This Friday is Bike to Work Day—a chance for people to hang up their car keys, get off the bus, and give cycling a try. To kick off our Bike Week line-up here at CityLab, I thought it might be fun and useful to take a more objective look at the state of biking in our cities and urban areas.
(Disclosure: I’m an avid cyclist who owns both serious road bikes and Dutch-style urban commuter bikes, including a refurbished Raleigh Tourist DL-1.)
Bicycles in cities are one of the very hottest of hot-button urban issues—right up there with gentrification itself, which isn’t unrelated, as we’ll see. On the one side are the urban biking optimists, who extol the benefits of bikes in cities and argue that two-wheeled transport is the source of many good things, from less pollution and energy use to more walkable neighborhoods and happier, healthier people. On the other are the urban biking pessimists who see bikes, bike lanes and bike share systems as symbols and promulgators of gentrification, housing unaffordability, and deepening urban divides.
For all the talk about how bikes are taking over cities and even getting the upper hand in the fabled War on Cars, the reality is that bikes are far and away the least common way people get to work. We seem to be arguing about bikes a lot more than we ride them. (Detailed data on more general bike use is skimpy, so we used commuting data compiled by the Census through their American Community Survey from 2011 to 2015 is the best we have—though the League of American Bicyclists has some data for city cycling from 2014).
Here’s a reality check: Even as the number of regular bike commuters has skyrocketed by more than 60 percent from 2000 to 2013, nationally, less than one percent of commuters (0.6 percent) bike to work at least once a week.
By comparison, 86 percent of commuters drive to work. Think of it this way: There are roughly 800,000 people pedaling to work across the United States, lined up against some 120 million Americans behind the wheel. The War on Cars, if it exists, is very asymmetric.
While workers in cities are more likely to cycle to work than those elsewhere, the reality is that just one percent of Americans who live in one of the 50 largest cities in the U.S. cycle to work. Young people cycle more than older ones. But still, just one percent of Americans between the ages of 16 and 24 ride their bikes to work. U.S. cycling also has a stubbornly persistent gender divide—with men more likely to ride their bikes to work than women (0.8 percent of men vs. 0.3 percent of women).
Of course, the number and share of people who cycle to work varies widely by city or metro area.
The table below shows the top 25 large metros (those with more one million people) by the share of commuters who bike to work and the location quotient for cycling commuters. A location quotient or LQ makes it easy to compare a metro to the national average. For example, an LQ of 1 means a metro is essentially at the national average, an LQ of 2 means it is twice as high, and LQ of 3 means it is 3 times as high, and so on.
|Rank||Metro Area||Share of bike commuters||LQ|
|2||San Francisco-Oakland-Hayward, CA||1.97%||3.29|
|3||San Jose-Sunnyvale-Santa Clara, CA||1.85%||3.09|
|5||New Orleans-Metairie, LA||1.16%||1.93|
|7||Minneapolis-St. Paul-Bloomington, MN-WI||0.96%||1.60|
|9||Los Angeles-Long Beach-Anaheim, CA||0.91%||1.52|
|12||Austin-Round Rock, TX||0.86%||1.44|
|13||Salt Lake City, UT||0.85%||1.41|
|14||Tampa-St. Petersburg-Clearwater, FL||0.79%||1.32|
|16||San Diego-Carlsbad, CA||0.67%||1.11|
|19||Miami-Fort Lauderdale-West Palm Beach, FL||0.62%||1.04|
|21||New York-Newark-Jersey City, NY-NJ-PA||0.58%||0.97|
|22||Milwaukee-Waukesha-West Allis, WI||0.56%||0.93|
|24||Grand Rapids-Wyoming, MI||0.50%||0.83|
Portland, Oregon leads with a location quotient (LQ) score of 3.92—meaning that the clusters of bike commuters are nearly 4 times more concentrated than in the other metro areas. There are three additional metro areas with an LQ score over 3, all in California (San Francisco, San Jose, and Sacramento). There are no metro areas with an LQ score between 2 and 3. And only 14 large metros—New Orleans, Seattle, Minneapolis, Boston, Los Angeles, Denver, Phoenix, Austin, Salt Lake City, Tampa, Washington, D.C., San Diego, Chicago, Philadelphia, and Miami—have cycling LQs greater than 1.
The largest share of bicycle commuters are in smaller and medium-sized metros, mainly college towns. Corvallis, Oregon, home to Oregon State University, has an whopping LQ of 14. Missoula, Montana (University of Montana) and the Colorado college towns of Boulder (University of Colorado) and Fort Collins (Colorado State University) have LQs of more than 7 times the national average share of bike commuters. California’s big college hubs of Santa Barbara and Santa Cruz follow closely, with LQs higher than 6. Other metros with high cycling LQs include Gainesville, Florida; Bloomington, Indiana; San Luis-Obispo, California; Madison, Wisconsin; Champaign-Urbana, Illinois; Iowa City and Ames, Iowa; and Ann Arbor, Michigan.
But what can say about how cycling relates to cities and metros area? Should cycling really be closely linked with gentrification and inequality?
To get at this, my colleague Charlotta Mellander ran a basic correlation analysis between the share of workers who cycle to work and key economic and social and demographic characteristics of metros. On one hand we have size and density, talent and education levels, and levels of technology and innovation; on the other, housing affordability, economic inequality, and segregation.
In the urban optimist corner, the share of commuters who cycle to work is positively and significantly associated with good things—higher density, more transit use, and less sprawl. Cycling is closely associated with density (.42) and use of public transit (.28), and it’s (strongly) negatively associated with the share of commuters who drive to work alone (-.59), an indicator of the deleterious costs of sprawl.
Cycling to work is also associated with higher incomes (.32) and higher wages (.37). That said, bike commuting is even more closely associated with the share of adults who have a college degree (.54), innovators (.40, measured by the share of residents who hold patents), and the share of workers in professional, knowledge, and creative occupations (.33). This would seem to suggest though that potentially cycling reflects the same sort of pressures that lead to gentrification. But here the data suggest that is really not the case.
When it comes to the perspective of the urban cycling pessimists, the evidence is more mixed. On the one hand, bike commuting is associated with higher housing costs (.44) and worse housing affordability, as measured by the housing-costs-to-income ratio (.43). On the flip side, bike commuting is not statistically associated with income inequality or income segregation, and it’s only weakly associated with other forms of economic segregation by education or occupation. Interestingly enough, cycling is negatively associated with the segregation of the rich—a most pernicious form of segregation, in which the wealthy wall themselves off from other groups. In other words, the affluent are less segregated in metros where a greater share of commuters cycle to work.
Cyclists may make convenient scapegoats for those who want to promulgate urban divisions, but there is little evidence that bike commuting is associated with urban inequality or economic segregation. In fact, cycling is one rare activity that may actually bridge some of our new urban divides. The two groups most likely to bike to work are the most and least educated: 0.9 percent of adults with a graduate or professional degree, compared to 0.7 percent of adults who did not complete high school.
The least advantaged groups are the most likely to use bikes to get to work—1.5 percent workers in households making $10,000 or less, 1.1 percent in households pulling in between $10,000 and $15,000, and 1 percent of those making between $15,000 and $25,000. Compare those numbers to the 0.5 percent of people hauling in over $150,000 who bike to work (a larger proportion than their middle-income peers). Of course, the latter group uses bicycles by choice—and have the means to live close to where they work. The former group rides as a matter of economic necessity.
It’s important to remember that this analysis covers only those who use their bikes to get to work and is for just one point in time. Our correlations point only to broad associations between variables; we can’t read anything in the way of causality from this analysis. That said, this admittedly crude analysis does suggest that bikes are associated with more good things than bad—no need to sound the bells about any New Urban Cycling Crisis just yet. That’s reason enough to get out and ride on Bike to Work Day.