Transit Observations from Philadelphia

I was in Philadelphia last summer for about five days. I have few observations as a pedestrian: I stayed in West Philadelphia, in the gentrifying zone radiating out of University City, and traveled to Center City, and both neighborhoods seemed intimately familiar to me as a (former) New Yorker. The street widths and setbacks looked very much like those of New York; West Philadelphia could easily be an area of Brooklyn. The difference to me was in the public transit rather than the pedestrian experience.

In New York, the subway is for everyone. The same is true of Singapore and Vancouver. In Philadelphia, it is not the case. The city is about 40% white and 40% black. On the trains I took, the Market Street subway and the Subway-Surface Trolleys, nearly everyone was black. A friend who lived in Philadelphia for ten years has observed the same on the buses, and adds that white people on buses tend to be college students.

But there’s more to the story. I think it’s a commonplace that in American cities other than New York, blacks ride public transit more than whites. What I think is more important is that whites tend to ride transit at rush hour. When I rode the trains in Philadelphia at rush hour, there was still a clear black majority on the streetcar or the subway car, but there were a fair number of whites. In the off-peak, I was at times the only white person on a streetcar that was filled to its seated capacity. The aforementioned friend says she thinks she saw the same, but as she rarely rode at rush hour, she is not sure.

It is not hard to come up with explanations for the difference. In Philadelphia, as in the typical Rust Belt city, the white population is quite suburbanized, much more so than the black population. It is also substantially richer. Both contribute to car ownership, and to driving in whenever traffic allows; since traffic is worst at rush hour, that’s when we see the most white people on public transit. The people who ride the trains and the buses outside rush hour tend to be urban residents who do not own a car, and in a city with the income distribution and racial dynamics of Philadelphia, they are predominantly black.

This injects a racial element into a lot of transit planning, especially for commuter rail. North American commuter rail is designed exclusively for suburban residents, who in Philadelphia and similar cities are usually white and at least middle-class. This is why it gets away with such poor off-peak service: hourly on most SEPTA Regional Rail lines, hourly or even every two hours on the MBTA, hourly on most branches of the New York commuter rail network. Although New York itself doesn’t have the typical Rust Belt city demographics, its suburbs have typical Rust Belt suburb demographics, so the situation is the same. The same is true of Boston, when one remembers that a huge fraction of its urban white population is in Cambridge and Somerville. Philadelphia is only where this racial division is the most obvious even on the subway.

Everything about North American commuter rail screams “you’re better than the hoi polloi who ride the subway”: the seating arrangement maximizing seating rather than standing space, the park-and-rides, the fares, the lack of fare integration with local transit, the schedules. Since peak-only suburban transit serves precisely the niche that the traditional white suburban middle class is comfortable riding transit in, it is necessarily segregated. Its riders even fight to keep it that way: witness for example the opposition in Stamford to developing the Metro-North station and moving the parking 400 meters away. This article complaining about parking lot waits is typical of the species; these complaints persist despite very high spending on commuter rail parking lots, for example in Hicksville.

The same transit agencies that fudge or make up numbers to avoid serving minority neighborhoods also ignore the possibility of improving off-peak service. Although off-peak service is cheaper to provide than peak service – it requires no new vehicles or infrastructure and fewer split-shift crews – the plans for service expansion typically focus on more peak capacity, despite often high crowding levels on off-peak trains. This is worst on commuter rail, but also affects subway and bus systems. In New York, the MTA’s crowding guidelines call for setting off-peak frequency such that the average train on each line will have 25% more riders than seats at the most crowded point of its journey. As anyone who’s ridden trains in Manhattan in the evening knows, trains are quite often much more crowded than this average. The MTA needs to keep its losses to a reasonable minimum, and on the core lines the off-peak frequency is not bad; but why keep claiming that trains only have 25% more riders than cars? The MTA is by comparison more honest about its capacity problems on the Lexington express trains, for example in the Second Avenue Subway environmental impact statement.

Many of the problems of American transit systems are directly traceable to the fact that the managers don’t often ride the trains, and their peer group is not the same as the average transit user. This is why we see little concern for off-peak service, and practically none with off-peak service on the whitest and more suburban form of transit, commuter rail. None of these managers of course intends to be racist or classist, but they unwittingly are.


  1. Stephen

    I suspect that the fact that white people ride buses in San Francisco is not unrelated to Muni making such strides (okay, compared to other US agencies…nothing Zurich wasn’t doing decades ago) in transit efficiency, what with PoP even on local buses now and the gradual movement towards dedicated lanes for the streetcars.

    Although my impression (never having ridden them) is that streetcars in Boston have similar demographics, and yet that city is regressing (I hear they’re not even letting people exit from the rear doors anymore), so maybe it’s not true.

    • Alon Levy

      My recollection is that there are many whites on the Green Line, but the only outlying branch of it that I’ve ridden is the D branch, the most suburban one. The trunk segments have everyone, same as in New York, but bear in mind that Boston is a rather white city once you include Cambridge and Somerville, and the subway stretches into even whiter Quincy, Braintree, Revere, Brookline, Newton, and Medford (and into heavily white/Asian Malden). There’s a noticeable race gap between people riding the A and B branches of the Red Line. Notably the only new rail lines, as opposed to replacements of old lines, are in whiter areas: the Red Line extensions, and soon the Green Line Extension. Meanwhile, the idea of railstituting the Silver Line gets sandbagged.

    • letsgola

      The streetcars in Boston serve areas that are mostly white. The B Line serves BU, BC, and Allston/Brighton, which is pretty white. The C Line serves Brookline, which is white (and rich). The D Line doesn’t really count as a streetcar but it serves Newton, also white and rich. The only one that plausibly serves (or served) black neighborhoods is the E Line, but it was cut back to Mission Hill about 30 years ago. The areas served (Mission Hill and Jamaica Plain) are sort of mixed n’hoods but are gentrifying. (Off topic, but the restoration of the E Line is an interesting case study on streetcars.)

      Then, of course, you have the Mattapan trolley, which does serve black areas (Ashmont, Mattapan) but also wealthy white areas in Milton. But at any rate, Boston as a whole has never had the same demographics as, say, Philadelphia or Washington.

      As Alon says the new rail lines (Red, Green) are in whiter areas. And don’t forget the Orange Line, which was relocated out of the heart of black n’hoods 30 years ago to serve Back Bay, Mission Hill, and Forest Hills…

      • Matthew

        CTPS did a system-wide passenger survey a number of years ago that could help answer questions about race and the MBTA:

        For example, look at chapter 11.4 of the Green Line PDF. According to their survey, between 70-85% of Green Line riders identify as white. Typically, the next two most numerous ethnic identifications are Asian or Hispanic. The “B” branch actually nudges out the “E” in terms of non-white ridership, at 28% vs 27%. Although there are many students on the “B” branch, there are also many students on the “E” branch, and Allston is a fairly diverse neighborhood if you look beyond the students. Also, the “E” branch competes with the 39 bus (cheaper) and the Orange Line (much faster).

        CTPS also asked a survey question about income but unfortunately they made the terrible mistake of not including enough brackets, so a lot of the nuance is lost through the problem of top-coding too much.

    • letsgola

      Addendum: when they were built, some of the Red Line & Orange Line extensions were to areas that are upscale now but considered pretty crappy at the time (e.g. Davis Sq). But they were white crappy areas.

  2. Adirondacker12800

    Congestion is too light and parking is too cheap for people outside of Metro New York to use mass transit off peak.

    • Rob

      Maybe, but it would be nice to find out. Even to Yonkers, the buses coming from the subway stop late at night.

      • Adirondacker12800

        .. late at night the congestion is much less and even in Manhattan street parking is easier to find.

  3. Eric

    You make lots of assertions here which are disconnected from and sometimes contradict each other.

    You imply that transit planners are biased towards the need of rich suburbanites – but also say the low frequencies of SEPTA commuter rail reflect lack of concern for riders, who in the case of commuter rail are those same rich suburbanites.

    And you note that subway ridership is overwhelmingly black (which BTW is to be expected from the demographics along its route), but don’t mention that the same subway is much more frequent and well-run than the commuter rail.

    As for the streetcars – they have all-black ridership because they go only to black neighborhoods (except the small gentrified area of West Philly). And their ride quality is horrible, because they are mixed-traffic streetcars and because they make very narrow turns in the tunnel. That’s legacy infrastructure, and I’m not sure how it could be changed except by segregating streetcar lanes or bustituting the whole routes, both of which are politically nonviable.

    As for the subway – having basically not expanded in the last 80 years, it too now serves almost all black neighborhoods.

    As for the white suburbs – I don’t see how commuter rail could greatly increase its frequency, the density just isn’t there. And if the residents refuse to add density, well, I don’t see how that can be blamed on the transit planners.

    • Alon Levy

      The low off-peak frequency of commuter rail mainly affects urbanites and non-rich people; rich suburbanites wouldn’t be riding it off-peak in the first place. This low frequency isn’t just for low-density areas: the Fairmount Line plans for Boston have trains coming every 40 minutes. Yonkers’ secondary stations have hourly off-peak service, and Yonkers’ main station gets two off-peak tph. In contrast, rich riders’ concerns with station parking, the commuter rail agencies pay a fair amount of attention to.

      Another part of the issue, which I didn’t mention in the post, is that high-income jobs are more CBD-centric than low-income jobs. Richer riders can walk from Grand Central or Suburban Station or South Station to their office, whereas poorer riders often can’t, and are more more likely to need to transfer to the subway. It makes it harder to ride commuter rail to your low-wage retail job on the Upper East Side than it has to be.

      My point about racial demographics isn’t just that blacks ride transit in higher numbers than whites. It’s that this is truer in the off-peak than in the peak. There’s a fair number of white people on the trolleys at rush hour. There are almost none in the afternoon off-peak.

      • Eric

        Commuter rail wouldn’t generally be useful for urbanites regardless of the frequency, due to the land use patterns in suburbs. That, and the focus of commuter/intercity rail on CBDs, are mostly the result of decisions made 50-150 years ago and not in the control of current transit planners.

        • Alon Levy

          But sometimes commuter lines do go through urban neighborhoods – Fairmount again, but also the SEPTA Main Line, and Metra Electric. In the case of Metra Electric, the Illinois Central had 20-minute frequencies, which Metra cut to hourly.

          • Z

            The Montclair Boonton line is interesting to me: the morning commute is mostly (certainly not overwhelmingly) white– but the reverse commuters are a much more diverse group of people, presumably a lot of MSU and hospital employees. And some of the trains are busy!

            However as a rich suburbanite I mostly use offpeak service, because the trains are less crowded and my job allows that luxury. Definitely not the only member of that species.

        • Stephen

          SEPTA’s Main Line (Paoli-Thorndale, R5, whatever you wanna call it) cuts through the heart of just about every single town in Lower Merion (Gladwyne, Bala Cynwyd, Penn Valley and Belmont Hills excepted). There are a fair amount of houses that are not within walking distance of the stations, but many are, as are virtually all commercial destinations. And yet, I can’t recall anyone but me and my brothers and my quite poor friend from high school riding it anywhere except into the city.

          In NYC, there’s even a lot of dense development (not dense by NYC standards, but dense by American standards) around the Atlantic Branch in what I call “Greater South Jamaica” (basically the African-American/West Indian neighborhoods south of downtown Jamaica, in places like Rosedale and St. Albans), which isn’t served by any subways (but is served by some quite high-frequency bus routes). I’ve never been there, but I presume that there’s lot of demand for travel between the various neighborhoods – to visit friends, go to church, shopping, work, etc. – and yet you’d be crazy to try to ride the LIRR anywhere but into Manhattan (or maybe downtown Brooklyn) given the fares and frequencies.

      • Adirondacker12800

        The slog on the 6 train gets them to there destination faster than taking commuter rail and then slogging on the 6 train from Grand Central or 125th.

      • Jonathan English

        I’m not sure why rich suburbanites wouldn’t be interested in off-peak service. If they ride it in the peak, there’s no reason why they wouldn’t ride it off-peak if service were better, either suburb-to-suburb or to the downtown for leisure trips or non-9-to-5 work trips. Plenty of wealthy people work long hours and they often like to go to the city centre for restaurants/theatre.

        The most powerful reason for the service patterns we see in the United States is assumption based on technology. A subway “must” run all day at a reasonable frequency, a mainline rail service is only for peak commuters, and bus service can run infrequently all day. These assumptions seem to govern much transit planning and are only rarely overcome.

        • Adirondacker12800

          Because they own cars and off peak the congestion is much less and parking is cheap.

          • Jonathan English

            Plenty of people around the world ride transit off-peak even if they are relatively affluent and own cars. If the only people we ever expect to ride transit off-peak are too poor to buy a car, then there will never be many people riding transit off-peak.

        • johndmuller

          One reason for not taking the commuter rail in the off peak (weekdays) is that the commuter lots are quite likely full, or require local residence permits or that there are too few non contract spaces available to assume there will be one for you. Regular parking spaces in the vicinity of the station may have 2 hour meters or other restrictions favoring the local merchants over commuter rail day-trippers. There is often little or no local public transit to the stations, not to mention at convenient times for your trip.

          Weekends and evenings work OK, provided that the schedule of trains is sufficiently convenient and/or runs late enough. OTOH, driving is also feasible during most of those times and may work out better for you depending on the details of your trip.

          Funding sufficient local bus service to get suburban commuters to the station is difficult to imagine, but may not be completely out of the question, especially if the commuter parking fees were raised sufficiently to pay for this (and the mayor and trustees were not planning on being reelected).

          • John Hupp

            It may be cheaper, but as someone else mentioned, suburban municipal transportation departments can make a lot of strange, populist decisions. If the bus service was run by the same regional agency as the commuter rail (instead of the municipality), then yes, that would be the obvious thing to do.

          • John Hupp

            Postwar suburbs can also be horribly auto-oriented, with extremely circuitous pedestrian routes, making them topologically unsuitable for public transit. The main exception you see with this are bus lines serving reverse commuters to large corporate campuses. (This was my experience in suburban Chicago growing up.)

          • Alon Levy

            Not all of them are. Long Island suburbs have street networks that buses can run on; it’s just that those streets are surrounded by endless parking lots.

          • John Hupp

            But those are mostly prewar suburbs, right? I guess suburbs served by commuter rail in the US tend to be prewar, at least immediately surrounding historic stations. I’m thinking of the far-flung 80s subdivisions that you get halfway between rail lines, like in outer suburban Chicago. People who are already a 15-minute drive from the train station and are a good 15-minute walk from an arterial where they could catch the bus, and then a 45-minute train ride into the city. Those people will be disinclined to add 50% to their commute time.

          • John Hupp

            Or like suburban Phoenix, where you could be in a gated subdivision with a 30-minute walk to the nearest Starbucks, in 105° heat. How do you convince people like that to take the bus? (Granted, those people wouldn’t have access to commuter rail in the first place.) I guess I’m trying to say that bus service is not a panacea for a lot of Americans, due to decades of really bad land-use planning.

          • Adirondacker12800

            Phoenix is a new urbanist’s wet dream. There is a strong street grid. Perhaps a bit too wide but every mile there’s a straight wide street that goes on forever and ever and ever. Philadelphia and Phoenix are vying for fifth and sixth largest cities in the country. Phoenix is 516 square miles and has a population density of 2,797.8 per square mile. Philadelphia is 134 square miles and has a population density of 11,379.5 per square mile.
            Carve new pedestrian walkways through the cul de sacs and it takes 20 minutes to walk to the bus stop where, because the population density is low the bus doesn’t come that often. And where the congestion is light or non existent. And parking is free and easy. People will drive.

          • David Alexander

            Wouldn’t it be cheaper to fund sufficient local bus service than to fund gigantic parking structures?

            The problem in Long Island’s case is that nobody really wants to take the bus, and suburban residents have considerable clout with their politicians so they’re going to demand why their tax dollars aren’t going to that parking lot or structure. I suspect in Long Island’s case, things are wonky given that the towns and villages control the zoning policy and railroad parking lots, but the counties runs the bus service.

            FWIW, I live two blocks away from a bus line that goes to my local LIRR station that has a large parking lot that sometimes fills up early leaving few off peak spaces until mid-afternoon. The bus routinely runs empty, but it suffers from no fare integration, hourly headways, and service that ends at 7 PM. I end up driving to Mineola and paying for parking at the garage there, but it’s still more convenient (and faster) than waiting for the bus and paying $5 for the cab ride back home when it doesn’t run. Unless the headways are reasonable*, it’s going to be faster for most people to drive to the train station in the mornings…

            *Maybe I’m spoiled from living in Queens for twenty years, but anything more than 5 minutes for a bus seems useless.

      • nei

        There’s a significant portion of NYC commuter rail riders who have to switch to the subway. Those working in Lower Manhattan, for example. There are also some white collar jobs in Midtown South (publishing houses are around 18-20th street, Google is in Chelsea around 14th street). A Long Island commuter working in the northeastern part of Midtown would have to take a subway or endure a half hour wlk. Those that take both are paying a rather steep commute fare.

        • Alon Levy

          Yes, this is true, but the bulk of commuter rail riders work in Midtown. There was a claim, I believe in the environmental impact statement of ESA, to the effect that 35% of LIRR riders work within walking distance of Penn Station versus 60% within walking distance of Grand Central. With these numbers and the projected ridership, the MTA built East Side Access.

          Now, to be fair, the original budget for East Side Access had a cost per rider that is low by US standards and semi-reasonable by European ones. It might have been about the same as the original projected cost per rider for Second Avenue Subway. Costs ran over massively, and this is more due to lack of cooperation between different railroads than due to racism.

          However, the reforms required to make commuter rail more usable by poorer riders, who owing to New York’s demographics are mostly nonwhite, are quite cheap as well. They just require picking more political fights, with established rider groups (who might object to the replacement of three-stops-then-express trains with local trains) and with unions (who will object to any on-board staffing reduction).

          • Adirondacker12800

            They didn’t use railroad when the neighborhoods were white. It’s faster to take the El or the Subway.

  4. Bedfordstreet

    I don’t want to argue with the overall points, which are good. But don’t oversell the racial disparities in Philadelphia transit. You rode on routes that serve the primarily-black neighborhoods of West Philadelphia. If you rode the El to the east, you would see many more poor white folks, as you travel through their neighborhoods.

    Don’t forget the overly suburban makeup of the SEPTA board that tilts the agency towards commuter rail: 2 from Philly, 2 from each of the 4 counties, 4 appointed by the state.

    • Stephen

      The El does serve a surprising number of middle-class/lower-middle-class whites in NE Philly…at least, surprising if, like me, you normally only ride the West Philly segment. Then again, sometimes it seems like half the white people who ride the El east of Center City are junkies who get off at Somerset…

      • JayinPhiladelphia

        What an ugly, and nonsensical, comment, Stephen. Yes, there are many white junkies here in Kensington, where I live, but I don’t think they’re exactly regular commuters.

        I don’t even know what to make of your first sentence, aside from the fact that you’re ‘surprised’ to learn that white people live in Fishtown, Kensington and Frankford, makes it quite obvious you’re not from around here. Some opinions should probably be best kept to oneself.

  5. Liam

    After riding in Philadelphia all my life, I had the opposite reaction riding the Red Line in Boston. “Why is everyone on this train rich and white?”

  6. Tom West

    Is this really a race thing, or is it an income thing? I wonder if rich blacks ride the subway more than poor whites….
    Or to put it nother way, is this actually “poor use transit more” coupled with “average income for blacks is lower than for whites”?

    • Alon Levy

      I’m fairly certain it’s a class thing rather than a race thing. In fact, my understanding of black middle-class culture is that it is strongly pro-car, viewing cars and the suburbs as aspirational the way whites did a generation ago; but at the same time, black middle-class areas like DeKalb County, Georgia are less NIMBY against subways connecting them with the city.

      However, it is much easier to tell someone’s race than class, and this is especially useful in a Rust Belt environment. There, people are mostly either white or black, so it’s easier to tell than in a heavily Hispanic city; and there’s also a tighter correlation between race and income.

      • JayinPhiladelphia

        Then doesn’t the class v. race thing collide against the point of your above post, though?

        Did you ride the El east of Berks? Did you ride a bus in Kensington, Bridesburg, Fishtown, Northern Liberties, Frankford, Tacony, virtually all of Northeast Philadelphia and most of South Philly?

        Making such sweeping empirical claims as you did in the post above seems clear you didn’t. You perhaps should have titled this post “Transit Observations from My Brief Stay in a Little Piece of Philadelphia, as a Means to Make Some Sort of Wandering Point.”

        And no, West Philadelphia could not “easily be a part of Brooklyn.” It’s a part of Philadelphia, and therefore does not suffer from nearly as much officially tolerated motor vehicle violence as New York does.

        • Alon Levy

          Yes, yes, I’m sure that in Philadelphia the cops are highly sympathetic to pedestrians who are run over by cars, unlike in every other American city.

          As for the white neighborhoods of Philadelphia, I didn’t visit them, but I don’t think you understood my post if you bring them up. I’m not (just) saying that the riders on the western half of the MFL are mostly black; I’m saying that there’s a noticeable difference between rush hour race and off-peak race. If everyone was black all the time, I’d believe it’s just neighborhood demographics.

          • Adirondacker12800

            It is neighborhood demographics. The rich people use cars to get around their neighborhoods and mass transit to get to work. The poor people use mass transit for both.

    • Eric

      It’s basically suburban/commuter rail, judging by land use patterns, not urban rail. So the white ridership is no surprise. It just runs more frequently than SEPTA commuter rail.

  7. John Hupp

    Here in Long Beach, white people do tend to be pretty racist or classist about the train. Despite the fact that the slow zones on the Blue Line are at the ends, where it runs like a streetcar, down the middle of streets, with more frequently spaced stops (and no signal priority, argh!), I have had white people tell me they would ride the Blue Line if only there was a more expensive express train that didn’t stop in the middle at the already widely spaced, largely grade-separated stops in Compton and Watts.

    My impression is that it would be economical to run an express train through downtown Los Angeles once they finish the Regional Connector, which would double the length of the line, except that the Regional Connector is a subway… It might be economical to build express tracks through Compton and Watts if the extension caused ridership to exceed the capacity of a double-tracked trunk line.

    We might hit capacity in the next decade due to induced demand given that Metro is buying new rolling stock to increase train frequency, but it probably wouldn’t be because more people from Long Beach are riding the train for two hours to get to Pasadena. More so if they extend the Green Line to Santa Ana or fork the Blue Line to Newport Beach. But it will never be economical to run express trains through Long Beach because the line dead-ends when it hits the ocean.

    • David Alexander

      I have had white people tell me they would ride the Blue Line if only there was a more expensive express train that didn’t stop in the middle at the already widely spaced, largely grade-separated stops in Compton and Watts.

      FWIW, I’ve met middle class Caribbeans who would ride the LIRR to Eastern Queens in order to bypass “ghetto people” who would board the E train at Sutphin Blvd. So it’s not just a racial issue, but a class issue as well, and I think the problem is exacerbated given the terrible reputation that Compton and Watts has in the minds of most people, and I’ll admit that these issues are partially why public transit has such a poor reputation, especially when compared to Canada.

  8. stevestofka

    Alon, I’d venture to guess that ridership demographics correlate to origin demographics.

    You mention you were coming out of West Philly, largely AA except for the University City area. I have noticed that nonblack Subway-Surface and West Philly el riders are correlated to the union of either (a) college students or (b) UC denizens working in CC. Similarly, the North Broad subway has a significant non-nonblack bias, with the largest nonblack population destined for Temple University.

    But contrast this with the decidedly mixed South Broad subway, which reflects the demographics of South Philly as a whole (read: largely nonblack), and is significantly more middle class; contrast this with the Frankford el, which runs through some of Philly’s poorest white neighborhoods (such as Port Richmond) and thus has decidedly mixed demographics at any time of the day. Indeed, riders originating at Berks and Girard are almost exclusively white or Puerto Rican–the closest AA neighborhoods have better connections to the Broad Street line.

    Similarly, the 9 and 27 bus route ridership–express buses from Manayunk/Roxborough to CC–reflects origin demographics: mostly nonblack, which blacks embarking largely at the connections facility Wissahickon Transfer. The same holds true for the grid routes largely serving nonblack origins, that is, for example, large chunks of South Philly’s bus network.

    I think you’re overracializing. It’s not that what you say isn’t true (especially for highly segregated cities, like Philadelphia); it’s just that what happens is that transit managers–or, perhaps more accurately, those who control funding–discriminate (at least outside the South) more on the basis of class rather than race. The racial component is secondary, resultant of AA’s general lack of upward mobility*, which is why it only appears relevant in segregated cities.

    Take Seattle as a counterexample. The metro has famously homogenous demographics, to the extreme that even Mediterranean ethnicities (Greeks, Italians, Spaniards, etc.) have complained they feel in the minority there; yet its mass transit management is otherwise indistinguishable from any other American city’s. The Sounder commuter rail uses similar equipment as e.g. Metra and Metrolink; the buses are essentially identical to Los Angeles’, Chicago’s, or Philadelphia’s. Transit segregation there occurs relative to class. This indicates to me that race is a red herring in these kinds of discussions–nothing more than a uniquely visible harbinger of class.

    Again, that does not mean I disagree with the conclusions of this post. In fact, if the goal of transit is maximum access, it would follow that the maximum investment would be in proven (i.e. high-ridership) areas regardless of race or class. That it is not speaks volumes of our cultural amorality.
    * I have discovered that African immigrants have better upward mobility than AA’s. Figuring out why that is would earn somebody a career in the social sciences.

    • Michael Noda

      I think you’re selling the race argument short; you’ve just done a good job of describing the legacy of Philadelphia’s sordid history of housing segregation. On the South Broad Subway, you’re right that ridership is well-integrated, but that ends at the top of the stairs, where the vast majority of white riders walk east and the equally vast majority of black riders walk west; that was the result of deliberate policies in the 20th Century. In Lower North Philadelphia, the “red line” between the white neighborhood (Fairmount) and the black neighborhood (Francisville) was 6.5 blocks away from Broad Street, not along it, so the ridership is much more homogenous. (I’ve lived in both places, as well as Manayunk/Roxborough; I can speak to the demographic realities of any of those places at arbitrary length.)

      • Steve S.

        My point is that these kinds of patterns occur regardless of race. Race pops as a consequence of Philadelphia’s own history, but the same patterns Alon describes occur in other, more integrated cities*, and in more demographically cohesive cities. This implies the root issue has to do with class (that is, classist resource allocation) and that injecting racial components into the argument misdirects. That is to say that a solution to the race issue requires a solution to the class issue, not the other way around.
        * And even more “integrated” cities tend to cleave into the same sorts of ghettos that South Philly and the River Wards do.

    • Nei

      Seattle has relatively decrease in transit use with income. At least during commute hours, where the buses get disproportionately used by downtown commuters.

  9. Zmapper

    Alon, as an anecdotal observation, do you recall what social groups the SEPTA in-vehicle advertisements were directed towards? When I was in Boston recently, I noticed that the Red Line advertisements were heavily aimed towards upper-class residents and college students (Icelandair and a programming company with the slogan “Code with a purpose” presumably aimed at MIT Computer Science undergrads), while Orange Line advertisements were aimed towards working class residents (Dunkin’ Donuts and associates degree programs). I didn’t use the Green or Blue Lines often enough to note their in-vehicle advertisements.

  10. David Alexander

    The same transit agencies that fudge or make up numbers to avoid serving minority neighborhoods also ignore the possibility of improving off-peak service.

    From what I’ve seen as one of those pesky minority residents that used to live in a minority neighbourhood, the partial problem is that the fare structure of the commuter railway network is prohibitive to those with low wages, and for those with high enough wages to pay for it, they’d rather drive into work, especially if it’s in a suburban location. It’s not just enough to build a station, but you need fare integration to ensure that it’s affordable for the riders.

  11. Steve Dunham

    This topic came up at the Virginia Association of Railway Patrons meeting in Richmond on March 29. It seems that racism may have played a large role in the decision by Virginia Beach not to have the light railway, the Tide, run into Virginia Beach, so the Tide terminates at the city line. However, the Tide has attracted a fairly wide demographic, and now Virginia Beach is interested again after all, along with every other city in the Hampton Roads area.

    As for the demographics of commuter rail vs. rapid transit passengers, the pattern is somewhat different in the Washington, DC, area, where I work. I commute on Virginia Railway Express (if only we had hourly off-peak service!) and Metrobus. White passengers are the majority, but not vast majority, on Virginia Railway Express and a minority on Metrobus, at least the routes I ride. However, Alexandria Transit buses carry a wide demographic of passengers, as does the Metro rail system. The Metro rail passengers are no doubt attracted by the federal transit subsidy, as a large proportion of rush-hour riders are federal employees.

    I’m not rich. I wish I could afford to live in the city.

  12. Michael Noda

    One thing you probably weren’t in town long enough to notice, is that West Philly’s more middle-class residents — who are disproportionately white — tend to get around by bicycle more than public transit; it requires a much bigger up-front investment than a pack of tokens, but is much cheaper in the long run, even assuming an alarming rate of theft. West Philly is flat enough and gridded enough that it’s easy to bike places even without much in the way of formal bike infrastructure, and the universities and hospitals are very bike-friendly. (It also helps that the gentrified neighborhoods are the ones that are closest in.) I don’t know what will happen to transit demographics once bikeshare comes online east of 52nd Street over the next couple of years, but it ought to be very interesting.

  13. tehgreengrass

    Interesting the relative lack of comments on the PATCO line from South Jersey into Philadelphia.

    “It’s basically suburban/commuter rail, judging by land use patterns, not urban rail. So the white ridership is no surprise. It just runs more frequently than SEPTA commuter rail.”

    No offense, but I think this trivializes the rail line way too much. Yes, it has a use case as commuter rail, but it also runs in the middle of the night as one of only four lines to run 24/7 in the whole country. Yes, it has white people, who live in the suburbs in greater number than black people, but also has a very mixed ridership; whether at rush hour or off peak, there are people of all races using it. Yes it runs more frequently than SEPTA, but not “just” more frequently; it is more frequent and more convenient to Center City than even some regional rail lines that exist wholly within the city limits (Chestnut Hills East and West, for instance).

    PATCO isn’t simply some sleepy suburban train into Philadelphia. It’s the only direct rail link into our city, and people learn how convenient it is from an early age. High schoolers use it, college kids use it, yes, commuters use it, senior citizens use it. People get to sporting events on it, they get to school on it (a nontrivial number of students get out in Camden for Rutgers, Rowan, and Camden County College in the morning), they go into town for dinner on it, they get home from the bars without having to drive on it. And to simply say the trains are frequent does it a disservice. There are 5 minute headways during rush hours. Off peak, headways are mostly in the 20-minute range. Even overnight, when there’s no track work, I believe they’re at 40 minutes, which is still more frequent than some SEPTA trains at rush hour.

    The line has run for the past 45 years as an absolute workhorse, despite the port authority treating it like crap. It may run in an “urban” setting (outside Camden, which obviously is urban), but the land use patterns around its stations are actually fairly dense for South Jersey. Even for mostly-white suburban towns, Collingswood, Westmont, Haddonfield, these are towns that grew up in the 20s, 30s, 40s, before the dominance of cars. They’re older, inner-ring suburbs whose land can’t just be generalized as the usual sprawling New Jersey suburbs. In fact these towns, like Collingswood, where I live, has seen a fantastic revival as a place to live over the past decade, not in small part because of the 12ish minute ride into the heart of Philadelphia.

    • Alon Levy

      I may be getting too conspiratorial here, but why does PATCO stub-end in Center City without good connections to the subway? I know that it’s conspiratorial to say that it’s because middle-class suburban commuters are interested in traveling to Center City but not so much to the rest of Philadelphia, so that target demographic cares less about connections. I know that the actual answer is almost certainly, “it’s a different agency, and if you want more answers, please sign a request form in triplicate.”

      • tehgreengrass

        It actually has great connections to the El and Subway! Its first westbound stop after the river is 8th & Market, where you can connect easily to the Market-Frankford El at and the Ridge Avenue Spur of the Broad Street Line, both of which also stop at 8th & Market. After that is 9-10th & Locust, which has no connections, but then at both 12-13th & Locust and 15-16th & Locust, there’s are fairly well signed underground connections to the Broad Street Subway’s Walnut-Locust Station. All of these connections are also announced at each station.

        The fact that it may not be apparent that it makes these connections is actually a point of intense ire of mine, and the reason that I spent time coming up with a possible map for PATCO to use in its stations – Basically, I think it is a horrible missed opportunity to better display the connections that the line makes with other transit options in the area.

      • Adirondacker12800

        It runs in what was a subway tunnel, ask the people who designed the subway why.

Leave a Reply to Alon Levy Cancel reply

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

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

Facebook photo

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

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.