Impact of the COVID-19 Pandemic on Subway Ridership in New York City

MTA Train

Impact of the COVID-19 Pandemic on Subway Ridership in New York City

The COVID-19 pandemic has had a profound and disparate impact on subway ridership. The emergence of the virus in New York City in March and April 2020 corresponded with a steep drop in subway usage. Citywide, April 2020 ridership was just 8.3 percent of what it was in April 2019. Since then, New Yorkers have slowly returned to the subway system.

While many New Yorkers and businesses turned to telecommuting to protect themselves from the virus, others have not had that luxury. As a result, ridership as a percentage of pre-COVID levels has remained higher in lower-income neighborhoods than in wealthy ones, though the disparity became markedly less pronounced beginning in the summer of 2022.

The accompanying interactive map demonstrates COVID-19’s disparate impact on the subway system across the City’s U.S. Census-defined neighborhoods. Select a neighborhood on the map to compare ridership figures to a number of socioeconomic indicators and to the City as a whole.

For the first two years of the pandemic, subway turnstile data published by the Metropolitan Transportation Authority (MTA) showed a correlation between median household income and subway ridership. Though not nearly as pronounced as in the earlier phases of the pandemic, neighborhoods with lower median household incomes still tend to have higher ridership as a share of 2019 levels when compared to wealthier neighborhoods.

In higher-income neighborhoods, residents are more likely to be employed in areas that can more easily adapt to remote work models, such as financial activities and business services. Ridership in these neighborhoods—and particularly in the City’s Central Business District—is also more dependent upon business, tourism and workers commuting into the office on a regular basis. Throughout most of the pandemic, as commuter levels remained low, stations in Lower Manhattan and the City’s Central Business District saw more profound ridership loss and were slower to recover ridership than stations in the outer boroughs. In neighborhoods where residents have been more likely to continue using the subway throughout the pandemic, common areas of employment are the health care and social assistance sector and the leisure and hospitality sector.

After a sharp drop driven by a COVID-19 resurgence at the end of 2021, ridership recovered to pre-Omicron levels quickly before slowing again in April and May 2022. Citywide ridership ticked up again in June 2022 and reached its highest post-pandemic level yet in September. This improvement is partly due to significant improvement in some of the Manhattan neighborhoods that have been the slowest to recover from the pandemic. 

Despite the improvement, half of Manhattan neighborhoods still trailed citywide ridership recovery in September 2022. Many of these neighborhoods, including Chelsea, Clinton, and the Midtown Business District, and Battery Park, Greenwich Village and Soho, historically account for significant portions of subway ridership. 

Historically, much of this slow return to Manhattan ridership has been driven by major station hubs, likely due to shifts in commuting to work. Times Square-42nd Street (70.4 percent), 34th Street-Penn Station (68.9 percent), Fulton Street (67 percent), and Grand Central-42nd Street (64.5 percent) all remained significantly below the citywide ridership recovery of 75.5 percent in September 2022. Until ridership returns more strongly to these major hubs or new ridership patterns are established, the MTA will continue to have difficulty reaching recovery projections for ridership and farebox revenue, like many systems across the country. The MTA revised its ridership projections downward in its July Financial Plan.

Other important stations in Lower Manhattan, such as Canal Street, Chambers Street, and Wall Street, also remain significantly below citywide numbers. However, there are signs that ridership return at these stations may be growing more rapidly in the fall of 2022 than it is in other parts of the City. Whether this is indicative of a longer-term trend, or whether the growth rate will stabilize as employers establish new in-person working standards remains to be seen. 

In contrast, nearly three dozen smaller stations, mostly located in the outer boroughs, posted ridership levels higher than they had in 2019 (though certain stations were closed or partially closed in September 2019, causing them to show greater improvement than they otherwise might). In July, the neighborhoods of Bensonhurst and Bath Beach, in southern Brooklyn, posted higher ridership levels than it had pre-pandemic, as did Greenpoint and Williamsburg in August and September, indicating that there are riders who are willing to return, and continue to rely on, the system as the pandemic recovery continues.

The accompanying tables identify selected transit hubs as well as the stations where ridership has been the most and least impacted by the pandemic.


Neighborhoods are identified by PUMAs (Public Use Microdata Areas).

Subway stations on PUMA borders were assigned to only one neighborhood.

Turnstile data is published by the MTA. OSC removed certain inconsistent data points to enhance the utility of the data. Four census neighborhoods do not contain any subway stations and thus were not included in the analysis:

  • Bayside, Douglaston and Littleneck;
  • Port Richmond, Stapleton and Mariner’s Harbor;
  • New Springville and South Beach; and
  • Tottenville, Great Kills and Annadale

There are five census neighborhoods in New York City where fewer than one-quarter of residents report using the subway system as their primary means of commute:

  • Queens Village, Cambria Heights and Rosedale;
  • Far Rockaway, Breezy Point and Broad Channel;
  • Co-Op City, Pelham Bay and Schuylerville;
  • Flushing, Murray Hill and Whitestone; and
  • Briarwood, Fresh Meadows and Hillcrest.

In three of these neighborhoods, the geographical area includes just one subway station. Data for these five neighborhoods were retained, though they are not considered to weigh heavily on the analysis.

Suggestions? Email [email protected]