Transit ridership dropped by 2.9 percent in 2017 despite a 0.7 percent increase in transit service (as measured in vehicle revenue miles). This isn’t big news to Antiplanner readers, but it’s a little more official with the release, earlier this week, of the 2017 National Transit Database. While we’ve previously looked at calendar year or July through June ridership numbers, the database uses the fiscal years of the individual transit agencies, which may range anywhere from July 2016 through June 2017 to January 2017 through December 2017, so the numbers won’t be exactly the same.
The full database also includes fares, costs, energy consumption, and other information not previously available for 2017. For example, transit used an average of 3,376 BTUs per passenger mile in 2017, a 2.3 percent increase from 2016. Greenhouse gas emissions per passenger mile also increased by about 1.0 percent. These increases, of course, are due to the increased vehicle miles combined with a 2.6 percent fall in passenger miles.
Transit’s 3,376 BTUs per passenger mile is just about tied with light trucks (pick ups, SUVs, full-sized vans), but well behind the average car. In 2015, cars used only about 3,030 BTUs per passenger mile and may have been even more energy efficient in 2017.
The 2017 database is distributed in the form of some two dozen spreadsheets, many of which are difficult to read without the context of information in other spreadsheets. As usual, the Antiplanner has summarized the most important data into one spreadsheet. The first 23 columns and 3,849 rows of the spreadsheet are raw data showing such things as trips, fares, costs, and vehicles used by each agency and mode of transit.
Columns X and Y are my calculations of energy consumption (in BTUs) and greenhouse gas emissions (in grams of carbon dioxide). The BTU calculations are based on standard factors used to convert gallons of gasoline and Diesel, kilowatt-hours of electricity, and so forth into a common energy measure. The conversion factors for greenhouse gas emissions from non-electrical energy sources are also standard.
For electricity, I relied on the Energy Information Agency’s calculations of CO2 emissions from electricity produced in each state. This is a bit problematic as electricity crosses state lines, so the numbers are probably underestimated for states (such as Oregon and Washington) that produce a lot of hydropower but still import coal-powered electricity.
Columns Z through AM in the spreadsheet calculate such things as average miles per hour (vehicle revenue miles divided by vehicle revenue hours), average vehicle occupancy (passenger miles divided by vehicle revenue miles), average number of seats per vehicle, and fares and costs per trip and per passenger mile.
For the last several years, the Federal Transit Administration has included rural transit in the database, which makes it much larger even though rural transit is insignificant. Previously, fewer than 1,900 rows were sufficient to cover all transit agencies and modes. Now it is more than twice that many. But the rural transit only adds 1.3 percent more trips to the total, and rural agencies aren’t required to estimate passenger miles, energy consumption, and certain other data.
Counting the urban agencies alone, transit ridership fell below 10 billion for the first time since about 2005. Although ridership was well below 10 billion trips per year in the mid-1990s, urban populations have grown, so transit trips per urban resident have fallen to the lowest in recorded history, 37.8. They will be even lower in 2018. Totals of all transit and urban transit only are in rows 3825 and 3826.
Rows 3830 through 3852 are totals/averages by mode, including such obscure modes as inclined planes and the Alaska Railroad, which isn’t really transit but is probably included in the database because it has received funding from the FTA. Rows 3851 and 3852 are totals for the various kinds of buses and the most important kinds of rail (commuter, heavy, light, streetcar, hybrid).
Rows 3855 through 3873 repeat the mode totals but count only those modes for which energy data are supplied. This is to make accurate estimates of energy consumption and CO2 emissions per passenger mile.
Finally, rows 3880 through 4368 include totals/averages by urban area. These urban areas are listed in an order assigned by the FTA after the 2010 census, which I think means in descending order by the population of those urban areas in 2010. Some population rankings have changed since then. The last row, 4369, includes all rural transit agencies (or at least all agencies for which an urban area isn’t assigned in the database).
The Pelvic and Urogenital Diaphragms As we become more appreciative of the abdominal area we become more aware of the options that are available to you. cipla viagra But if you are the kind of person who worries best viagra for women about medication side effects and unintended consequences, you need to consider those that come with these drugs. Pitch black colour shilajit is mainly used for doing penile sildenafil pills massaging. purchase levitra opacc.cv It causes problems such as depression, erectile dysfunction, minimised ejaculate volume, etc. Based on 2016 transit data and 2015 energy data for automobiles, I’ve been saying that transit is more energy-efficient than the average car only in New York, San Francisco-Oakland, Portland, and Honolulu. Although the Department of Energy still has not issued auto energy data for 2016, much less 2017, it is likely that this statement will remain true for 2017. Of course, the reason why only four urban areas can be below the average is that one of them is New York, where 40 percent of all transit ridership takes place.
I have to concede that transit in San Juan, Puerto Rico is also more energy efficient than the average car, but while Puerto Rico is definitely part of the United States, it has a very different economy with a much lower rate of auto ownership. To be accurate, I need to say that transit is more energy efficient than the average car in only four urban areas within the fifty states.
Transit in just two more urban areas beat the light truck average of about 3,380 BTUs per passenger mile: Chicago and Atlanta (and Chicago only beat it by 4 percent). If you want to save energy, drive a plug-in hybrid (especially if you live in a region that gets a lot of its electricity from non-fossil-fuel sources).
In another disturbing trend, the average number of passengers on board buses declined from 9.6 in 2016 (and 11.1 in 2014) to 9.0 in 2017. The average number aboard a rail car declined from 29.5 in 2014 to 28.9 in 2016 to 28.3 in 2017. While this is predictable given declining ridership and increased service, it suggests transit agencies are out of touch with their customers.
Of the top 50 urban areas (including San Juan), ridership fell in all but four: Phoenix, Seattle, Kansas City, and New Orleans. Jarrett Walker, the transit expert who appeared on a couple of panels with the Antiplanner earlier this week, makes the point that to increase transit ridership you have to increase transit service. But increased service cannot guarantee increased ridership and even when it does there are usually diminishing returns, which means that a large (and expensive) increase in service may be needed to get a small increase in ridership.
Transit agencies in Tampa-St. Petersburg increased vehicle miles of service by 10.0 percent in 2017, but ridership still dropped by 7.5 percent. Orlando increased service by 5.2 percent to get a 4.8 percent drop in ridership. Nashville increased service 4.5 percent and lost 2.3 percent of its riders. These are just a few of many such examples.
Seattle achieved its 2.2 percent increase in ridership with a 2.6 percent increase in service, while Phoenix got a 6.9 percent ridership increase from a 10.6 percent increase in service. New Orleans increased service 4.9 percent to get a measly 0.2 percent increase in ridership. Kansas City increased service by 16.8 percent to get a 10.2 percent increase in ridership — but all of the increase came from free rides on the city’s new streetcar. Not counting the streetcar, service increased 9.8 percent but ridership fell 2.8 percent.
Walker has helped transit agencies redesign their transit systems to provide better service to neighborhoods where ridership is highest while trimming service where ridership is lowest. While he achieved some success in Houston, it didn’t show up in 2017, as ridership there fell by 2.1 percent. While transit agencies everywhere should consider Walker’s ideas, I suspect they will produce a one-time increase in ridership after which ridership will return to a downward trajectory.
(By the way, the forums featuring the Antiplanner with Jarrett Walker are now downloadable as videos or podcasts. You can choose between Monday’s one-hour forum or Tuesday’s two-hour forum, which differ mainly in the length of the question-and-answer period, but thelonger forum also allowed the two speakers to respond to each other’s presentations.)
There’s not much hope for the transit industry if the only way for it to increase or maintain ridership is through massive and continuing increases in service. The 2017 data merely provide more evidence for what the Antiplanner already suspected: at least in most parts of the country, transit is dying.
Along with the 2017 database, the Federal Transit Administration updated its historic time series, which includes ridership and operating cost data back to 1991, capital costs back to 1992, and fares back to 2002. The most important spreadsheets in this series are table TS2.1, which has operating costs, fares, trips, passenger miles, and vehicle revenue miles by transit agency and mode, and table TS3.1, which has capital spending by agency and mode.
Warning: The FTA updates urban area numbers with every decennial census, but it hasn’t corrected the numbers for transit agencies that disappeared before the 2010 census. So if you try to add data by urban areas, you will need to see if any former transit agencies in your urban area are associated with the wrong urban area number.
