top of page

Why Most People Drive when They Could Take Transit

One of the most popular terms in planning is “automobile dependence”. The premise asserts that most people in most places in the United states are involuntarily dependant on their cars – with no other option, and that in an ideal world, in which everyone would be happier and more productive, people would ditch their cars and hop on a bus or a train to get around.

The idea is so ingrained in the minds of many planners that a simple search for “automobile dependence” on popular urban planning news site returns nearly 2,500 articles on the subject. “Transit-oriented development” also turns up nearly 22,000, a relatively astonishing figure since the term “urban planning”, the profession for which the entire site is designed, returns only 25,000.

The assumption that people would prefer to take public transportation entails both a condemnation of the general automobile orientation of our current cities and a challenge to future cities to envision planning and designs that rejects the automobile on behalf of transit. Many planners and new urbanists, in fact, advocate that the world would be better off if transit were at the foundation of our cities. According to new urbanists, cities should aspire toward TOD models "with transit as a foundation of mixed-use neighborhood structures served by thoroughfares while still making daily uses walkable". That idea of walkability is a popular one, and I know few people who wouldn’t prefer the ability to get a drink a short walk from home. Similarly, “mixed-use” is a relatively harmless idea as well, though in reality jobs have been dispersing throughout metro areas, closer to where people live, for most of the history of cities.

The bigger idea behind transit-based city planning, however, the idea of “remedial” planning, creating growth boundaries and positioning uses around communal citywide transportation, leaves much to be desired. According to the CNU Lexicon, TOD creates the efficient spacing of light rail such that stations could be connected by a freebar bus system. The assertion has inspired many cities to pursue transportation policy initiatives backing TOD, pushing spending on light rail and buses, and re-zoning entire municipalities under the idea that if only people had access to transit they’d take it. But both in theory and in practice, this kind of remediation bears more negatives than positives.

People can access transit, but they aren’t taking it

One of the problems with the presumption that designing around transit would be more optimal – and with the shift in policy and spending that it incurs – is that most cities America already have a remarkably interconnected and accessible transit system – with plenty of housing options for those who want to be reliant on transit. According to a 2005 ABCNews/Time magazine/Washington Post poll 59% of Americans say they could take transit to get to work – if they wanted to – and this figure is closer to 70% in urban areas. Access to transit is abundant, and cites are continuously expanding these figures behind well-coordinated and widely-publicized campaigns to increase transit accessibility, separate from the zoning and other hurdles of TOD. Chicago, for instance, has already embarked on an initiative to up its transit accessibility to 75% by 2040 as part of its “Goto 2040” plan.

Despite all of these efforts, transit use nationally languishes below 5%. In cities like Portland, where transit has been prioritized for decades, transit’s share of total commuting is a languid 7%, down from 12% in 1980. In the few areas where it’s increased, the uptick is virtually negligible in the context of all commutes and miles traveled.

So why aren’t people taking transit?

In overwhelming numbers, people are choosing not to take transit. While the ABC poll found that 59% of people could, theoretically, take transit to work, just 4% of poll respondents said they actually do; and just 10% say they ever take transit for any facet of their weekly needs. Census figures indicate that these figures have not changed much since 2005.

The biggest reason people choose not to take transit is that even in big, transit-optimal cities, transit is overwhelmingly slow and inconvenient. 93% of people nationwide think driving is more convenient, and the average transit commute takes 60 minutes, as opposed to the average driving commute, which is 26. 56% even think transit is more expensive given their job prospects and housing costs. In Chicago, despite its “Goto 2040” ambitions and widespread transit accessibility, transit’s slow, linear nature (with frequent stops) means that just 6.3% of the region’s jobs (as researcher Wendell Cox concluded) can be reached within a reasonable 45-minute journey on transit.

This is especially important in light of an across-the-board trend in American cities, wherein regardless of population or density average commute times are remarkably static at around 25 to 30 minutes. There is only so much time in the day, and for most people there are diminishing returns on how far they are willing to travel to go to work. The result is that, in general, people’s choices about where to live are dramatically limited by how long it will take them to get there. And in the vast majority of cases people’s job and home choices are far, far broader if they use a car as their primary mode of transportation. Simple math illustrates that transportation that is twice as fast actually allows a person access to an area four times as large over the same time period.

The manifestation of this limiting trend is that commute times are remarkably static despite the differences in population between the 28 largest metro areas. While population varies widely, from populations around 300,000 to nearly 10 million, average commute times among the largest 28 metropolitan areas vary much more narrowly, from only around 25 to 35 minutes. This cap places a limit on people's living conditions and their job access, limiting the scale of economies as labor markets.

Cars provide access to greater numbers of jobs in nearly every city than transit. Denser locales with greater transit close the gap through density and clustering of jobs, but in no instances do they close the gap all the way. People aren't taking transit because it doesn't provide them quick access to jobs in nearly the same way cars do.

Would a transit-based model work better?

Many transit analysts blame the fact that transit is inconvenient on decades of highway subsidization that has “forced” cities to be built in ways that are hostile to ideal transit models. In reality, however, even an idealized conceptual multi-modal transit system offers nowhere near the timely coverage areas as freely operated automobiles.

Even with 100% of the jobs concentrated no further away from any resident than in an accessible city center (an unrealistic and dubious notion given the relative dispersion of America’s urban jobs - downtowns rarely house more than about 10 percent of regional jobs), even the most idealized TOD-based city with unlimited funding for transit lines can only accommodate about a 10-mile radius within which all residents could get to transit in a 10-minute walk and get to their job in 30 minutes (the empirically observed limit across the nation on average commute time tolerances).

By contrast, using a car for personal mobility can allow a person to cover 12 and a quarter times the area of that entire hypothetical city in 30 minutes. In other words, even at 10 times higher density a TOD would still offer substantially less mobility to people on transit than if they took a car – even to a single destination. This is due to both the speed and directional freedom a car provides as opposed to the fixed collectivism of public transit.

This hypothetical example, by the way, is backed by the empirical data that demonstrates an even more extreme condition even when analyzing the most transit-rich cities we’ve ever conceived. Per the American community, an aggregated analysis of 30-minute commute sheds in American cities shows that around 57 percent of regional jobs can be accessed in most cities by car within around 30 minutes. In some metropolitan areas this number reaches as high as about 73 percent.

By contrast, per data from the University of Minnesota, for the average person, public transit offers access to no more than 1 to 3 percent of jobs in even the most transit-rich of cities. Even in New York City, among the densest and most transit-rich cities in the world, for the average resident only around 200,000 jobs are accessible via transit within this 30-minute timeshed. This is fewer jobs accessible within 30 minutes via transit in New York than even job-light cities like Memphis or Grand Rapids provide access to via the car. This is an astounding differential.

The reason for all this, and the reason transit will never support the kinds of broad economies on which America depends, is that it imposes limits on where jobs can be, where “else” you can go, as well as when you can get there. It’s the inherent limitation of collectivism. The freedom of time and movement provided by personalized mobility is such an astounding contributor to personal economic (job) freedom that the comparison between collective modes of mobility and personalized ones is hardly comparable. To put it bluntly, to build an economy primarily based on collective mobility (transit) that resembles the levels of freedom and prosperity available in the United States would be all but impossible, and certainly not desirable. This is why relatively high-speed modes of personalized mobility innovations like Uber, scooters, autonomous vehicles, even bicycles continues to flourish while transit, no matter how much investment, languishes. Hybrid models that leverage collective travel for cost savings while providing greater freedom of movement than fixed transit - like minibus services and Skytran - are also outperforming the fixed transit models.

Of course, the faster the mode of personalized commuting the better. As researcher Alain Bertaud has pointed out, the benefits of personalized mobility become exponential the faster the speed. Cars move people about 10 times as fast as pedestrian and 3 times as fast as bikes on a straight line, but the exponential multiplier generated by freedom of movement squares these ratios. Within 30 minutes a car can access 10 times as much area as a bike, and 100 times as much as a pedestrian.

Ending the myth of automobiles dependence

The reality is that “automobile dependence”, in the way it is most often used, is a misleading term. Certainly, if the world ran out of oil tomorrow, the United States would be dramatically affected, at least in the short term – as would every developed nation in the world. Cars account for 86% of passenger miles traveled here, and without them an enormous majority of people would need to find a new way to get to work. And for the few people falling in the age brackets where individuality is possible but driving is not, mobility can be challenging without transit.

But the claim that these statistics reflect any level of widespread personal “dependence” on automobiles (and that that’s a bad thing) – is not backed up by the evidence in most cases. In many ways, it’s akin to claiming Americans are now “dependant” on Apple’s iphone. The reality is that while many people choose to have iPhones, most people could quickly adapt to other brands, and that for those who have bought iPhones they’ve done so because they felt it increased their productivity or quality of life. Widespread use is not the same as dependency.

Any good urbanist will point out the value of cities to the world – they are the embodiment of the power of the economy of scale to create growth and improve life. So why would we limit our economies of scale by imposing a form of development limited by transportation systems that aren't conducive to scale?

Most Americans know the answer to this question – which is why, regardless of policy, they continue to choose to drive in their cars.

Contact Capital Frontiers at:

bottom of page