Why Public Transit?

Anti-Transit Arguments Are Mostly Rhetoric, Lacking Facts

Over the years, we have grown weary of the many rhetorical, generally fact-free arguments against transit. These include the ubiquitous “You’ll never get people out of their cars,” “People won’t ride transit,” or a favorite of rail transit opponents, “There isn’t enough patronage potential to justify rail.” Such arguments are almost exclusively rhetorical in nature, lacking specific facts and examples.

(Note: some links are missing as we rebuild the Publictransit.us website)

The more sophisticated transit opponents such as Wendell Cox (www.publicpurpose.com) and Randal O’Toole (www.ti.org) regularly cite “pseudo-facts” such as the “decades of declining market share” for transit compared to automobiles. The Cox/O’Toole line of reasoning conveniently omits the fact that the act of driving is heavily subsidized by a vast array of non-driving activities, including “free” parking provided by retail and commercial real estate, the cost of private garages and driveways covered by mortgage payments, and so forth.

The annual cost to the American economy for motor vehicle accidents and parking alone has been estimated as between $300 billion and $700 billion+ annually by reliable government and academic sources by the mid-2000’s. These costs would be equivalent to a $2.00 to $5.00 tax on each gallon of motor fuel.

Lack of Proper Context

Opponents of rail transit, in particular, are eager to report the general success of “bus rapid transit” (BRT) in Latin America (e.g., Curitiba, Caracas, Quito, etc.) However, this information almost always lacks context and caveats. For example, under U.S. conditions where automobiles are a ready alternative for most people, U.S. travelers simply will not tolerate the level of crowding experienced by Latin American urban dwellers. The only exceptions to this rule are a few large U.S. cities such as New York, Philadelphia, Boston, and San Francisco where auto ownership rates are well below the U.S. average.


BRT advocates who tout bus rapid transit’s undeniable success in Latin America also fail to mention that much lower wage rates make BRT economic to operate at much higher volumes than it possible under U.S. conditions, where wage rates make rail transit feasible generally around 5,000 daily passenger miles per (two-way) route mile (see links below). Therefore, many BRT claims are simply propaganda unless this essential context has been provided. This general lack of hard information led to our original research, Special Report No. 1: Bus Rapid Transit in Curitiba, Brazil: An Information Summary. Our research has led us to conclude that BRT actually makes sense in a much wider context than the higher traffic density corridors where most advocates claim BRT as a substitute for rail: in the (literally) hundreds of transit corridors in U.S. urban areas where the volume of transit ridership justifies capital expenditures to improve service reliability and operating speeds, but is insufficient to justify investment in rail transit, e.g., streetcars, light rail, commuter rail, and so forth.

Generally these potential BRT corridors would attract substantially less than 5,000 daily passenger miles per (two-way) route mile (see Special Report No. 2 link below). Many of these potential BRT corridors are currently served by inadequate, infrequent local bus service, but adequate capital investment to improve the level of service, operating speed, and service reliability could generate a much greater volume of total transit usage (even if well below the “threshold” for rail transit).

Three Decades of Missing Follow-up Analysis

Another motivation for the website was that during the past 25 years, we were unable to find any follow-up research after publication of a landmark transit study funded by the Carter Administration’s Urban Mass Transportation Administration (UMTA, now FTA), Urban Rail in America (Pushkarev, Zupan and Cumella, 1980). We also could not find any research to a classic transit patronage forecasting paper, “Preliminary Transit Patronage Estimation for Small Urban Areas via Transit Service Factor” (Dennis Neuzil, P.E. Traffic Engineering, August 1975). In the former case, the new Reagan Administration dropped any follow-up research on Pushkarev et al’s work. There has apparently been little academic or practitioner interest in simple statistical models useful for predicting transit patronage during the past 30 years; we found almost no U.S. research on this important topic.

Regarding transit traffic density as explored by Pushkareve et al, we wanted “hard numbers,” not simply rhetorical arguments and propaganda. Therefore, to quantify the question “how many people would ride transit, where and under what conditions?” we undertook many hundreds of hours of research to develop the data we now present in our “traffic density” section, and to follow up on the pioneering work by Pushkarev et al. The results are presented in Special Report No. 2: Traffic Density Thresholds for Rail Transit: A Retrospective.

The analysis presented in Special Report No. 2 confirmed the prescience and accuracy of Pushkarev et al’s original analysis and predictions. Pushkarev et al’s minimum recommended thresholds for rail transit were remarkably close to independent analyses by Japanese and Western European scholars, e.g., the figure of 4,000 to 5,000 daily passenger miles per two-way route mile. Their conclusions were also remarkably close to criteria reviewed and used in the 1949 bankruptcy case of the Pittsburgh [PA] Railways. Pushkarev et al’s predictions of rail transit patronage and traffic density also proved conservative for the several new rail transit lines recommended in their program that actually were built since 1980. In a remarkable coincidence, by 2004 almost the exact mileage of new rail transit lines recommended by Pushkarev et al had actually been constructed, albeit a number of new routes in cities not originally anticipated.

In a given travel corridor without any unusual constraints, typically a minimum of 15% to 20% of travelers will use high performance transit, given origins and destinations served directly by the transit line. This percentage increases significantly to a range of 40% to 60% “transit share” when significant constraints exist, such as high downtown parking rates or parking limits, constrained transportation capacity due to geographic constraints, etc. The major reason transit does not attract such high percentages of total trips within such a corridor is due to limited transit coverage, the fact that many trips have origins and destinations outside the transit corridor, and only lower performance transit, e.g., slow, surface buses, provides such connections.

Rail transit is generally economically feasible when transit travel volumes exceed 5,000 daily passenger miles per (two-way) route mile, and assuming capital investments are generally no more than $4,000 to $5,000 per daily passenger mile. To be justified, commensurately more expensive transit lines require commensurately higher travel volumes. In some site-specific cases, investment in rail transit–particularly streetcars–can be justified when specific development projects and strategies are supported. The best examples include the $60 million investment in the Portland Streetcar that “kick-started” more than $1 billion investment in Portland’s Pearl District.

The $90 million investment in the “Tacoma Link” streetcar supports development in all of downtown Tacoma. In one case, a private Indiana hospital chain could easily justify a $30 million investment in a short automated shuttle between its two downtown Indianapolis locations despite traffic density of less than 2,000 passenger miles per (two-way) route mile, mainly due to travel time savings and a commensurate increase in productivity for its expensive medical staff.

Followup Analysis of Simple Ridership Forecasting Models    

Special Report No. 3: The Transit Field of Dreams: If You Operate It, Will They Come? follows up on the transit patronage forecasting work by Neuzil and others.