Residents used WMATA relatively equally to commute from 2009 to 2015, but almost every income group dropped transit in 2016 during SafeTrack. Ridership got even worse in 2017. That’s according to a DC Commute Times analysis of U.S. Census Bureau data.
Commuters Within DC Use Transit Mostly Equally
According to census data on commuting methods, at least a quarter of commuters in every income bracket use transit to get around town. In 2017, 37.7 percent of commuters between $25,000 and $35,000 used transit. $15,000 - $25,000 was slightly behind at 37.3 percent. Those were the two groups most likely to use transit, while $10,000 to $15,000 was the least transit-using group at 26.9 percent.
The median income in the DC region is about $77,000 for an individual or $110,000 for a family. That means the average DC commuter fits into the $75,000 or more census category. In 2017, 31.9 percent of people in the top income category transit. That fits right into the middle of all the income groups. DC’s wealthiest workers don’t stand out regarding commuting habits.
Going back to 2009 with annual data shows consistently stable ridership between income brackets in each year. In 2017, the median commute share was 31.8 percent with a standard deviation of 3.7 percent. The standard deviations between commuting shares go down over time, suggesting a convergence of mode choice preferences as things like ridesharing have popped up in DC. These data show poor or rich people aren’t that much likely to use or not use transit than any other income bracket.
WMATA Lost Riders of All Incomes During SafeTrack
Looking at the commuting data across years within each income bracket shows almost every income group drop transit. The drops occur between the 2015 and 2016 years of data, which points to the service disruptions that started in 2016 as an external shock to people’s habits. In 2017, the data show these people have not gone back to transit despite the end of SafeTrack. People continued to leave transit in 2017. Not a single income bracket increased transit usage in 2017.
DC commuters making less than $10,000 dropped to 32.8 percent transit in 2016 compared to 40.6% the year before. Even more of them left transit in 2017, falling to 30.2 percent transit use. Professionals between $65,000 and $75,000 dropped to 31.3 percent in 2016 from 40 percent in 2015. These folks left transit even more in 2017, down to 30.3 percent.
The only income bracket that stuck with transit in 2016 from 2015 were folks earning $25,000 to $50,000. They both used transit more in 2016 by a few percentage points. However, both groups dropped transit in 2017. Transit use by the $35,000 to $50,000 cohort cratered to 31.7 percent from 41.6 percent. This trend in middle incomes suggests there was a middle group of commuters who stuck with transit through 2016 but abandoned it as more alternatives arrived in the form of dockless bikes and scooters.
Comments About the Data
Data for these comments come from the American Community Survey, conducted continuously by the Census Bureau. Specifically, the commuting data come from Table B08119 “Means of Transportation to Work by Workers’ Earnings in the Past 12 Months.” The Census breaks transit out from bicycle and taxi. “Transit” for our purposes here is rail and bus, excluding bikeshare, Uber/Lyft, scooters, magic carpets, etc.
The income data are in dollars of the year they were collected. 2016 figures and before are not adjusted for inflation. I hope to modify the data in the future, but you can ballpark the effect of inflation by assuming more people rise into the higher income categories as we get closer to the present day.
My analysis includes only commuters within the District of Columbia—I don’t capture Virginia, Maryland and West Virginia commuters within the DC metropolitan area. A metro area is a more representative bucket of people to survey. For instance, you may find Kansans and Missourians who work in Kansas City. However, ACS doesn’t have 2017 data broken down by metro area.
Commuting data comes annually broken down by congressional district, ZIP code and census tract. In a full analysis, I would aggregate to the DC area by adding up these smaller geographic areas. But that takes a lot of math, and I’ve spent all afternoon lining up column headers in Excel between CSV files that don’t line up between years.
A critical limitation of the data is that it’s just two variables—income and share who commute—in an environment with many other factors that could explain commuting habits. Higher-income people may be able to afford to live closer to transit, which probably makes a commuter more likely to use public transit. This commuting by income chart does not meet the standard of full analyses that hold all other variables constant using statistics. With the smaller geographic units like zip code, you could control for income variation and other factors.
I hope to work with these data further and conduct a more thorough analysis.