What is The Expected Impact of the COVID-19 Pandemic (and others) on U.S./World Transportation?

May 25th, 2020

COST Commentary: This post is a more extensive projection of the impact of the current pandemic on U.S. and worldwide transportation. The previous posting focused on the implications of the current pandemic on transportation in the Austin Region: Texas’ 2019 Transit Ridership Decline is Consistent with the Total United States Decline The article below further substantiates the previous postings conclusions by predicting the “most profound change will be the number of people working at home.” The Work-at-Home projections in the previous Austin Region posting are are conservative compared to the projections below.
Transportation After the Pandemic
By The Antiplanner | May 19, 2020 | Policy brief

Most people living through this pandemic have wondered, “What will change after COVID-19?” The transit industry in particular is worried about whether it will get back its lost riders, while airlines are just hoping to survive long enough to recover. While a lot of uncertainties remain, some things are less uncertain than others. This paper will focus on what is likely to happen in the first year or two after the various stay-at-home orders are lifted and the economy begins to recover.

1. More People Will Work at Home

The most profound change will be number of people working at home. The American Community Survey reported that more than 8.2 million people, or 5.3 percent of the nation’s workforce, worked exclusively at home in 2018. The share was much greater in some areas: 8.6 percent of Colorado workers and 15.4 percent of Marin County, California workers worked at home.

Yet this is only a small fraction of the number of people who could work at home. The Bureau of Labor Statistics estimates that 29 percent of American workers have jobs that would allow them to work at home. Another study estimates the number is as high as 34 percent. A little more than 30 percent of American workers have white-collar jobs, and what these studies are saying is that most white-collar workers could work at home while only some blue-collar workers could do so.

Global Workplace Analytics, an organization that focuses on “the future of work,” goes even beyond that, estimating that 56 percent of American workers could work at home at least part of the time. It predicts that, after the pandemic, 25 to 30 percent will work at home multiple days of the week.
photo 1 for Randal's article
Working at home can increase productivity and save both employers and employees money. Photo by Andrew Neel.

Working-at-home was already growing before the pandemic, increasing from under 6 million people in 2010 to well over 8 million in 2018. The pandemic has become a giant experiment in working-at-home: one survey found that 88 percent of employers big enough to have human resource departments are having some or all workers work at home.

The experiment appears to be successful. More than half of workers working at home during the pandemic report that their productivity has increased due to fewer distractions and time saved from commuting, while only a quarter feel their productivity has declined.

Employers have another reason to keep people working at home: Global Workplace Analytics estimates that they can save $11,000 a year for each person working at home half of the time, while the employees themselves would save another $2,500 to $4,000 a year. Obviously, they could save even more if employees work at home full time. Each percentage increase in the share of people working at home would save well over $10 billion dollars per year.

Twitter has already let many of its employees know that they will be allowed to work at home permanently after the pandemic is over. Based on all of this, it seems reasonable to expect that, on any given day after the pandemic, the share of people working at home will at least triple (not all of whom may be working exclusively at home — two people working at home half time equals one less daily commuter). This means that the share of people commuting to work is likely to decline by more than 10 percent.

2. More Will Move to Lower Densities

People have been moving into lower density areas since at least the end of WWII, and — aside from the interruption caused by the Depression and War — really for several decades before that. Demographer Wendell Cox shows this trend has continued through 2019. Despite claims that there is a pent-up demand to live in dense cities, a 2018 Gallup poll found that 40 percent of Americans who lived in cities aspired to live in lower-density areas while more Americans aspired to live in suburbs and rural areas than actually lived there.

The pandemic will accelerate this trend. A recent Harris poll found that 27 percent of Americans, including 39 percent of city residents, are considering moving to less-dense areas because of the coronavirus. Younger people who are supposedly most attracted to the cities are the ones most likely to want to move, and some already have.
Photo 2 for Randal's Article
Automobiles have liberated the many Americans who want to live in areas with large yards and access to open space. Photo by FutureAtlas.

Jobs will move to low-density areas too. Why should a business pay downtown rents when half of its employees are Zooming to work? As Joel Garreau noted in Edge City, businesses tend to locate their headquarters based on where their CEOs want to live, and higher-income people are already leading the charge out of the cities.

Cities aren’t going to lose 39 percent of their populations overnight. After 9/11, the Wall Street Journal reported that many banks were leaving Manhattan. Despite this, Wendell Cox estimates that the number of jobs in lower Manhattan grew by 10 percent between 2000 and 2015.

What is really going to accelerate the trend to decentralize is the increase in the number of people working at home. This is because most people’s commutes are limited by a time budget: they don’t want to spend all their time traveling. Average commutes today are about 25 minutes long so someone who commutes five days a week spends 250 minutes a week commuting. If they drop to commuting three days a week, they can spend 150 minutes commuting. If they drop to two commutes a week, they can spend 100 minutes on their commutes.

Lower densities may or may not insulate people from infectious diseases. But they do give people access to much more affordable housing. The increase in working at home after the pandemic is going to lead to a new burst of growth in suburban and exurban areas.

3. Oil Prices Will Stay Low

From 2011 through mid-2014, nationwide gas prices hovered around $3.50 a gallon for regular gasoline. Then they suddenly dropped and since then hovered around $2.50 a gallon. The pandemic brought them down to around $1.70 a gallon.

There is now a worldwide glut of oil on the market, a glut partly caused by a price war between Saudi Arabia and Russia. While economic recovery after the pandemic will consume much of the glut, it won’t stop the price war. Moreover, recovery won’t be instantaneous, so prices are likely to remain below their pre-pandemic average for at least a year, and perhaps longer. These lower prices will influence people’s travel and location decisions, encouraging more people to drive and move to lower-density areas.

4. Transit Will Lose 25-35% of Riders

Based on all the above considerations, transit in a post-pandemic world is likely to carry at least 25 percent fewer riders than it did before the pandemic. This will happen for several reasons.

First, the white-collar workers who are most likely to work at home are also most likely to have been transit commuters before the pandemic. In 2018, a higher percentage of people who earned $75,000 or more per year commuted by transit than any other income class. Based on this, increasing the share of people working at home from 5 to 15 percent will reduce the number of people commuting by transit by at least 10 percent. This is going to especially impact rail transit, which tends to serve mostly white-collar workers while bus transit serves mainly blue-collar workers.
Photo 3 for Randal's Article
Will people dare ride transit without masks? Photo by Eneas De Troya.

Second, taking 10 percent of commuters off the road will have a visible effect on congestion. As economist Anthony Downs realized, people respond to congestion by changing their travel times, routes, or modes of travel. When something happens to relieve congestion, some of those people change back, which he called a triple convergence. If just 5 percent of those who change their habits in response to reduced congestion are transit riders who decide to drive instead, transit will lose another 10 percent of its commuters. This will impact both rail and bus transit.

Third, transit works best bringing people from dense suburbs to denser downtowns. As jobs and people move to more remote areas, transit becomes increasingly ineffective. The acceleration of decentralization will probably reduce transit commuting by another 5 percent.
Fourth, low fuel prices will also encourage people to drive rather than ride transit. Finally, will all of the people who simply decide that, between infectious diseases, crime, and other problems, transit is not worth the hassle. Based on these considerations, a 25 percent decline in future transit commuting below pre-pandemic levels seems conservative; 35 percent is quite possible.

According to table 9a of the National Household Travel Survey, as of 2017 work-related trips made up 40 percent of transit ridership. But people who give up using transit for commuting are probably also going to give up using it for shopping, social, recreational, and other purposes. Thus, we can expect to see per capita transit ridership to be at least 25 to 35 percent lower after the recovery than it was before.

5. Driverless Cars Will Be Delayed

The various companies developing driverless cars are each following one of two pathways. First is the Tesla model, in which new cars are sold with driver assist features that are gradually improved through software (and possibly hardware) upgrades until they are fully driverless. Second is the Waymo model, in which cars are placed in ride-hailing service that can only operate in specific zones. Eventually, with the revenues earned from ride hailing, the zones would be expanded to include the entire country and at some point, such cars would be sold to the public.

Before the pandemic, I would have bet that the Waymo model would win the race to have fully driverless cars. COVID-19, however, has thrown a monkey wrench into this business plan. Taxi and ride-hailing drivers can sanitize their cars after each use. They can even do the sanitizing in front of squeamish customers to ensure that they are comfortable riding their cars. But a driverless car, by definition, has no one to do such sanitizing.

Waymo and other companies following this model, including Ford and General Motors, will have to anti-viral materials in auto interiors, UV lights, or other autonomous safeguards that will assure potential ride hailers that their vehicles will be free of infectious diseases. This is likely to delay the widespread introduction of driverless ride hailing.
Photo 4 for Randal's Article
Unless manufacturers can develop a system of autonomous sanitizing of auto interiors, dreams of driverless ride hailing may be one of the victims of the coronavirus. Photo by Dllu.

Ford, which previously promised to begin mass producing driverless cars in 2021, has already announced a delay until 2022 due to the coronavirus, and that still assumes the ride-hailing model. When driverless cars are finally available, they are going to speed up the decline of mass transportation, but the pandemic may have added several years to that time.

6. Air Travel Will Take a Hit but Survive

Keeping people six feet apart from one another requires giving each person on a plane close to 60 square feet of space, yet airlines have been profitable only by packing people into as little as 7 square feet of space per person. Several airlines have already gone bankrupt or shut down completely due to the effects of the pandemic.

However, unlike urban transit, for long-haul traffic the airlines do not face competition from a mode that is faster, more convenient, less expensive, and less likely to transmit infections. While the airlines may have to make some efforts to ensure customers that interiors are sanitized after each flight, the demand for long-distance air travel is not likely to decline enough to put the industry at risk.
Photo 5 for Randal's Article
Nearly empty planes allow passengers to socially distance themselves but aren’t economically sustainable for the industry. Photo by Mx. Granger.

Short-distance air travel is different. Most airlines offer short-distance flights as feeders into their long-distance flights. Some, however, cater to business travelers making short hops. Before the pandemic, for example, Alaska offered hourly flights between Seattle and Portland and many if not most of the people on those flights were just going between those two regions.

Such short-distance air travel faces the same competition as public transit, namely private automobiles. Some experts believe that driverless cars, when they become capable of intercity travel, would significantly reduce the short-haul airline market. Public response to the coronavirus might make that happen even before driverless cars are available. Thus, there will be fewer short-haul flights and the ones that are left will all be feeders into long-haul routes.

7. Amtrak Will Lose 10-20% of Riders

Amtrak could benefit from a decline in short-haul air travel. Amtrak trains offer more space per passenger than commercial airliners, and generally aren’t as full. But they still don’t offer six feet of social distancing, even when operating at their usual 50 percent of capacity, and Amtrak can’t guarantee that all interiors are sterilized at all times.
As a result, Amtrak is likely to lose more riders to automobiles than it will gain from the airlines. It probably won’t lose as many customers as urban transit because many of the people who ride Amtrak are part of a niche market that may be more loyal than transit riders.

Like the airlines, Amtrak’s losses will likely be greatest in its short-haul routes. Train trips in the Northeast Corridor, for example, average about 160 miles, while those on state-supported day trains average about 125 miles. The same people who would switch from short-haul airline flights to driving are likely to switch from Amtrak to driving.

Solve Today’s Problems Today

People’s memories are short. Once the pandemic is over, many will return to their pre-pandemic routines, using various forms of mass transportation, eating in restaurants, and staying in hotels just like they did before.

Some changes, however, will be irreversible. The biggest one will be more people working at home. The second, which partly follows from the first, will be more people and jobs moving from dense cities to lower-density suburbs or exurbs. These changes, in turn, will significantly reduce transit ridership and contribute to declines in short-distance travel by plane and train.

The future is always uncertain, but these projections reaffirm the prescription I have always given to transportation planners: solve today’s problems today so that the future will be best able to solve whatever problems it faces. This means not committing resources to mega-projects, whether highways or mass transportation, that may not be needed on the future.

New roads will be needed in regions that are growing and expanding. Unless they involve road pricing, which is politically controversial, attempts to relieve congestion by building new capacity in existing urbanized areas may be extraordinarily expensive. In these areas, transportation agencies should focus on microprojects, such as traffic signal coordination and fixing traffic bottlenecks, that can produce rapid and positive results.

Legislators and appropriators should take this opportunity to rethink subsidies for transportation. Transit agencies and Amtrak would like people to believe that everyone deserves to have access to their forms of transport even if they rarely or never use it. When ridership goes up, they demand more subsidies to carry the loads. When ridership goes down, they demand more subsidies to make up for the lost fare revenues.

The reality is that people move to low-density areas knowing that by doing so they give up access to some services that are only available in high-density areas. As more people respond to the pandemic by moving to lower densities, taxpayers should not be required to pay to provide those services.
Photo 6 for Randal's Article
Despite transport subsidies that are heavily biased towards Amtrak and transit, these forms of travel make a trivial contribution to the nation’s transportation system: Amtrak carries about 1 percent as many passenger miles as the airlines; transit about 1 percent as many as autos. This chart is an update of the 2017 chart in policy brief 20; see that policy brief for how these numbers were calculated.

Transportation subsidies have grown unchecked for too many years. In 2018, subsidies to Amtrak were about twenty times greater, per passenger mile, than subsidies to air travel, while subsidies to transit were about 80 times greater, per passenger mile, than highway subsidies. To a large degree, both Amtrak and urban transit are obsolete forms of travel that are slower and more expensive than the alternatives, which is why users are not willing to pay their full costs.

The best thing Congress and the states can do is to end subsidies to all forms of transportation and allow private companies and public agencies to provide a level of services that can be funded out of user fees. This will lead to a transportation system that is efficient, resilient, and that meets that needs of all Americans.

About ‘The Antiplanner’
The Antiplanner is an economist with forty years of experience critiquing public land, urban, transportation, and other government plans.

Imagine: Significant Congestion Relief & Higher Quality-of-Life With Lower Cost & Taxes for all Austin Citizens.

May 24th, 2020

COST Commentary:

Recent events with the COVID-19 virus have highlighted the potential of accelerating several current trends, with possible significant, rapid & positive impacts on Austin area’s mobility. These trends have existed for many years in the U.S. and Austin, but have been developing more slowly in the past. Key trends include:

1. Work-at-Home is Growing: This is one of the defined “modes of commuting” in the U.S. census. It has been the fastest growing mode in the country and in Texas for several years. In 2017, work-at-home exceeded the Public Transit mode of commuting to work in the U.S., for the first time. In the Austin urban area, the Work-at-Home mode has grown much faster than many other cities. The chart below shows this major growth of Work-at-Home over the past 18 years while transit commuting to work has been trending down since 2015. Total transit in Austin has reduced over the past 20 years. Work-at-Home was roughly the same as transit commuting in 2000 and in 2018 it was 4 times the Public Transit Work Commute mode. An extremely high percentage of the Work-at-Home results in a car off the roads during peak traffic periods, as we have seen with the increased work at home, and loss of work, during the COVID-19 impact.

Austin Urban Area Work-at-Home & Transit

2. Public Transit Commuting is Declining: Transit use has been declining in the U.S. for years. Texas has less total transit ridership than it had 20 years ago while spending several Tens of Billions of dollars, primarily on light rail, in major cities to increase transit ridership. This includes the addition of Dallas’ longest light rail in the Nation. It has all failed: Transit use has declined; congestion has increased; and cost of living continues its rapid growth. Of the four major Texas cities, Austin has had the largest population growth (85%) and the greatest decrease in transit ridership (-16%), during the past 20 years. Austin is still in its reduced transit ridership trend after recently giving more than 2 million free rides a year to all K-12 students and redesigning the bus routes to be more efficient. More can be found about Texas and Austin declining ridership in the post just prior to this one on the COST web site. The COVID-19 situation has, of course reduced current Austin transit ridership by more than 60%, even with total free fares for the past two months. Even with the negative trend, the Austin area has significantly higher transit ridership per citizen than Dallas. San Antonio, with its bus only system, spends far less on transit, per person and has the highest ridership per citizen, but its ridership has also declined. One senior industry expert projected a 25-35% permanent reduction in future transit ridership due to the COVID-19 pandemic.

Note: Please see the previous post regarding Texas transit ridership decline.
Texas’ 2019 Transit Ridership Decline is Consistent with the Total United States Decline

3. People are moving to the Suburbs with Lower Density: This is a trend of several years in major cities and has been accelerating. Some predict this trend will accelerate faster with pandemic concerns. Work-at-Home increases and affordability issues will also support this suburb movement trend as people can find more affordable housing and do not have the cost of automobile commuting. There is also likely to be an increase in movement to the suburbs due the fact that about 80% of an average city region’s new jobs are being created in the suburbs and not in the central core.

The coronavirus epidemic has dramatically increased the number of Work-at-Home people while significantly decreasing the number of Transit commuters. If these trends continue, the current Austin/Cap Metro plan, Project Connect, would be a devastating waste of citizens’ tax dollars and result in lower quality-of-life for all. These trends and the current and approaching technology trends will be transformational, creating “The New Mobility,” including less congestion, less road building, less parking and less transit, to name a few. This is all part of “The New Mobility” trend to achieve “on demand” mobility where trips are from doorstep to destination, cost-effective and in acceptable times.

The good news is that The New Mobility can result in major positive impacts on all citizens:

The Problem

The spending of many billions of dollars on light rails throughout the country, including Texas, has failed to achieve the advertised goals of reduced congestion, improved environment and providing effective, equitable mobility to serve the greater good of all citizens. The result has been rapidly growing costs for taxpayers/citizens to support and highly subsidize a rapidly decreasing population choosing transit. This transit population has moved from primarily low-income citizens with no alternative to those who are middle income and choose transit, as a convenience, because it is highly subsidized by taxpayers.

The Solution

We are at the tipping-point of heading in the direction of resolving mobility deficiencies with truly little taxpayer costs compared to the current Austin/Cap Metro Project Connect’s outdated and obsolete plan which will burden this community with generations of added costs while not achieving mobility goals to improve citizen’s quality-of-life. Current, known technologies and pandemic events are creating greater recognition and are bringing “The New Mobility” more rapidly forward to provide many “free” benefits for all. “The New Mobility” will increase the ability of people to travel where they wish, when they wish and cost-effectively, in a reasonable time. It will significantly reduce congestion; reduce the cost of additional roadways, transit, parking and vehicle fuel; lower the cost-of-living; and provide greater quality-of-life for all.

Austin needs to pause, step back and thoroughly evaluate “The New Mobility” and better understand the role each mobility segment will play in the very near future. It would be unconscionable to “double-down” on Project Connect’s massive, unsustainable rail transit plan to serve a minuscule portion of the citizens, when a many times less costly plan is automatically developing to improve mobility for every citizen. For example, a rapid growth of Work-at-Home citizens could easily double the relatively small 8.6% (2018 census) of the Austin area Work-at-Home folks. This increase is a minimal expectation, but will, alone, substantially improve congestion. There are numerous other technology advancements which will also reduce roadway vehicles such as drone delivery to reduce delivery vehicle services, etc. Costs/taxes for roadways, transit, parking, vehicle fuel and others will be substantially reduced.

Austin is a Leading Innovative and Visionary City. Let us continue to move Forward, Not Backwards with light rail, based on 200 year-old, obsolete technology.

Texas’ 2019 Transit Ridership Decline is Consistent with the Total United States Decline

May 7th, 2020

by Jim Skaggs, COST Team, 4-07-2020

This is a companion article to the previous posting “U.S. and Texas 2019 Transit Ridership Declines Predict Transit’s Dim Future,” presenting additional detail regarding transit ridership in Texas’ four largest cities. Texas’ major cities’ transit decline is consistent with other major U.S. Cities., but has a longer declining trend than most.

Texas’ Transit ridership is less today than it was 20 years ago. During this period, Austin has experienced the largest percentage transit ridership decline of 16% and has been the fastest growing region with an 85% population increase. The total ridership decline for the 4 largest cities (Dallas, Houston, San Antonio and Austin) has been 12% as indicated in the second graph below.

Please note: Dallas made a significant change in the way its counts/reports bus ridership in 2019. This resulted in a 30% ridership increase from 2018 to 2019. It is clear this increase is due to the counting procedure. First, no city, under normal, continuing operations has ever increased ridership this magnitude in one year. Dallas media reports indicate a ridership decline in 2019 versus 2018. They did not go back and historically change all years to have a common comparison. The new counting, apparently, did not affect light rail ridership which declined 2.4% in 2019 versus 2018. Until this year’s formal report to the ‘American Public Transportation Association’ (APTA), Dallas’ bus ridership had been reported slightly higher than Light Rail for some time. The reported 2019 bus ridership is 30% higher than 2018 and is 37% higher than light rail in 2019. DART, Dallas’ transit agency, has already approved a plan to improve its bus route structure by 2022, as Houston completed in 2015 and Austin did in 2019. Therefore, with incomplete facts, but good indications, I left the 2019 bus ridership the same as it was in 2018, which does not consider the decline reported, in non-specific numbers, by Dallas media. This results in Dallas having almost the same total light rail and bus ridership as 20 years ago and the the least decline of the 4 cities. This is after Dallas has spent billions of dollars on rail and creating the longest light rail in the U.S., in an attempt to increase rail ridership when population increased more than 50%.

Although Dallas spent many billions of dollars more than the other 4 major Texas cities to create the longest light rail in the Nation, Dallas’ transit ridership per-capita (per-citizen population) is less than all 4 cities. Houston, Austin and San Antonio have heavily weighted bus transit, all with significantly greater transit riders per capita. Even with Dallas’ major increase in 2019 bus ridership, due to their modified counting system, Austin per-capita ridership is 35% higher than Dallas and therefore more cost-effective without light rail. San Antonio is, by far, the most cost-effective transit system and is the only one of the 4 cities with zero rail transit. San Antonio collects a 0.5% sales tax to fund transit, only one-half of the 1% sales tax collected for transit in the other 3 cities.

As noted in the chart below, Houston experienced a huge ridership decline when they embarked on light rail. A small portion of this decline was recovered when Houston completed the redesign of its bus route system in 2015. These bus route improvements have been accomplished in several cities, including Austin (2018), and have typically resulted in delaying the ridership decline with short term ridership gains of a small percent and then continuing the decline. Austin’s new bus transit route structure opened in 2018 and the impact was very small in 2019. We will see!

Austin has experienced the longest term, reasonably consistent ridership decline of all the Texas cities. Adding the commuter rail in 2010 resulted in little movement in the trend. The slight up-tic in 2019 is a bit of a deception by Cap Metro. The actual improvement in the bus route structure is even smaller than the small increase that the chart indicates. At the same time Cap Metro opened the improved routes, it changed a key policy and allowed K-12 students to ride free. This resulted in more than 2 million free K-12 riders in the first year of operation. While the media coverage indicated a significant number of students were switching from school buses to Cap Metro free-ride buses, we do not know the proportion.
4 City Ridership
The next chart reflects the total ridership and population of the 4 major Texas cities. As shown the total population increase for the 4 regions was 7.2 million or 57%. With the population increase and the ridership decline, the Ridership Per-Capita declined a whopping 44% over the past 20 years.
4 Cities Total Ridership_Population Master
The Future: New Mobility

Recent events with the COVID-19 virus have highlighted the strong potential of accelerating a number of current trends in the U.S. Two of these trends have been moving somewhat slowly for several years, but their acceleration could very likely have significant, positive impacts on mobility.

1. Growing Work-at-Home trend: ‘Work-at-Home’ is one of the defined “Journey to Work” modes of commuting in the U.S. census. It has been the fastest growing commute mode for several years and recently exceeded the Public Transit mode of commuting to work in the U.S. In Austin, the Work-at-Home mode has grown faster than average, and it is more than double the Public Transit mode. A very high percentage of the ‘Work-at-Home’ citizens result in a car off the road.

2. Declining Public Transit: On the other hand, Transit use has been declining in the U.S. for years. Texas has less transit ridership than it had 20 years ago while spending several Tens of Billions of dollars in major cities to increase transit ridership, including the addition of Dallas’ longest light rail in the Nation. Of the four major Texas cities during this 20-year period, Austin has had the largest population growth (85%) and the greatest decrease in transit ridership (-16%), even with the benefit of millions of free rides for K-12 students.

The coronavirus epidemic has dramatically increased the number of ‘Work-at-Home’ people while significantly decreasing the number of Transit commuters. What is the lasting impact of these trends? The Project Connect plan was very obsolete before the virus and a moderate increase of these two trends could result in an even more tragic, devastating waste of citizens’ tax dollars. These trends and the current and approaching technology trends would be transformational, creating The New Mobility, including less congestion, less road building and less parking, to name a few. This would result in lower cost, faster commuting; greater affordability; lower taxes; and, lower cost mobility with greater access to opportunities and offerings. One small example: Just doubling the Work-at-Home population in Austin would substantially reduce congestion on the roadways, reduce the need for additional road capacity and educe travel times.

Austin needs to pause, thoroughly evaluate the rapidly arriving “New Mobility” and better understand the role each mobility segment will play in the very near future. It would be unconscionable to “double-down” on Project Connect’s massive rail transit plan to serve a minuscule portion of the citizens, when a several times less costly plan is very possible and will improve cost-effective mobility, better meeting needs of all citizen. Let’s move Forward, not Backwards.

U.S. and Texas 2019 Transit Ridership Declines Predict Transit’s Dim Future (Before the COVID-19 VIRUS CRASH)

April 25th, 2020

COST Commentary: Randal O’Toole’s Article, below, was published after the official 2019 ridership data was released and prior to the onslaught of the COVID-19 virus which has produced the most devastating impact on transit in modern history. The current impact on COVID 19 on transit and consideration of the future implications will be discussed in upcoming articles on this site.

Focusing on local Cap Metro/Austin area transit, there are some very Dishonest, Deceptive and Disparaging aspects. This also will be covered in later articles. The summary is: Cap Metro and most of the uninformed media are describing local transit ridership by preaching the term” “17-month streak of increasing boardings.” What does this mean to you? It implies that there is an increasing trend in ridership over 17 months. No, the facts are that there have been 17 months where each months ridership is greater than the same month in the previous year. In other words: if ridership was up a few people in the first of these 17 months from the same month a year ago and remained up by an average of that same number of people over the 17 months, they would describe it the way they have, even though the ridership number in the 17th month could be the same as it was in the first month with no growth at all from the first month. This deception is only a “tip of the iceberg.” In mid-2018, Cap Metro opened the newly revised, “more effective,” bus route structure which had been developed over a year with significant cost. Many cities,including Houston did a similar restructuring. Austin and each city experienced a slight ridership increase which is expected after many years of minor response to the community’s rapidly changing transit transportation needs. It is not reasonable to expect this one-time route improvement to be a continuing upward trend as implied by Cap Metro. Studies of several cities which redesigned their bus route have shown that they typically increase a little in the beginning but continue in the decline a little later. However, Cap Metro has also been very Deceptive in the shadows. It did not explain the total ridership impact of a major policy change about the same time as the new routes opened in mid-2018. This policy provided free ridership to all K-12 students and more than 2 million rides were recorded in the first year. With this “free Ride” increase, any increase in total transit ridership is very small, due to the routing upgrade. And, as shown by other cities, the slight increase Is temporary very unlikely to extend the increase.

The Bottom Line: Cap Metro”s ridership is still less today than it was 20 years ago and this long declining trend is very likely to continue and accelerate in view of the predicted impacts of coronavirus and possible future pandemics. Meanwhile, the Austin region’s population has grown more than 80% in the past 20 years resulting in the transit ridership per-capita dropping steeply. This has resulted in taxpayers paying huge subsidies for every transit rider as transit fares are less than 10% of the operating costs (excludes large capital costs). Transit ridership is less than 2% of the daily passenger trips (less than 1% of the daily passenger miles traveled). The trend is that lower income citizens are being driven beyond the range of transit, resulting in this segment being a smaller percentage of the reducing ridership and middle/higher income riders are increasing their ridership percentage as a CONVENIENCE alternative, subsidized by taxpayers. Austin and Cap Metro are now proposing a massive transit upgrade, named ‘Project Connect”, for $10 Billion. This system is totally outdated and ineffective; it will not improve congestion and will very likely make it much worse; it will severely limit future citizen generations’ access to funds necessary to meet their primary needs due to major long term debt and tax increases to pay for an expensive, ineffective light rail system, the most costly public transportation mode in the world. The City of Austin has never fully explained a responsible plan as to how it will pay for the system and its huge annual operating costs. The tax and debt increases would be very large for many years. In any event, the congestion will become worse, citizens’increased taxes will be a large burden, cost-of-living will have large increases for everyone, and all citizens will experience a lower quality-of-life.

Reminder: This posting of 2019 transit ridership does not include any of the major negative effects on public transportation which numerous experts are predicting, resulting from the COVID-19 virus impact. Cap Metro’s ridership has been down more than 60% during the past several months. The City needs to step-back; fully reconsider the future needs/roll of public transit; and, the massive, destructive impact on citizens and the entire community of this ill-advised, obsolete Project Connect transit plan which would serve a minuscule portion of the population at the expense and degradation of quality-of-life for all.
2019 Ridership Data Reveals Transit’s Dim Future

By The Antiplanner | February 11, 2020 | Policy brief

Thanks to a late-year surge in New York subway ridership, nationwide transit ridership in December 2019 was 3.0 percent greater than December 2018, and ridership for 2019 as a whole was 0.1 percent greater than in 2018, according to data released last week by the Federal Transit Administration. Take away the New York City subways and nationwide ridership fell by 1.5 percent in December and 1.2 percent for 2019 as a whole.
Ridership New York
Ridership is growing on New York subways but declining almost everywhere else.

New York City subway ridership (not including PATH trains) grew by a phenomenal 14.7 percent in December and 3.6 percent for the year as a whole. While subway ridership peaked in 2014, it rose in 2019 to the second highest in its history. Its post-World War II peak was only about 2.0 billion trips a year compared with 2.7 billion in 2019.

The rest of the nation’s transit industry didn’t fare so well. Ridership in both December and 2019 as a whole fell in 32 out of the nation’s 50 largest urban areas, from 2018 to 2019. When compared with 2014 ridership, 2019 ridership fell in 44 out of the 50 largest urban areas.

Even New York-area ridership, other than the subway, is doing poorly. Not counting the subway, New York-area ridership fell by 3.6 percent in December. In fact, non-subway ridership in the New York area peaked in 2005 and has slowly but steadily declined since then. In 1995, the subway accounted for less than half of the region’s ridership; in 2019 it was 63.5 percent.
Transit Trends by Mode

Rail ridership surpassed bus ridership in 2017, more due to the decline of the latter than the growth of the former. “All other” mainly includes ferries and demand response but also includes automated guideways (people movers), monorails, ferries, and other minor forms.

Transit agencies should particularly worry about the future of their bus systems. Transit buses, including commuter buses, trolley buses, and bus-rapid transit as well as conventional buses, carried fewer riders in 2019 than in any year since 1939. According to appendix A of the American Public Transportation Association’s latest transit fact book, buses carried 75 percent of all transit riders in the country in 1960 and 65 percent in 1990. As of 2019, it was down to 47 percent.

The growth of New York subway ridership doesn’t mean that heavy rail makes sense anywhere else, as it is declining in almost every city that has it other than New York and Washington.

Light rail is also doing poorly, having lost more than 4 percent of its riders in both December and in 2019 as a whole. Streetcars are also in decline despite the fact that most of them are fairly new. Hybrid rail, meaning Diesel railcars operated on light-rail or commuter-rail schedules, grew in 2019 but only because of the opening of a new line in the San Francisco Bay Area in late 2018. In addition to New York City, heavy-rail ridership grew in Washington, but it declined in Atlanta, Baltimore, Boston, Chicago, Cleveland, Los Angeles, Miami, Philadelphia, and San Francisco. Commuter-rail ridership is growing slowly, mainly due to growth on New York’s Long Island Railroad and the opening of new lines in Denver, Salt Lake City, and Seattle.
This next section is about spreadsheets for the data.

The Big Six

Outside of New York (which sees 44 percent of all transit rides in the country), the six biggest transit urban areas are Los Angeles, Chicago, Philadelphia, Washington, Boston, and San Francisco-Oakland. Together, these six areas are responsible for 27 percent of U.S. transit ridership. These six areas plus New York are 71% or total U.S ridership.
Urban Area Transit Ridership
None of the six biggest transit markets after New York are healthy.

Yet transit is hardly healthy in any of these areas. Ridership in Los Angeles and Washington peaked in about 2008; Chicago in 2012; and the others in 2013 or 2014. San Francisco’s transit is doing best, but still lost nearly 4 percent of its riders since 2014. The others all lost at least 10 percent and Los Angeles more than 20 percent. If transit can’t hang on in Chicago, whose downtown is dense in both jobs and residences, then it is really becoming a one-urban-area industry.

Seattle: The Fading Hope

As transit ridership declined in most other urban areas, it grew in Seattle, which gave many in other transit agencies hope that some sort of formula could be found to rescue their systems. Some credited Seattle’s ridership growth to new light-rail and commuter-rail lines, but as shown in the chart, commuter rail is practically insignificant while light rail is only a small player in the region’s transit.
Seattle Area Ridership by Mode
Despite spending billions on rail, the vast majority of Seattle transit riders are still bus riders.

In fact, the region’s ferry system carries more riders than light rail, which should be embarrassing to rail supporters as Washington’s 22 ferries (not all of which serve the Seattle area) typically cost about $100 million each while Seattle has so far spent several billion dollars on light rail. Meanwhile, commuter buses carry far more people than commuter rail at a far lower cost. (Note that a decline in motor bus ridership in 2012 is due to the reclassification of some bus routes as commuter buses.)
I’ve noted before that Seattle’s growing ridership is mainly due to growth in the number of jobs in downtown Seattle. Indeed, from 1999 through 2018 the correlation between downtown Seattle jobs and Seattle-area transit ridership is almost perfect at 0.97. Amazon is now looking elsewhere to locate its employees, which may be why Seattle transit ridership declined by 0.9 percent in 2019.

Death Spirals

Two factors that contributed to declines in transit ridership after 2014 were ride hailing and gasoline prices. The growth of ride hailing has been emphasized by many transit agencies seeking to tax Uber and Lyft for their own benefit. But at least as important was a dramatic drop in gasoline prices; in some markets, the average price per gallon of regular gasoline fell by 50 percent. This encouraged people to switch from transit to driving: the United States had 8.7 million more workers in 2018 than 2014, but the number of workers who lived in households without a car actually declined, as did the number of transit commuters.
Urban Area Transit Ridership (1)
Ridership in many urban areas began to decline long before 2014.

While both of these are factors in declining transit ridership; in some urban areas, ridership began to fall long before 2014. These include Atlanta, Baltimore, Cleveland, Sacramento, Milwaukee and Memphis, among others. As shown in the chart, declines in several of these areas began after the 2008 financial crisis, while others began even earlier. In some of these areas, such as Cleveland and Milwaukee, the declines probably relate to a decline in downtown jobs; in others, such as Atlanta and Baltimore, they are probably due to cuts in bus service made to help finance rail.
Cleveland Area Trips and VRM by Mode
VRM: Vehicle Revenue Miles
Cleveland transit is clearly in a death spiral, as cuts to service follow declines in ridership, leading to more service cuts.

Bus Reconfigurations

Of the six major urban areas that did not lose ridership in the last five years, four of them did Jarrett Walker-style reconfigurations of their bus systems, including Columbus, Houston, Richmond, and Seattle. Richmond, which redesigned its bus system in 2018, saw a 14.9 percent increase in ridership for the year. However, Houston, which introduced its redesigned bus system in 2015, saw only a 0.3 percent increase in ridership for the year. Numbers like these lead me to conclude that such reconfigurations are worth doing but are not a long-term solution to transit ridership decline.
Urban Transit Ridership 2 - Houston, etc

Redesigning bus networks often, but not always, leads to increases in ridership, but such increases may not be sustained in the long run.

The Transportation Research Board recently released an evaluation of bus network redesigns, including case studies in five different urban areas. The paper noted that this is “the hottest trend in transit” and that many agencies are considering such redesigns in response to declining ridership. Yet, in some cases, the paper found, the best it did was to slow the decline. Baltimore’s, for example, is still declining, and while Bloomington’s ridership has grown since it implemented its new system in 2016, its ridership still has not reached the level it was in 2014, just two years before the reconfiguration.

Rail Disasters

Los Angles, Dallas, Phoenix, Baltimore, St. Louis, San Jose, and Charlotte are just some of the urban areas that have spent heavily building new light-rail lines only to see it blow up in their faces as bus ridership sank more than light-rail gained new riders. The introduction of light rail into these regions, most of which were never suited for rail transit, is nothing short of disastrous.
L.A. Area Trips and VRM
VRM: Vehicle Revenue Miles
Rail may be a boon for Los Angeles’ construction industry but it is a disaster for transit riders and transportation in general.

Whenever it opens a new light-rail line, Los Angeles loses five bus riders for every light-rail rider it gains, and even light-rail ridership declines in years that it doesn’t open new lines. As the chart shows, the region’s loss in bus riders is paralleled by, and largely the result of, a loss in bus service.
Portland-Area Ridership by Mode
Portland opened a major new light-rail line at the end of 2015, resulting in a slight increase in rail ridership accompanied by a big drop in bus ridership.

The New York Times once called Portland “the city that loves transit,” but what it meant was “the city run by a light-rail mafia that loves to spend taxpayers’ money on transit.” (The term “light-rail mafia” was first used in print in a 2004 newspaper article that is no longer on line but that I’ve posted here.) Portland has a hybrid-rail line that carries so few passengers it doesn’t even show on the chart. Portland opened a light-rail line near the end of 2015 that cost $1.5 billion, yet it barely registers as a blip on the chart, and buses lost more riders than rail gained that year. Both bus and rail have declined since then.
D-FTW Jpeg
VRM: Vehicle Revenue Miles
Like so many other urban areas, Dallas-Ft. Worth lost more bus riders than it gained rail riders when it spent billion on commuter-rail and light-rail construction.

Dallas spent billions of dollars building more miles of light rail than any other city in the country, but its bus ridership has fallen far more than rail gained. (The increase in bus ridership shown for 2019 is due to a change in counting methods, not an actual increase.) The Dallas-Ft. Worth commuter rail line carries an insignificant number of passengers and a new commuter line opened a year ago at a cost of about a billion dollars is carrying just 20 percent of its expected riders.

Non-Rail Urban Areas

The six largest urban areas that haven’t built any rail transit (other than, in a couple of cases, a short streetcar line) saw mixed results in 2019. Three saw ridership grow; three saw it decline. There certainly isn’t any indication in the data that rail transit significantly boosts ridership in the short or long run. Instead, all it does is boost the tax burden on local residents.
Urban Area Trasit Ridership(3)
Ridership in 2019 grew in half of the six largest urban areas that don’t have rail transit, which is a better record than for the areas that have rail transit.

Transit’s Dim Future

Despite the growth in New York City subway ridership, the future for the rest of the transit industry is dim. Building rail transit usually does more harm than good to a transit system. Bus reconfigurations can give transit a one-time boost but will not be a long-run savior. Crowding jobs into downtown Seattle helped transit in that urban area, but the policies that led to that were self-limiting due to strife over housing affordability, homelessness, and taxes.The reality is that gasoline is cheap and autos provide people access to far more jobs and other economic opportunities than transit. This means more people are buying cars and becoming less dependent on transit. Instead of mourning the loss of transit riders, we should celebrate that those people now are likely to have better jobs, housing, and other economic benefits.

Cars are also getting more fuel-efficient, electric cars are getting more affordable, and transit already uses more energy and emits more greenhouse gases per passenger mile than the average car in all but a handful of urban areas. Thus, there is little justification for trying to get people out of their cars and onto transit, which in turn means there is little justification for the tens of billions of dollars of annual subsidies American taxpayers give to the transit industry.

Transit’s Energy Savings Are A Myth in Texas and the U.S.

January 11th, 2020

COST Commentary: This recent article by Randal O’Tool is summarized in his final conclusion for the article below: “Thus, driving a car or light truck is more greenhouse-gas-friendly than transit in 90 of the nation’s 100 largest urban areas (and all but a handful of the smaller ones).

If transit does not improve traffic congestion or the environment in Texas cities, what is the reason Austin, Dallas, Houston and San Antonio are spending many billions of dollars to increase transit and its ridership. The result has been decreased transit in all four cities over the past 20 years while population has grown a total of approximately 150%. These wasted Taxpayers dollars are subsidizing each average transit rider in Austin with more than 100% of the trip’s costs as reported by Cap Metro. This is due to Transit agencies reporting trip costs without including capital cost of major investments in buses, rail transit equipment and other special transit capital.

Austin has the strongest population growth of 170% in the past 20 years and the largest decline in transit ridership of 20%. It is obvious that our leadership is not serving the greater-good of the citizens of Austin.

Urban Transit Is An Energy Hog

By The Antiplanner | December 24, 2019 | Policy brief

Transit is often touted as a way to save energy. But since 2009 transit has used more energy, per passenger mile, than the average car. Since 2016, transit has used more than the average of cars and light trucks together.

Automobiles and planes are becoming more energy efficient each year. But the annual reports of the National Transit Database reveals that urban transit is moving in the opposite direction, requiring more energy to move a person one mile in each of the last four years.
Randal Energy Graph 1
Transit has been less energy efficient than the average car since 2009. Light trucks (vans, pickups, SUVs) may soon become more efficient than transit as well. 2018 automobile data are not yet available; 2017 numbers are estimated from this report; prior years are from the Transportation Energy Databook.

The reason for this is simple: ridership is declining, but transit agencies aren’t proportionately reducing miles of transit service. As a result, the average occupancies of buses and other transit vehicles has declined in every year since 2013. While transit agencies may be purchasing more fuel-efficient vehicles, the increase in average efficiencies per vehicle mile can’t make up for the loss in passengers.
Randal Energy Graph 2
Transit occupancies have steadily declined since 2013. Bus” includes commuter bus, rapid bus, trolley bus, and conventional bus (which the FTA calls “motor bus”). “Rail” includes commuter, heavy, light, and hybrid rail and streetcars, but not monorail or automated guideways. “All” includes all transit, not just bus and rail.

These numbers are based on the National Transit Database, which reports the number of gallons of Diesel fuel, gasoline, natural gas, and other fuels as well as the number of kilowatt-hours of electricity that are used by transit systems across the country. I’ve converted these numbers to British thermal units (BTUs) using standard factors, such as that a gallon of Diesel fuel has 138,500 BTUs.

For electricity, I also took into account the fact that two-thirds of the energy used in a power plant is lost in generation and transmission. In other words, in order to deliver 1 kilowatt-hour (3,412 BTUs) of energy to a customer, an electrical system must consume the equivalent of 10,236 BTUs of fossil fuels or other energy at the power plant. Electric motors tend to be more efficient than internal combustion engines, but when the losses from generation and transmission are accounted for, the efficiencies are about the same.

Energy Consumption by Mode
The calculations show that ferries and streetcars use huge amounts of energy per passenger mile, as do automated guideways (i.e., people movers), which aren’t shown in the chart but average even more energy per passenger mile than ferries. Buses and light rail are well above the average automobile.
Randal Energy Graph 3
Ironically, the most energy-efficient transit mode–van pools–is the one that is based on conventional automobiles rather than large buses or railcars.

Commuter and subway/elevated trains (heavy rail) appear to be more efficient, but this is largely because commuter- and heavy-rail numbers are dominated by New York where occupancy rates are high. Commuter rail lines in such regions as Dallas-Ft. Worth, Miami, and even Philadelpha use far more than the average amount of energy per passenger mile, as do heavy rail lines in Baltimore, Boston, Los Angeles, and Miami. Perhaps the biggest surprise is the DC Metrorail, the nation’s second-most heavily used rail system, which consumes almost 25 percent more energy per passenger mile than the average light truck used in 2017.

(We apologize for the slight technology format glitch in this display of data. It will be corrected soon.)

Energy Consumption by Urban Area
The numbers for individual urban areas are even worse for transit. Among the largest 100 urban areas, transit is more energy-efficient than cars only in New York, San Francisco-Oakland, and Honolulu. Transit in Atlanta and Portland is less energy-efficient than cars but more than the average light truck. Just about everywhere else, transit is a real energy hog. The adjacent table has numbers for the 54 urban areas. Among smaller urban areas, Stock- ton (which is the 102nd largest area) appears to be more energy efficient than cars, but only because the Altamont Commuter Express is attributed to Stockton.

Even where rail transit appears to be more energy efficient than driving on an operational basis, this doesn’t account for the energy costs of construction. Urban roads carry far more passengers over their lifetimes than rail lines, so the energy cost of construction per passenger mile is much higher for rail transit. Rails must be rebuilt about every 30 years, which also requires large amounts of energy. Heavy use of steel and concrete also has a high greenhouse gas cost.

Greenhouse Gases
Though transit is less energy efficient than cars, it emits slightly fewer greenhouse gases per passenger mile than the average car. Transit was actually worse than the average car as recently as 2010, but by 2014 it had reduced its climate footprint by 25 percent.

It accomplished this partly by converting from Diesel to other fuel sources, originally biodiesel but more recently compressed natural gas. In addition, the nation’s electric industry has converted from heavy reliance on coal to heavy reliance on natural gas. Both of these changes reduced greenhouse gas outputs per unit of energy. Since 2014, however, declining transit ridership increased greenhouse gas emissions per passenger mile by about 7 percent.
Randal Energy Graph 4
The main transit energy trend over the last decade has been the replacement of Diesel fuels with compressed natural gas, which paralleled the electric industry’s conversion from coal to natural gas.

Calculations of greenhouse gas emissions are straightforward for most fuels since burning a gallon of gasoline, Diesel, or natural gas results in specific outputs of carbon dioxide. For electricity, I presumed that the electricity used by a transit agency is generated by a the combination of power sources used in the agency’s state, as reported in the Department of Energy’s State Electricity Profiles. Even if a transit company claims that it buys renewable energy, the reality is that electricity is fungible, and renewable energy consumed by a transit agency means less renewable energy for someone else.

While transit scores better than automobiles overall, this is only because of New York, which produces some 44 percent of transit riders and whose electricity profile claims to emit less than half the national average of carbon dioxide per kilowatt-hour. However, New York doesn’t generate enough electricity to satisfy its needs and must import some, and the greenhouse gases attributable to imported electricity is unknown.

Two-thirds of all states are net electricity exporters, and some major exporters such as Texas and Wyoming generate most of their electricity with fossil fuels. Many of the importer states, including California and New York, generate most of their electricity from non-fossil-fuel sources, but their imports are probably more dependent on fossil fuels.

For a sensitivity analysis, I assumed that electricity brought into net importer states was generated by the national average of fuel sources. Under this assumption, electric-powered transit generated 22 percent more greenhouse gases in California, 15 percent more in New York, and about 7 percent more in Massachusetts, Maryland, and Virginia, while Washington DC transit generated 17 percent less greenhouse gases. For the most part, these numbers aren’t big enough to fuss about, especially since we can’t accurately estimate the mix of sources of energy that is imported into the various states. The greenhouse gas emissions shown in the adjacent tables are based on state electricity profiles with the caveat that the actual numbers in California and New York are probably higher while DC is probably lower.

Based on the state profiles, transit is more greenhouse-gas-efficient than cars nationwide, but it is more efficient than cars in only seven out of the nation’s 100 largest urban areas. Further, transit is more greenhouse-gas-efficient than light trucks in only three more urban areas. Thus, driving a car or light truck is more greenhouse-gas-friendly than transit in 90 of the nation’s 100 largest urban areas (and all but a handful of the smaller ones).

The results of my calculations of energy consumption and greenhouse gas emissions for each transit agency, mode, and urban area are in my 2018 Transit Database summary spreadsheet. For details on how to use this spreadsheet, see last week’s policy brief.

The Nation and Austin have a Loosing Struggle with Ineffective Transits’ Trend of Increasing Costs and Decreasing Ridership.

December 24th, 2019

COST Commentary: In a 20 year declining ridership trend, Austin is reporting a slight gain in transit ridership in 2019. This gain is primarily due to the the result of the one-time impact of a major system route revision in mid 2018, called Cap ReMap. This is the first such route revision in many years. It is very unlikely to regain the ridership of 20 years ago based on similar experiences in several major cities. Meanwhile, Austin’s Cap Metro transit operating costs have been on a significant up-trend of about 150% in the past 10 years. This has resulted in increasing taxpayer subsidies for each rider.

Below is an article by Randal O’Toole and posted on his Antiplanner blog. It is a summary of the Nationwide trend of transits’ increasing costs and declining trend in ridership. This is also a period of increasing population. Austin’s four major cities have increased population an average of 150% in the past 20 years and all four cities have less transit ridership today than 20 years ago.

Costs Up, Riders Down: 2018 National Transit Database (NTD)

By The Antiplanner | December 17, 2019 | Policy brief

Taxpayers spent nearly $3.75 billion more subsidizing transit in 2018 than the year before, yet transit carried 215 million fewer riders, according to the latest data released by the Federal Transit Administration. The increase in spending didn’t even translate to an increase in service, as transit agencies provided 44 million fewer vehicle miles of service in 2018.

n percentage terms, subsidies rose by 7.4 percent while ridership fell by 2.1 percent and vehicles miles of service fell by 0.9 percent. These numbers are from the 2018 National Transit Database, a series of 30 spreadsheets summarizing the annual performance of all of the nation’s transit agencies that have received federal support (which is nearly all of them). Numbers in the database are based on each agency’s fiscal year, so may not exactly agree with calendar year numbers calculated from the monthly updates.

Total transit ridership in 2018 was lower than any year since 2006. Bus ridership has plummeted to be lower than any year since 1940, when streetcars still carried almost half of all of the nation’s transit riders.

The industry has not responded to declining ridership by reducing its costs. Instead, operating costs grew by $1.9 billion (4.0%), despite the decline in vehicle miles of service. Expenditures on capital improvements, that is, expansions of existing systems, grew by 7.9 percent or close to $500 million. The vast majority—84 percent—of these capital improvements were for some form of rail transit.
Cost Up-Riders Down - Chart 1
Most of the increase in operating costs was due to labor.

The biggest increase in operating costs was fringe benefits, accounting for nearly $700 million while operator salaries and wages were responsible for $177 million. “Service costs,” defined as “labor and other work provided by outside organizations,” accounted for $282 million of the increase, while the growth in non-operator salaries and wages was $58 million. Fuel costs accounted for $110 million of the increase and utilities almost $90 million. But clearly, labor was responsible for most of the growth in operating costs.

Transit’s State of Poor Repair

The database also keeps track of capital expenditures for existing systems. In the past, I’ve called this “maintenance” but it should more properly be called capital replacement. Changing the oil on your car is maintenance; buying your first car so you don’t have to depend on transit any longer would be a capital improvement; buying a new car to replace one that is worn out is capital replacement.

The Department of Transportation also just released its latest report on the status of the nation’s transportation infrastructure. The report found that the transit industry has deferred capital replacement for so long that it has a $98 billion backlog in 2014 dollars, which would be more than $106 billion in today’s dollars. In order to erase its backlog in twenty years, the report concluded, the industry needed to increase its spending on capital replacement to $18.4 billion a year ($19.6 billion in 2018 dollars).

In fact, the industry spent $14.9 billion in 2018, a $1.4 billion increase over 2017. Most of that increase was probably spent on activities aimed more at slowing the decay of transit infrastructure than at reversing that decay. At the 2018 rate, it will take more than 75 years to bring transit systems to a state of good repair.

Among capital costs (including both improvements and replacement), the biggest increases were for renovations of passenger stations, replacement of signaling systems (which should include positive train control for existing rail lines), and replacement of vehicles. Even though the recent status report says that transit’s fixed guideway systems (meaning, for the most part, rails) have a $23 billion backlog, the transit industry reduced spending on guideway rehabilitation by $131 million, even as it increased spending on construction of new guideways by $217 million.
Cost Up - Riders Down - Chart 2
The biggest increase in capital replacement spending was for stations, not guideways, suggesting the political agencies are more inter- ested in visible improvements, not ones needed for safety.

The industry could eliminate its backlog in less than 20 years if it stopped building new transit lines and spent all of that money, about $6.7 billion in 2018, on capital replacement. But transit agencies are more interested in empire building than in keeping their infrastructure in good repair, an attitude reinforced by Congress’ willingness to spend $2 billion a year providing matching funds for new rail transit lines.

Even better than replacing obsolete rail lines with more obsolete rails would be to replace worn-out rail systems with buses. Buses don’t require dedicated stations, sophisticated dedicated signaling systems, or dedicated guideways. Buses have shorter lifespans than rail vehicles, yet they cost so much less than railcars that the long-run cost per bus seat mile is much lower than for railcars. In some places, such as Manhattan, buses may not be able to replace trains, but in most places buses can move more people per hour than rail lines taking up the same amount of real estate.

Modal Results

We’ve already seen calendar year results from the monthly updates to the National Transit Database. However, those numbers were preliminary (and were missing in a few cases) while the numbers in the final 2018 spreadsheets will become “official” even though the fiscal years of the agencies in the database are not all the same.

The database shows that transit ridership declined in 2018 for almost every major mode of transit. It fell for commuter buses, rapid buses, trolley buses, and regular buses, with buses in total losing more than 99 million rides. Heavy rail alone lost another 92 million riders and light rail lost 38 million. While ridership grew by 0.6 percent for both commuter rail and hybrid rail, those gains were so small that rail transit lost a total of 126 million riders. Rail also lost more than buses in percentage terms, with rail ridership falling 2.6 percent compared to a 2.0 percent decline for buses.
Cost Up - Riders Down - Chart 3
Costs per trip increased for almost all major modes of transit.

Although vehicle miles of service declined, they didn’t decline by as much as ridership, with the result that costs per rider and per passenger mile significantly grew. Costs per trip grew by 7.1 percent by bus and by 9.0 percent for rail, with much of the increase for rail due to the added spending on capital replacement.
Cost Up-Riders Down - Chart 4
Subsidies per passenger mile grew even faster than costs per trip.

Subsidies grew even faster as taxes covered an increasing share of the cost of transit. Subsidies per passenger mile grew by 9.9 percent for rail and 10.8 percent for buses. At a 62 percent increase, subsidies grew fastest for trolley buses, followed by 40 percent for hybrid rail and 31 percent for rapid buses. In the case of trolley buses, most of the increase was for capital replacement, while for hybrid rail and rapid buses most of the increase was for capital improvements.

Urban Area Results

Ridership declined in 2018 in 40 of the nation’s top 50 urban areas. On a percentage basis, the worst hit were San Juan (-29.5%), which was due to Hurricane Maria; Milwaukee (-12.2%), and Cleveland (-10.9%). In numbers, the worst were New York (-60.5 million), Los Angeles (-23.6 million), Chicago (-14.8 million), San Juan (-10.3 million), Boston (-10.2 million), Miami (-8.9 million), Atlanta (-7.8 million), Baltimore (-7.6 million), San Francisco-Oakland (-6.3 million), and Philadelphia (-5.8 million).

Seattle has been celebrated for defying trends and gaining transit riders, but ridership in 2018 was only 0.6 percent more than 2017. Houston, which reformed its bus system, gains 2.5 percent more riders. Ridership gains were also enjoyed by Denver (6.9%), San Antonio (7.2%), Indianapolis (3.2%), and Columbus (2.7%). Pittsburgh and Providence gained less than 1 percent each.

Reliable data on downtown jobs is difficult to obtain as the definitions of downtowns can vary from person to person. Wendell Cox did an inventory of downtown jobs in the nation’s 50 largest cities in 2000 and a second inventory based on 2010 data using a consistent method of defining downtowns. Of the 47 urban areas on both lists, downtown job numbers fell in 29 and grew in 18.

Some of the largest declines were in cities that are losing transit ridership. Cleveland and Milwaukee each lost 15 percent of their downtown jobs. Chicago lost 7.6 percent and Los Angeles 5.0 percent. New York, which enjoyed large gains in transit ridership in the 2000s, gained 14 percent more downtown jobs. It is likely that transit increases and declines in more recent years are also due to changes in downtown job numbers.

Transit Speed

Los Angeles Metro blames the loss of transit riders on slowing transit speeds due to congestion. This leads transit advocates to argue that transit buses “deserve their own lanes” in order to boost speeds and increase ridership. This is a ironic considering that the increase in traffic congestion in Los Angeles and many other places is largely due to policies that spent most transportation dollars building rail transit lines rather than improving roadway capacities.

In fact, the database offers some support for the claim that ridership is affected by transit speeds. Average transit speeds can be roughly calculated by dividing vehicle-revenue miles by vehicle-revenue hours. By this measure, transit vehicles average 15.08 miles per hour in 2018, down from 15.16 miles per hour in 2017 and 15.20 miles per hour in 2016. Los Angeles bus speeds averaged 10.4 miles per hour in 2018, down from 10.5 in 2017 and 10.6 in 2016.

With ridership and speed data going back to 1994, the correlation between Los Angeles Metro bus speeds and ridership is a respectable 0.65. Counting all transit in the country, with data going back to 1991, the correlation is even higher at 0.78. Of course, correlation doesn’t prove causation and there may be other factors at work affecting both speeds and ridership. In addition, this measure of transit speeds is crude: if a bus reaches the end of its route waits 10 minutes before starting on the return trip, that ten minutes would be included in the calculation of miles per hour. Thus, speeds could appear to be increased or decreased simply by reducing or increasing the wait or dwell times between vehicle trips.
Cost Up - Riders Down - Chart 5
Average speeds of most modes of transit declined slightly in 2018, but speeds of motor bus (the FTA’s perplexing name for conventional bus service) and light rail both slightly increased.

In any case, rather than make congestion worse for non-bus riders (which means the great majority of people in every American city not named New York) in order to make it better for buses, it would make more sense to fund programs that would relieve congestion for everyone. This is especially true because simply having dedicated lanes doesn’t make buses much faster, as most of their time is spent picking up and dropping off passengers. The 2018 database reveals, for example, that supposedly “rapid buses,” many of which use dedicated lanes, go an average of 10.1 miles per hour, compared with 12.0 miles per hour for regular buses.

The Free-Transit Movement

Due to increases in average transit fares, total fare revenues grew by $44 million or 0.3 percent despite a loss in transit riders. The increase in fare revenues, however, only offset about 1 percent of the total increase in costs. As a result, fares now cover just 23.6 percent of the costs of transit and subsidies to the transit industry grew from $50.5 billion in 2017 to $54.3 billion in 2018.

The decline in the share of transit costs paid by users is a good thing, according to California Senator Scott Wiener. “Transit agencies brag about high fare-box recovery,” Wiener tweeted recently. “High fare-box recovery is bad. It means tax $ isn’t supporting transit. It means high fares that lower ridership & harm low income ppl. The goal is low farebox recovery. Transit is a public good & should have taxpayer support.”

Wiener’s November 21 tweet was written in support of the free transit movement, which was promoting a fare strike a few days later on November 29. Yet this movement in general and Wiener’s tweet in particular is based on numerous fallacies.

Most important, transit is not a public good, at least, not in the economic sense of the term. A true public good meets two requirements: it is non-excludable and non-rivalrous. That is, everyone benefits whether they pay for it or not (non-excludability) and one person’s consumption doesn’t reduce the availability to another person (non-rivalrous). Some economists class public goods as an example of market failure and suggest that government support is needed to supply such goods.

Ironically, the article linked above uses a lighthouse as an example of a public good, but Nobel prize-winning economist Ronald Coase showed that the first lighthouses were provided by private insurance companies. Other articles use public parks as examples of public goods, but parks are both excludable—just build a fence—and rivalrous—they can easily be filled to capacity with people. This shows there is a lot of confusion about what is and is not a public good.

There shouldn’t be any confusion about transit, however. If I sit in a transit seat, you can’t sit there too, showing that it is rivalrous. Putting gates on the entrances to transit stations and doors on the entrances to buses makes transit excludable. Thus, transit doesn’t meet either requirement for being a public good, much less both of them.

Wiener could mean something else when he uses the term “public good,” but the only definition I can think of would be that transit is a public good because it is currently provided by public agencies. But just because something happens to be supported by tax subsidies today doesn’t mean it deserves those subsidies or that they should continue forever.

Some of the respondents to Wiener’s tweet claimed that “everyone benefits” from transit, so everyone (except perhaps the transit riders themselves) should pay for it. On one hand, it is easy to show that many non-transit riders don’t benefit from transit at all. On the other hand, it’s possible to argue that “everyone benefits” from everything, whether they use it or not, but that doesn’t mean that society could work if everything were free. For example, it could be argued that everyone benefits from movies in movie theaters because the people watching the movies aren’t robbing banks or starting arson fires, but if we insisted that movies be free, we’d get a lot lower quality movies.

Wiener’s claim that charging people a significant fraction of the cost of transit would “harm low income ppl” is also specious. If some farmers produce organic food and generously sell it for less than it cost them to grow it, are they actively harming people if the prices they charge are still too high for some to pay? Clearly, the answer is “no.” The “harm” may be that some peoples’ incomes may be too low for them to afford essential goods and services, but the remedy for that is not to target one or two of those goods and services and subsidize them but to find out why those peoples’ incomes are low and to fix those problems.

The important thing that Wiener and the backers of the free-transit movement fail to understand is that prices provide critical signals to both users and producers indicating what something is worth. If prices are low, users will want more but producers will provide less. If prices are high, producers will make more but some users might be discouraged from purchases. If transit ridership is declining even though three-fourths of the cost of transit is subsidized by taxpayers, that’s a signal that transit really isn’t worth much to potential travelers.


The latest data show that ridership is declining despite increased spending on transit. The reason for that is that transit is not capable of competing against other modes of travel. Rather than trying to figure out how to “save transit,” people who care about mobility, low-income people, and the environment should worry about making sure that what replaces transit does so economically, safely, and with environmental sensitivity.

The 2018 National Transit Database includes more than 30 spreadsheets that are sometimes difficult to understand. The Antiplanner has collapsed the most useful data into a single spreadsheet showing ridership, passenger miles, service, fares, costs, and other data. The raw data for every transit agency and mode are in rows 1 through 4320 and columns A through Y.

Columns Z and AA are the Antiplanner’s calculations of energy consumption and greenhouse gas emissions that will be discussed in detail in next week’s policy brief. Columns AB through AQ are calculations of such indicators as miles per hour, average vehicle occupancies, fares per trips/passenger mile, and costs per trip, passenger mile, and vehicle revenue mile. Summaries by mode are in rows 4327 through 4368. Summaries by urban area are in rows 4380 through 4870.

One item that is questionable is column Y, the miles of rail transit. The data in the spreadsheet providing this information was entered inconsistently, so check the numbers in this column before quoting them. Fortunately, these numbers aren’t needed for any later calculations.

National Transit Ridership Down in October 2019. Cap Metro (Austin) Reported a 6.6% Gain, But Did It?

December 21st, 2019

Cost Commentary: Transit continues to loose ridership throughout the Nation as reported by Randal O’Toole in the article below, from his Antiplanner Blog. As noted, the New York region had 44.2 % of the total U.S. transit ridership in 2019, thru October. As Randal states “transit is simply irrelevant in most of the rest of the country.” This applies to all of Texas where transit ridership in its four major cities is less today than 20 years ago, after spending many tens of billions of dollars to increase transit ridership. Austin has had the most percentage decline in ridership of the for Texas cities, in the past 20 years. Real life’s situation totally contradicts Austin/Cap Metro’s joint effort, called “Project Connect,” which has developed a massive transit upgrade plan. Austin is moving toward a vote on the initial segment of the plan in November 2020. The cost is not yet announced by Austin, but the initial segment could cost up to $10 billion and the total plan could be more than double this, exceeding $20 billion.

As indicated in the next posted article: As ridership has declined over many years, transit costs have continued to increase. This, of course, places a growing burden on all Austin area citizens to highly subsidize the less than 1% of citizens who will use the planned new transit. Many trends in technology, some in current operations and several rapidly advancing, will dramatically change public transit as we know it today. The old concept of “fixed/dedicated” lanes for trains and buses are becoming more obsolete each day and will be totally obsolete before Austin/Cap Metro can complete “Project Connect.” Project Connect can only serve the interest of City Politicians, Transit Executives and Contractors which crave the large expenditure of taxpayer funds to develop ineffective systems.

Austin was one of the few cities to report a transit ridership gain in October, 2019, compared to October 2018. As reported, this 6.6% increase was primarily due to the major route restructuring, called ‘Cap ReMap’ opened in mid 2018. However, facts are not clear on this. About the time Cap Metro opened its new route structure, it implemented a new policy of allowing K-12 school students to ride free. Many students have abandoned school buses to ride free Cap Metro buses. Cap Metro’s own report indicated ridership of over 2 million of these student riders in the first year, from mid-2018 to mid-2019. The question is: How many of these free riders would have ridden Cap Metro anyway. If these new free riders were a major portion of the total K-12 ridership, there is a very small ridership increase due to the Cap ReMap program. These small gains do not compensate for the 20 years of ridership decline of about 20%, the largest, current 20-year decline of the four major Texas Cities. These route updates have been implemented in numerous cities and do achieve short term ridership gains, but do not weaken the numerous reasons that have outdated Cap Metro’s Project Connect. Cap Metro’s reporting of these increases primarily focused on a few routes where more significant gains were accomplished, instead of the total ridership. Cap Metro must communicate with transparency and total integrity, and, eliminate deceptive reporting so citizens can fully understand the broad, likely, devastating implications of this transit path before being asked to support a Proposition of many billions of dollars with a follow-on commitment of many, additional billions of dollars. This is a multi-generation commitment which will dramatically limit the ability of future generation to address key needs of their time resulting in lowering their quality of life.

Please note below that many of the largest transit ridership declines are in cities with much greater density than Austin. Los Angeles is the most dense urban area in the U.S. and is one-third more dense than the New York City region. L.A.’s transit ridership has been in a major decline for many years. If Transit is failing in the most dense urban area, why do our leaders believe transit will succeed in the Austin region which has less than One half the L.A. density and 8-9 times less urban population?

October Transit Ridership Down 1.6% in October, 2019

By The Antiplanner | December 16, 2019

The nation’s transit industry carried 1.6 percent fewer riders in October 2019 than it did in the same month in 2018, according to the latest monthly data release from the Federal Transit Administration. Ridership fell for light rail, hybrid rail, and most kinds of buses, but grew for commuter rail and heavy rail. October had the same number of work days in 2018 and 2019, so the decline in ridership can’t be blamed on a difference in work days.

Ridership declined in 31 of the nation’s 50 largest urban areas. The numbers show an increase for Dallas-Ft. Worth, but that’s due to a change in the method of counting bus riders in Dallas, so in reality ridership probably declined in 32 of the nation’s 50 largest regions. Porn.

In terms of percent, the biggest drops were in New Orleans (-17.1%), Louisville (-12.6%), Phoenix (-11.8%), Boston (-10.3%), and Virginia Beach-Norfolk (-9.9%). In actual numbers, the biggest declines were in Boston (-3.6 million riders), Chicago (-2.8 million or -5.2%), Los Angeles (-2.3 million or -4.7%), Philadelphia (-1.4 million or -4.3%), and Atlanta (-1.0 million or -7.9%). Phoenix, San Francisco Oakland, Minneapolis-St. Paul, San Juan, and Cleveland all lost more than 200,000 riders.

New York transit ridership grew by 0.1 percent, which wasn’t enough to offset the declines elsewhere. The decline in most of the rest of the country while ridership grows in New York represents a continuation of a trend since at least 1991. In that year, the New York urban area accounted for 33.5 percent of all transit riders. So far in 2019, the region has accounted for 44.2 percent of all riders. Transit advocates use the growth in New York to justify subsidies everywhere, but the truth is that transit is simply irrelevant in most of the rest of the country.

As usual, you can download my enhanced spreadsheet that provides annual totals in columns HP to IG; modal totals in rows 2163 through 2173; transit agency totals in rows 2180 through 3179; and totals for the 200 largest urban areas in rows 3190 through 3391. The spreadsheet is about 10 megabytes in size.

On the same day that it posted the October update, the Federal Transit Administration also posted the complete 2018 database, which consists of 30 spreadsheets with information on costs, fares, transit vehicles, employees, energy consumption, and more. Tomorrow’s policy brief will analyze the database in detail, while next week’s policy brief will look at energy and greenhouse gas emissions calculated from the database. The FTA also posted the latest annual time series, with operating costs, riders, and other data going back to 1991, capital costs back to 1992, and fares to 2002. The December 31 policy brief will analyze that in detail.

New, Expensive Transit and Density Do Not Boost Transit Ridership

November 30th, 2019

Does Transit Capital Spending Boost Transit Ridership?

By Randal O’Toole, The Antiplanner | November 26, 2019 | Policy brief

Does spending a lot of money on transit improvements boost transit ridership? Since 1992, Dallas-Ft. Worth and Houston have each spent about ten times as much money on transit improvements as San Antonio and Austin. Transit systems in all four urban areas carry fewer riders today than they did in 2000. While Houston ridership has grown since 2012, it is because of a low-cost restructuring of its bus system, not because of transit capital improvements (e.g., new light-rail lines).

To find out whether it is generally true that spending more on transit can generate more riders, I gathered data for more than 100 of the nation’s largest urban areas. The not-so-surprising result is that spending more on transit improvements doesn’t do much to increase ridership. Moreover, the data indicate that urban areas that spend a lot on transit capital improvements don’t grow faster and may grow considerably slower than areas that don’t. Finally, the numbers show that increasing urban densities may have once had an effect on transit ridership, but doesn’t seem to anymore.

Houston and Dallas-Ft. Worth are bigger than San Antonio and Austin and so their transit systems carry more riders, but spending billions on rail transit did not noticeably affect ridership.

Data Sources

The data I used to reach these conclusions come from the Census Bureau and Federal Transit Administration. For each decennial census, the Census Bureau identifies the size and population of each urban area, including the central city or cites in that area, suburbs, and unincorporated areas that have more than about 1,000 people per square mile or are otherwise developed. The Census Bureau also estimates how many people use transit to get to work in each of the urban areas.

Since 2005, the Census Bureau has published annual estimates of population numbers and how people get to work based on the American Community Survey, an annual survey of about 3.5 million households. I gathered population numbers for 1990 through 2010 and estimates for 2018, and transit’s share of commuting from the 1990, and 2000 censuses, and the 2010 and 2018 American Community Surveys.

In addition to measuring the growth of urban areas, the Census Bureau sometimes redefines them, merging some and splitting others. In 2000, the Miami, Fort Lauderdale, and West Palm Beach urban areas were merged into one. Seattle and Tacoma urban areas were also merged. The San Francisco urban area lost Concord, Livermore, San Rafael, and Vallejo, though San Rafael was added back in 2010. The Los Angeles urban area lost Mission Viejo, Santa Clarita, and Thousand Oaks. To keep data comparable over time, I added the numbers for urban areas that had been merged or would be separated.
In addition, three urban areas in Colorado–Boulder, Denver, and Longmont–and three in Utah–Ogden, Provo-Orem, and Salt Lake City–are each served by one transit agency. To keep data comparable, I added the numbers for these together.

The Federal Transit Administration’s National Transit Database reports capital expenditures by transit agency and mode for every year from 1992 through 2017. This “capital” spending actually combines capital improvement–that is, construction of new transit facilities–with replacement of existing transit infrastructure and equipment. Since transit agencies that use modes of transit that require lots of infrastructure will need to spend money both building and replacing that infrastructure, I didn’t attempt to separate these numbers.

The FTA numbers sometimes miss spending on early stages of capital improvements. It appears that if an urban area builds a new mode of travel, the FTA sometimes neglects to report capital expenditures until the mode becomes operational. For example, Portland spent $166 million on its commuter-rail line, which opened in 2009, yet only $5.6 million appears in the database, all of which was spent after 2009. This may be a mistake in the database, but it appears to have happened in other cities as well.

In addition, some transit agencies pay other railroads to run commuter trains on their lines. Though the other railroads may use some of this money to make capital improvements or replace existing infrastructure, these costs are counted as operating costs, not capital costs. Thus, capital costs are underreported for commuter rail.

The National Transit Database also has a file showing transit ridership and other operating data for every year from 1991 through 2017. Ridership numbers are based on each agencies’ fiscal years, and a plurality if not a majority of agencies have fiscal years that end September 30. For 2018 numbers, I used October 2017 through September 2018 numbers from the database’s monthly ridership updates. While the latest update also has October 2018 through September 2019 data, i.e., F.Y. 2019, I decided to stop with 2018 so the data would be comparable to the census data and because a few transit agencies were late in reporting the most recent ridership numbers.For 1990 ridership data, I used a spreadsheet from the 1990 National Transit Database that isn’t posted on the FTA web site. This reported transit trips, passenger miles, and other data for each transit agency and urban area. Unfortunately, Winston-Salem’s transit agency neglected to report data that year, so I used 1991 ridership for that urban area.

Naturally, I combined the FTA data for urban areas that had been merged or separated by the Census Bureau in 2000. In addition, if you download the FTA historic database, you need to carefully go through it to ensure that urban areas are assigned the correct identification number. The urban area numbers are based on their population ranking in each decennial census, and if the ranking changes, then the numbers can change. If a transit agency disappears or is absorbed by another agency in one decade, the urban area number assigned to that agency is the number in the decade it disappeared, and the FTA never goes back to fix them.

After reviewing the data, I deleted San Juan, Puerto Rico and McAllen, Texas from the dataset as I don’t have reliable 1990 numbers for the former and McAllen didn’t even have a transit system in 1990. I was also missing some 1990 data for most urban areas with fewer than than 380,000 people in 2018. This left 101 urban areas ranging from New York to Durham, North Carolina.

For those 101 urban areas, I was able to find or calculate:

1. The population of each urban area in 1990, 2000, 2010, and 2018 and the annual population growth rate in the intervening periods;
2. The land area of each urban area in those years (the land area in 2018 will be nearly the same as 2010 as the Census Bureau makes only trivial adjustments between decades);
3. The change in population density between each of those years;
4. Capital expenditures, adjusted for inflation, in each year from 1992 through 2017, which I summed into three groups: 1992 through 2000, 2001 through 2009, and 2010 through 2017;
5. Per capita transit ridership in 1990, 2000, 2010, and 2018 and the annual change in per capita ridership in the intervening periods;
6. Transit’s share of commuting in 1990, 2000, 2010, and 2018 (only available for about 60 urban areas for 1980 and 1990) and the change in transit’s share in the intervening periods.

Per Capita Capital Expenditures

Per capita spending on public transit improvements ranged from $4 a year in Columbia, South Carolina and Augusta, Georgia to $350 a year in the New York urban area. Fifteen urban areas spent more than $100 per year, all of which have extensive and/or expensive rail systems. Eleven other urban areas with some form of rail transit spent between $50 and $100 per year.
Most urban areas that spent less than $50 per person per year have no rail other than a streetcar line (whose capital cost may not be included in the National Transit Database). Exceptions were Nashville, which spent little opening a commuter-rail line, and Orlando, which has spent more than $50 per capita since it started building its commuter-rail line but less than that before it had commuter rail. Norfolk-Virginia Beach and Buffalo both have light-rail lines but also spent less than $50 per year due to the shortness of those lines.

New York transit riders come closer than most to covering operating costs with fares. But the region spent $162 billion on capital costs, mostly capital replacement rather than new construction, over 26 years, none of which was recovered by fares. This made New York the most expensive urban area in terms of per capita capital costs.

To compare capital costs with outputs such as ridership, I used Excel’s correlation function. A correlation of 1.00 is perfect; a correlation of 0.00 means no relationship. In practice, correlations of any two sets of 100 random numbers can frequently be as high as 0.10, so anything below that can also be considered random. Correlation does not prove causation, but lack of correlation indicates lack of causation.

Because New York is so different from other urban areas, with four times as many downtown jobs, much greater central city population density, and the nation’s most extensive rail transit system, much of what is true about New York has no applicability to other urban areas in the United States. I tested correlations both with and without New York, but in most cases did not find a significant difference.

Capital Spending & Ridership

First, I tested the correlation between per capita capital spending in each decade with ridership at the end of that decade (2018 in the case of the most recent partial decade). These correlations turned out to be high at around 0.7 to 0.8.

There is a strong correlation between capital spending (which includes replacement of existing capital equipment) and transit ridership.

This is an example of correlation not proving causation. Instead, the urban areas with the highest per capita ridership were those with large downtown job concentrations and relatively dense residential areas. These include Boston, Chicago, New York, Philadelphia, San Francisco, and Washington. These urban areas also happen to have legacy rail transit systems that require lots of spending on replacing existing infrastructure and equipment, which the National Transit Database considers a capital cost.

Capital Spending & Ridership Growth

What we really want to know is whether an urban area can increase transit ridership by spending more on transit, which usually means building new rail lines. To answer this question, I compared capital spending in each decade with the growth in transit ridership in that decade. Since there may be a lag period between capital spending on ridership growth, I also compared capital spending in each decade with the growth in ridership in the following decade.

The correlation between capital spending and growth in transit ridership is negligible.

In every case, the correlations were low. In only one case was the correlation greater than 0.10. Subtracting New York from the mix reduced the correlations even further.

Capital Spending & Population Growth

Rail advocates often argue that spending money on rail transit stimulates urban growth. Actually, they argue that it stimulates development along the rail lines, but the implication is that it also stimulates growth. After all, if it doesn’t stimulate growth, then all the rail line is doing is influencing the location of new development that would have taken place without the rail line. The difference is crucial because rail advocates also argue that the increased tax revenues from the new growth can help pay for the rail line, and if there is no net new growth, then there will be no net increase in tax revenues.

If there is a correlation between transit capital improvements and population growth, it is weak and quite possibly negative.

I compared per capita capital spending with population growth in each decade. In case there is a lag effect, I also compared capital spending with population growth in the following decade. The correlations were low, though not as low as between capital spending and ridership growth. However, a majority of the correlations were negative, suggesting that more capital spending slows population growth.

Capital Spending & Transit’s Share

In most urban areas, transit’s share of commuting is low and declining. But the correlation between per capita capital spending and the change in transit’s share of commuting is moderate, between 0.3 and 0.6 in most cases. I suspect this is another case where the two variables–per capita capital spending and the change in transit’s share–are not a causal relationship but are related to a third variable, in this case the growth in downtown jobs.

The correlation between transit capital spending and the growth in transit’s share of commuting seems to be mainly due to a few urban areas with large downtowns and, in most cases, legacy rail transit systems that require much expensive capital replacement.

The positive correlation seems to be mainly due to the older urban areas with high downtown job numbers and legacy rail systems: Boston, Chicago, New York, Philadelphia, San Francisco, and Washington. They are joined by Seattle, whose downtown has seen amazing job growth in the last decade. However, other urban areas that spent a lot on rail, including Dallas, Houston, and Denver, have seen transit’s share steadily fall, while transit share fell in two out of the three time periods in Baltimore, St. Louis, and San Diego. Interestingly, transit’s share of commuting grew in Phoenix and Orlando until they built rail, when it fell.

It is worth noting that several urban areas, including Boston, Chicago, Philadelphia, Portland, San Francisco, and San Jose, were able to increase transit’s share of commuting between 2010 and 2018 despite a decline in per capita transit ridership in the same period. This is likely because ride hailing services such as Uber and Lyft are taking more non-commuting trips away from transit than commute trips. This is a continuation of trends that began in the 1920s when automobiles first became affordable to a majority of American families.

Density & Per Capita Ridership

Urban planners fervently believe that they can boost transit ridership by increasing population densities. Since my data set includes populations and land areas, I was able to test this. As usual, I tested changes in density in each decade with changes in per capita ridership in that decade as well as in the next decade.

If increasing population densities ever had an effect on transit ridership, that effect is far weaker today.

The results were mixed. Increasing densities were associated with increasing per capita ridership in the 1990s and 2000s, but negatively associated with them in the 2010s. Further, increasing densities in either the 1990s or 2000s had no effect on the growth of per capita ridership in the 2000s or 2010s. These results suggest that, if it ever was true that increasing densities could increase transit ridership, it isn’t true anymore.


The amount of money transit agencies spend on capital improvements has almost no effect on ridership or regional growth. If anything, regions that spend more on transit improvements grow slower than ones that spend less. Capital improvements may have a small effect on transit’s share of commuting, though the real effect is most likely from the growth of the number of downtown jobs and the fact that regions with growing downtowns have a lot of rail transit that requires capital replacement.

This analysis also found some indications that factors that once influenced transit ridership have less of an influence or no influence today. Urban areas were once able to increase ridership by increasing their population densities, but that no longer appears to be true.

What it really comes down to is that, outside of New York and six other urban areas, transit is a negligible factor in transportation. In six of those seven urban areas (not including Seattle), most money must go to capital replacement, not expansion, which is a side effect of those regions’ reliance on expensive forms of transit.

Transit is in decline in most of the nation. Just spending more money on transit is not going to change this. Transit agencies and cities that want to increase ridership need to find more cost-effective ways of doing so than building expensive transit improvements.

For those who are interested, the data I collected to produce this policy brief is available in a downloadable spreadsheet.

Austin is Planning for an Unattainable Transit Fantasy

November 27th, 2019

Cost Commentary: This article by Randal O’Toole is the summary of a very thorough analysis of Austin’s transit plan for the 2020 election. It presents further strong evidence that the plan is not responsible because similar plans for public transit have never been achieved by any City in the U.S. and this plan is also contrary to transit trends of reducing ridership throughout the major Texas Cities and other similar cities from coast to coast.

You may review additional articles regarding transit, prior to or after this posting, for additional confirming information.

Planning for an Unattainable Fantasy

By Randal O’Toole in ‘The Antiplanner’ | November 5, 2019 | Policy brief

Austin is one of the fastest-growing cities in America, and the city of Austin and Austin’s transit agency, Capital Metro, have a plan for dealing with all of the traffic that will be generated by that growth: assume that a third of the people who now drive alone to work will switch to transit, bicycling, walking, or telecommuting by 2039. That’s right up there with planning for dinner by assuming that food will magically appear on the table the same way it does in Hogwarts.

Austin planners say that 74 percent of Austin workers drive alone to their jobs. In this, they are already behind the times, as the 2018 American Community Survey found that 75.4 percent of Austin workers drove alone (that’s for the city of Austin; the drive-alone share in the the Austin urban area was 77.0 percent). The 2018 survey was released only a month before Austin’s latest planning document, but even the 2017 survey found that 75 percent of Austin workers drove alone. You have to go back to the 2016 survey to find 74 percent drive-alones. So while Austin planners are assuming they can reduce driving alone from 74 to 50 percent, it is actually moving in the other direction.

Planners also claim that 11 percent of Austin workers carpool to work, an amount they hope to maintain through 2039. They are going to have trouble doing that as carpooling, in fact, only accounted for 8.0 percent of Austin workers in 2018.

Chart 1- Austin Fantasy
“Today” is what Austin planners say today’s commute shares are, which appears to be based on 2016 numbers. 2018 shows commute shares from the 2018 American Community Survey while 2039 shows Austin’s targets.

Planners hope to increase telecommuting from its current 8 percent (which is accurate) to 14 percent. That could be difficult as they have no policy tools that can influence telecommuting.

Planners also hope to increase walking and bicycling from their current 2 and 1 percent to 4 and 5 percent. Walking to work is almost always greater than cycling to work, so it’s difficult to see how they plan to magic cycling to be greater than walking. This is important because cycling trips are longer than walking trips and so have more of a potential impact on driving.

Finally, planners want to increase transit from 4 to 16 percent. In fact, transit carried just 3.24 percent of workers to their jobs in 2018, down from 3.62 percent in 2016. Changing from 4 to 16 percent is a an almost impossible 300 percent increase; changing from 3.24 to 16 is an even more formidable 394 percent increase. Again, reality is moving in the opposite direction from planners’ goals.

When reading this plan, my first question was, “has anyone ever been able to reduce driving alone to work from roughly 75 to 50 percent?” And the second question was, “has anyone ever been able to increase transit’s share by 300 to 390 percent?” Of course, I had similar questions about the projected quintupling of cycling and other parts of the plan, but those were the two big ones. We can answer these questions by looking at changes in commuting in various cities and urban areas between 2000 and 2018, which is approximately the amount of time in Austin’s planning period.

Austin’s Plan

Austin planners offer a list of strategies and projects that are supposed to produce major changes in transportation habits. For the most part, the strategies are similar to those used in many other cities.

For example, the carpooling strategies include Commute Solutions, a web site that allows people to find potential carpoolers; Smart Trips, another web site; Movability, a web site for employers; vanpooling; and similar programs. All of these programs assume that people are actively looking for carpooling partners. The reality is that the vast majority of carpooling is “fampooling,” that is, family members riding together to work. Carpooling has declined because family sizes have declined, so there are fewer opportunities for fampooling.

Austin’s “active transportation” (meaning walking and cycling) strategies include new sidewalks, pedestrian and bike trails, a Safe Route to School program, and similar programs. Again, communities all over the nation are using similar programs. Safe Route to Schools, for example, is a federal grant program that has given money to cities all over the country.

Austin’s transit strategies include adjusting traffic signals to give priority to transit vehicles, transit incentives including discounted transit passes and a frequent-rider program, new park-and-ride stations, and of course Project Connect, Capital Metro’s dream of high-cost, “high-capacity” transit routes. (The term “high-capacity” is in quotes because some modes that Capital Metro calls “high-capacity,” such as light rail, are in fact low-capacity transit.) Again, many other cities have used signal priority systems, discounted transit fares, and high-cost transit systems to attract riders.

To see how well these programs have worked, I looked at journey-to-work data published by the Census Bureau. From 1960 to 2000, the decennial census asked a sampling of people how they got to work. Since 2005, the Census Bureau has done an annual American Community Survey asking people, among other things, how they get to work. The most recent American Community Survey data are from 2018.

Since Austin is proposing to change people’s transportation habits by 2039, or 20 years in the future, I compared data for 2000 with 2018, which is close to 20 years of change. I looked at the data for 262 of the nation’s largest cities and 208 of the nation’s largest urban areas and posted a spreadsheet with these data so you can see what happened in your city or urban area.

I first looked to see which areas saw the biggest declines in the share of workers driving alone to work and/or the biggest increases in the share taking transit to work. Then I tried to determine what caused those changes and whether Austin’s plans are likely to produce similar results.

Reducing Drive-Alone Share

Between 2000 and 2018, the share of workers driving alone to work increased in 53 percent of major cities and 54 percent of urban areas. In most of the places where driving alone declined, it fell by less than 3 percentage points. Among central cities such as Austin, driving alone fell by more than 9 percentage points in only two: Seattle, where it fell by 12 percentage points, and San Francisco, where it fell by 10. It also declined by 19 percentage points in Seattle’s suburb of Bellevue and by 12 percentage points in San Francisco suburbs San Mateo and Mountain View.
Chart 2 - Austin fantasy
In this and the next two charts, “before” is 2016 for Austin and 2000 for the other central cities; “after” is Austin’s 2039 target and 2018 for the other central cities. The other cities shown are the ones that saw the greatest decline of driving alone between 2000 and 2018.

Among major urban areas, driving alone declined by 10 percentage points in Livermore (which is really a suburb of San Francisco but is counted as a separate urban area by the Census Bureau), 6 in Seattle, Concord (another suburb of San Francisco), Danbury, and Ann Arbor, and 5 in Flagstaff, San Francisco-Oakland, Rochester, Albany, and Boston.

The fact that driving alone fell by much more in cities such as Seattle and San Francisco than in their urban areas suggests that a sorting process is taking place, where people who prefer not to drive move to the cities while people who prefer to drive are sorted into the suburbs. The result is that city programs that attempt to reducing driving may have a negligible effect when the urban areas are considered as a whole.

In Seattle, the main factor changing commuting habits has been the tremendous growth of jobs in the downtown area, a result of Amazon, Microsoft, and other high-tech companies building new downtown high-rise office buildings. According to the Downtown Seattle Association, downtown Seattle had 244,000 jobs in 2000 and 314,000 in 2018. Today, Seattle may be the only major city in American that has more than half of its jobs downtown. Since hub-and-spoke transit systems work particularly well for downtown workers, increasing downtown jobs increases transit’s share of commuting. Downtown San Francisco also has the fourth-largest concentration of jobs in the United States.

Only about 20 percent of jobs in the city of Austin (and less than 10 percent in the Austin urban area) are located in downtown Austin. Thus, Austin would have a difficult time replicating Seattle’s results.

Austin’s plan for reducing the share of people driving alone to work involves reducing parking and road diets (converting auto lanes to bike or bus lanes). They call this “managing demand” as in “managing parking supply to reduce demand” or “manage congestion by managing demand.” But creating a shortage of something doesn’t change demand; all it does is create frustrated travelers. Many cities and regions have tried similar programs, yet no city or urban area has been able to reduce driving-alone’s share of travel by 24 to 26 percentage points in the last eighteen years, as Austin hopes to do.

Increasing Transit’s Share

Between 2000 and 2018, transit’s share of commuting grew in 43 percent of the nation’s major cities and 37 percent of the nation’s major urban areas. Among central cities, the biggest increases in transit’s share of commuting took place in Albany (4.6 percentage points), Seattle (4.0), New York (3.5) and San Francisco (3.3).
Chart 3 - Austin Fantasay
As with the previous chart, the central cities shown saw the greatest changes in transit’s share between 2000 and 2018, yet none came close to Austin’s target.

On the other hand, transit’s share declined in many cities and urban areas that have invested heavily in transit improvements. Transit’s share declined by 5.0 percentage points in Atlanta, 2.6 in Denver, 2.1 in Houston, and 1.6 in Dallas. Transit’s share also declined in each of these urban areas.

Among cities where transit’s share was about 3.2 percent in 2000, which is what Austin’s was in 2018, only one — Kalamazoo, Michigan — saw a large increase in transit’s share, and that was only 50 percent more than what it was in 2000. Phoenix and Charlotte both had 3.2 percent shares in 2000, invested heavily in light rail, and saw transit’s share nonetheless decline by 2018.

Austin’s dreams are also contradicted by recent ridership trends. Austin, Dallas-Ft. Worth, Houston, and San Antonio are some of the fastest-growing urban areas in the United States, having collectively gained 50 percent more people from 2000 to 2018. Yet the transit systems in all four urban areas have lost 4 to 22 percent of their riders. Per capita ridership has fallen by 33 to 58 percent, with Austin transit suffering the largest losses and per capita losses. Thus, Austin’s assumption that it can increase transit’s share by more than 12 percentage points, or by 394 percent, in the next 20 years appears highly unrealistic.

Maintaining Carpooling

Many cities have carpooling programs like the ones planned or used by Austin, yet since 2000, carpooling has declined in 92 percent of major American cities and 90 percent of major urban areas. As noted, most carpooling is fampooling, so unless family sizes increase, carpooling is likely to decline.

Increasing Walking and Cycling

A slight majority of cities and urban areas saw cycling’s share of commuting increase, but only a few — 5 percent of cities and 33 percent of urban areas — saw walking increase. Even where increases took place, they tended to be small.

Among central cities, cycling grew by 3.6 percentage points (which is close to Austin’s target) in Portland, 3.1 in Washington, 2.4 in New Orleans, 2.2 in San Francisco, 1.9 in Seattle, 1.6 in Denver, and 1.5 in Minneapolis. In other central cities and most suburbs it grew by less than 1.5 percentage points. Walking increased by 6.2 percentage points in the city of Boston; 2.9 in Portland, Maine; 2.3 in Washington; 1.9 in San Francisco; 1.6 in New York; and 1.0 in Seattle.
Chart 4 - Austin Fantasy
When taken together, Austin’s goals for increasing walking and cycling together are the one set of targets that appear to be attainable.

Among major urban areas, cycling increased by 1.9 percentage points in Santa Barbara, 1.4 in Portland and Anchorage, 1.0 in Madison, 0.9 in San Jose, and 0.8 in New Orleans and San Francisco-Oakland. Walking increased by 4.0 percentage points in Flagstaff; 2.8 in Santa Cruz; 1.6 in Portland Maine; 1.4 in Ft. Collins; 1.1 in Seattle; and 1.0 in Boston.

Austin’s goals for walking and cycling appear to be the only ones that seem attainable, yet even they will be difficult. Moreover, a lot of the increase in walking and cycling in various cities seems to be coming out of transit’s share, not the drive-alone share. Notice from the first chart that Austin itself has seen an increase in walking and cycling at the expense of transit, while drive-alone’s share has also increased.

Austin’s numbers are also unrealistic in that planners assume they can increase cycling’s share of travel to be greater than walking’s share. This is significant since, as previously noted, cycling trips tend to be longer than walking trips and so are more likely to have an effect on total driving. Nationwide, cycling’s share of commuting exceeds that of walking in less than 4 percent of major cities and less than 3 percent of major urban areas.

Increasing Telecommuting

Between 2000 and 2018, people working at home grew in 95 percent of major cities and 98 percent of major urban areas. Most of this growth took place without any city or regional policies promoting it, and the most growth appears to have taken place in high-tech cities and urban areas such as the cities of Berkeley and Palo Alto and the Raleigh and Austin urban areas.

While working at home may increase further, Austin’s assumption that it can increase working at home by 6 percentage points to 14 percent appears unrealistic. Only two suburbs — Highlands Ranch, Colorado and Scottsdale, Arizona — had that high a share of people working at home in 2018. The highest shares in any central cities are Portland and Atlanta, each of which are under 10 percent.

Realistically, only certain jobs are amenable to working at home. Most working-class jobs must be done either in factories or on site. Many office jobs, such as jobs in the banking, insurance, and similar sectors, require regular face-to-face contact. Most education, health care, retail, and wholesale jobs must also be in non-residential locations. The number of people working at home may grow, but the amount of that growth is beyond Austin’s control.


Planners have developed two main approaches to transportation. One is to estimate how people will travel and then provide and maintain the infrastructure to allow them to do so as efficiently and safely as possible. The other is to imagine how you wish people would travel and then provide the infrastructure assuming that to happen. The latter method is likely to lead to misallocation of capital resources, increased congestion, and increased costs to travelers.

Austin’s plan is firmly based on this second approach. The city’s targets of reducing driving alone by a third, maintaining carpooling at an already too-high number, and increasing transit by 394 percent are completely unrealistic. No American city has achieved similar results in the past two decades and none are likely to come close in the next two decades.

As discussed in a previous policy brief, one of the biggest factors in commute patterns is the number of downtown jobs. Many people think population density is a major factor, but among urban areas the correlation between urban population densities and the share of commuters who use transit is only about 0.4. Meanwhile, the correlation between the number of downtown jobs and the share of commuters who use transit is nearly 0.9.

So it is not surprising that the cities that have seen the biggest reductions in driving alone and biggest increases in transit commuting — Seattle and San Francisco — have hundreds of thousands more downtown jobs than Austin. Even they haven’t seen changes as great as Austin is fantasizing for 2039. In short, Austin needs to go back to the drawing board and develop a plan that is based on how people will actually travel in 2039 and not one based on how planners wish they would travel.

Best 13 Page Summary of Rail Transit Ever Written: From Greatness to Outdated & Obsolete in 200 Years

October 8th, 2019

COST Commentary: The paper below by Randal O’Toole is an excellent summary of U.S. rail transit, beginning as an early, effective transit mode in Boston in 1838 (19th Century) and ending today as rail transit has become outdated and obsolete as a transit mode in the 21st Century. This is particularly important as Austin and Cap Metro consider the future role of of transit in Austin by their ‘Project Connect’ collaboration. The outdated rail and new technologies that exists now and that are moving rapidly into transportation/mobility considerations will dramatically change the future of human mobility. This all seems to have been given little attention/consideration by the ‘Project Connect’ plans which are being developed between the City and Capital Metro.

I recommend careful reading of the paper below which is only 13 typed pages, excluding references. It encapsulates the strong basis for eliminating rail and dedicated transit lanes as alternatives to address Austin’s future mobility challenges. It can be all summarized in the question: Why should all taxpayers highly subsidize the use of public transit serving less than 1% of the passenger miles traveled, which includes an increasing percentage of higher paid riders. If one truly analyzes and considers this, along with the horrible, resulting tax burden which will limit opportunities for future generations, there is no way the eventual proposed 2020 transit bond will be approved by voters in Austin.

Total U.S. Public transit ridership has declined 6.9% in the past three years (2016-2018) and is about equal to ridership in 2005, 13 years ago. [Please see a later posted article to view 2019 ridership: U.S. and Texas 2019 Transit Ridership Declines Predict Transit’s Dim Future (Before the COVID-19 VIRUS CRASH). The national transit decline in this article is somewhat similar to the transit ridership in the 4 major Texas Cities (see post following the later 2019 U.S. Transit Decline post): All four Texas cities had less transit ridership than in 1999, nineteen years ago (1999-1918). This is a profound statement made by almost all citizens throughout the Nation: They have chosen not to ride public transit because it does not meet their mobility needs. Limiting trips to public transit’s access creates a degraded quality-of-life and limits citizen’s access to the region’s opportunities.

Please consider this posting and the ‘Cost Commentary’ content in more recent, previous postings listed below:

Austin’s Major 2020 Transit Bond and “2018 Census Data Show Transit in Decline”
Facts and Data Evaluation of Austin’s Proposed Transit System Indicates Continuing Death Spiral
Austin & Cap Metro’s New Transit Plans Will Result in Huge Wasteful Cost Impacts On All Taxpayers and Increasing Congestion, Reducing Quality-Of-Life For All Citizens
Transit declines as people rapidly increase car ownership
Why Cities Are Abandoning Light Rail Transit In Major Public Transit Decline
Mass Transit is collapsing everywhere
New Austin & Cap Metro Transit Plan will waste billions of taxpayer dollars and totally fail.

There are many prior postings on the COST web site: www.costaustin.org

We will continue to expand content regarding this critical transportation issue. I got extra energy from the this 13 page summary. The number 13 is my “lucky” number, adopted at age 11 in Burnet, Texas after winning a cake-walk. It was also my uniform number in high school sports; Football Captain, All City, All Conference and third-place team in Washington state where I lived at the time.
Six Problems with Rail Transit and Six Myths of Rail Transit

by Randal O’Toole, January 4, 2017

Rail transit has long had a presence in American cities. The first commuter trains served the suburbs of Boston in 1838. The first successful electric streetcar opened in Montgomery, Alabama, in 1886. Chicago opened the first electric elevated train in 1895, while New York opened the first electric heavy-rail subway line in 1904. Electric-powered commuter trains date to 1906.

Through the mid-20th century, private transit companies served the vast majority of American cities. After World War II, these companies operated profitable, if declining, businesses in the face of rising automobile ownership. A handicap was that transit companies were considered public utilities and highly regulated. They had to seek government permission for route changes, fare increases, and other service changes. By 1950, buses were recognized as a less expensive, more flexible, and safer transit mode than streetcars or most other types of rail transit.

The beginning of the end for private transit came in 1964 with the Urban Mass Transit Act. The act promised federal capital grants to public agencies that took over private transit companies. Within a decade, the private transit industry was virtually wiped out, replaced almost completely by tax-subsidized public agencies.

Today, city governments that are frustrated with automobiles and congestion are turning to the 19th century technology of rail transit for relief. But pumping subsidies into rail transit is based on a nostalgic view of the past, and it is not economically or environmentally sound. It will not solve America’s congestion woes.

The Department of Transportation’s Federal Transit Administration (FTA) has an annual budget of $12 billion, most which is spent on subsidies to state and local governments.(1) Through these subsidies and related regulations, federal policymakers play a major role in shaping urban transportation choices.

Transit funding is not a proper function of the federal government, and it distorts state and local decision-making. Federal funding encourages state and local governments to pursue high-cost and less-efficient transportation solutions — in particular, rail transit. Outside of a few hyper-dense cities in the world, rail transit is a luxury for the few paid for by everyone. Commuter trains and subways may be necessary to keep Manhattan moving, but that does not mean that the rest of the nation should subsidize them. Outside of New York City, rail transit makes little economic sense.

The federal government should end its transit subsidies, and American cities should focus on finding economically sound and consumer-driven approaches to easing congestion. Policymakers at all levels should work to revive private transit options for cities, and they should allow consumers to make transportation choices in a neutral and competitive market environment.

History of Urban Transit

Early urban transit ventures were privately financed. However, because many of these ventures used public rights-of-way, companies often had to obtain franchises from city councils. But other than rights-of-way, transit companies received no subsidies or other public support through the end of the 19th century.

The first popular public transit was the omnibus, a horse-drawn wagon with seats for passengers.(2) New York City saw its first omnibus in 1827. Then came the horse-drawn railcar in the 1830s, which would be used in more than 500 American cities. In the same decade, the first steam-powered commuter trains started carrying suburban workers into Boston. The first successful elevated transit line was built in New York in 1871.

Electric streetcars arrived on the scene in the late 1880s. They were much more efficient than previous forms of travel and would be adopted by more than 1,000 American cities and towns. Other innovations included interurban rail lines, electric-powered elevated rail lines, and subway lines, which were first installed in New York City in 1904.

Until this point, urban transit was privately financed and unsubsidized. All this innovation to improve convenience and reduce costs came from private entrepreneurs. But soon after the turn of the century, governments began to intrude. Government-owned streetcar lines were opened in Bismarck, North Dakota, and Monroe, Louisiana. And New York City took over the previously private Staten Island Ferry.

Private transit companies faced several financial difficulties. Many streetcar lines were built by real-estate developers to attract people to their housing projects. A developer would subdivide land on the city fringe, build a streetcar line from the development to downtown, and sell lots and homes. The profits on the real-estate development paid for the capital cost of the streetcar line. Transit fares covered only the operating cost. That worked fine for a few decades; but when the time came to replace the streetcars, rails, and other equipment, the companies often lacked the capital.

One way to raise funds was to increase fares. But governments regulated fares, and proposals to raise fares were regularly rejected by city councils and public utility commissions. This left many transit companies with aging streetcar fleets in precarious financial positions.

Progressive-era politicians saw the public takeover of transit companies as a solution. San Francisco was the first major city to operate its own streetcars starting in 1912. New York City started operating and acquiring subway lines in 1932. In 1938, Chicago obtained the first federal grants to support construction of a publicly owned rail line.

In cities where transit remained private, electric power companies often worked to consolidate streetcar lines under one owner. That gave rise to concerns about monopoly. In 1935 Congress ordered power companies to divest their transit operations. Since transit was already struggling due to the rise of the automobile and the Depression, this mandate put many companies on the brink of bankruptcy.

One solution was to convert streetcar systems to bus systems, which did not require as much infrastructure. Of the hundreds of American cities served by streetcars in 1910, at least 230 either went out of business or converted to buses by the end of 1929. Another 300 converted during the 1930s, and then 100 more in the 1940s.(3)

Fifty American cities still had streetcars in 1949. But by 1967, only Boston, Cleveland, El Paso, New Orleans, Newark, Philadelphia, Pittsburgh, and San Francisco still had streetcars, while New York and Chicago were the only other cities to still have other forms of rail transit. The conversions from rail to buses were made for efficiency reasons, not monopolistic reasons, as often claimed.(4)

By the early 1960s, all the rail transit systems except one had been taken over by public agencies, but the vast majority of bus systems were still private. That changed quickly when Congress started to make capital grants available to public agencies — but not private companies — that operated or acquired transit systems. Within a decade, all but a handful of transit systems were taken over by tax-subsidized public agencies.

Congress did not pass the Urban Mass Transit Act of 1964 in order to provide mobility to low-income families who could not afford cars. Rather, Congress was reacting to proposals by various railroads to discontinue interstate commuter trains serving Boston, Chicago, New York, and Philadelphia.(5) At the time, these four urban areas plus San Francisco had the only commuter trains in America. Urban leaders argued that the commuter trains were essential to maintaining jobs in downtown areas.

The Urban Mass Transit Act was designed to provide federal support for interstate commuter trains. But politics quickly broadened that mission to providing federal support to mass transit in every state and metropolitan area. Since then, about $160 billion has been spent on federal rail subsidies, and the result has been a monument to the folly of federal intervention into a properly local and private activity.

Six Problems with Rail Transit

1. Cost

The most important thing to understand about rail transit is that it is very expensive. The Government Accountability Office has shown, for example, that buses running on arterial streets can provide service as fast and frequently as light rail at a lower operating cost and for about two percent of the capital cost.(6) Outside of a few very dense places such as Manhattan, Tokyo, and Hong Kong, there is little that trains can do that buses cannot do faster, better, more flexibly, and at a lower cost.

The typical light-rail project today costs about $160 million per route mile, although one project in Seattle cost well over $600 million per mile. Heavy rail typically costs about twice as much as light rail: A recent extension of the Washington Metrorail system cost almost $300 million per mile, for example. Commuter-rail typically costs $5 to $10 million per mile.

Freeways typically cost much less than rail.(7) The Fort Bend Tollway Authority built a four-lane freeway on the outskirts of Houston, complete with interchanges and over- and underpasses, for $2.4 million per lane mile. The Colorado Department of Transportation widened Interstate 25 through the heart of Denver, which required numerous overpasses, at a cost of $19 million per lane mile. Urban freeways cost more than suburban ones, but counting urban and suburban areas together, the average cost is less than $10 million per lane mile.

Rail advocates claim that rail lines can move as many people as several freeway lanes, but to make that claim they use a double standard: comparing full railcars with the average occupancy of commuter automobiles. In fact, like automobiles, the average transit vehicle carries far fewer people than its capacity. Most rail cars and buses carried an average of less than one-sixth of their capacity in 2014.(8) Even sport utility vehicles do better than that.

When calculated using full automobiles and full trains, a single freeway lane can move more people per hour than most light-rail lines, while four freeway lanes can move more people than most subway or elevated lines. When calculated using automobiles and trains with average occupancy rates, a single freeway lane moves several times as many people as a light-rail line, and two freeway lanes move more people than the busiest subway lines in the United States.(9)

2. Cost Overruns

Rail transit projects are notorious for cost overruns. According to the Federal Transit Administration (FTA), federally subsidized rail projects built between 1980 and 2015 had overruns averaging 50 percent.(10) Moreover, there has been no tendency for estimates to improve, as overruns since 2010 have been greater than in previous decades. By comparison, a study by Danish planning professor Bent Flyvbjerg found that overruns for North American highway projects averaged just 8 percent.(11)

Here are some examples of over-budget rail projects:

• In 1998, Phoenix proposed building a 13-mile light-rail line for $390 million, or $30 million per mile.(12) Completed in 2008, the final cost of the 19.6-mile line was $1.41 billion, or $72 million per mile.(13)
• In 2000, Charlotte, North Carolina, estimated that a light-rail line would cost $331 million.(14) The final cost turned out to be $427 million.(15)
• In 2004, the first 12-mile leg of the Dulles rail project in Virginia was projected to cost $1.5 billion.(16) The final cost when completed was $2.9 billion.(17)
• In 2004, Denver’s Regional Transit District persuaded voters to support a $4.7 billion rail transit system. The later estimate was that the system will cost 68 percent more at $7.9 billion.(18)

While cost projections are not an exact science, Flyvbjerg believes that persistent underestimates of rail construction costs result from “strategic misrepresentation, that is, lying.”(19) Planners deliberately lowball estimates in order to gain project approval. Once the project is approved, they develop more realistic estimates, add expensive bells and whistles, and respond to political pressures to lengthen the proposed project.

Many of the original estimates for transit projects are made by consulting firms that expect to receive contracts for engineering and construction later on. As such, these firms have an incentive to make projections that will gain approval, both by underestimating the costs and overestimating the benefits.

In one example of strategic misrepresentation, Parsons Brinkerhoff (now known as PB) compared a proposal to bring rail transit to Madison, Wisconsin, with improvements to bus service. To its dismay, the company found that bus improvements alone attracted more riders than bus improvements combined with rail transit. Later, PB admitted that it crippled the bus alternative, making it appear that rail transit was needed to boost transit ridership.(20) When the government agency that hired PB presented the results to the public, it never mentioned the bus alternative at all, making it appear that rail transit was the only way to attract people to transit.(21)

Despite the record of cost overruns, transit agencies often claim that they finish their projects “on budget.” For example, after adjusting for inflation, Denver’s Southwest light-rail line cost 28 percent more than its original estimate.(22) The city’s Southeast light-rail line went 59 percent over its original estimate.(23) Yet Denver’s Regional Transit District insisted that both projects were “on budget,” based on the deceptive notion that project costs matched the final budgeted amounts — not the earlier and lower estimates.(24)

3. Rehabilitation

Transit agencies generally go heavy into debt to fund rail projects by issuing long-term bonds. But the costs do not end when the bonds are paid off: rail lines must be completely replaced, rebuilt, or rehabilitated about every 30 years. Except for the right-of-way, everything — cars, tracks, roadbed, stations, electrical facilities — must be replaced or upgraded.

The first Washington, D.C., Metrorail line opened in 1976. In 2002, just 26 years later, the Washington Metropolitan Area Transportation Authority estimated that it needed $12.2 billion — roughly the cost of constructing the original system — to rehabilitate the system.(25) It did not find this money, so the system has suffered frequent breakdowns and service delays.(26) In 2009, an accident due to maintenance failures killed nine people. In 2015, smoke in a subway tunnel caused by maintenance failures killed one more. Although the Metro delayed many trains to undertake maintenance work in 2016, the agency now says that it needs $25 billion to completely restore the system over the next 10 years.(27)

The heavy rail transit systems in Chicago, San Francisco, Boston, and New York also face fiscal crises from high rehabilitation costs. Rehabilitating light-rail lines is also expensive. The first modern light-rail lines, including those in Buffalo, Portland, Sacramento, and San Diego, are now about three decades old and will need major work. Nationwide, the FTA says that urban transit systems have a deferred maintenance backlog of $86 billion.(28)

Rehabilitation costs do not increase the capacities of transit systems, and thus should be considered maintenance costs. But the FTA allows transit agencies to count rehabilitation as a capital cost. The significance is that when rail advocates claim, as they often do, that rail lines cost less to operate and maintain than buses, they are ignoring these long-term maintenance costs.

Many of the same agencies that cannot afford to maintain their existing systems are nonetheless embarking on expansions. New York’s MTA is spending $16.8 billion building an eight-mile Second Avenue Subway. Washington’s Metro is spending $6.8 billion building a rail extension to Dulles airport. San Francisco’s BART is planning to spend more than $6 billion for a line to San Jose. Boston is spending $2.3 billion extending light rail to Medford, Massachusetts. Chicago is extending several of its commuter-rail lines.

The disconnect between planning and budgetary reality has become typical of rail-transit agencies. Rail transit fares do not come close to paying the operating costs of systems, much less the costs of line rehabilitation or new rail construction. If transit agencies cannot afford to maintain their existing lines, it makes no sense for them to build new ones.

4. Ridership

To justify spending billions of taxpayer dollars on rail, advocates often claim that middle-class automobile owners will not ride a bus. In fact, transit ridership is more sensitive to frequency and speed than to whether the vehicles run on rubber tires or steel wheels. “When quantifiable service characteristics such as travel time and cost are equal,” say researchers, “there is no evident preference for rail travel over bus.”(29)

The real problem is what happens to overall transit system ridership when agencies face soaring costs and must choose between keeping the trains or buses running. Having built rail, many agencies feel compelled to cut more efficient bus service, which in turn causes overall transit ridership to stagnate or drop.

In the late 1970s, Atlanta began building a heavy-rail system. By 1985 it had 25 route miles and ridership had grown to 155 million trips per year. Since then, the Atlanta urban area population has doubled, rail miles have also doubled, yet ridership has fallen somewhat. In 2014, it was just 137 million. While rail ridership has grown, that growth has been at the expense of bus ridership.(30)

In the early 1980s, Los Angeles maintained low bus fares, and between 1982 and 1985 ridership grew by more than a third. Then it built rail transit, which suffered huge cost overruns. In response, the Los Angeles Metropolitan Transportation Authority (MTA) raised bus fares and cut service, leading to a 17 percent drop in bus ridership by 1995. The NAACP sued, arguing that the agency was cutting service to minority neighborhoods in order to finance rail lines to white middle-class neighborhoods. The court ordered the MTA to restore bus service for 10 years, forcing it to curtail its rail plans. Bus ridership rebounded to above its 1985, pre-rail level, while rail ridership stagnated. As soon as the court order expired, MTA cut bus service and started building new rail lines, and ridership has fallen again.

When St. Louis opened its first light-rail line in 1993, it was hailed as a great success because system ridership, which had shrunk by nearly 40 percent in the previous decade, started growing again. But when St. Louis opened a second line in 2001, doubling the length of the rail system, rail ridership remained flat and bus ridership declined. By 2007, total system ridership was no greater than it had been in 1998.

As of 2003, about half of the urban areas with rail transit had ridership declines compared to the mid-1980s. The remaining areas enjoyed increases in ridership, but at rates slower than increases in driving and, in most cases, slower than population growth.(31)

Many areas with bus-only transit systems did far better. From 1983 to 2013, ridership on bus transit systems in Austin, Charlotte, Houston, Las Vegas, Louisville, and Raleigh-Durham all grew as fast, or faster, than automobile driving. The 2010 census revealed that the number of commuters taking transit to work in urban areas with rail transit declined overall, while the number in bus-only urban areas increased.(32)

5. Land Use

Buses are flexible and can be easily rerouted when travel patterns change. Rail lines take years to build and are much harder to re-rout. While bus systems can respond to rider demand, rail systems must generate their own demand. This leads transit agencies to promote intrusive land-use regulations that mandate or subsidize high-density developments close to rail stations, so-called transit-oriented development.

Instead of customizing transit systems to serve American cities as they exist today, transit agencies want to rebuild the cities to suit the kind of transit service the agencies want to provide. There are many problems with this aspiration, not least of which is the fact that transit carries less than 5 percent of travel in all American urban areas except New York, and it seems absurd to design cities around such little-used transportation systems.

Another problem is that land-use policies have, at best, marginal effects on people’s transportation choices. Transportation technologies do influence land uses: cities built before 1890, when most people walked for most of their travel, were much denser than cities built between 1890 and 1930, when streetcars were popular. Those cities, in turn, were denser than cities built after World War II, when autos were dominant. But this process is not reversible: building cities to pre-auto or pre-streetcar designs will not lead people to significantly reduce their driving.

According to Robert Cervero, an enthusiast of transit-oriented development, such developments do increase transit ridership, but this is “due to residential self-selection — i.e., a life-style preference for transit-oriented living.”(33) In other words, people who prefer transit over driving choose to live in such areas, but that does not mean that the urban design has influenced their travel habits.

Also, the market for transit-oriented development is very limited: surveys find that four out of five Americans would prefer to live in single-family, detached houses than townhouses, condominiums, or apartments.(34) Once the one-in-five demand for multifamily is met, cities like Portland and Denver have had to resort to huge subsidies to entice developers to build more high-density developments, and the evidence indicates that the people who live in those developments drive just as much as people elsewhere.

6. Capacity

Rail transit is often described as “high-capacity transit.” Yet its capacity is not that high in reality. Heavy-rail lines, such as subways, are the highest capacity transit. If people jam in tightly, the “crush capacity” of a heavy-rail car is about 180 people. An eight-car heavy-rail train can therefore carry about 1,440 people. If rail systems can manage to move one train every minute — and only a few American rail lines can support such traffic — the line can move 86,400 people per hour. This is how rail advocates are able to claim that a rail line can move as many people as several freeway lanes. The capacities of other forms of rail transit are much lower.

However, few places really need such high-capacity transit. People look at the freeway during rush hour and think, “If only we had a rail line, all those people could ride.” But the reality is that all the people in those cars have different origins and destinations. They may all be in this freeway corridor now, but a rail line will only serve a tiny number of them because most do not both live and work near a rail station.

What cities really need is flexible-capacity transit: transit capable of moving large numbers of people where needed while also economically moving small numbers where demand is lower. Rail transit, with its very high fixed costs, cannot qualify. Rail-transit agencies try to provide flexible capacities by having buses feed into rail stations, but transit agencies lose riders every time they ask people to change vehicles.

Buses can provide flexible-capacity transit. They can run on city streets in the suburbs where demand is low, move to high-occupancy vehicle (HOV) or high-occupancy toll (HOT) lanes where demand is higher, and if demand warrants, use exclusive bus lanes. While the capacity of exclusive bus lanes may not be quite as high as heavy rail, buses, unlike the trains, can easily start from different origins and diverge to different destinations, allowing more people to make their transit journeys without transfers.

This means that light rail and commuter rail really make no sense anywhere. Cities that need low-, moderate-, or high-capacity transit can meet those needs with buses. Only areas that need ultra-high-capacity transit might need to consider heavy rail, and outside of New York City no place in the United States truly needs such high capacities.

Six Myths of Rail Transit

Rail transit is not cost-efficient, and it generally fails to attract significantly more riders than improvements in bus service. Yet rail advocates have many other arguments for why cities should build rail transit systems. Here are six popular, but erroneous, reasons that are often given for building rail lines — six myths of rail transit.

1. Economic Development

Rail advocates say that rail transit stimulates economic development. They often point to a streetcar line built in downtown Portland, Oregon, that supposedly stimulated $1.5 billion worth of new development. In fact, much of that development was subsidized with hundreds of millions of dollars of tax-increment financing and other developer subsidies. It is not likely that the streetcar alone would have generated the developments without the subsidies.

Does any rail transit stimulate new development? In the mid-1990s, the FTA asked this question of Robert Cervero, who is a strong proponent of transit-oriented developments, and Samuel Seskin, who works for Parsons Brinkerhoff, the consulting firm that has had a hand in many major rail transit projects over the decades. Despite their bias in favor of rail transit, the authors found that “urban rail transit investments rarely ‘create’ new growth,” and at best, it may “redistribute growth that would have taken place without the investment.”(35)

In other words, rail transit is at best a zero-sum game benefiting some property owners at the expense of others. The property owners who mainly benefit are those who own land downtown because most rail systems are hub-and-spoke systems that focus on downtown. That explains some of the political calculus behind rail: downtown property owners benefit a lot, so they lobby for rail. Other property owners each lose a little, so they have little incentive to lobby against it.

2. People Will Not Ride Buses

Rail advocates claim that buses will not entice middle-class automobile drivers out of their cars, but that rail can. In reality, improvements in bus service can attract as many new transit riders as rail construction — and at a far lower cost.

A survey found that the median income of Washington, D.C., Metrorail riders was 47 percent higher than Metrobus riders, and that 98 percent of rail riders owned an automobile compared with only 80 percent of bus riders.(36) For those who want to get middle-class commuters out of their cars, this is a Pyrrhic victory, considering that the fossil-fuel-generated electricity used to power the Metrorail system emits more carbon dioxide per passenger mile than the average SUV. Meanwhile bus riders have faced repeated service cuts so that the transit agency can keep the trains running.(37)

In the late 1990s, the San Francisco Bay Area Metropolitan Transportation Commission — which spends two-thirds of the region’s transportation funds on transit systems which carry only four percent of the region’s passenger travel — was considering where to invest.(38) One possible project was an extension of BART trains to San Jose, a line so expensive that taxpayers would spend nearly $100 per trip. Another project was improved bus service in Richmond, a city with a heavy population of low-income minorities. Bus improvements, planners estimated, would cost just 75 cents per trip.(39) The commission decided to support the rail line and not the bus improvements. In the commission’s judgment, spending $100 to attract one high-tech worker onto a train was more important than spending the $100 to attract 133 low-income workers onto buses.

A factor behind such decisions is that middle-income individuals are more likely to vote than lower-income individuals. Getting voters to support the idea of transit as a daily necessity — rather than something used only by those who cannot afford cars — is part of the rail agenda. Thus, the goal is not so much to get a large number of people out of their cars, but to get enough people out of their cars to create some middle-class transit dependency and generate support for transit funding.

3. Rails Avoid Congestion

Rail advocates say that an advantage of rail is that it is not slowed by congestion, as are cars and buses. But that begs the question why a small number of people deserve a heavily subsidized transit system allowing them to avoid congestion, while everyone else who is paying the subsidies for rail has to sit in traffic.

More importantly, the premise that we need rail because buses cannot avoid congestion is incorrect. Most major urban freeway systems have high-occupancy vehicle (HOV) lanes or high-occupancy toll (HOT) lanes. HOT lane tolls are adjusted to ensure that the lanes do not become congested. An urban area with a network of HOV or HOT lanes as a part of every freeway could allow buses to avoid congestion throughout the region.

Such HOV or HOT lanes cost less than rail lines, and the congestion relief they provide benefits everyone. Bus riders benefit because buses traveling at freeway speeds go faster than light rail (which averages about 20 miles per hour) or heavy-rail trains (which generally average no more than 40 miles per hour). The car drivers using HOV or HOT lanes benefit because they also avoid congestion. People who do not use the lanes also benefit: HOT lanes built parallel to a freeway in southern California draw a third of the traffic off the freeway.

In general, rail transit costs more to build and operate and does less to relieve congestion than new HOV or HOT lanes with bus rapid transit. With a well-designed highway system, buses can not only avoid congestion as well as trains, they can do so at a far lower cost and be put into service almost immediately instead of after a decade of planning and construction.

4. Political Equity

Rail advocates sometimes say, “Let’s build one line and see how it works.” But then the construction of one line often creates political momentum for additional lines. Every urban community wants a share of federal rail dollars, no matter how poorly rail might actually serve their community. The federal grant-in-aid system creates major distortions in urban transportation decisions — cities are steered into less efficient transit choices.

Consider the transportation improvements in a corridor between downtown Denver and Denver International Airport in the 1990s. A study showed that rail transit would cost more to build and operate and less to relieve congestion than new freeway or HOV lanes.(40) Nonetheless, planners picked rail transit.

Suburbs in the Denver area agreed to support the Regional Transit District’s multibillion-dollar rail plan if all the rail lines were built at once. Suburban officials realized that the plan would probably suffer major cost overruns (as proved to be the case) and that whatever line was last on the schedule would probably never be built. In response to the cost overruns, the transit agency proposed to build just four of the six planned lines — a proposal that was naturally rejected by officials from the suburbs whose lines would be left out.(41)

Rail advocates have applied the equity argument to neighborhoods as well as cities. San Francisco is spending $1.6 billion on a 1.7-mile transit tunnel as a part of a project to extend light rail to the Bayshore neighborhood.(42) According to the FTA, Bayshore transit ridership is already very high, but a main reason for building this line is “to achieve a goal of equity with other communities currently served by rail.”(43) By the FTA’s reasoning, if one neighborhood is served by an unaffordable transit line, equity demands that every other neighborhood get equally unaffordable service.

5. It Works in Europe

Americans visit Europe and ride the London subway, the Paris metro, or trams in Italy, and come home wishing the United States had more such transit systems. However, the United States is not Europe: our population densities are lower, and our incomes are higher, so fewer people are likely to ride transit even in dense areas.

Also note that rail transit in Europe is not necessarily efficient. As of 2007, at least 150 European urban areas had some form of rail transit, compared with 30 in the United States.(44) Europe spends several times more money subsidizing those rail lines than the United States spends on transit.(45)
European rail lines may be convenient for tourists, but the average European rarely uses them. In 2004, the average American traveled 87 miles on rail transit; the average European just 101 miles.(46) This difference hardly commends Europe as an example of successful rail transit. Moreover, the share of European travel on rail transit declined from 1.4 percent in 1980 to 1.1 percent in 2000. Meanwhile, the share of European travel using automobiles increased from 76.4 to 78.3 percent.(47)

6. Protecting the Environment

Many people take it for granted that rail transit is good for the environment. The reality is that rail transit uses about as much energy and emits about as much pollution per passenger mile as automobiles. Most transit systems are actually brown compared with the latest cars. The Washington Metro rail system, for example, uses more energy per passenger mile than the average car, and generating electricity to power the system emits more greenhouse gases than the average sports utility vehicle.(48) To the extent that rail transit might save any energy at all, the financial cost of getting a few people to ride trains and drive less is huge.

People who sincerely want to save energy and reduce pollution should focus on making automobiles more environmentally friendly, not on trying to get people to ride rail transit. Economist Charles Lave called this the “Law of Large Proportions,” meaning “the biggest components matter most.”(49) In other words, Americans travel 60 times as many passenger miles in urban areas by automobile as by transit, so a small investment in reducing the environmental effects of automobiles will go much further than a large investment in transit.

Private Transit Solutions

Reforms should remove federal subsidies and related regulations from the transit equation. Federal intervention creates numerous perverse incentives for state and local governments, including:
• Cities are encouraged to build inefficient rail lines because more than half of all federal funds are dedicated to rail transit.
• Transit agencies are encouraged to find the most expensive transit solutions because federal rail construction funds are an open bucket — first-come, first-served.
• Innovative solutions are by-passed, and high costs are guaranteed because of the requirement that transit agencies obtain the approval of their unions to be eligible for federal grants.
• Local transit agencies have strong incentives to claim success with their projects no matter how badly they fail because of the requirement that agencies must refund federal grants if projects are cancelled.
• Federal rules impose a transit planning process that is biased in favor of higher-cost transit projects, and the process allows agencies to systematically low-ball cost estimates and overstate potential ridership.
• Federal subsidies have been mainly directed to capital costs of local transit, not operating costs. That has led to agencies to favor expensive rail over less expensive buses and favor larger buses when smaller ones would do the job.
• Federal regulations distort the flow of funding to the most efficient solutions, such as rules that tie the distribution of transportation funds to air quality planning.

These factors and others have promoted less efficient transportation solutions than would have been favored without federal intervention. I have discussed these problems elsewhere at length.(50)

If we removed the federal government from the picture, state and local governments would rethink their urban transit financing. One issue is that the average American transit agency gets only a third of its operating funds, and none of its capital funds, from fares. This means that transit officials are less interested in increasing transit ridership than they are in persuading politicians and taxpayers to give them more money. Increased ridership is actually a burden on transit systems: even though transit vehicles are, on average, only one-sixth full, they tend to be fullest during rush hour, when new riders are most likely to use transit.

Today’s government-owned rail transit systems make no financial or transportation sense. They only work because few people use them and everyone else subsidizes them. Because rail transit costs at least four times as much, per passenger mile, as driving, it means that if everyone rode today’s rail systems instead of automobiles, cities would go bankrupt trying to keep the systems running.

Yet urban transit does not have to be expensive, and it does not even have to be subsidized. The United States has several completely unsubsidized transit systems that work very well. One is the Atlantic City Jitney Association, whose members own their own buses and operate routes scheduled by the association.(51) Rides are $2.25 each and cover all major attractions in the city. Unlike most government-owned transit systems, the jitneys operate 24 hours a day, 7 days a week, and receive no subsidies from any government agency. Such jitney service is illegal in most other American cities because it would compete against government monopoly transit agencies.

Another unsubsidized transit system is the públicos, or public cars, of San Juan, Puerto Rico. Públicos are independently owned and operated buses that typically seat 12 to 15 passengers. At least six companies operate públicos and they provide both urban and intercity service. Fares vary depending on the length of the ride, but in 2014 they averaged less than $1.50. Although públicos compete against a public bus system and a recently built heavy-rail line (whose cost rose from a projected $1.0 billion to $2.2 billion), the públicos carry more riders each year than the public buses and trains combined.(52)

A third unsubsidized transit system is the NY Waterway ferries, which connect multiple points in New Jersey and Manhattan. Founded in 1986 by Arthur Imperatore, NY Waterway offers a service that none of the government transit agencies in the region thought to provide.(53) Passengers arriving in New York City can take NY Waterway buses to and from various points in Manhattan at no extra charge. Although the company accepted a federal subsidy in 2001 to temporarily replace subway service between New Jersey and the World Trade Center after 9/11, it is otherwise funded entirely out of fares.(54)

Public transit agencies encourage people to believe that if their subsidies disappeared, people without cars would lack any mobility. In fact, private transit would spring up to take the place of government transit, and it would be superior to government transit. It would be more likely to offer door-to-door service, operate during more hours of the day, and provide more limited or nonstop services to popular destinations.

American taxpayers can no longer afford costly and inefficient government transit systems, particularly rail transit systems. Federal subsidies ought to be
eliminated and local governments should open transit to private and entrepreneurial solutions to relieving congestion.
Randal O’Toole (@antiplanner) is director of the Transportation Policy Center at the Independence Institute (@i2idotorg), a free market think tank in Denver, and a senior fellow with the Cato Institute (@CatoInstitute) in Washington, D.C. He is author of the book, “Romance of the Rails: Why the Passenger Trains We Love Are Not the Transportation We Need” and many hundreds of previous books, Policy Papers and articles.
1 Budget of the U.S. Government, Fiscal Year 2017, Analytical Perspectives (Washington: Government Printing Office, 2016), Table 29-1.
2 A timeline of early transit history is in 2015 Public Transportation Fact Book(Washington: American Public Transportation Association, 2015), p. 51.
3 “List of Streetcar Systems in the United States,” Wikipedia, tinyurl.com/7za8zb. Accessed December 2016.
4 Randal O’Toole, Gridlock (Washington: Cato Institute, 2009), p. 137.
5 George M. Smerk, The Federal Role in Urban Mass Transportation (Bloomington, IN: Indiana University, 1991), pp. 60–61.
6 Government Accountability Office, “Mass Transit: Bus Transit Shows Promise,” GAO-01-984, September 2001, p. 4.
7 Examples from Randal O’Toole, Gridlock (Washington: Cato Institute, 2009), p. 60.
8 Calculated by dividing passenger miles by vehicle revenue miles from data in 2014 National Transit Database (Washington: Federal Transit Administration, 2016).
9 Passenger miles per track mile calculated from data in 2014 National Transit Database (Washington: Federal Transit Administration, 2016). Passenger miles per freeway lane mile calculated from data in “Highway Statistics 2014,” U.S. Department of Transportation, www.fhwa.dot.gov/policyinformation/statistics/2014, Table HM-72. Freeway occupancies assumed to average 1.67 people per vehicle as estimated in Summary of Travel Trends: 2009 National Household Travel Survey (Washington: U.S. Department of Transportation, June 2011), Table 16.
10 For a summary of the data and links to the FTA reports, see Randal O’Toole, “Rail Transit Cost Overruns,” The Antiplanner, January 19, 2015. And see Randal O’Toole and Michelangelo Landgrave, “Rails and Reauthorization,” Cato Institute Policy Analysis no. 772, April 21, 2015.
11 Bent Flyvbjerg, Mette Skamris Holm, and Søren Buhl, “Underestimating Costs in Public Works Projects: Error or Lie?” Journal of the American Planning Association 68, no. 3 (2002): 285.
12 “Central Phoenix/East Valley Light Rail” (Washington: Federal Transit Administration, 1998), tinyurl.com/837vda.
13 “Central Phoenix/East Valley Light Rail” (Washington: Federal Transit Administration, 2004), p. 1, tinyurl.com/9vsa2o.
14 “South Corridor LRT” (Washington: Federal Transit Administration, 2000), tinyurl.com/7rcxwq.
15 “South Corridor LRT” (Washington: Federal Transit Administration, 2005), tinyurl.com/8x6dwd.
16 Federal Transit Administration, “Baseline Report on Major Project Monitoring of the Dulles Corridor Metrorail Project,” July 27, 2007, Table 1.
17 Paul Duggan and Lori Aratani, “At Last, the Silver Line Is Ready; Metro Says Passenger Service Will Start July 26,” Washington Post, June 23, 2014.
18 “Summary of Changes to FasTracks Program: Attachment 1” (Denver, CO: Regional Transportation District, 2008), p. 2, tinyurl.com/4kodgc.
19 Bent Flyvbjerg, Mette Skamris Holm, and Søren Buhl, “Underestimating Costs in Public Works Projects: Error or Lie?” Journal of the American Planning Association 68, no. 3 (2002).
20 Parsons Brinckerhoff, Transportation Alternatives Analysis for the Dane County/Greater Madison Metropolitan Area (Madison, WI: Transport 2020, 2002), pp. 7-6, 10-2, and 10-22.
21 Transport 2020, Transport 2020 Oversight Advisory Committee (OAC) Summary Report(Madison, WI: Transport 2020, 2002), p. 21.
22 Booz Allen Hamilton, Managing Capital Costs of Major Federally Funded Public Transportation Projects (Washington: Transportation Research Board, 2005), p. 55.
23 Final Report: Southeast Corridor Major Investment Study (Denver, CO: Colorado Department of Transportation, 1997), p. 4-24.
24 “A Message from RTD General Manager, Cal Marsella” (Denver, CO: Regional Transit District, 2005).
25 “America’s Transit System Stands at the Precipice of a Fiscal and Service Crisis” (Washington: Washington Metropolitan Area Transit Authority, 2004), p. 1.
26 Lyndsey Layton and Jo Becker, “Efforts to Repair Aging System Compound Metro’s Problems,” Washington Post, June 5, 2005. And see Lena H. Sun and Joe Holley, “Aging Equipment Blamed in Metro Incidents,” Washington Post, August 28, 2007.
27 Max Smith, “Metro Needs $25 Billion for Basics — And That’s if Everything’s Been Properly Taken Care Of,” WTOP News, November 29, 2016.
28 Statement of Carolyn Flowers, Acting Administrator, Federal Transit Administration, before the U.S. House of Representatives, Committee on Transportation and Infrastructure, Subcommittee on Highways and Transit, May 24, 2016.
29 Moshe Ben-Akiva and Takayuki Morikawa, “Comparing Ridership Attraction of Rail and Bus,” Transport Policy Journal 9, no. 2 (2002).
30 All ridership numbers are from the National Transit Database, Federal Transit Administration, various years, www.transit.dot.gov/ntd.
31 Randal O’Toole, “Rail Disasters 2005,” American Dream Coalition, June 2005, p. 6.
32 American Community Survey 2010 (Washington: Census Bureau, 2013), Table B08301. And see 2000 Census of Population and Housing (Washington: Census Bureau, 2002), Table QT-P23.
33 Robert Cervero, “Transit Oriented Development’s Ridership Bonus: A Product of Self-Selection and Public Policies,” University of California Transportation Center, Berkeley, 2006, p. 1.
34 National Association of Realtors, “The 2011 Community Preference Survey,” March 2011.
35 Robert Cervero and Samuel Seskin, An Evaluation of the Relationship between Transit and Urban Form (Washington: Transportation Research Board, 1995), p. 3.
36 Kytja Weir, “Survey: Metrorail Users More Affluent, Better Educated,” Washington Examiner, May 17, 2009.
37 For example, see Kytja Weir, “Area Bus Riders Face Service Cuts,” Washington Examiner, March 31, 2009.
38 Metropolitan Transportation Commission, Final Transportation 2030 Plan (Oakland, CA: MTC, 2005), p. 35.
39 Bob Egelko, “Inequity in Funding Discriminates against AC Transit Riders, Plaintiffs Claim in Suit,” San Francisco Chronicle, April 20, 2005.
40 Kimley-Horn & Associates, East Corridor Major Investment Study Final Report (Denver, CO: Denver Regional Council of Governments, 1997), pp. 37–39.
41 FasTracks Annual Program Evaluation Summary: 2008 (Denver, CO: Regional Transportation District, 2008), p. 3.
42 “Third Avenue Light Rail Phase 2 — Central Subway, San Francisco,” Federal Transit Administration, 2012, pp. 1–2.
43 “Bayshore Corridor,” Federal Transit Administration, 1996.
44 Randal O’Toole, Gridlock (Washington: Cato Institute, 2010), p. 77.
45 Randal O’Toole, Gridlock (Washington: Cato Institute, 2010), p. 77.
46 Randal O’Toole, Gridlock (Washington: Cato Institute, 2010), p. 77.
47 Key Facts and Figures about the European Union (Brussels: European Union, 2004), p. 52.
48 Randal O’Toole, “Does Rail Transit Save Energy or Reduce Greenhouse Gas Emissions?” Cato Institute Policy Analysis no. 615, April 14, 2008, pp. 4, 8.
49 Randal O’Toole, Gridlock (Washington: Cato Institute, 2010), p. 78.
50 Randal O’Toole, “A Desire Named Streetcar,” Cato Institute Policy Analysis no. 559, January 5, 2006.
51 www.jitneyac.com.
52 Randal O’Toole, Gridlock (Washington: Cato Institute, 2010), p. 80.
53 N. R. Kleinfield, “Trucker Turned Builder: Arthur E. Imperatore; Creating Shangri-La on the Hudson,” New York Times, January 4, 1987.
54 Sascha Brodsky, “Many Routes to Ferry King’s Success,” Downtown Express, July 17, 2002.

©2007 Coalition On Sustainable Transportation