The story of commuting to work during the Covid-19 pandemic is one of disruption. As we socially distanced, shifted to remote work, and followed public health guidance, many people found new ways to work – most often from home. With the release of 1-year estimates on how people commuted to work in 2021 from the American Community Survey, we can begin to quantify the shift to working from home and the impact of that shift on biking, walking, and transit and how they compare to other types of commutes.

2020 was a year of maximum data disruption: the annual American Community Survey was not able to produce 1-year estimates for biking, walking, and transit commuting. The data we did get for 2020 — 5-year estimates that reflected data collected between 2016 and 2020 — gave some indication of the disruptive shift, but now we have more specific data – and it shows dramatic changes.

Please check out our data on to see the changes in the following tables:

There are some big takeaways when looking at data from before the pandemic and during the pandemic. Based on 2019 vs. 2021 1-year estimates, more than 18 million people changed to working from home, shifting from all other modes except “taxi cab, motorcycle, and other means.” The shift to working from home has decreased the number of people commuting by other modes, including about 14.5 million teleworkers who no longer drive to work alone. Biking, walking, and transit saw similar dramatic shifts to fewer people using them to get to work in most states and cities.

While the overall data shows decreases in bike commuting, there are several bright spots including New York City – which showed a continued increase in bike commuters even as other major cities like Washington, D.C., saw steep decreases. 

It would be premature to declare the “bike boom” over in light of this new data, which reflects the widespread disruption felt across the country. We must also keep in mind that this data about the commute to work represents less than 10% of all biking and walking trips.

The disruption of the Covid-19 pandemic brought on a mode shift more dramatic than advocates for reducing car commuting could have dreamed about. The big question now is whether people will return to commuting to work and if they will change how they get to work when they do. The 14.5 million fewer people driving alone to work in 2021 compared to the pre-pandemic year of 2019 present a huge opportunity to realize mode change if they do not continue to work from home. Changes to commute behaviors are often most likely when a person is establishing a new routine. People who shifted to working from home will establish new commute routines if they return to the office, whether due to a new office location, new job, or new schedule that involves fewer commutes to the office each week. These changes to routine are an opportunity to help people choose a way to get to work other than driving alone and make that their new routine.

The commute to work is often one of the longest trips that people take in a typical day. According to data from the 2017 National Household Travel Survey, the average trip to work is about 12 miles, compared to 7-8 miles for errands, visiting family, and going to church (Table 5b). Trips to and from work make up about 20% of trips (Table 10b) but due to their regular pattern of peak travel each day, they have an outsized role in transportation planning, engineering, and policy. 

For biking trips, commute to work data is the only federal data we get. It is valuable data because it comes out annually (usually), and is available for many different geographies allowing us to compare cities, states, and even census tracts with the same data nationwide. However, we know that commutes to work do not represent 90% of bike trips, limiting its insight. There is a profound need for more data about biking trips for other purposes and biking travel generally. 

While many cities, and some states, have invested in bicycle counts and other forms of bicycle monitoring we still lack a good annual national picture of how often people bike, why they bike, and how bike trips change over time. Relying only on the commute to work leaves out young people who do not work, older adults who are retired, and others who do not work, in addition to the people who do work but use bikes for other purposes. Having only one limited annual federal data source for understanding biking puts our understanding on shaky ground and limits our ability to craft and evaluate policy changes. Covid-19 disruptions put this data fragility into relief and hopefully will spur renewed federal interest in better data, because it is desperately needed.

If the shift to working from home is durable, then there is an opportunity to focus more on planning and engineering for short everyday trips that are shorter distances and easier to do by biking or walking. Overall, 39.4% of all trips are 2.5 miles or less and those are trips that people may choose to do by bike if we build safe and connected places to bike for short trips.