavourable outcomes for riders and drivers alike. We can see this by comparing Bangalore and Hyderabad, two cities with different price regulations.

What are price caps?

Bangalore and Hyderabad are similar cities, but differ primarily in how prices are controlled. In Hyderabad, there are no price caps on rides. In Bangalore however, prices for the most common vehicles are limited to between 12 and 24 Rs/km. This means that even when our dynamic pricing algorithm would prefer to set prices higher or lower to balance the marketplace, it cannot raise them above 24 Rs/km or go under 12 Rs/km. Since the marketplace cannot adjust to changing conditions, supply and demand are frequently mismatched. In 2015, our team described how an outage of dynamic pricing caused such a mismatch, leaving riders stranded on New Year’s Eve. The price caps in Bangalore are like having an outage every weekday rush hour.

What happens at a price cap?

When the price cap binds, there are many riders looking for a ride but few drivers available to provide them. The most immediate consequence is a rise in the ETA (expected wait time for a car to arrive after requesting it) for two main reasons. Drivers are likely to all be busy, forcing riders to wait for drivers to become available. Furthermore, with relatively few available drivers, they are spread thinly across a geographic area and so the average distance between a driver and the pickup location increases. It becomes much more likely that a driver is sent to a distant location, when in a more balanced, uncapped market that trip would have been fulfilled by a closer driver.

These changes to ETAs have additional consequences. Riders who are frustrated with long ETAs begin to cancel trips they have requested, meaning they do not reach their destination and drivers waste time en route to cancelled pickups. Drivers also begin rejecting more dispatches since they are so far away, causing them to spend more time waiting around. If enough drivers reject a trip, the rider request may also expire without being fulfilled. These effects can become a vicious cycle as drivers who stay on the road spend more time en route, while others decide not to drive because of their lowered earnings. Both effects further reduce the number of available drivers and make the situation worse. Overall, we see the share of requests that complete decrease sharply because the market becomes unbalanced at the price cap.

How are drivers and riders impacted by a price cap?

The price cap decreases the number of trips that are completed, which harms both riders and drivers. For drivers, the price cap causes them to earn less than they would otherwise. Because of the price cap, drivers complete fewer trips as they spend more time en route to far-off pickup locations. Price caps that regularly bind will just as regularly reduce opportunities for drivers to earn.

For riders, the rise in ETAs is a sign of decreased service quality. The corresponding rise in cancellation rates indicates that many are dissatisfied when the price cap binds, preferring to not travel or seek another mode of travel when they must wait too long to receive a ride. A lucky few riders who manage to get rides at the price cap pay less than they would otherwise. However, the decrease in completed trips means that overall, many riders are impacted as fewer riders are able to travel to their desired locations.

Conclusion

In summary, price regulations create a vicious cycle that disrupts the marketplace. Without dynamic pricing, an undersupply of cars becomes persistent. With low supply relative to demand, ETAs rise and the number of completed trips falls. This lowers driver earnings and further reduces driver supply, restarting the cycle. Dynamic pricing is necessary to maintain a balanced marketplace, and neither riders nor drivers stand to benefit from capping prices.