The first thing students are taught in any introductory microeconomics course is that the price of something, let’s say chauffeur services, and the quantity of it that consumers want is depicted by a negative-sloping demand curve. The difference between what consumers are willing to pay for a chauffeur ride (the demand curve) and what the chauffeur asks (the market price), is what is known as the consumer surplus. The bigger the consumer surplus, the better for society.
But even though the idea of consumer surplus is used in many applications, measuring it has always been problematic. That is because, in the real world, demand and supply move together, and it is therefore difficult to establish the exact shape of a demand curve.
That was, until Uber. A team of economists (including Steven Levitt of Freakonomics fame) recently published an NBER Working Paper that uses almost 50 million UberX ‘consumer sessions’ to identify a demand curve for taxi services, and then calculate the consumer surplus that these services generate. A ‘consumer session’ is basically when someone logs onto the UberX app and requests the price for a proposed trip. The consumer either accepts the price and wait for an Uber driver to pick them up, or they don’t, and find alternative transport.
What makes Uber unique is that its prices vary according to demand (for its services) and supply (the availability of drivers). This unfortunately also means that it is not possible to simply calculate a demand curve when the price increases by 10%, because the increase might be the result not of greater demand by consumers for Uber trips, but of lower supply (having fewer drivers on the route). The research team use a clever trick to get around this. Say the algorithm predicts that the price must increase by 1.249. This is then rounded down to 1.2 for the consumer. Other times the algorithm suggests the price must increase by 1.251, but the app then rounds this up to 1.3. It is this discrete difference when the price is essentially the same which the authors exploit using regression discontinuity analysis.
If this sounds very geeky, the results are worth the wait. First, the authors find that demand is quite inelastic (around 0.5). This means that if prices increase by 10%, demand will only fall by 5%. Second, they compute the dollar value of consumer surplus in Chicago, Los Angeles, New York and San Francisco to be roughly $2.8 billion annually. This is more than six times Uber’s revenue in those cities. Put another way, for each $1 spent on an UberX ride at the lowest price, the authors estimate that the consumer ‘receives’ $1.57 in extra surplus. In short, Uber generates massive benefits for society-at-large.
Why does Uber generate so much consumer surplus compared to normal taxi operators? Another NBER Working Paper, written by Judd Cramer and Alan Krueger, suggests that it is because of the higher capacity utilisation rate of Uber drivers: “UberX drivers spend a significantly higher fraction of their time, and drive a substantially higher share of miles, with a passenger in their car than do taxi drivers.” There are four reasons for this. First, Uber’s better matching technology (an app that anyone can download on their phones). Second, the larger scale of Uber’s usage in comparison to taxi companies. Third, highly inefficient taxi regulations which limit the number of routes or time periods taxi drivers can operate. Fourth, Uber’s flexible labour supply model and pricing model which match supply with demand.
South African regulators have had varied responses to Uber’s entrance in the local market. There has been opposition from the taxi industry, sometimes violent. Proponents of Uber, on the other hand, often highlight the entrepreneurial and job-creating opportunities the service creates.
What this research shows, though, is that Uber’s main benefit is the massive surplus it generates for consumers. According to the Levitt research team, one day’s worth of consumer surplus in the four US cities they analyse is worth about $18 million. “If Uber were to unexpectedly disappear for a day, that is how much consumers would lose in surplus.”
Aside from this consumer surplus, Uber services create many positive externalities, from lower congestion and pollution levels to semi-skilled employment to, perhaps more tenuously, greater social interaction and cohesion – I’ve had some fascinating conversations with Uber drivers, and know of one driver that was offered a scholarship by a client. But, most importantly, when regulators and policy-makers debate the pros and cons of Uber and other such services that will almost certainly appear in the next few years, it is worth remembering one of the basic principles of introductory economics: the immense benefits society derives from the additional consumer surplus.
*An edited version of this first appeared in Finweek magazine of 20 October.
