A civil debate: Are driverless cars good for cities?
February 16, 2017
With experts predicting widespread autonomous vehicle (AV) adoption in the not-too-distant-future, many policymakers, designers, and ordinary citizens are left scratching their heads, uncertain of what to expect and how to prepare.
We asked John Eddy, who leads Arup’s research into the impact of AVs on the built environment, and Ryan Falconer, head of the firm’s Canadian transportation consulting business, to stage a mock debate about the risks and rewards of a driverless future. With John adopting an optimist stance and Ryan channeling his inner critic, we explored the consequences of AV deployment for three important issues: active transportation (i.e., walking and biking), suburban sprawl, and social equity.
Let’s start with active transportation. John, will AVs make our cities better for pedestrians and cyclists? How will they encourage people to get out of cars and use active transportation, universally acknowledged to be better from the standpoint of public health and greenhouse gas emissions?
John: There’s no question we’ll be able to create a much safer environment for active travel. Shared autonomous vehicles will need less parking and road space than today’s cars, so we’ll be able to devote more space to pedestrians and cyclists.
And although the price of car rides should drop with driverless cars, I don’t think that’s going to stop people from using active transportation. There are multiple reasons why somebody walks or rides a bike, and a lot of it comes down to personal preference.
Ryan: I think the convenience of autonomous vehicles will disincentivize people from using active transit. I’ll link that very strongly to one of the points we’ll get to momentarily: sprawl, which will discourage people from taking active transport because of the travel distances involved.
The convenience of autonomous vehicles will disincentivize people from using active transit.
And I’ve got real concerns from a safety perspective. I see a future with autonomous vehicles gradually deployed in our cities, sharing the road with manually driven vehicles. This will create significant issues for bikers and pedestrians, who won’t be able to tell the difference. Driverless cars will react faster to someone in their path than a human driver could because of their risk-processing capabilities and situational awareness. People will get used to that and take more risks around AVs — but they might confuse the two kinds of vehicles.
John: Let’s not assume the vehicles will be indistinguishable. But even if they are, today’s drivers cover a very broad spectrum of skill and reaction time, yet their vehicles are indistinguishable. Defensive driving, cycling, and walking are rooted in anticipating the habits of the worst actors. Driverless vehicles will add to that spectrum of skill, and defensive cyclists and pedestrians will adjust without forgetting the worst actors.
Ryan: Yeah, but in a dynamic traffic situation people may have only a second to process that information. They may get it wrong. And when they do, all eyes will be on the autonomous vehicle market, just like they were following the Tesla Autopilot accident.
Let’s jump to sprawl, which also has massive implications for public health and climate change. How will AVs affect land use patterns?
Ryan: Automated vehicles will communicate constantly with one another, allowing them to travel closer together and move much more efficiently and reliably. They’ll also be able to travel faster. Passengers will be able to travel farther within the same amount of time.
With long commutes less of an issue, people might move farther away from the city. Urban living has become more popular in recent years, but there’s certainly an enduring demand for the suburban lifestyle.
John: Transportation isn’t the only factor in suburbanization, though. It’s a complex issue. For one thing, space considerations are very real in people’s decision making. Even for businesses — in some of the major tech companies, the square footage per employee went down to 125 square feet. But people weren’t happy, so it’s back to around 175.
The same is going to happen in cities. Right now a lot of people, particularly the young and the old, are tending towards an urban lifestyle, but they’ll generation out of that.
For me, sprawl is really a regulatory issue and a pricing issue. People often move farther out to find cheaper housing; they drive until they qualify for a mortgage. But this doesn’t reflect the real costs involved. The community as a whole has to build the infrastructure required for sprawl — roads, power lines, and the like — and maintain it over time.
We’re seeing recognition that this system needs an overhaul. VMT, or vehicle miles traveled — taxing people based on how far they drive — is becoming part of the regulation discussion in many states.
Ryan: These changes are happening fast, and I’m not confident that regulation can keep up at all, much less in a way that takes all the important nuances into account and helps to create what we might define as optimal outcomes.
For example, right now a reasonable proportion of VMT taxation is linked to infrastructure cost, but also to carbon consumption. So in a future where all automated vehicles are electric and they’re powered by renewable energy, perhaps governments will think, “Okay, we no longer have to charge VMT tax.”
John: The interesting thing is that the discussion about policy has already started, though. I do believe there is an opportunity to get ahead of it and test different pricing options to appropriately acknowledge the full cost of transportation.
Ryan: I’ll say one thing — there’s a stark difference between omnipotent governance and actual governance. The drop in your property value because of a new policy is a very tough thing to swallow. Politicians won’t want to anger voters across entire geographical regions.
And finally, social equity. Will driverless cars make the divide between rich and poor in our cities better or worse?
John: San Francisco’s highest auto ownership rates occur in some of the more economically challenged neighborhoods. There are a variety of reasons for that, not the least of which is less-than-ideal access to public transit in those areas.
Let’s say you work nights and buses in your area run very infrequently between midnight and 6am. Buying a car might be your most attractive option, although the cost amounts to a significant percentage of your income. Having access to a low-cost automated ride whenever you need it could improve your quality of life.
Ryan: But if your night job is driving that bus, your quality of life could suffer. Today, driving a bus or a taxi, an Uber or a truck is a solid employment option for many different kinds of people — people without college degrees and immigrants, for example. With the introduction of autonomous vehicles, these jobs will likely vanish.
The second issue is pricing. Some transit agencies may partner with the private sector to provide autonomous vehicle service or even replace their own bus fleets with smaller autonomous vans. So you’re potentially looking at a different price point. I could see a future where, instead of paying a couple of dollars for a connecting bus trip and a free or discounted transfer to a higher-level, higher-capacity route, people have to pay $5 for a connecting automated Uber trip — and they’re doing that twice a day. That’s a considerable cost for people of limited means. Benefiting people with lower incomes would require some very carefully thought-out legislation and new business models.
John: In the case of buses and smaller vans, the removal of the driver in the front seat could significantly drop the transit agencies’ operating costs. Yes, it’s a negative for those folks, but the savings could help provide a higher level of service to the community as a whole.
On the price-point issue, we can’t jump to the conclusion that automation will cause more financial problems in these communities than it solves. Making it easier for people to get from A to B at convenient times could let them take advantage of better options for employment, childcare, and other really critical considerations.
So I agree with you that we have to do more work to determine what kind of system could improve these people’s lives.