In 40 years, will self-driving cars send us packing for the suburbs?


At the end of 2013, the journal Cityscape put the following statement to contributors and asked their opinion of it: “In 40 years, the average person will live closer to her neighbors and farther from the ground than she does today.” This is a critique of one response. More to come…

Most urbanists will tell you that we’ll be living at higher densities sooner than you think. Nathaniel Baum-Snow, an economist at Brown University, is one of them. He cites a handful of reasons why he thinks we’ll all be living closer to our neighbors 40 years from now, including commute times, declining fertility rates, and stalled highway construction. Baum-Snow makes valid points, but many of his assumptions presume that the world 40 years from now, at least technologically, will look similar to today. Given the last 40 years, I’ll be surprised if that happens.

Self-driving cars are one of the biggest threats to the future of cities, and widespread adoption could single-handedly undermine one of Baum-Snow’s most compelling arguments—commute times, which he says will be a driving force behind increasing urban densities. As incomes have risen for many city dwellers—those in the top 50 percent, at least—the value of their commute time has also risen. Given Marchetti’s constant, which says that commute times tend to hover around 30 minutes each way, snarled traffic will force the wealthy out of the suburbs and back to cities. In a sense, we’ve already started to see that.

But self-driving cars could reverse that trend. As people’s commutes are freed up for other tasks, including work, they’ll stretch their daily trips, once again allowing them to live where they want. And as we’ve seen, people want to live where they have more space.¹ Compounding the problem is the fact that most early adopters are likely to be wealthy, the same people Baum-Snow says will be looking to drive less.

Working in cities’ favor, Baum-Snow adds, are declining fertility rates. Between 1967 and today, birth rates have fallen from 0.122 births per woman of childbearing age to 0.065, nearly a 50 percent drop. With people having half as many children, the need for space should decline. But as we’ve seen, that’s not necessarily the case. Between 1973 and 2012, median home sizes have grown from 1,525 square feet to 2,306 square feet, the same time that fertility was declining. In that same time period, median household income has risen almost $10,000 when adjusted for inflation, a nearly 25 percent increase.²

Then there’s the siren song of the suburbs—school quality. Suburban and small town school districts frequently outperform their urban counterparts. Baum-Snow notes that the quality of public schools in big cities has stabilized, at least, but that isn’t quite the same as improving. Plus, even if they do improve, urban schools will have to overcome the stereotype wrought by decades of poor performance.

(There is a simple fix, of course—invest in public education at all levels. Early childhood programs have shown great promise at preparing children of all backgrounds for full-time school, and education is the best chance many of children have to break free of poverty.)

Given these realities, I don’t think the future points to downtown, as Baum-Snow does. It’s true that city centers are the hot place to be, but density numbers don’t reflect the newfound enthusiasm. And trends in technology could shatter the many cities’ recoveries.

Despite that, I think many of us—at least, those of us not in the top few percent—will be living closer together. Not downtown, but in the sprawling inner suburbs that will be indistinguishable from the rest of the city, slogging through long commutes between home—which was convenient to the old job we got laid off from—and our new jobs on the wrong side of town.

  1. Home sizes briefly halted their inflation during the recession, but they’ve started to rise again.
  2. On top of that, interest rates are lower today than in the 1970s. That could change 40 years from now, of course.

Photo by Michael Colburn.


Baum-Snow, Nathaniel. 2013. “Changes in Urban Population Densities Over the Next 40 Years.” Cityscape 15(3).

Related posts:

In 40 years, will we live in cities in the sky?

Marchetti’s constant, or why the 30 minute commute is here to stay

Population density fostered literacy, the Industrial Revolution

Thanks to the support of readers like you, Per Square Mile remains independent and ad-free.

If you enjoy what you just read, please consider supporting the site with a monthly recurring donation as a Sustaining Member.

You can also support Per Square Mile with a one-time donation.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

  1. “But self-driving cars could reverse that trend. As people’s commutes are freed up for other tasks, including work, they’ll stretch their daily trips, once again allowing them to live where they want.”

    Utter nonsense. Travel time is still not-with-the-kids time. And robotaxis don’t change the geometry of congestion (on any meaningful timescale: don’t talk to me about close packing, because that’s a 22nd century thing).

    America’s urban renaissance is not a flash in the pan, but a rediscovery of public space: that a nice street to walk down is actually an important part of a full life. And then all the economics that come with that: the nice street has active frontages, which need sustainable businesses, which need a surrounding customer based that is there both mid-week and evenings/weekends. Plus the infrastructure costs-to-taxpayers ratio: those miles of suburban road and piping and wires aren’t free.

    America’s urban renaissance is not a flash in the pan. Rather, in the curve of history, the mid-20th century experiment of suburban sprawl was the temporary blip, that has been shown to fail to bring economic prosperity or quality of life improvement.

    1. I’m not saying it’s a good idea, just that it’s a threat to the urban renaissance. Acknowledging the potential for technology to change what we assume is the status quo (or current trend) isn’t “utter nonsense”, it’s good planning.

      1. Good planning is planning for fewer cars. We all know you get the cars you plan for. Robotaxis change nothing about what constitutes good planning (and, as below, make it politically easier).

      2. Mode choice is a function of infrastructure and economics, not of technology and certainly not an endogenous preference.

        The Dutch don’t rent bikes in Atlanta, just as Georgians don’t rent SUVs in Amsterdam.

  2. In fact, robotaxis will be an urbanist-environmentalist boon by making sprawl repair politically easier. Cheap robotaxis (and solar-fueled EVs are already lifetime cheaper: robotaxis, with higher utilisation and no driver, will be cents per mile) and robobuses will take the wind out of loss of parking complaints, kick-starting the virtuous cycle of stroad-to-boulevarding and mid-rise-mixed-use liner buildings where once were parking lots.

  3. Smart cars make open up the possibility of taxing people proportionally to their road usage.

    the utility of small, driverless electric vehicles promise to make dense urban spaces much quieter, cleaner, safer, and thus vastly more appealing than many of them are now where the sound of roaring engines carries on all day and all night.

    Widespread adoption of autonomous taxis will also reduce the need for excessive amounts of parking space, which opens up a lot of room for infill development, which is beneficial when making the argument for growth limits.

    Self-driving cars with all the other forms of automation trickling in will have many people underemployed. They won’t be able to afford the suburbs whether they want to or not.

  4. Will self-driving cars be faster, cheaper to buy or operate, or drive on cheaper roads? if not, they won’t make a significant change to American society.