Monday, December 25, 2017

The costs and benefits of transport aggregators

The beneficial effects of transport aggregators like Uber stand out, especially given the experience of consumers with business as usual taxi operators. They are much more convenient (lower wait times, ease of hailing, doorstep access etc) and far cheaper, which in turn induces more people to use such services.

Peter Cohen et al used granular surge pricing data to estimate that the overall consumer surplus generated by UberX service in the US in 2015 was $ 6.8 bn, and that each dollar of consumer spending generates about $1.6 in consumer surplus. Chungsang Tom Lam and Meng Liu used Uber and Lyft data for New York City to show that platform users gain 72 cents per dollar spent on these platforms, with 64% of the welfare gains coming from dynamic pricing.

But these studies are confined to the consumer side. How about the externalities - the impact on the urban transport eco-system? In particular, the costs of congestion are well documented, and how much does the addition of aggregators worsen it?

CityLab points to a just released report by Bruce Schaller who found that during 2013-17 the number of taxi/aggregator vehicles in the Manhattan Central Business District (CBD) rose 59% on weekdays, and the number of such vehicles in the same area in late afternoon doubled to over 10,000 vehicles, and though taxi trips declined, total passenger trips rose 15%. He shows that aggregators contributed to a 36% increase in the amount of miles traveled by for-hire vehicles in the CBD, with lengthier trips and more "deadheading" (cars traveling without passengers). The result of all this was an 18-19% reduction in average traffic speeds.
Another study of Regina Clewlow and Gouri Shankar Mishra of University of California, Davis used data from comprehensive surveys in seven US cities in the 2014-16 period and came to similar conclusions. They found that ride-hailing services led to an average 6% reduction in use of bus services, and that 49% to 61% of ride -hailing trips would not have been made at all, or by walking, or biking or transit. 

There is an even more damaging dynamic at play. Ridership of the New York mass transit system has been declining in recent years, on the back of poor service quality and safety concerns. Schaller's research suggests that aggregators are amplifying the decline, especially by drawing the more affluent passengers off trains into cars. As City Lab points out, this can trigger a death spiral - "fewer transit riders means less revenue and demand for improved transit" and a poor quality mass transit system used by the less well-off. This dynamic applies to cities in any developing or developed country.

In simple terms, the assessment of such innovations are about whether the private benefits from them are commensurate with their social costs. The former are amenable to being quantified and often rigorously too. In contrast, the latter are very difficult to capture and have long-drawn general equilibrium effects. The resultant propensity to under-estimate the latter causes an exaggeration of the benefits of such innovations.

More fundamentally, the biggest urban transportation challenge is to achieve the modal shift away from private vehicles to mass transit systems - increase the share of public transport and decrease the share of private vehicles. Therefore, any innovation or disruption that improves the efficiency and thereby increases the use of private vehicles, as transport aggregators like Uber appears to be doing, fails the first-order test of social benefit.

No comments: