Who bikes?

Workers in low income households rely more on bikes for commuting, but the data show people of all income levels cycle to work

There’s a lot of hand-wringing and harrumphing about the demographics of cycling. Some worry that bike lanes cater to higher income, spandex clad commuters, and are yet another signal of gentrification.

Workers in every income category bike, but bike commuting skews toward lower income households.

In response, experts hasten to point out that workers from low income households are the ones who be more reliant on cycling to get to work.  (That’s correct, by the way–as we’ll see in a second).  But in the process of trying to make a point, they’ve exaggerated the case.  Here, for example, is a recent tweet repeating a claim made in a CityLab article earlier this year:

That’s a pretty strong claim.  Anne Lusk’s article says data for that claim comes from a 2015 CityLab article, written by then-staffer Eric Jaffe.  Jaffe published some Census data on bike commuting by income group, and offered two charts.  The first chart shows the fraction of persons biking and walking to work by household income (lower income households are more likely to to walk or bike to work than higher income households).  A second chart shows the number of persons commuting by bike, with data aggregated by income groups in multiples of 50,000 (i.e. 0 to $50,000, $50,000 to $100,000, etc).  While Jaffe’s original article shows that lower income households are more likely to cycle to work than their higher income counterparts, it actually doesn’t support the Lusk twitter claim about the largest number of commuters being from households with under $10,000.

Let’s take a look at the most recent census data, from the five-year American Community Survey for 2013-2017.  We’ve used the data from the IPUMS website, because that let’s us tabulate data by our own custom income ranges.  We’ve narrowed our look to persons aged 25 to 64 who reported commuting to work. For this study, we look at household income bins of $10,000 each ranging from zero to $250,000.  Our first chart looks at the share of commuters traveling by bike in each income group.  (As is common in such data, we’ve excluded persons who work at home from these tabulations).

Data in this table are arranged from low incomes (at the bottom) to highest incomes (at the top).  Data show each range of $10,000 in income; and we’ve truncated incomes for those with more than $260,000.  The data show that those with the lowest levels of income are the most likely to rely on bikes for commuting. Almost 1.6 percent of commuters income incomes with less than $10,000 commute by bike, nearly three times the national average of about 0.5 percent.  Those with incomes of 10,000 to $20,000 are more than twice as likely as other American workers to commute by bicycle.  In general, the share of workers who cycle to work declines with increasing income, up to about $100,000 in household income, and then increases modestly as income rises.  (That increase in biking is definitely modest: those with incomes of $150,000 or more are universally less likely to commute by bike than those with incomes of $20,000 to $30,000).

Our second chart show the number of bike commuters in each income bin. As with the US income distribution, the number of persons in higher income categories gets smaller and smaller as household income rises above $50,000. (The high number of persons in the above $260,000 income category reflects the fact that we’ve consolidated all the bins above that dollar amount).  When we divide bike commuters into $10,000 household income bins, the modal number (most frequent category) is $20,000 to $30,000.

So the claim in the City Lab article, that “the single biggest group of Americans who bike to work live in households that earn less than $10,000,” simply isn’t true. About 30,000 of the more than 800,000 regular bike commuters in the US live in households with incomes of less than $10,000.  Every income group up to $100,000 or more has more cycle commuters in that income group than the under $10,000 group.  Based on the statistics presented in Eric Jaffe’s 2015 CityLab article, a claim about “the most households” holds for households with incomes under $50,000, but not for the category under $10,000.  But more importantly, it’s pretty arbitrary how you define income groups:  should it be $10,000 increments, $50,000, or some other measure.  Which of these arbitrarily defined groups is “largest” tells us more about where someone has chosen to draw the lines than it does about who is cycling.

A better way to look at the distribution is to consider the median income of bike commuters, relative to other commuters.  The median bike commuter had a household income of about $72,000 according to this data series, meaning about half of all bike commuters have household incomes less than that amount and half have more.  To put than in context, the median car commuter had a household income of about $82,000.  So, on average, bike commuters live in lower income households than car commuters.  For reference, the median bus commuter lived in a household with an income of about $62,000.

On average, bike-riding skews more toward lower income households, but it turns out that workers in every household income category are bike commuters.  Lusk’s article–and others–make the excellent point that we need to build cycling infrastructure in a way that is inclusive for a range of income and social groups. Accurate data can help make that case.

Joe Cortright

,