The Ups and Downs of Streetcars

Originally published 2/20/2009

In this post I look at a couple arguments for streetcars in Houston, backward-looking and forward-looking. I find one lacking but the other with potential. Next I look at some different network topologies, and finally I criticize the Skoda vehicles which have become the de facto face of modern streetcar operations in the U.S.

The First Wave

The first wave of electric streetcar construction lasted from about 1885 to 1920, primarily as a means of selling off “suburban” lots that would otherwise be too distant from the city core. By 1925 automobiles were common enough that electric traction was no longer required to sell new developments, and streetcar expansion tapered off. From there, American streetcar companies can be roughly divided into two groups.

The first group invested in infrastructure, purchased new equipment, modernized existing equipment, cut underperforming routes, found ways to reduce labor costs, and generally created operational efficiencies that allowed them to sustain themselves well in the 1950’s. Perhaps the best known of these lines was the the Oakland and East Bay Area’s Key System, which consolidated a hodgepodge of local-service streetcar lines into a regional network with several express lines operating direct to San Francisco via the Bay Bridge. The demise of the Key and other systems like it had less to do with profitability and more to do with political and corporate intrigue. The widescale purchase of streetcar systems by a GM-backed consortium intent on creating a captive market for motor buses (satirized in “Who Framed Roger Rabbit”) is a common sticking point for transit advocates.

The second group of streetcar systems made minimal investments in the physical plant and generally ran the 1885-1920 systems as-is until they fell apart. And it is into this second group of systems which Houston’s streetcar network falls. Steve Baron’s Houston Streetcar History Pages has the details. Houston’s last cars were ordered in 1927. The entire system was gone by 1940. By contrast, the Key System ran until 1958, as did Chicago’s trolleys. The busiest lines of Los Angeles’s Pacific Electric lasted until 1961, and Washington DC operated trolleys until 1962. Closer to home, Dallas lasted until at least 1960 while El Paso’s modern PCC cars looked very healthy in 1958. Meanwhile Houston’s Bellaire line was gone by 1927, a victim of “frequent derailments caused by worn-out track.” Safe to say that Bellaire was a money-loser, even at the height of the roaring 20’s.

Arguments For and Against

A commonly-heard argument for streetcar is “Houston had streetcars once, so it makes sense to bring them back.” The above history destroys that argument. Houston’s streetcar extensions were an ancillary component of various real estate plays, and were not maintained after the lots were sold. At a time in history when comparable cities were modernizing their fleets with PCCs, Houston was closing down its last lines.

It’s easy to conclude that Streetcars won’t work in Houston. And indeed, this might be the case if Houston was the type of slow-growth, overly-zoned city whose inner neighborhoods look nearly the same as they did at the height of the streetcar era. But H-town is dynamic. It changes. Large swaths of inner Houston look nothing like they did in 1940, and in some places newer, denser development outnumbers the original housing stock 4 to 1.

Although streetcars aren’t justified by the historical record, they may be justified by present and future growth patterns. Which means if we’re going to do any kind of successful analysis on potential routes, we need to junk the “restoration” mentality and start from scratch. Forget about “bringing back” service to Heights Boulevard – that neighborhood has seem comparatively low-density redevelopment, and its residents have typically opposed condos, townhomes, or anything that isn’t a restoration or imitation of a bungalow. But what about Washington? Montrose? The East End? Where should we build, if at all? To answer this question it helps to look at what the objectives of a streetcar line are.

Network Design and Topology

Streetcar isn’t really any faster than the bus – streetcars may use dedicated lanes and priority signal timing, but this can be done with buses too. Rather, most of the benefit of a streetcar system derives from its permanence. There’s something reassuring about a transit line that can only run on a certain route. Streetcars won’t stop one block over at peak periods, nor will every streetcar between 3:25 and 5:05 bypass your stop to serve some community center – the kind of things common to bus routes. And trains offer a nicer ride than even the most modern bus. The phenomenon of rail bias results: replace a bus with an equivalent train, and you’ll see an increase in ridership. (The inverse is also true, as the recently-dieselized transit companies of the 1950s found out).

The other derived benefit of streetcar’s permanence is its effect on new development. Streetcar represents a substantial investment in the provision of comfortable, frequent transit service to a district, and helps to promote denser development by increasing access to the city and decreasing the need for parking. Thus the ideal streetcar line serves a neighborhood that has already seen significant density increases (proving that the market will support dense development there) and has room for even further increases in density, which can be enabled by the streetcar.

Another criteria applies to Streetcar in the Bayou City. Because Houston is polycentric, any streetcar line must be double-ended. With full buildout of Metro’s light rail lines, we’ll have a core transit network linking all of inner Houston’s major employment centers. By providing interchange with this network, as well as future commuter rail services, a single streetcar line can serve riders bound for a myriad of destinations. Enough theory; let’s look at some applications.

(click image for larger version)

The shaded areas represent neighborhoods in the box created by 10/45/59/610 which have seen significant residential density increases in the last few years. These are the places that are “hot” right now. (Denser areas which haven’t seen much recent activity are not shaded.) From a future development perspective, these areas are a prime target for streetcar service. Washington Avenue seems like a possible choice, so let’s look at some prospective routings:

(click image for larger version)

Here’s a basic streetcar line running down Washington from Downtown to roughly the WOW Roundabout. It serves the entire Washington corridor, providing access to hip new housing and entertainment options. But look more closely… this line is only useful for getting downtown. It’s reasonably convenient if that’s where you want to go. It’s okay if you want to head for the Medical Center, since that only requires a single transfer to the Red Line LRT. But what about other destinations? Greenway Plaza requires two transfers – one Downtown, one at Wheeler. And you can forget about Uptown – not even the most transit-dependent individual will try to get from Shepherd/Washington to Westheimer/Post Oak by going through downtown. No, they’ll just take the bus.

In truth, by not connecting to anything at its outer end, the Washington line shown above is probably squandering 40 to 70% of its potential ridership. So let’s look at a couple options for fixing that.

(click image for larger version)

Extending the line to the Northwest Transit Center provides a direct connection to:

  • Metro Uptown LRT
  • Express bus service to Energy Corridor
  • Express bus service to 290

The Galleria is now a single transfer away, and unlike Downtown (where a transfer is often a several block walk) we can put the Streetcar and LRT platforms right next to each other. The route to Greenway Plaza is about the same distance, but it only requires one transfer. Looking at this particular route in context, one almost wonders whether it should actually be part of the LRT system, with Washington Line trains operating Downtown-Washington-Uptown-Hillcroft and University Line trains operating Eastwood-Richmond-Uptown-NWMall.

(click image for larger version)

A cheaper option is to stick a commuter rail stop in the middle of Rice Military and connect our streetcar to that. In this case the streetcar is essentially providing a local service over a corridor where the CRT is running express; this sort of tiered service is already used in SF (Third Street LRT / Caltrain) and Chicago (various “L” lines / Metra). This line is still inconvenient for Uptown or Greenway, but serves other markets; you could live off Washington and work at NASA, for example.

The same concepts apply when we look at other corridors. What about West Gray?

(click image for larger version)

Nice corridor, but it doesn’t go anywhere besides Downtown. Most other destinations still require those two dreaded transfers. So why not connect it to the University Line?

(click image for larger version)

Running it down Hazard keeps the trolleys out of heavy Shepherd traffic, and doesn’t duplicate METRO’s extremely nice 26/27 Orbital Bus. A route on Shepherd stays on arterials, but would probably require some serious ROW acquisition for bypass lanes at intersections. (please don’t take the Taco Bell away.) Shepherd also squanders ridership potential because half of the effective walking radius is in deed-restricted River Oaks, as opposed to the mostly-unrestricted (and thus developable) Montrose.

(click image for larger version)

Jogging west on Westheimer to Kirby serves West Ave and the other new developments in Upper Kirby. But what’s important is that all these routes interchange with the University Line, providing a single-transfer ride to Uptown and Greenway.

Once you start to think in terms of transfers and the larger system you can analyze other routes using the same methodology. Montrose can work if the south end connects to the Main Street and University Lines and the north end connects to either Downtown or the NIT. Dowling works almost by definition, since it can connect to the Harrisburg Line at the north and the University Line at the south. A Jackson Hill / Dunlavy route can work if it there’s a Washington/Katy LRT line available to provide its northern interchange point.

The common denominator: Well-connected routes will always perform better than isolated ones. And its corollary: any Streetcar lines must be secondary to regional LRT lines. The LRT system can succeed without streetcar connectivity, but the reverse is not true.

The Limitations of Skoda

I’ve spent considerable time in a number of West Coast cities and have had several occasions to ride the Tacoma LINK and Portland Streetcar, both of which use the Skoda vehicles that now appear in literature promoting any new streetcar proposal. Portland’s brightly-painted, modern-looking cars make great visual copy – but you need to ride them for yourself.

Because the Skoda vehicles are noisy, clunky, and exceedingly slow. The traction motors make an awful screeching noise upon acceleration, and the suspension transmits every imperfection and joint in the rails, as well as a considerable amount of track noise. They are nothing like the smooth-running LRVs which we’ve grown used to riding in Houston and other American cities. Even the restored PCC cars operating in places like SF and Kenosha have quicker acceleration and a smoother ride than the Skoda vehicles. And that’s 1940’s technology.

So what options exist for a potential Houston streetcar network? Plenty. In the US, the popular conception of an LRV is a 104.5″x90′ vehicle roughly compatible with the specifications of the Frankfurt U-Bahn, the source of the first LRVs produced by Siemens in North America. But the U-Bahn is actually a suburban metro system, more comparable to the DC Metro. Almost all Euro-spec trams are closer in size to the 8′x70′ vehicles in use in Portland and Tacoma, and some are even smaller. Several of these manufacturers exist in some form in the U.S.:

  • Siemens: Manufactures the Avanto vehicles used by METRORail. Also sells the variable-length Combino Supra to a number of European customers
  • Kinki Sharyo: Manufactures Dallas’s LRVs. Also manufactures streetcar-sized LRVs for Hiroshima and Manila
  • Bombardier: Manufactures those ubiquitous double-decker commuter rail cars. Also produces an entire line of Euro-spec trams
  • Kawasaki: Builds a sizable chunk of the New York City Subway’s rolling stock at its plant in Yonkers. Also produces a number of streetcar-sized trams for Japanese customers
  • Alstom: Collaborated with Bombardier on Amtrak’s Acela train. Sells the Citadis tram globally.

Any of these companies could provide an attractive Streetcar to Houston, as could others; “buy American” requirements contain a one-city loophole, which allows a single agency to purchase foreign-built vehicles. For instance, Las Vegas inaugurated its BRT system with buses from France-based Irisbus. Houston could do the same with any number of manufacturers worldwide, perhaps enticing them to set up manufacturing in the Bayou City. After all, New Orleans produces its own streetcars in-house, perfect replicas of the historic stock but air-conditioned and ADA compliant. Houston could manufacture its own streetcars too.

Whether or not Streetcars make sense for Houston depends on a number of factors, including development patterns, progress on the core LRT system, and the extent of future LRT/CRT lines. Given continued economic growth and a continuation of present demographic trends, some Streetcar service seems likely. And if this is indeed the case, wouldn’t it make sense to keep Houston’s streetcars… in Houston?

4 thoughts on “The Ups and Downs of Streetcars”

  1. Never rode a Skoda car, but the Hiroshima streetcars you show ride like a dream. And I swear the rails in many parts of Hiroshima haven’t been maintained in decades. Plus, they’re BIG.

  2. It would be nicer if the shaded map images could be even bigger. It’s really hard to read the shaded name designations in such a small pic. Or even the medium pics on Flickr.

  3. So, what are you saying? To me, these streetcars you’re posting images of (like the Skoda) look very much like the Houston light rail that runs through Main. To me, it really looks like the based on the routes you’re showing, theses lines don’t really go anywhere and could be better serviced by a bus. Perhaps dedicated bus lanes, then?

  4. I’m pretty sure they used to be bigger, but flickr killed the large images. The originals for this are on an ancient desktop computer with no internet access, which I primarily keep because my laptop can’t run AutoCAD LT. Next time I’ve got that thing fired up I’ll try to remember to upload the full-sizers to imgur.

Comments are closed.