So let me guess: it’s gonna run in a circle, right?  A July 2011 press release from FTA gives a succinct overview of what are some ill-conceived notions about streetcar in America.  The document is the public go-ahead to build a streetcar system for Atlanta with a $47.6 million TIGER II grant awarded in October 2010.  “The electric streetcar will run 2.6 miles through the heart of Atlanta’s business, tourism and convention corridor. Its planned 12 stops will provide access to residential, cultural, educational and historic centers, including the Dr. Martin Luther King Jr. National Historic Site and the Georgia Aquarium.” 

Atlanta is in my backyard.  It is the nearest major city to where I live, it is where my brother lives, and it is the FTA Region IV headquarters and so I am a frequent visitor.  The aquarium is terrific, but the destinations listed above could be called Places Atlantans Do Not Go.  In my visits to Atlanta, which include FTA trainings in downtown hotels, I have consistently been amazed at how resistant to street activity the south’s largest city is, the Coke Museum and Olympic Village notwithstanding.  “This streetcar project will give people the option to leave the car at home and get to where they need to go in downtown Atlanta,” U.S. Transportation Secretary Ray LaHood said in the release.  FTA Administrator Peter Rogoff echoed, “Atlantans will be able to get to work, students will be able to get to the university and patients will be able to get medical care without having to spend a dime on gas.”

That’s optimistic.  Of Atlanta’s 38 metro stations, 36 have parking, and mostly ample parking at that.  MARTA is an odd beast in that it serves a metropolitan area of 28 counties but the trains run in only three, primarily in DeKalb and Fulton Counties but with a tiny bit in Cobb.  In Fulton county 30.7% of households live at or below the poverty line; in DeKalb 29.5% do.* Therefore the population best served by MARTA’s geographic coverage are the urban poor but its stations are designed for suburban commuters and the capacity of its system is not limited by the number of rail cars but by the number of parking spaces it has.  The destinations on the proposed streetcar are places neither suburban nor urban commuters are likely to go, but the route design does follow an established pattern in U.S. streetcar building.

Transit agencies have focused on attracting more downtown commuters really since mass transit was invented even though employment patterns have changed drastically.  As a result over decades now transit agencies have worked to increase transit’s share of an ever-decreasing source of passengers, streetcar being one such strategy.  Thompson and Brown’s extensive work, including “Where Transit Is Growing: Surprising Results,” Journal of Public Transportation, 2005, has demonstrated transit works best in the decentralized systems (aka multi-destinational and grid networks) that are found in the west. 

The Brookings Institute study Missed Opportunity: Transit and Jobs in Metropolitan America, May 2010, confirms the problem of downtown-oriented (aka radial or hub-and-spoke) systems: “The typical metropolitan resident can reach about 30 percent of jobs in their metropolitan area via transit in 90 minutes.”  That is, few jobs in a lot of time because systems focused on serving downtowns are only effectively serving jobs in downtowns but the CBD has not been the growth market for employment in nearly a century and has not had a majority of regional jobs in any American urban area for generations.   So why blow $50 million on a circulator in a particularly weak downtown?

Again, Rogoff: “In addition to the convenience this streetcar will provide to the millions of tourists and conventioneers who visit Atlanta each year, it will be a lifeline for thousands of people who live near the new line.”  We can ignore the last clause—when their lives depend on it people living downtown, whether high rise dwellers or urban poor, can get a bus or drive a car to the Martin Luther King Historic site—but the tourist point raises a question.  Can tourists fuel a streetcar system? 

In a word, yes.  The cable car in San Francisco is often bypassed by locals for trolley buses and other nearby transit because tourists so completely fill the vehicles that visitors can be seen hanging from the rails outside the cars.  The F-Market Line in the same city uses historic PCC cars and has a greater mix of commuters, but is still a tourist draw.  This is the busiest streetcar line in the country with 5,555,980 passenger trips in 2005 as seen in table 1 (I made this table a few years ago but still find 2005 a good base year because of the hurricanes in 2005 and 2006).  The “but” here is that San Francisco is a special case: a dense, historic, highly walkable city on a bay with lots and lots for tourists to do and high quality public transit to get them where they want to go.

The experiences of the Tampa TECO Trolley and Little Rock River Rail are probably more predictive of what the Atlanta streetcar will do.  Both of these are expressly focused on tourists with vintage reproduction cars, 8 mph speeds, and stops at tourism sites; to drive home the point that this is not a transit project the Tampa TECO line does not make its first run until 11:00 am!  Unfortunately neither the TECO line nor River Rail operate in anything approaching a tourism Mecca like San Fran and the ridership numbers for both prove it: 565,002 trips in Tampa and 154,734 in Little Rock (2005).  But the notion that a transit project should serve any one group is silly to begin with, whether tourists, Richard Florida’s famous creative class (young, tech savvy professionals.  See Jarrett Walker’s short post on humantransit.org here  http://www.humantransit.org/2011/07/whos-leading-the-non-car-renaissance.html), or even the urban poor.  This is simply not a recipe for successful transit.

What seems not to be understood by cities and the consultants hired to design and build streetcar systems is that when we got rid of streetcars last time the Germans imported the idea, lengthened the trains, and applied the essential principles of successful rail transit: shorten dwell times by using off-vehicle payment (or other means not involving the driver) and multiple doors for both entry and exit, dedicate right-of-way for rapid travel, serve dense land uses, and provide connections to the greater transit system.  In other words take the characteristics that make heavy rail (subway) so efficient and put them on the street for a fraction of the cost.  Strassenbahn worked and Americans re-imported the idea calling it light rail starting with the San Diego Trolley in the 1980s and LRT has replaced heavy rail as the way emerging great cities (Houston, Seattle, Charlotte, etc.) realize their rapid transit dreams.  When the streetcar craze hit we ignored all of this.

Downtown circulation was not part of the streetcar playbook during the turn of the century decades when streetcar enjoyed its greatest success.  Back then routes were straight line components of grid or radial systems but every system built since the reawakening began in the late 1980s has been a downtown circulator of some sort.  Not only do these new systems ignore the fact that downtowns are no longer the employment centers they once were, they also have melded the worst aspects of rail (lack of maneuverability, expense) and bus (slow, pay the driver, stop frequently, use only the front door for boarding) to create gold-plated transit projects that are functionally useless for serious transit patrons (as opposed to tourists). 

The Get-‘Em-Downtown mentality has much to with this but part of the blame has to fall on the Portland streetcar simply because it was the first modern system and because it was so successful (2,587,033 trips in 2005). But Portland too is a special case, a city with a downtown whose utility is more like San Francisco’s than any southern city and a place that is famously multimodal.  Portland has a 30% transit-to-work mode split to Atlanta’s 3.7% (Transport Politic), for example.

The streetcar project I would like to see is a modern car on a straight line, with dedicated ROW, limited stops (every half mile instead of quarter mile), off-vehicle payment, and using all doors.  If that sounds like light rail it should.  Streetcar infrastructure is much less expensive than LRT.   Seattle recently completed a streetcar and LRT within the same year and gives us a good comparison: the 1.3-mile South Lake Union Streetcar at $39 million per mile and the 14-mile light rail at $193 million per mile.  There is no reason a streetcar cannot be used as light rail on the cheap except capacity; a streetcar is more like an articulated bus than a true train (engine pulling multiple cars) and has an advantage in that its size and construction demands allows it to fit in smaller urban spaces as examples in France and Portland, where a car goes right through Portland State University, have shown.  Streetcar would not be a good fit for a ten-mile alignment in a city of 4 million, but a five-mile route in mid-size city would be ideal.

The die, however, is already cast in Atlanta where the system will be built, ribbons will be cut, and photos taken.  Once the politicians stop riding the ridership numbers will drop and that is when the howling starts:  $47 million for this?  That wasn’t mentioned in the press release. 

*Based on 4 person household, all data 2009