Tuesday, January 4, 2011

Beyond thick and thin markets for transit

In their extremely useful paper on private bus transit (PDF), Daniel Klein, Adrian Moore and Binyam Reja give us the concepts of "thick" and "thin" markets for transit: "Another distinction of fundamental importance is whether ridership on the transit route is potentially heavy enough to sustain the cascade of jitneys in the absence of scheduled service," they write on Page 39.

The Flushing-Chinatown van run is an example of a thick market. You could say that the subway provides an anchor, but it's a very inconvenient one: if you give up on the van you have to walk three blocks north to the subway, and then take that to another subway that might not go exactly where you want to go. These vans really function without any anchor.

The vans from Bergenline Avenue to the Port Authority are an example of a thin market. The New Jersey Transit 156 and 159 buses provide scheduled service in case the vans don't come; I've used them myself. They are run by NJ Transit, but they could probably be run at a profit by a private company.

Thick marketRidership can sustain a jitney cascade without an anchorFlushing-Chinatown
Thin marketRidership can sustain a jitney cascade with an anchorBergenline-Port Authority

This doesn't capture other possibilities, though. What about a line in some far-flung sprawl suburb where car ownership is very high and roads are wide? For example, Dutchess County Loop A. A line like that is never going to make a profit as long as those conditions persist. To exist, it must remain subsidized until gas prices rise above eight dollars a gallon. Let's call it "No market."

There's also a scenario in between Loop A and the Bergenline Avenue buses, where jitneys can run with an anchor, but the ridership doesn't provide a profit for the anchor, so it has to be subsidized. The Bergenline Avenue buses that go across the George Washington Bridge may be an example of that. Let's call it a "Wafer-thin market."

I would argue that most of the routes chosen by the TLC for their pilot program are wafer-thin markets. Due to competition from parallel routes and private cars, they cannot support profitable transit by themselves, but they may be able to support a jitney cascade if they have a subsidized anchor.

The B39 route across the Williamsburg Bridge had potential to be a thin market. If it were better marketed, it might have worked. The only one that had the potential to be a truly thick market was the B71, if it had been extended through the Brooklyn-Battery Tunnel. It could have provided Red Hook and Carroll Gardens residents with a one-seat ride to Financial District job sites, possibly being time-competitive against the subway. I guess we'll have to wait to see that one tried.

Here's our new table:

Market typeDescriptionExisting examplePotential TLC route
Thick marketRidership can sustain a jitney cascade without an anchorFlushing-ChinatownB71 through Brooklyn-Battery Tunnel
Thin marketRidership can sustain a jitney cascade with a profitable anchorBergenline-Port AuthorityB39
Wafer-thin marketRidership can sustain a jitney cascade with a subsidized anchorBergenline-GWBB23, B71, Q74, Q79
No marketRidership cannot sustain jitney service, even with a subsidized anchorDutchess Loop A

Notice that I've still put the Q79 up as a wafer-thin market. The TLC has put that route out for bid, but I really don't think there's enough ridership to support it without some kind of subsidized anchor. It's great that Bob Friedrich is still fighting for service on Little Neck Parkway, but his best bet is to try to find some funding to subsidize the service. It's just not going to work without subsidies unless there are other drastic changes in the neighborhood.


busplanner said...

There is an issue (dilemma might be a better choice of words) with both thin and wafer thin markets as you describe them.

The cascade of jitneys diverts some ridership and revenue from the anchor. For sake of argument, let us assume the jitneys allow some reduction in service by the anchor but not proprotionate to the total number of fares removed from the anchor.

This offers riders a better frequency of service in the corridor. But it weakens the anchor. As cost pressure rises on the anchor, the anchor cuts service; but not necessarily on the route with jitney competition (as that route is still well patronized and may be either less profitable than previously or only marginally unprofitable). Instead, the anchor cuts service drastically or completely on another route that could not exist without a great subsidy and would never support a cascade of jitneys.

Thus, jitneys can improve service on a specific route while weakening the overall network.

Cap'n Transit said...

Busplanner, did you read the Klein, Moore and Reja paper? They have very specific recommendations for dealing with this problem in thin markets.

busplanner said...

Cap'n - I see nothing in Klein et al that addresses the point I was trying to make. If the route is part of a network and the anchor becomes less profitable because of the cascade of jitneys, the entity running the anchor route will have less available revenue to cross-subsidize the thin or very thin markets where there is no jitney competition. The entity will then further reduce or eliminate the very thinly supported service.

Separately, I have some concerns about the curb right arguments presented by Klein et al for a variety of reasons even if one confines consideration to a single route (the efficacy of enforcement being the greatest one).

Also Klein et al acknowledge that jitneys do not have to accommodate passengers who they choose not to accommodate. Thus, in the thick route scenario where the jitneys remain after the anchor disappears, that leaves some members of the public wishing to use a transit service unserved.