Graphic detail | Taxis v Uber

Substitutes or complements?

By THE DATA TEAM

WHAT black cabs are to London, yellow taxis are to New York. From the charming checker cabs (pictured) of the 1950s and 60s to the workhorse-like Crown Victorias of the 1990s and 2000s, yellow cabs have been a fixture of the city’s streets, and a surefire way to inform viewers in the early scenes of a film that it is set in the Big Apple. But ever since Uber, the near-ubiquitous car-hailing app, arrived on the scene in 2011, the taxi fleet’s days have seemed numbered. Catching a cab requires standing on the street until an available one happens to drive by, giving the driver directions in case of an unfamiliar destination and rummaging around for cash or wrangling with a finicky credit-card machine at the end. In contrast, Uber lets passengers summon a vehicle from a smartphone, informs them when it is outside, feeds the destination into the driver’s navigation software and lets the rider walk out immediately upon arrival.

There is ample circumstantial evidence of the damage Uber has wrought on New York’s yellow-cab industry. The average price of one of the city’s 13,771 medallions (licenses to drive taxis) has fallen from an average of $1m during the summer of 2014 to $690,000 over the past three months, an aggregate loss of some $4 billion of value. Earlier this year city officials tried to come to traditional taxis’ aid, by threatening to cap the growth of “for-hire” cars (the category that includes Uber vehicles) on the grounds that they were exacerbating congestion—though the government temporarily backed down after Uber launched an aggressive public-relations campaign against it, and agreed to delay the proposed limits until the completion of a traffic study in November.

Nonetheless, hard proof that Uber is the primary cause of yellow taxis’ woes has been hard to come by. Not all Uber passengers would otherwise have hopped in a cab for a given ride—some might have taken the subway or bus, walked, biked or stayed at home. Moreover, Uber is not the only new source of competition for taxis. In 2013 the city both expanded the fleet to include apple-green “boro taxis”, which mainly serve areas outside downtown Manhattan, and launched the Citi Bike bicycle-sharing programme, which is now the largest of its kind in America.

Only now has the picture begun to clear up. In response to the city’s now-postponed plan to limit its growth, Uber recently published six weeks’ worth of data on its service in New York, revealing the number of active drivers and rides at each time of day. And on August 5th the Taxi and Limousine Commission (TLC), the industry’s regulator, released figures on every yellow-taxi journey from 2014 to June 2015 (the totals for 2013 had already been distributed). The new statistics allow for a preliminary answer to the hotly debated question of whether Uber is merely filling the gap in an artificially under-served market, as the company claims, or whether its army of part-time drivers is displacing full-time incumbents, as the taxi industry contends.

The data will provide strong talking points for both sides. The best news for the Uber camp is that the advent of its service has coincided with a significant increase in the total number of rides in New York. Although Uber has not shared statistics for 2013, a leak to Business Insider last year revealed an average of 140,000 Uber trips per week in the city during December 2013. Assuming a steady compound growth rate during the past two years, that suggests that there were 333,000 Uber rides in June 2013. Adding that to the 14.4m yellow-cab trips that month yields a total of 14.7m. In contrast, by the same month of this year, the combined sum for Uber plus traditional taxis was 15.8m. This 7.5% increase in two years makes clear that the market is not zero-sum, and that the benefits enjoyed by Uber and its customers and drivers have not come entirely at yellow cabs’ expense.

However, the figures also suggest that the majority of Uber’s growth has come from substituting for taxis rather than from complementing them. While Uber expanded approximately tenfold over the past two years, from a bit over an estimated 300,000 rides in June 2013 to 3.5m in June 2015, yellow cabs’ hail volume has fallen by 2.1m during the same period. As a first approximation, this implies that just 35% of the growth in Uber rides during this 24-month period has been in addition to the preexisting market demand, leaving 65% that has replaced trips that would otherwise have gone to taxis. (A complete analysis would need to account for the growth trend in the industry before Uber arrived, any expected changes in demand from the overall state of the municipal economy and the impact of Citi Bike and boro taxis, among other factors, and could yield a substantially different ratio.) A further indication that Uber bears significant responsibility for the drop in cab hails is that the decline is most pronounced in the middle of the night, when passengers place the greatest value on Uber’s advantages in convenience and comfort. Taxi rides between 11pm to 5am have fallen by 22% since June 2013, whereas trips at all other times are only off by 12%.

On the surface, a 30% crash in medallion prices over a one-year period might seem like an overreaction by investors to a 14% decline in ridership. But given the industry’s high fixed costs—even when cabs are stuck without passengers, their owners and drivers still have to pay for fuel and the amortised price of buying and maintaining the car—that drop in revenue probably comes almost entirely off the bottom line, representing a much bigger decline in profits. Moreover, medallion prices are forward-looking just like shares of companies are, and presumably reflect expectations that the bite Uber has taken out of taxi earnings is likely to grow. And the industry’s economics could get even tougher if Lyft, Uber’s smaller rival that recently set up shop in New York, decides to set off a price war in the Big Apple, as it previously did in San Francisco. Even if the city government does wind up imposing limits on for-hire cars, medallion prices probably have plenty of additional room to fall.

UPDATE: The print version of this article includes a geographic breakdown of the data and an additional chart. Read it here.

Discover more

Three reasons why oil prices are remarkably stable

Can it last?

Why America is a “flawed democracy”

EIU’s index plots the country’s democratic decline since 2006


Five charts compare Democrats and Republicans on job creation

Our analysis of the past eight American presidents