Can the Densities of Some Neighborhoods Be too Low for Transit to Work?
Streetcar tracks in New Orleans. Credit: Ray Devlin
In cities across the United States, you can find examples of “streetcar suburbs”—enclaves of mostly single-family homes built between the turn of the century and the 1930s. These are often good-looking, tree-lined places full of heterogeneous character and history, in many ways so different from contemporary suburban sprawl.
One thing most of these these now in-town communities share with more modern suburbs, however, is a full-throated dependence on the private automobile, ironic considering the “streetcar suburb” appellation.
This raises an important question: How did these places transform from transit-needy to car-obsessed? Are the many streetcar lines being planned and built across the country destined to be abandoned once again thanks to low ridership?
It’s worth looking back at the old streetcar suburb model to understand what went wrong.
Unlike many of today’s trolleys, which are being built with the explicit purpose of generating medium-to-high density housing around stations, most streetcar suburbs were from the beginning to be made up principally of single-family or two-family homes. Developers often helped sponsor rail line construction through these neighborhoods to encourage buyers; other times, electric utilities established amusement parks at the end of the corridor to spur ridership.
This approach, however, did not produce high enough users to save private transit systems all over the country from bankruptcy and eventual government takeover. The nation’s overall transit ridership has decreased from about 160 average annual trips per person in 1920 to about 29 annual trips per person in 2000.
That’s partially a result of the vast increase in car use, but also a consequence of the decline in transit service caused first by the replacement of streetcars with buses and then later by the gradual phase-out of many bus operations.
Some have suggested that the switch from transit to cars originates in a conspiracy by General Motors and others to get people to drive cars in their place, but the more depressing reality is probably that most neighborhoods served by streetcars were not dense enough to allow for efficient, frequent service into them over the long-term.
Unlike inner-city districts with their medium and high-rise buildings, streetcar suburbs are characterized by low densities, little neighborhood retail within walking distance, and very few accessible jobs, three significant factors that make them difficult to adapt to transit. In other words, while they may have been built with streetcars in mind, they transitioned to the automobile age naturally.
Commuter rail, operating at much lower frequencies, still plays a role for such older suburbs at the outskirts of large cities, but that transit service is only possible with a very large downtown business district.
Why is the fate of the streetcar suburb so essential to the fate of today’s American cities? Because their difficulties retaining transit service suggests that communities of similar densities will not be able to support frequent, accessible transit service. Are today’s streetcar networks going to suffer a similar fate if they’re built in places with inadequate population?
Aaron Renn argued last week on the Urbanophile that metropolitan areas with populations of less than about two million inhabitants don’t necessitate the kind of high densities urbanists often promote. Citing the example of Columbus, Ohio, Renn suggested that because these regions are small enough in area to make commuting from one end to the other by car possible within a short amount of time, creating dense, walkable neighborhoods focused around a “huge, packed, downtown core” is not absolutely necessary.
In some ways, his argument rings true: for those driving private automobiles, neighborhoods like the former streetcar suburbs may be ideal. For businesspeople hopping from one side of the region to the other (“to lunch”), driving in medium-sized cities works fine.
On the other hand, for everyone else—the young, the old, the poor, the sick—such neighborhoods provide no alternatives. You can’t easily walk to school or to the store or to the senior center when you live in a streetcar suburb. Nor can transit operators provide adequate service, since densities are too low to make frequent buses possible.
The State of Washington considered a law last year that would have mandated up-zoning areas around transit stations to encourage higher densities at rail and bus stops. Though this legislation ultimately failed, its goal was on point: If you’re developing a transit system, you better make the neighborhoods around it adapt to using it. Otherwise, you’ll get a system with low ridership, clearing the way for the eventual dismantlement of the service and a reduction in transportation alternatives for the population.
Ultimately, if you believe in the mobility benefits brought on by public transportation, you have to also believe in increasing densities to support those services. The lesson of the streetcar suburb suggests that not taking that second step could eventually mean the failure of the first.
Yonah Freemark is an Urban Leaders Fellow, sponsored by the Rockefeller Foundation. He writes the Grassroutes column for Next American City. He also writes The Transport Politic blog. Contact him at yonah@americancity.org
For an Increasingly Urban Nashville, the Waterfront is the Place for Change
In New Haven, a fence separates city from suburb
The Cleansing Fire
Imagining a More Sustainable Suburbia
Stephen Smith in Washignton, DC on Tue, Apr 20, 2010 at 5:37pm
Saying that the streetcar suburbs failed on their own merits because the GM conspiracy is wrong is fallacious in that it doesn’t take into account all of the other possible reasons for their decline. For example, there’s one school of thought that believes that they were economically feasible, but that the government regulated them out of existence through measures like fixing fares at 5 cents despite wartime inflation, higher legally mandated labor costs on trolleys as opposed to buses, not to mention the widespread subsidization of the roads, which were at that time not mostly funded by user gees, but rather out of general revenues (which, perhaps not coincidentally, were in large part paid by “traction magnates”). Oh, and did I mention zoning? The neighborhoods could easily have been retrofitted to take advantage of higher transit capacities (adding stories here and there, bulldozing entire blocks and building larger buildings, buldozing corners of blocks for more retail/commercial).
Here are links to papers dicussing some of the aforementioned government interventions against the streetcars (I can’t figure out if a href HTML tags are okay and there’s no comment preview, so I’m just going to list them):
http://books.google.com/books?hl=en&lr;=&id=P4TrXitjuU8C&oi=fnd&pg=PR7&dq=streetcar+companies+five+cent+fare&ots=7vzrZXo8jd&sig=AKyzdYA5WiDWDHcPNvbY7Uj6o6Q#v=onepage&q=streetcar companies five cent fare&f=false
http://www.jstor.org/pss/3113171
http://www.jstor.org/pss/2119941
http://www.jstor.org/pss/1063661
http://www.marthabianco.com/juh.pdf
Amanda in Stamford, CT on Tue, Apr 20, 2010 at 9:02pm
You say that streetcar suburbs lacked easily-accessible jobs, but by definition a streetcar suburb was adjacent to a denser center of employment where nearly all the workers from the suburb traveled for work every day. Now, we might have more workers per household, but the likelihood is much lower that their jobs are located along the original streetcar line or in the historic center. In fact, the rise of the two-worker household makes it much more likely that the household has located itself halfway between two rather-long commutes to suburban workplaces, removing transit as a real option even when the home has bus or rail nearby.
Peter Smith on Tue, Apr 20, 2010 at 11:54pm
Can the Densities of Some Neighborhoods Be too Low for Transit to Work?
Yes.
This has been another edition of Simple Answers to Simple Questions.
Jon Reeds in London on Wed, Apr 21, 2010 at 8:04am
Is the point at issue not whether light rail transit is viable in low-density towns and cities, but at what point does competition from driving to work etc, undermine it economically? Most of these systems were built in a world where the car didn’t provide viable competition and they worked very well. But from the 1920s onwards motoring became a major player. And if commuting by car is environmentally unsustainable in a low-carbon world, then what does “viable” come to mean in this context? That will depend on how much commuting is made by other modes and if the answer is very little, then streetcar systems would be viable at low-densities. That’s not an argument for building at, or sustaining, low densities as plainly sustainable transport becomes viable much sooner at higher densities.
Certainly here in the UK, very few low-density 20th century suburbs got tramways, but the old, compact cities did away with every last system (except Blackpool) in the 30-40 years up to 1962. Bus and car competition played a part, but this was primarily policy-led.
Light rail transit can certainly be viable at very low densities in the right circumstances. Rural Belgium developed a vast network of rural electric tramways in the early 20th century and the last of it survived into the 1970s.
But, as I say, these are not arguments against densification.
Jon
Michael Holbrook in Cleveland on Wed, Apr 21, 2010 at 9:10am
The streetcar suburb layout was just an extension of the cities themselves, and it worked very well until people abandoned it for a variety of reasons. But inherent structural failure of the TOD concept was not one of those reasons. It worked just fine and always will. Most of the jobs and walkable retail are found downtown, where all the transit lines converge.
No there’s not a department store “walkable” on every block in a streetcar suburb, nor is there sufficient employment for everyone who lives there… instead there’s a transit line going downtown. That’s the whole point. Unlike today’s setup, the idea was never to have people do everything within a couple miles of home and never leave that area. Hence the transit. People used to go downtown for things, all kinds of things. Transit oriented means downtown oriented. Downtowns are retail hubs by definition. Rip out the trolley lines and watch your downtown collapse. Go ahead, try it. Oops… we already did.
JamesR in Bronx, NY on Wed, Apr 21, 2010 at 9:11am
I would argue that a lot of the types of communities to which the author refers are the worst of both the urban and suburban worlds. They’re often too dense to make driving and parking easy, but not dense enough to support effective transit. By transit, what I mean in this context is fast-headway rail or Bus Rapid Transit designed to serve more than just a low income population with no other options. In other words, choice riders. Yeah, many of these towns may have a heavy rail connection to the regional CBD, but the ridership of these systems is never maximized because of the omnipresent “last mile” connectivity problem from the station to the homes, coupled with the reality of inadequate parking at the rail stations themselves.
As far as examples of these types of communities, Northern NJ, lower Westchester County, NY and Nassau County, NY come to mind. Again, too dense for an easy, laid-back suburban lifestyle, not dense enough for real urbanism and effective transit. Upzoning and rezoning for mixed use in these communities is basically off the table due to very strong NIMBYism in these areas, but a computerized jitney-type shuttle service could do a lot to solve the last mile issue. Once people are in their cars in the morning, they are a lost cause as it is easier to just drive the rest of the way rather than transfer between modes. The key is to keep them out of their cars during their commute all together.
jfruh in Baltimore, MD on Wed, Apr 21, 2010 at 11:42am
One huge difference between yesterday’s streetcars and today’s is that yesterday’s were expected to be profit-making operations. Would those streetcars have gone out of business if operated under today’s environment, where 50-60 percent farebox recovery is considered a good showing and some form of tax/fee-supplied subsidy for transit is taken as a given?
I live in what I think of as an archtypical streetcar suburb—Charles Village, in Baltimore—but perhaps its a bit denser than what you’re discussing here? The housing stock is mostly two- and three-story rowhouses, and many of the latter have now been transformed into 3-flat apartment buildings. One thing we have is two low-level commercial corridors on either edge of the neighborhood, with restaurants, shops, and even a small grocery store. I don’t know if there’s an American English name for this kind of thing, but the British “high street” fits pretty nicely in my mind. Are these not typical features of other streetcar suburbs?
(There’s also a couple of full-sized grocery stores just beyond the neighborhood limits, and within walking distance, but I imagine that this is something of an anomaly in modern good-sized cities.)
Alexis on Wed, Apr 21, 2010 at 1:30pm
The other thing that’s not noted by this article is that streetcar suburbs used to be much more dense, not because the housing stock was different but because more people lived in each house. Families were larger and some of the neighborhoods of this type had other relatives, servants, nannies, etc. in the households as well. These days, with Mom, Dad, and one or two kids (and sometimes just one or two adults) the neighborhoods are far less dense than they once were.
JH in Toronto on Wed, Apr 21, 2010 at 4:10pm
Toronto’s streetcar suburbs never lost their streetcars (most of them, anyway). The streetcars back then were single cars, not the longer light-rail proposed today. These mondern lines ALWAYS raise property values and spur development along their corridors.
Streetcar suburbs were fairly walkable neighbourhoods, I’m not quite sure why you believe they aren not, and do have commercial strips nearby. I’ve been living in 3 over the last 4 years and I can walk to essentially everything I need in 5 minutes. The entire reason these neighbourhoods were built was because of the streetcar, before that it was necessary to walk to work, but the streetcar made living a bit farther from work and taking transit feasable.
They were transit cities back then, far more functional than the cities we have to navigate today.
Evan on Wed, Apr 21, 2010 at 11:15pm
Agreed, JH. I lived in Toronto in a streetcar suburb—walkable, aesthetically-pleasing, high property values, tonnes of commercial and community services within walking distance, and so on. The streetcars, as far as I know, are profitable and still running (efficiently and regularly) to this day.
Besides, a streetcar suburb like Riverdale in Toronto (provide your own US equivalent) achieves as high or higher densities than high-rise neighbourhoods in our car-oriented cities. Density not about packing people in to apartment blocks (the Tower in the Park fails). Effective population and service densities can be achieved in so many forms.
Free parking, laughable gov’t investment (what percentage of transportation money goes to roads and highways vs. tracks?), cheap oil, and, yes, big business are three major contributors to public transit’s downfall.
Rail is still the best motorized way (least expensive, most efficient, least detrimental to the environment, aesthetics, culture, public health, etc.) to move goods and people—as far as I can see/experience/read/hear/figure, at least—with exceptions to the rule.
aover on Thu, Apr 22, 2010 at 12:54pm
As someone who lives in a streetcar
suburb (in Columbus no less), I have to call you out on your analysis.
I can easily walk to all of the things you mention and more (grocery
store, senior center, corner store, Asian markets, multiple schools,
parks, boutiques, restaurants, bars, coffee shops, barbers, banks,
etc.). Everything I need is within a few blocks walk. In fact I can do
all of my Christmas shopping without a car and without leaving the
neighborhood. I can easily get to work without a car (bus or bike are
both options I use). It is served by the highest capacity and most
frequent bus line in the whole city. All of this was made possible by
streetcars and it continues to work because of the density that the
streetcar helped to develop.
Sure, the amusement park and zoo are gone and the old streetcar barn
is a KFC, but that hardly negates the impact that the streetcar had
on the neighborhood. It continues to thrive today, and is one of the
most attractive places to live in the city. In fact, it one of the
three residential areas of the city that held its own in property
values during the downturn. Those outerring ‘burbs that are truly car
dependent lost upwards of 20%+ in value.
So your next time you might want to put away the broad brush.
Streetcars helped make my neighborhood (and many other inner-ring
Columbus neighborhoods), and if they were put back they’d still work
today.
JP in LA on Thu, Apr 22, 2010 at 7:20pm
I don’t know what the ridership numbers are, but the 101 and 102 trolley lines to Media and Sharon Hill outside Philadelphia serve low density suburbs, and most stations are located in single family home neighborhoods far from any density. They do a decent job of connecting towns and neighborhoods to schools and other institutions, commercial areas, and the city. It should be noted that in addition to these trolleys, Media is also served by the regional interurban rail line, the R3.
I believe streetcars can serve areas of low density, serving mainly to connect neighborhoods and other transit lines such as heavy rail. This is one argument for bringing back streetcars to Washington, DC, which already has an excellent subway.
In northern San Diego County, the Sprinter has very high ridership numbers, exceeding initial projections. It is a high-frequency light rail transit line that connects low density suburban the communities. The line links to the Amtrak, Metrolink, and the Coaster for trips to San Diego or LA. In addition to linking communities, it provides transportation for two regional colleges.
So I don’t buy the argument that transit needs areas of high density housing and other development to survive.
Michael Holbrook in Cleveland on Fri, Apr 23, 2010 at 10:16am
I’m fairly certain that Cleveland sold all it’s streetcars to Toronto. Glad you like them. Wish we could get them back. Lakewood, Ohio, a streetcar suburb of Cleveland with duplex-style housing arranged along twin commercial corridors whose streetcars are long gone, still boasts America’s highest population density between NYC and Chicago… just a little short of 10k/sqmi per the last estimate I saw, and considerably higher than the density of Cleveland itself. Cleveland’s density currently ranks below several of its own (formerly) streetcar suburbs, due to all the demolition and decay within the city proper. Our light and heavy passenger trains are still jam packed every day. I’m not aware of any real-life situation in which this article’s thesis would apply.
jon on Fri, Apr 23, 2010 at 11:27am
couldnt you say they were dense enough at the time given that streetcar line construction was cheaper, operating costs were much less and despite the single family housing along the route, those adjacent residents all used the service and made multiple trips a day by streetcar (whereas now only a handful use the service and make at the most 1 trip a workday via transit)?
Alon Levy in New York on Sat, Apr 24, 2010 at 4:44pm
Yes, at very low densities, transit won’t work. But the threshold is lower than you think. Calgary, with 1,400 people per km^2, has developed a good light rail system, whose per-capita ridership approaches that of the New York metropolitan area. Sendai (with 1,300 people/km^2) and Sapporo (with 1,700) have busy subways, whose per capita ridership is close to that of New York City. Zurich’s suburbs, which range from 1,000 to 2,000 people per km^2, are connected to the city by a modern commuter network, with high frequencies throughout the day and good connections to the local streetcar network.
Bear in mind that in all of the above examples, higher density clearly would increase usage. Calgary seeks densification, especially in its downtown; Sendai and Sapporo lag behind denser Tokyo and Osaka in subway usage; the Zurich suburbs have much lower transit use than Zurich itself. But at any given density, an American city would have much lower transit use than a Swiss, Japanese, or even Canadian city, due to poor planning and neglect, which lead to high operating and construction costs and suboptimal route choices.
d.p. in Seattle, WA on Thu, Apr 29, 2010 at 2:24pm
Yonah, I’m sorry to say that it you had done even a bare minimum of historical research on Streetcar Suburbs, you might realize that the conclusion you draw—that such places are inherently too low-density for transit, holds no water.
In fact, the true lesson of Streetcar Suburbs is that medium-density, pleasant, contiguous, and aesthetically pleasing single-family neighborhoods can not only support transit, but can support highly frequent transit when that neighborhood chooses to (or must, as in pre-automobile times) use transit.
That’s why, before the car conquered America, streetcars ran through areas that we might consider relatively low-density near-constantly! In fact, lower density cities (Kansas City, Seattle, Winnipeg) and tiny cities (Missoula, Muncie) easily supported dozens of streetcar lines with extremely high frequencies. Commercial development spread linearly along the lines themselves, but the residential areas they serviced were often of extreme low density. The frequent and reliable (though usually slow) trolley service more than compensated for the distances walked from home to stop.
Are you even aware of the Interurban trolleys that criss-crossed this country? Even the most rural areas could still muster a tipping-point demand to support trolley service back when that was the primary mode of transit.
It was the migration to an automobile-centric mentality that wiped out demand for functional transit in Streetcar Suburbs—it had nothing to do with density levels itself.
And there remain obvious contemporary counterexamples to your theory. In prototypically suburban Brookline, Massachusetts—one of the earliest Streetcar Suburbs, and one with plenty of automobile usage today—3 trolley lines are, to this day, packed to the gills all the time!
Diana on Sat, May 22, 2010 at 4:53pm
Streetcar suburbs are very well suited to transit because they are highly walkable and adjacent to denser main streets (“high streets” as another commenter mentioned). Even if you don’t run a streetcar line directly through the residential areas, put one on the main street and people will use it.
If there is a lot of driving in streetcar suburbs today, it’s certainly NOT because they are inherently car-oriented. It’s because transit is long gone, and residents may be isolated from other parts of the city, even if they have access to some retail close to the neighborhood.
Streetcar suburbs on the West Coast are thriving. Only one thing is missing: the actual streetcar.
Brent in Toronto on Mon, May 24, 2010 at 11:41pm
The issue as I see it is not so much the abandonment of streetcars, but simply the changes that have occurred in the surrounding city. In the golden age of streetcar suburbs, there were two scales of development. Most local everyday needs were within walking distance of homes—smaller and more frequently-spaced facilities such as stores, schools, libraries, etc.—whereas larger regional draws and employment centres were downtown, easily accessed by streetcar. Since that time, there have been two significant land use trends that have impacted transportation:
a) Decentralization—downtown may still remain the primary centre for jobs, and possibly for major retail and entertainment facilities, but a large proportion of residents now commutes to areas outside of the downtown that are usually less conducive to transit ridership (whether because of poor transit service, or because the surburban workplaces offer ample free parking). If you assume 80% of the workforce in a given streetcar suburb used to commute downtown, whereas that number has now slipped to 40%, that’s cut ridership on that streetcar (or bus) service to downtown in half. (Potentially leading to the textbook downward spiral of ridership loss leading to service cuts leading to further ridership loss.)
b) Change from numerous local to fewer regional facilities. (I am sure there is a better planning term for this.) A streetcar suburb originally would have supported several mom-and-pop hardware stores serving the immediate local community. Most are now gone, replaced with a few centralized big boxes that are 10 to 20 times the size and, by necessity, designed to be driven to. I believe the most critical examples are grocery stores and pharmacies (and, arguably, related specialized stores such as bakeries). This has happened in some communities to a greater extent than others; my Toronto neighbourhood still has small, local grocery stores, hardware stores, pharmacies etc. within walking distance, while when I lived near downtown Hamilton I would have had to drive to buy groceries. This doesn’t necessarily reduce transit ridership on its own (those trips would mostly have been walking) but can result in a change in transportation lifestyle—if a neighbourhood no longer supports everyday trips locally by foot, it encourages (or requires) higher levels of auto ownership and use.
d.p. in Seattle, WA on Tue, May 25, 2010 at 3:35am
Brent, I just want to say how incredibly cogent your writing is!