When Too Much Ride-Hailing Competition Harms Consumers

Economic theory suggests that more competition is nearly always a good thing for consumers as it gives them more options. Research from MIT suggests, however, that for ride-hailing, this might not always be the case as too many vehicles results in congestion.

The study shows that ride-sharing has the potential to significantly add to congestion, and so there is probably an optimal size of the fleet for any given metro area.

“What this shows is that by not coordinating ride-hailing companies, we are creating a huge amount of additional traffic,” the researchers explain. “If cities were to use a platform to coordinate ride-hailing, we could reduce overall congestion and traffic in cities all over the world.”

Traffic costs

The researchers gathered anonymized taxi data to understand where people requested rides from. The data covered New York, Singapore, Vienna, San Francisco, and Curitiba. This allowed for a huge variance in the volume of rides taken, from 150 million in New York down to 300,000 in Vienna.

This data was used as a proxy for the total ride-hailing demand in each city, with the researchers then modeling the flow of traffic needed to ensure journeys were optimally completed. The model also examined the ways in which firms competed independently of one another to allow the team to isolate the impact of adding a new firm to the market.

The analysis revealed that adding a single ride-hailing firm to any territory had varying effects on the number of cars that should be deployed in order to successfully meet demand. For instance, in Manhattan, any new entrant would only increase the quantity of vehicles by around 3% due to the saturation of the market already. In Curitiba, however, a new entrant would increase supply by 67%.

“We think it’s positive to have multiple providers,” the researchers explain. “But if they are not coordinated, there is a price to pay, so to speak.”

On demand

The study suggests that the biggest factor affecting the supply of vehicles is the density of demand from passenger vehicles, although the average traffic speed also plays a major role. For instance, in New York passengers tend to be clustered together, so additional firms won’t really change the number of vehicles too much. This isn’t the case in Curitiba, where passengers are more spread out.

“If there is very dense demand, even if you do not coordinate, you still have a good pool of vehicles to draw from [nearby], and efficiency is still pretty good,” the researchers explain. “If you are in a city without that density of demand, non coordination costs a lot. The other factor is the speed of traffic. In a noncoordinated market, you might require a vehicle that is further away. If the traffic speed is high, that might be okay, but if the traffic speed is low, it might be very inefficient to serve that customer.”

So what is the best approach? The researchers hypothesize that the ideal scenario would be one in which there is a single ride-sharing platform that is used both by all consumers and all driving firms.

“Certainly this doesn’t mean arguing for less competition,” they explain. “We can combine competition and efficiency by using a common platform. It’s just a matter of regulation by the cities. And these are heavily regulated markets, so we’re not arguing for anything new.”

While such a notion seems fanciful given the huge status the likes of Uber and Lyft enjoy, the researchers point out that they largely exist for things such as micromobility and bike-sharing services. It’s perhaps fair to say, however, that such services weren’t starting from a point of having multi-billion dollar businesses that you’re asking to cede control of their platforms.

Facebooktwitterredditpinterestlinkedinmail