Can Microtransit Serve As A Public Provider?

There has been some criticism of microtransit. Here I dissect the different points.
By Ethan Finlan | Dec 03, 2018 |
By Ethan Finlan | Dec 03, 2018 |

The concept of “microtransit” appeared about four years ago to much hype. Today, it finds itself on the defensive. While it remains largely popular in the abstract, influential figures in the transit planning realm have pushed back, doubting whether the microtransit experiments that transit agencies have pursued are worthwhile. Because microtransit operations can carry inherently fewer riders per vehicle than most buses, the argument goes, they're less efficient, and increase, rather than decrease, vehicle miles.

The charge is led by Jarrett Walker, the influential transit consultant whose work in Houston helped the city reverse ridership loss. In February, Walker devoted a week on his Human Transit blog to exploring the benefits and costs of microtransit. His conclusions, in sum, were that: 1) microtransit is less efficient because, on average, pilots thus far have carried fewer riders per vehicle than the worst performing bus routes, 2) it is mostly a rehash of “dial-a-ride” services, which already existed, and 3) because microtransit users tend to be higher income, any attempt to subsidize them risks devoting scarce resources to elites.

Walker is a smart analyst, and this is a critique that warrants attention. Must we conclude that microtransit is doomed to inefficiency, or that any attempt to shift subsidies from underperforming bus routes to flexible services is a waste of money?  The answer may not be as simple as “yes” or “no,” and in fact depends on what the goals of these services are. To find the answer, let's explore his three points.

Does microtransit move less people than fixed-route services?

In the abstract, the answer to this question is yes. Walker demonstrates this in his post on the purpose of microtransit, comparing modes of transport by trips made each service hour. At the upper end of the spectrum, a frequent, grade-separated subway line moves more than 200 people in this time frame, while subsidized “paratransit” services focused on accessibility carry the fewest (0-2 hourly). Microtransit came out ahead…of the latter option, carrying 0-3 per service hour. Looking specifically at the question of “coverage” – where the priority is not sustaining high rider levels but rather providing service to an area not conducive to walking – Walker claims similarly dismal results, because on average, low-ridership buses carry more than 10 passengers per hour. He concludes that if a “microtransit” service actually accomplishes this throughput, it is in fact a “fixed route.”

One problem with this conclusion is that most microtransit services are relatively new. Granted, some have failed, the most prominent being Bridj. It was one of the original microtransit services, marketing itself as “Uber for buses” and eventually operating in four U.S. cities, including Kansas City, where it received a subsidy from the regional transit authority. Bridj closed its doors last year, although it was bought out by an Australian consortium which has since revived the service there.

Many transit advocates pointed to this failure as evidence of private microtransit being unviable. However, where Bridj failed, other companies are growing. Chariot, which operates on a similar model, while prioritizing low fares and efficient service over amenities, continues to operate in New York and San Francisco, while also doing various contracted services. Via, a service modeled after Israeli shared-taxis, operates in four cities. Uber and Lyft continue to experiment with moving more passengers per vehicle. Put simply, it’s too early to declare microtransit a failure.

More to the point, let’s consider the question of whether lower ridership on average routes with “flexible” characteristics renders microtransit inefficient, relative to “coverage” services with buses. Having established that a low-ridership conventional bus moves more people currently per hour, the next question is whether the 15-20 people on this hypothetical bus are being served well. Just because they ride the service doesn't prove this, since coverage route ridership tends to concentrate on people who can’t drive, either due to a physical handicap or poverty. For those who can get cars, they are likely to move to cars, particularly with access to auto loans. This shift has been cited as a likely cause of decreasing transit use in Los Angeles.

More flexible systems, by contrast, may still offer better value for the rider than a lower-frequency coverage bus route. Riders are likely to aggregate destinations, such as a suburban office park or a mall, that is difficult for an arterial-serving bus route to reach without distorting its route. Further, microtransit may even improve such bus services, by providing first/last mile connections. In that case, service hours may not be the best metric, but rather, total people moved throughout a service day. Continued use of microtransit for coverage purposes may well render higher total ridership, while still mitigating a larger increase in VMT that would come from riders shifting to private car ownership.

What is Unique About Microtransit?

The criticism that microtransit is only another form of dial-a-ride services may apply to smaller agencies. These programs have existed since at least the mid-1980s, and relied on eligible riders contacting the agency for pick-ups at their location. Microtransit, like transportation network companies, works on an app-based ride hailing model.  Angie Schmitt of Streetsblog recently criticized microtransit on these grounds – that the “best services” are more akin to these existing programs.

Yet really, this is not a criticism. With smartphones becoming dominant, naturally transit agencies should try to utilize them. If smartphones make the process easier than calling a dispatcher – and possibly needing to wait on hold due to over-subscription – microtransit has improved the “dial-a-ride” concept.

Moreover, there's a distinction between pilot programs and services that are established organically, as was the case with the original microtransit operations (and their predecessors, jitney services). Pilot programs, such as the one being tried in Arlington with Via, and those cited in an Eno Center report mentioned by Walker, tend to cover geographically-limited areas, and rely on a single provider. Organic services, by contrast, often see competition – Chariot competing with Uber ExpressPOOL, for instance – and reflect market demand.

Such competition plays the crucial role of identifying routes with concentrated demand. Again, Walker is correct that in dense urban environments, the more passengers carried per service hour, the more efficiency. Microtransit operators – who concentrate in dense areas – would be the very ones who achieve these efficiencies. In a less-regulated environment, these operators would scale their services to become true fixed routes over time, as the impacts of agglomeration dictate travel patterns. Even if they don't, the data can still be used by public agencies.

But the problem is that cities instead over-regulate microtransit, so that we can't really know how well the industry would do. An example is when New York City prevented a private operator from providing replacement service on cancelled bus routes in Queens.

Do microtransit subsidies just subsidize higher-income riders?

This is also a valid concern - what if money is being used for microtransit services that disproportionately benefit the rich, and take away from coverage services?

This problem could be addressed by providing subsidies on the demand side, and letting providers enter and exit the market as demand warrants. That said, this may prove more challenging for lower-density markets, as companies would have less incentive to enter them. Such a "voucher" experiment is still worthwhile, though.


In conclusion, microtransit's popularity speaks to the inefficiency of public transit in America. People with means are willing to pay higher fares to avoid a public system that doesn't serve them. Whether microtransit can be efficient as a broad public provider is more complicated, but right now, the answer isn't yet a decisive yes or no. The way forward for transit agencies is not to scrap their pilot microtransit programs, but to use them to explore benefits and drawbacks, as a path to providing the best mobility possible.