Why San Francisco may not be the most useful reference point for the Uber-ification of local transport

By Michael Carney , written on September 22, 2014

From The News Desk

With taxi industries and local regulators looking to understand the implications of a ride-sharing dominated local transportation market, San Francisco is held up as the prototype. After all, the Bay Area is home to Uber, Lyft, and Sidecar, and thus was the initial market for each of these services, which are now seven-, five-, and four-years-old, respectively.

This makes data presented at the San Francisco Municipal Transportation Authority’s board meeting last Tuesday of particular interest. According to the latest study, rides in traditional taxis have declined by 65 percent in the last 15 months, from an average of 1,424 monthly rides per taxi in March 2013 to just 504 in July of this year. The study also reveals, worryingly for those who cite accessibility and rider discrimination as problems with these new services, that the number of pickups in wheelchair-accessible taxis fell from a high of 1,378 last March to just 768 this July.

This shouldn’t be all that surprising, given the increased competition. But it’s nonetheless revealing as to the impact these on-demand services are having on the for-hire transportation market.

But San Francisco may not be a great prototype for other markets. For one, the state of the local transportation in the city has been abysmal for years. SF taxis are regarded by residents and visitors as some of the worst among major US cities in terms of cleanliness, reliability, and service. The city’s BART and Muni rail systems aren’t much better across any of those measures. That is to say, San Francisco residents have had more reason than most to look for local transportation alternatives.

Secondly, the Bay Area is, for obvious reasons, ground zero for technology early adopters. And as such, services like Uber, Lyft, and Sidecar, like Yelp and eBay before them, naturally find large and eager customer communities in the region. Consumers in Indianapolis, IN and Columbus, OH, which rank just above and below San Francisco in terms of population among US cities, are unlikely to see the same level of fervent uptake.

For example, I’ve been in Chicago twice in the last three months and while I took Uber exclusively around the city, I was definitely in the minority based on anecdotal evidence from conversations with those in my social circle and drivers on the service. Cabs and local rail options still rule Chicago, although Uber is making a growing dent in the market.

Of course, Uber in particular, and Lyft and Sidecar to a lesser degree, are now well-known brands and ride-sharing in general is a familiar concept to many consumers before the services even reach their home market, which could positively affect adoption in new markets. But San Francisco still offers a unique set of circumstances so it could be misleading to extrapolate data coming out of this region and expect things to play out similarly elsewhere.

Uber and its investors have made a compelling case for why the company’s low-cost, low-friction service actually increases the size of the for-hire transportation market. Again, turning to anecdotal evidence of people I know in major ride-sharing markets like San Francisco, New York, and Los Angeles, this appears to be accurate. Uber investor Bill Gurley, of Benchmark Capital, recently wrote that his friends use Uber three-times more often today than they used taxis and black cars two years ago. Assuming these observations are accurate, it’s yet another complicating factor in assessing the impact of Uber, et al on for-hire transportation markets.

There is no question that the ride-sharing and on-demand transportation in general is a disruptive force on traditional taxi and limo services. The reality is that each market is unique in terms of rider demand and quality of existing options. So while SF taxis may have seen ride volume decline by 65 percent of late, this is likely more of a directional indication rather than an exact model for what can be expected elsewhere.

[Image via UCSF]