Streetsblog posted an article yesterday that quotes and attempts to refute the Antiplanner by claiming that increasing subsidies to transit agencies actually makes them more efficient. The article, written by former Strong Towns staffer Kea Wilson, misinterprets both the Antiplanner’s quote and the meaning of efficiency.
Does reducing the share of transit costs that are covered by farebox revenues increase efficiency? Photo by AgentAkit.
Transit systems get more efficient when they are more heavily subsidized, Wilson asserts. How can this be true? Efficiency is economically defined as “when all goods and factors of production in an economy are distributed or allocated to their most valuable uses and waste is eliminated or minimized.” Before the pandemic, transit agencies were typically spending four times as much money moving someone a passenger-mile as automobiles. That sounds pretty inefficient to me and increasing subsidies even more is likely to be even more inefficient.
Wilson quotes me as saying, the “great experiment of socializing public transit has failed” because it “failed to boost ridership or stabilize the industry.” That’s certainly true: despite operating subsidies increasing from less than $18 billion a year in 1990 to $26 billion in 2006 to more than $41 billion in 2019, transit ridership in 2019 was the lowest since 2006, and transit trips per urban resident were the lowest in recorded history, which includes data back to 1890.
The real point, however, is that my quote said nothing about efficiency, as it was focused more on ridership than efficiency. If transit subsidies don’t increase ridership, I was saying, then what good are they? When I do focus on efficiency, I point out that the inflation-adjusted subsidy per transit trip have grown steadily from nearly zero in 1960 to about $1 per trip in 1980 to $2 in 1990 to $3 in 2015 and well over $3.50 in 2019. That’s not efficient.
I also say that increasing subsidies makes transit agencies more dependent on politicians than on transit riders, so the agencies design their transit systems to please politicians, not to gain new riders. This often means building megaprojects that ignore changes in transportation patterns that have taken place in the last century.
Wilson goes on to argue that more heavily subsidized transit systems are more efficient because they have higher ridership loads. That’s a weird definition of efficiency. Higher ridership loads might make a transit system more energy efficient if all agencies used the same vehicles, but if an agency had to spend a lot more to get that ridership, it is not economically efficient.
In support of her claim, Wilson cites a new paper from three non-economists that “applies new metrics to old problems.” In other words, the paper attempts to redefine what is meant by efficiency.
I am reminded of Lewis Carroll, who quoted Humpty Dumpty as saying, “When I use a word, it means just what I choose it to mean — neither more nor less.” By defining efficiency to mean something completely different from economic efficiency, the authors of the paper and in turn Wilson can say that transit critics such as the Antiplanner are wrong to criticize wasteful spending on urban transit.
The article is behind a paywall so I’ll have to take Wilson’s word for what it says. Wilson presents a chart from the paper showing that urban areas that spend more per capita on transit have higher vehicle loads. The first thing to note is that rail vehicles are much larger than buses and therefore capable of higher vehicle loads. Thus, urban areas that have lots of rail transit are likely to have higher vehicle loads.
A $5 million railcar can carry 150 people while a $500,000 bus can only carry about 60, but that doesn’t make the railcar more efficient. For example, the Salt Lake City and Richmond urban areas have about the same population and 2019 per-capita subsidies in rail-heavy Salt Lake City were more than five times greater than in bus-only Richmond, but Salt Lake transit vehicles carried only 56 percent greater average loads. Spending all that extra money (even more when capital costs are counted) didn’t make Salt Lake transit more efficient; it made it far less efficient.
The second thing to note is that the data in the chart compare very different urban areas, ranging from New York to Minneapolis-St. Paul, at one point in time. If we want to compare the effects of subsidies on efficiency, we need to look at the effects subsidies have on ridership and other factors over time. As I pointed out in the paper quoted by Wilson, despite increasing subsidies to transit over the past 60 years, per capita transit ridership has declined over time.
Beyond this, the chart shown by Wilson has only fourteen urban areas. From a transit point of view, the United States has three kinds of urban areas: New York; regions with well over 200,000 downtown jobs; and everything else. The chart shows New York well above all others, then six urban areas clustered together, five of which are in the above-200,000 downtown jobs category, and then seven more urban areas clustered together, all of which have well below 200,000 downtown jobs. This last cluster doesn’t show any particular correlation between subsidies and vehicle loadings.
In other words, correlation doesn’t equal causation. There are clearly factors other than just transit subsidies that affect vehicle loadings, including the size of transit vehicle, population densities, and density of downtown jobs. Those other factors, in turn, can also influence subsidies: voters in urban areas whose transit systems have more downtown jobs and carry more trips per capita are more likely to support subsidies. This doesn’t mean that, for example, Houston, Miami, or Phoenix can make their transit systems more efficient by increasing per capita subsidies.
Wilson shows two other charts from the paper, neither of which consider subsidies per transit trip, the share of operating costs covered by fares, or other factors that I would consider important for an efficiency analysis. Which reminds me of a quote from another fantasy: “You keep using that word; I do not think it means what you think it means.”
Have you researched personal rapid transit?
https://milpitasprt.com/
Cyrus992,
Yes. PRT is a joke. We already have something better. It is called automobiles.
Any form of transportation that requires dedicated infrastructure won’t be able to compete with cars and planes, the two main forms of transportation we use today.
Traffic capacity per line vs. a six-lane arterial/collector route? Lesser need to yield to other cars and of course pedestrians.
Private personal costs? Personal responsibilities? Impacts on energy, social, urban beauty, and health? Not to mention lesser need to build so much asphalt for travel and parking.
Me thinks they have Poor understanding of Efficiency.
In 2018 before Pandemic Transit carried 0.1% of US modal split but received 54 Billion dollars.
By 2022 Transit carried barely more Half or 0.05% of Passenger miles and gotten nearly 100 Billion.
Transit uses virtually Half the nations transportation dollars for 1/200th the total passenger miles. Outside New York and a few cities it’s the most useless public industry government pays for.
Transit doesn’t follow population densities unless said populations are very mobile and work concentrated.
Transit also doesn’t work in distributive model. Transit works modestly well in either extremely dense urban conglomerations (Hong Kong, NYC, Tokyo) or very specific districts with job concentrations (NYC financial center, Washington DC, etc)
Subsidies will not fix transit, because there’s no incentive to spend the money on most essential aspect, Capital costs and decades of deferred maintenance. On the other hand, if they were required to spend All their new budget increases solely on capital refurbishment…….but that wont happen.
California passed a State bill, eliminating it’s long practiced mandate on farebox recovery of Caltrains and rail transit systems. prior to, these systems were required to meet a minimum farebox recovery (35%) to comply for federal funding in their affairs. Since removed since advent of Construction of HSR in California. If California had no faith their trains could make any meaningful revenue, why’d they build High speed rail. It’s an extortion racket in reverse.
California then: LOOK at our economy, give us money for High speed rail it’s a great investment
Now: Were’ Broke/poor give us money for high speed rail as investment.
Thanks for your comments. One correction: Transit’s share before the pandemic was 0.9 percent, not 0.1 percent. In 2021 it was 0.5 percent, not 0.05 percent. It is now about 0.75 percent. You were off by one decimal point but your point is still accurate: it’s a lot of money for not much return.
Rather than throwing money away on public transit that only the elites use, the government should reinforce our highway guardrails, sadly out-of-date with today’s safer, larger, more environmentally friendly automobiles: https://slate.com/business/2024/02/car-safety-guardrails-bloat-electric-vehicles.html
Personal rapid transit” tries to take all visual/privacy aspects of ones own automobile with the intensive independent/dedicated infrastructure of public transit systems…….
Personal Rapid Transit bills itself as using ultra lighter vehicles that eliminate the needed safety weight penalty because they’re controlled and segregated from conventional traffic. With an average vehicle occupancy rate of 1.67 to carry a population movement daily of 200,000; of 50-100 vehicles a day are needed. That requires a comprehensive infrastructure system to hold up to the weight load of that many vehicles.
PRT comes off as a particular kind of crank. Silicon Valley failures who push/whore technology.
Transportation innovations are economic/social game changers if they change the dynamics and permit new widely accessible transportation access.
WHile ago I argued a “Netflix” style mobile transport system, using vans/shuttles instead of buses owned private vendors. they can transport kids/students without need parents, own vehicles.
Traffic capacity per line vs. a six-lane arterial/collector route? Lesser need to yield to other cars and of course pedestrians. The cost is about 1/10th of light rail so we can build plenty of them.
Private personal costs? Personal responsibilities? Impacts on energy, social, urban beauty, and health? Not to mention lesser need to build so much asphalt for travel and parking.
Rather than respond to LazyReader (who apparently is a lazy reader who hasn’t bothered to understand PRT), I will address the authors comment:
“PRT is a joke. We already have something better. It is called automobiles. Any form of transportation that requires dedicated infrastructure won’t be able to compete with cars and planes, the two main forms of transportation we use today.”
What about PRT is a joke? Yes, we have automobiles, but they have various downsides like death rates, congestion, pollution, and excessive use of limited resources. PRT is safe than cars, does not have congestion problems because it has separated infrastructure (like ferries, air planes, and rail lines), uses electricity for fuel, and shares limited resources. So, in congested or high-density areas, a widespread PRT network can compete with cars on the basis of trip time, cost, and convenience. Learn more at https://milpitasprt.com/technology/
.
Competing with planes is something cars only do for shorter distances (under a few hundred miles). While commuter and intercity rail compete in that space, they tend to be slow due to frequent stops. PRT cabs traveling at 40 mph can match the average speed of BART commuter rail because they don’t stop at intermediate stations. The ITNS PRT design that the Milpitas PRT demonstration project will use can travel at 80+ mph, making it twice as fast as commuter rail and faster than intercity rail – at a fraction of the cost. Learn more at https://milpitasprt.com/
Hi, Everyone! This is my first blog comment ever so I may not be familiar with the protocol. I am one of the co-authors of the article being discussed and while I appreciate the blogger’s honesty about not actually reading the article before presenting his take on it — given that he opens that take with a concern about his words being misinterpreted by a blogger, I would have expected a bit more care in presenting our words.
We absolutely do use regional average vehicle load as our efficiency metric, but we are in no way making an unqualified case for operating subsidies. Quite the opposite.
Our goal is to offer some new metrics to foster better transit policy discussions. Check it out at this free link (https://rdcu.be/drHU6) and assess for yourselves if we succeeded.