New Haven Should Follow Houston’s Lead on Transit. Wait, What?

Note: A slightly different version of this piece this ran in the New Haven Independent on 5/27/14, under the title “Houston Could Point the Way for City’s Buses.” 

(Opinion) Yes, New Haven’s bus system is broken. Hope can be found in, of all places, the Lone Star State.

The critiques of some CT Transit riders—made in this Independent article—are pretty standard fare for transit riders in all but the most transit-rich cities: The buses don’t come reliably on time, off-peak frequency is terrible (and weekend service essentially nonexistent), buses don’t always go to where the jobs are, and the route structure, relying on transfers at a central point (the Green) is incredibly inconvenient for many riders. These are all very valid, and accurate, criticisms.

Meanwhile, state transportation Commissioner James Redeker insisted (in an Independent interview; click on the video to watch it) that “a bus service is scheduled based on demand” and that New Haven’s bus system runs on a fine balance of resource allocation and response to demand, and that sparse off-peak service and the Green-centric route structure are both results of that calculation.

Is New Haven’s transit system actually the best it can be, demand considering? I’d argue no.

Assuming that Commissioner Redeker’s contention that resources for urban transit are severely limited is accurate—and considering CT Transit’s recent fare hike, which seems to have been the result of a legislative raid on dedicated transit funds, I see no reason to dispute the argument that statewide political will for increasing transit bus funding is sparse—New Haven has to find a way to improve its bus operations more or less within the bounds of current operational funding. Luckily, last week also provided a high-profile example of a city which is doing exactly that—and in an unexpected place.

Houston, of all places (yes, the city with no zoning code and massive urban sprawl) is currently playing host to a much-anticipated (among the transit and urbanist world) revamp of its bus system. The regional transit agency hired consultant Jarrett Walker, a favorite of transit purists everywhere, to redesign its system around goals of increasing transit reach and ridership. As with many of his projects, Walker’s plan for Houston is built around the concept of a frequent network, a system of buses and light rail which are assured to come every 15 minutes or less, every day of the week, at all hours during which the system operates. The new plan will place 111 percent more riders within a 1/2 mile of frequent buses, and 55 percent more jobs. You can see a mesmerizing .gif of the transformation below:


The magical part? Houston is doing it with no new operating funds. Operational funding for the system will stay at exactly the same level that it had before. That shouldn’t seem possible, but, as Walker explains on his blog:

• That’s how much waste there was in the existing system.  Waste in the form of duplicative routes, and due to slow meandering routes created due to a few people’s demands.

• Hard choices are proposed about expensive service to very small numbers of people.  The plan devotes 80 percent of Metro’s resources to maximizing ridership, which all of these frequent lines do, and only 20 percent to providing access to people living in expensive to serve places. Currently only about 50-60 percent of resources are devoted to services where high ridership is a likely outcome.

This shift in focus will have negative impacts on small numbers of riders who rely on those services, but these were small numbers indeed. (About 0.5 percent of existing riders end up over 1/4 mile of service, and most of them are just over that threshold.  Often, their longer walk is to a better service, a tradeoff that most people are willing to make in practice.)

In other words, Houston was able to make some easy decisions (to cut inefficient or wandering routes), but also had to make some hard ones (retrenching service from areas that aren’t densely inhabited and/or are difficult to serve on a logical route).

The numbers of people negatively affected are small, if Walker is to be believed, but any city following this model of resource allocation will clearly irritate some people. The key is to make sure that the burden of cuts doesn’t fall on lower-income communities, and to carefully consider whether the demands of certain vocal segments of the population make sense in a regional context.

In addition, some riders might find that their former one-seat ride now requires a transfer, since the system is less core-centric and more gridded, but in theory even they should end up saving time. And again, this transformation is planned with no new operational funding.

It’s not clear to me how exactly the financials of the Houston revamp are to be worked out such that it does not require new funding, but given Walker’s recommendations in his book, I imagine that it consists of:

• The aforementioned pruning of costly, low-ridership routes.

• Projecting that higher ridership means better farebox recovery ratios (the percentage of operating costs paid directly by rider fares, which according to Redeker is around 25 percent in New Haven—not far off from the average in a city this size).

• Eliminating costly “split shifts” for drivers, where drivers get a premium for driving during the busier morning and afternoon peak hours, and not in between.  By employing drivers on standard eight-hour shifts and giving them something to do during the day, the transit agency can actually save money even if it is paying for more hours overall. (This, of course, assumes that labor agreements allow such an arrangement.)  Renegotiating labor agreements, of course, is often a very, very politicized question, which brings me to my final point about how to go about bringing a Houston-style approach to New Haven.

Implementing a frequent network in New Haven would require two things above all: political will and changing the mindset of transit operations from trying to serve demand to trying to induce it. The political question is notably tricky in a situation where the city does not control its own transit system. New Haveners from top to bottom can and have shown support for a revamped, improved transit system, but the city lacks the financial capacity to fund improvements itself, and is in any case dependent on the willingness of  state-run CT Transit to commit to following the city’s lead.

That dependency on CT Transit also means that shifting the agency’s conservative mindset (and I assign no blame there—transit agencies are captive to, and have their tone set by, a political system that values transit little) will require sustained political pressure from both the Harp administration and New Haven’s state and potentially even federal lawmakers.

Capital-intensive, sexy transit projects like improving Metro-North’s New Haven Line,  building out the New Haven-Hartford-Springfield commuter rail, or even the New Britain-Hartford CT Fastrak busway offer concrete rewards both to transit riders and to politicians who can point to newly poured concrete and newly laid rails and show off what they have accomplished. Doing something as simple as reworking the way a city’s already-existent local bus system works, and making the buses come more frequently, offers no such flashy rewards. It just requires a lot of hard work, and focus on the issues from both the grassroots and the leadership. And that’s hard.

At the same time, though, New Haven’s transit is stuck in a downward spiral where there is only an attempt to match service to demonstrated demand, no attempt to build ridership or move people from cars to transit. If Houston, of all places, can bring together a transformatively cheap and simple transit plan, why can’t a city that was built around horsecars and still longs for its lost trolleys?

5 thoughts on “New Haven Should Follow Houston’s Lead on Transit. Wait, What?

  1. Yes, but.
    1. New Haven doesn’t have an underlying grid the way Houston does. It has a downtown grid, but go just a few blocks outside the Green and the street network becomes more hierarchical, which makes it harder to create a frequent grid. Jarrett mentions Sydney as one city where a frequent bus network is more problematic for this reason. For example, try to come up with a coherent bus route that connects Fair Haven or points east with Hamden or points north.
    2. New Haven conversely has a possible rail line to look into reactivating: the Farmington Canal. It conveniently parallels Dixwell, the busiest bus corridor. It is grade-separated in most of downtown, although it requires a new connection of a few hundred meters at its southern end.
    3. Presumably, pruning low-ridership routes in New Haven means killing the meanders, and instead concentrating frequency on the core routes? The issue there is that often the meanders are only at the outer end of the line. Maybe they consume more operating-hours than I imagine, but they’re not necessarily splitting frequencies.
    4. Does New Haven even have major job centers outside downtown? Hamden has a couple shopping centers, but at least on OnTheMap, I don’t see anything like the secondary job clusters of Houston.

    • Thanks for the feedback, Alon! Some thoughts:

      1) True, New Haven doesn’t have a coherent grid in the way that Houston or Chicago does, but the street layout makes more sense than one might think. A true “grid” system might not work, but the introduction of a few “crosstown” lines is something Independent readers have been begging for, and I think there are at least two or three that could work. I’m planning a follow-up post to this one on that exact topic, but for now there are two primary routes I’m thinking about: one from West Haven (Savin Rock) via the West Have M-N station, Campbell, UNH, Forest Road, Central, Westville Village, SCSU and Arch to Dixwell, and another from Howard Avenue in City Point via Sherman Parkway, Hillside, Edwards, Humphrey and Lombard to Fair Haven. You could also run something on the East Shore from East Haven or Lighthouse Point to Montowese and/or North Haven, but that’s not a very dense area. Mainly, “gridding” the system would consist of straightening out the primary routes and running additional frequencies.
      2) You and I had this discussion on Twitter once. I think the Farmington Canal segment would be great as LRT, but people have been fighting for decades to get a trail there–I don’t see them giving it up anytime soon. The only way I could see that happening is if crime concerns on the trail get way out of hand (Newhallville is a pretty rough neighborhood). There have been a bunch of muggings along the trail, and there was one particularly publicized case where an elderly Yale professor who was working on a model home in the area was attacked. I don’t think it’s enough to convince people to give up on the trail (nor should they for that reason), but it’s the only way I see the ROW becoming open for transit. A BRT-lite system on Dixwell is much more likely.
      3) Most, but not all, of the meanders are at the outer ends of the line. Check out the end of the Z line near Howard Avenue. There’s also a ton of pruning to be done in suburban areas–basically nothing outside of the New Haven city limits, Dixwell and Whitney in Hamden, downtown West Haven and East Haven, the Route 1 corridor, and maybe Foxon Road deserves transit service. And some of the frequency splitting is problematic–Whalley, for example, has buses every 10-15 minutes during rush hour as far as Westville Village (a tolerable frequency, though I hear those buses are very, very crowded), but the line splits there, and there is still plenty of ridership potential Westville Village–Amity (a distance almost as long as the downtown-Village segment).
      4) I’ve found the RPA jobs access map really good for this: . Absolutely true that most of the jobs in New Haven are concentrated downtown, but there are a few secondary centers: St. Raphael’s hospital (which is currently inaccessible by bus from the Hill, without going through the Green, even though it’s basically adjacent), Science Hill and “Science Park” (next to the old Winchester factory) which are near downtown but not in it, UNH, some manufacturing in Fair Haven, etc. The fundamental problem–and this is not really something that transit can address–is that very few of the downtown jobs are actually held by people who live in New Haven. The latest city plan had it that 77% of jobs in New Haven are held by people living outside the city. That’s a huge issue, and it also explains the massive amount of parking Downtown, which I know you’ve (rightly) criticized in your posts about New Haven.

      • The lack of a grid makes these crosstown routes really awkward, though. They involve turns, side streets, and realignments between major crosstown streets that don’t align with each other – for examples, the Edwards-Humphrey and the Howard-Sherman turn.

        The West Haven route also has to detour to serve the Metro-North station. This is a predictable consequence of the fact that the station was placed without regard for the preexisting bus network. No biggie, it’s only $100 million.

        The pruning is a lot more productive, yeah. I didn’t remember this, but even the D is horrifically branched. A lot of those routes just look like night circulators, going everywhere without much speed or frequency.

        Re commuters from outside the city, some of them are within transit range. As of 2000, 32% of New Haven workers live in New Haven; 22% live in Hamden, West Haven, and East Haven; 11% live in the Shore Line East towns; and 5% live in the New Haven Line towns, of whom about half live in Milford. We can discount commuter rail for now, but presumably a fair number of people in Hamden could be induced to take the D and J. I don’t think BRT-lite would be it (Farmington rail might), but higher frequency could help.

  2. Yeah, the crosstown routes are awkward, but it’s not, for the most part, horrible; and it’s not like anyone’s going to be riding them for the entire length. I’m thinking of my proposed routes as the equivalents to a couple of crosstown/circulator routes that were added here in Albany in CDTA’s last service revisions, the 100 ( and 138 ( They were both asked for by lower-income communities to give better access to major job centers (especially hospitals), and both have been reasonably popular, especially the 100. I think they’d do even better if they ran more than every half hour.

    The West Haven station siting is unbelievable. It’s not a huge detour for a bus on Campbell, but it is something, and there is land available where Campbell passes under the tracks for a station. Just awful.

    The branching is the worst part of the New Haven system, both for frequency and for legibility. It really, really has to go. The problem with Hamden is that it has some fairly dense areas and some really sprawly ones. The relatively dense areas are already decently served by transit (though because of branching, not as efficiently as possible). For those areas, I think simple increases in frequency and reliability would be the best incentives to gain transit users. Unfortunately, the buses wander off into the sprawly areas which don’t really deserve service. Ultimately, the goal for Hamden should be to redevelop the commercial areas along Dixwell into a denser, mixed-use kind of environment; the New Haven area needs to add housing, and you could add a lot in that corridor, replacing strip malls that have really struggled in recent years. Farmington Canal LRT would be great for supporting that, but if you can’t wrest the ROW from the trail people, then bus lanes on Dixwell would work too.

  3. Pingback: Sioux City: Straightening Buses and Getting Route-Length Right | Pedestrian Observations

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s