How will driverless cars affect our cities?

By Issi Romem

Issi Romem has a PhD in economics at the University of California, Berkeley, with a focus on urban and real estate economics. He has consulted for the Bay Area Council Economic Institute on matters involving transportation, real estate and the regional economy. Soon he will be joining the team at OnPoint Analytics.

Mar 25, 2013 | Smart Cities | 23 comments

Driverless cars are an imminent reality

Google is the most conspicuous developer of autonomous vehicles, but it is hardly alone in pursuing this venture. Most automakers are competing to introduce their own driverless cars to the public, and are doing so piecemeal, system by system. The components of the upcoming driverless car are being introduced into current models as ever more elaborate mechanisms to aid the driver, such as self-parking features and automated collision avoidance systems. Recently, a group of researchers at Oxford University developed a self-driving system which can be installed in existing manually driven vehicles, and whose cost is hoped to fall as low as 150 dollars within a matter of years.

Driverless cars will dramatically affect urban form, in two ways

Many anticipated consequences of driverless cars have already received attention on this blog and elsewhere, such as their impact on the mobility of the elderly, on taxis and car sharing services and on the future of the car industry. A crucial aspect which has escaped attention is the impact of driverless cars on urban form, which I anticipate will follow two broad predictions:

  • Cities will greatly expand, again: Faster and more efficient transportation will convert locations that are currently too remote for most users into feasible alternatives, abundant with space. Like suburban rail in the early twentieth century and the mass consumer automobile that followed, driverless cars will generate a gradual, but dramatic expansion of cities.
  • Buildings and parking will be uncoupled, freeing up valuable land: After dropping off passengers, driverless cars will independently seek parking (or their next car-share customers) and they will show up for the return ride at the tap of an app. As soon as driverless cars are common enough, the demand for adjacent parking will dwindle and parking lots in areas where land is sufficiently valuable will be ripe for conversion to other land use. As parking in high-value areas is thinned out or altogether purged, the micro-structure of cities will change – you guessed it – dramatically!

Why will cities expand?

Driverless cars will make it less “costly” for people to travel a given geographic distance, partly because they will be free to engage in other activities while travelling, but primarily because of reductions in travel time. Unlike human drivers, autonomous vehicles will follow optimal routes given real-time traffic conditions without fail. More crucially, as soon as suitable roads such as freeways (or lanes thereof) are declared off limits to manual driving, driverless cars will travel – safely – at much higher speeds than we do today. Gains in efficiency will follow from coordinated traffic management protocols, too. Once vehicles communicate with each other traffic through intersections and merges will flow much more smoothly than permitted by today’s traffic signals, stop signs and merging lanes, leading to substantial gains in travel time (a partial, human-mediated step in this direction is explored in this article).

If people currently forego affordable, spacious dream homes because the associated commute is too long, a technology that condenses the time needed for commuting along the same route – and allows doing so in the back seat – will make those homes more agreeable. Similarly, businesses whose location depends chiefly on access to appropriate labor or clientele will find that potential locations which are currently too remote will become feasible. It will still be crucial for them to sit “close” enough to their talent pools or their customer base, but because what matters for “closeness” is travel time rather than geographic distance, these firms will be able to reap the benefits of more remote locations without giving up “closeness.”

How far will cities expand?

The extent to which cities expand will be determined by the extent to which travel times are reduced. The more efficient traffic flow becomes the broader the geographic range in which living and working becomes feasible.

Will we ever hit a point at which people are no longer interested in the extra space offered by more distant locations? This is unlikely. Today swimming pools and three car garages are common in suburban homes, but who would have imagined that possible before the advent of the mass consumer automobile? Perhaps the current equivalent is the wish voiced by some home buyers – typically just beyond the urban fringe – that neighbors’ homes be out of sight. That seems like a lot to ask in today’s suburbs, but it could well become the norm looking forward.

When will this happen?

Most estimates suggest that the arrival of the fully self-driving car on the consumer market will occur within a decade. Provided that it will be possible to install these systems in existing manually driven cars – much as hands-free cellphone devices can be installed today – then there will be no need to wait for the entire stock of cars to gradually be replaced, and a much faster process of adoption will ensue. The speed of the process will be determined by people’s willingness to give up the driver’s seat, and by the adaption of the legal environment, first to permit driverless cars and then to secure them an exclusive right of way (a separate lane on the freeway). Google and the automakers will go to great lengths to ensure that legal barriers are removed and that the driverless car is adopted quickly. The devotion of a separate right of way may be a more challenging feat, but it will be difficult to reject in light of the gains it will offer.

Following these developments, the gradual process of city expansion will take place over many decades, much as the ramifications of the mass consumer automobile continue to play out almost a century after its arrival.

Is this good news or bad?

Ultimately, the accelerated drift of the city past the current metropolitan fringe implies sprawl on an unprecedented scale. This is unwelcome news for those readers who, like this author, share a romantic view of dense urban life. But there is good news as well.

In his 1991 classic, “Edge City”, Joel Garreau wrote that it is “the suburban home with grass all around that made America the best-housed civilization the world has ever known.” If the widely spaced mansions of the future are to today’s suburban home what today’s suburban home is to yesterday’s urban tenement, then we are in for a glorious improvement in our material welfare. But this grates the city lover’s ear and there is good news for city lovers, too.

The uncoupling of buildings and parking

Once most people stop driving manually, there will be a far less compelling need for buildings and parking to be adjacent. This does not mean that all parking lots will be converted to other land use – the total need for parking will only be reduced if other developments like increased car-sharing take off. But it does mean that parking lots on the most valuable land will be available for infill development. Driverless cars will gladly navigate to abundant off-site parking that will substitute for the lost parking on less valuable land.

The places in which infill development takes place will become denser and more walkable. The busiest suburban shopping districts will probably be among the first to see their parking built upon, as will clusters of suburban office towers which often spread out over vast areas. In so doing these areas will attain a more urban feel.

Of course the broader environment will remain suburban, but the local clusters of walkable density we have today – primarily old town centers engulfed by sprawling metro areas – will be joined by a new breed born of formerly pedestrian-free suburban centers and infill development upon parking. Given that the overwhelming majority of dense walkable areas in this country were built before World War II, a new generation and breed of walkable locations is rather exciting.

And what about the carbon footprint, you ask?

Traveling greater distances at greater speeds will require more energy. Full stop. Car sharing will not undo this in spite of reducing the total number of cars, because car sharing essentially only does away with the time cars spend parked.

Under the pessimistic premise that each car continues to emit greenhouse gases at current rates, the effect of driverless cars on urban form spells out a magnified carbon footprint. But technology is not stagnant. Today’s gasoline powered cars are already far more efficient than they were even a decade ago, and the ongoing transition to electric vehicles means that the energy needed for traveling greater distances at greater speeds will no longer need to come from fossil fuels. Instead, cars can be powered by any source of energy used to produce electricity, including more sustainable alternatives.

Contrary to the intuition that associates rapidly advancing sprawl with environmental disaster, persistent progress in sustainable energy could ultimately dissociate the suburban lifestyle from the greenhouse gas emissions it implies today, severing an important link between sprawl and climate change. The crucial question in this respect is whether the greening of our energy will precede the brunt of our cities’ future spatial expansion or not.


Leave your comment below, or reply to others.

Please note that this comment section is for thoughtful, on-topic discussions. Admin approval is required for all comments. Your comment may be edited if it contains grammatical errors. Low effort, self-promotional, or impolite comments will be deleted.


  1. You will find a different analysis of potential effect on cities at

    There will be pressures in both directions, both for expansion and contraction.

  2. Quite interesting.
    I agree that the technology for driverless cars will be ready in a decade or so, but am not so sure that people and our legal system will be ready/willing by then to start adapting. And i wouldn’t expect exclusive driverless lanes for quite some time – a matter of $$ and infrastructure demands.
    The overall thesis is sound though, and quite thought-provoking. Driverless cars could be a huge influence on increased urban density for all the reasons given.

  3. I wrote about this topic, with a slightly different conclustion at:

    I do think the idea that cities will sprawl much more than what they are now is an unfounded fear. I think it’s pertinent to ask, especially in the American context, how much more can cities really sprawl? And, history has shown that commute times generally do not expand much beyond the 20-30 minute average time limit regardless of transportation mode.

    At any rate, it’s an interesting debate.

  4. Fear of sprawl? How does my living 50 miles form the city center affect your life at all. Sounds more like urban elitists trying to tell other people how to live their lives.

    • Paul,

      While I agree that you living 50 miles from a city doesn’t affect my life (or any others) directly, if you work in the city and drive there everyday, it would have an indirect affect on EVERYONE because you (hypothetically) have a greater carbon footprint. Your living choice and need to drive a min of 100 miles a day probably has a greater negative impact on the environment thus indirectly affecting us all. This is the issue.

      Now multiply your situation out by millions of people and it DOES becomes an environmental problem whether you want it to or not. Personally, I don’t like living in cities and am a fan of suburban sprawl (I grew up in the country) but I can only support it if sprawl can be done in a sustainable and environmentally responsible way. Today, sprawl = more driving and more driving = worse for planet. Tomorrow that may change. I’m rooting for Tesla.

    • Depends…if you work in that City, or have a company dependent on that City’s existence, and your 50 mile away taxes don’t support it, you are a leech…that’s how.

  5. Missing from this article is any mention of the impact on safety. As virtually all death and disability connected to transportation are the result of driver error, particularly driving under the influence, aggressive driving and driving too fast for conditions, automated vehicles should make transportation much safer.

  6. “Unlike human drivers, autonomous vehicles will follow optimal routes given real-time traffic conditions without fail. ”

    The Travelling salesman problem is still a NP problem, that implies that obtaining “optimal routes” in growing complexity maps with a mix of human and non-human drivers will remain a dream… routes will remain approximative and fails will still happen…

    BTW, what’s the difference between a car you can’t control and a (public transportation) train with a private room? the second option is less expensive.

    • The differences are convenience and implementation.

      Cars could be phased in one at a time, either by adding on the system or as new cars replace older cars. The train would require the construction. While it sounds easier, finding the land required to build the track isn’t exactly an easy process in an urban area, nor is it typically affordable.

      Assuming both are implemented, the train doesn’t magically hover and take you to your location. As was implied with the reduced need for adjacent parking, humanity is usually lazy. The train being further from the point is similar to parking needing to be close. If the walk is “too far,” the commuter will take other options.

  7. Hi, I’m another grad student at Berkeley but from transportation engineering. I don’t agree with these predictions and I wrote about it here.

    The gist is that much of our travel time is consumed traversing queues of other vehicles rather than distances. Queue time depends very much on constraints built into our road design. AV’s will raise capacities surely, but the effect will be muted unless we redesign roads for AV’s, which we probably won’t do if past policy is any guide. Moreover, if cars will always be in motion, there are major gains to locating nearby other origin-destination pairs. If you live in the super-exurbs and rely on AV’s then you will pay for deadheading the vehicle to your house.

  8. I also disagree and wrote about it here

    Sprawl is a subsidised policy choice, not a natural result of technology (cheap oil, car loans). Similarly the recent urban renaissance isn’t about high oil prices, but about the rediscovery of the joys of compact walkability. People like living in places where they can walk to the store, where they can see other people. In surveys, Americans overwhelmingly say they want to live in this kind of ‘small town’ neighborhood.

    E-robotaxis will be the catalyst to let aging boomer nimbies join with Gen Y, and provide the political support needed to retrofit 20th century sprawl into compact, mixed-use, walkable communities.

  9. Lots of disagreement here.

    I just want to chip in and suggest that your analysis doesn’t take into account that a whole bunch of land used for parking is suddenly going to be freed up for residential/commercial construction.

    In my judgment, this newly freed-up land will help constrain further urban sprawl.

    • And…. read the article before leaving a comment….


      I have no idea how I missed such a big chunk of text like that.

  10. No thanks. I’ll stick with my bicycle. I don’t want a big house with a swimming pool either. This seems like a future for the rich 1%, what about the other 99%?

  11. I largely agree with the assessment of the opposing trends that will be at play as a result of autonomous vehicles. Our extensive research indicates that in Southern California, autonomous vehicles will enable sprawl, but will also facilitate concentrations of density in multiple centers. The autonomous vehicles will coexist, and be compatible with, an increase of mass-transit options, and other forms of mobility (bicycles, pedestrians, mobility-scooters). Overall, we foresee a greater range of affordable choices for how to live, with better viability for various levels of density. You can read more in our (cool) new book, The Car in 2035: Mobility Planning for the Near Future.

  12. Some interesting thoughts here. Curious if you have any ideas about how much faster cars might go in an ideal world and if say people are willing to commute 30 minutes how far out that might mean they can live.

    • I think the answer is both “it depends” and “no one knows”. The increase of efficiency would be different according to the type of road – freeway, arterial, residential – and the amount of other types of traffic (pedestrian, non-autonomous, bicycle). The efficiency is best where autonomous vehicles are the only type of vehicle in a lane, such as a designated freeway lane. I don’t think the answer can be summed up by a single percentage.

  13. This will work wonders in the last mile connectivity and can help in mass urban transport being the way forward in future

  14. Once we eliminate the risk of accidents, surely we can introduce lighter weight motor vehicles. Presently, 95% of the fuel used by a car is required to move the car around.

  15. Real Estate Values. If professionals can start their day’s work from the moment they step into the car in their garage, then they will be wanting to live farther and farther from the city’s center. The car will become more than an efficient car, it will become an efficient office, equipped not just with telecommunications, but with multi-media devices allowing for visual communications around the world. Your one to two hour commute will be the most productive part of your day.

  16. Great article! Your question is very interesting and it is reflecting the current situation in major cities. My city is too crowded with the growing number of cars and it makes the air getting worse, needing the best solution for this situation.


Submit a Comment

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

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Read more from the Meeting of the Minds Blog

Spotlighting innovations in urban sustainability and connected technology

Closing the Water Access Gap in the United States

Closing the Water Access Gap in the United States

Today, over 2 million Americans are living without access to clean, running water. The newly released ‘Close The Water Gap’ report by DigDeep and the US Water Alliance pulls back the veil on America’s hidden water crisis.

This is the first-ever comprehensive look at indoor water access across the United States, and its findings are explosive: Race is the strongest predictor of vulnerability. In six states (plus Puerto Rico), progress is actually backsliding. More than 44 million Americans are served by water systems with recent violations of the Safe Drinking Water Act.

The Link Between Climate Change & Water

The Link Between Climate Change & Water

When thinking about conserving water, we should also be focusing on how more efficient water use correlates with energy savings. Studies show that when households participate in water savings programs, they also conserve energy and reduce strain on the power grid during peak demand periods while saving consumers money on their utility bills.

Water utilities can also dramatically increase their energy efficiency and reduce overall energy usage by adopting locally based solutions. For many municipal governments, drinking water and wastewater treatment plants are typically the largest energy consumers, often accounting for 30 to 40 percent of total energy consumed. Overall, drinking water and wastewater systems account for approximately two percent of energy use in the United States, adding over 45 million tons of greenhouse gases annually.

Using Data to Reduce Public Health Risk

Using Data to Reduce Public Health Risk

Addressing the impact of heat on health is well-aligned with MCDPH’s vision and mission “to make healthy lives possible” by protecting and promoting the health and well-being of MC residents and visitors. The climate has significant impacts on our community’s health. Through extensive surveillance and community surveys, we have demonstrated the importance of local public health data to increase buy-in from new and existing partners and obtain funding to address this significant public health issue. We encourage other health departments to consider the power of data and collaboration as they seek methods for protecting the public’s health from a changing climate.

Share This