The current issue of the Journal of the American Planning Association includes vigorous discussion of an old question: Do compact, mixed-use forms of development reduce the number of miles people drive, and if so, by how much? The issue is framed around an analysis by Mark R. Stevens suggesting that the impact of land use on vehicle-miles traveled (VMT) is small.
This debate has been going on for decades, without cease and without apparent resolution. And one imagines that it might go on for decades more.
We have, however, long since blown past the point where the problems with our transportation and land use system (including, most urgently, global warming, but also many others) can be solved with the kinds of tactical interventions that absorb most of the oxygen in these debates: a different block configuration here, a transit stop there. Those things are important, especially at the local level, and are pieces of the overall puzzle, but simply heaping tactical interventions atop one another does not amount to creating a strategy to confront our most urgent problems.
There are important findings in Stevens’ analysis that should inform tactical approaches going forward, but the far bigger questions relate to how we understand America’s 21st century landscape, and how we might unite an array of tactical approaches, each with limited power on its own, into powerful strategies for moving the nation in a more sustainable direction.
Tactics: Build More, Build Closer
The two tactical-level findings in Stevens’ paper of greatest import arise from his assessment of the impact on VMT of residential self-selection and metropolitan extent.
Because studies like this one focus on the question of whether the physical characteristics of a neighborhood affect travel behavior, they ideally should correct for all other factors – including the possibility that people living in a certain neighborhood might drive less not because of the size of the blocks or proximity to transit, but because they are people who don’t like driving, and would drive less than average regardless of where they lived.
Stevens’ paper suggests that residential self-selection works in an unintuitive way – that it magnifies the impact of community form on VMT. This is important. As Stevens writes, “One possible explanation for these findings is that persons with preferences for housing close to downtown or in dense urban neighborhoods have a difficult time finding such housing, so they can be forced to live in a less-dense, more auto-oriented neighborhood than they would prefer.”
The second important insight is that the most powerful single tool for reducing VMT in Stevens’ analysis is reducing residential distance to downtown. As Stevens explains, “a household would be expected to drive roughly 32% fewer miles if its distance from downtown decreased by 50% (e.g., if it moved from 10 miles to 5 miles from downtown).” Contra Stevens’ headline claim that the potential VMT reductions from land-use change are “small,” a 32% reduction in per-capita VMT would be a sizable down-payment toward decarbonization of the transportation system and many other societal benefits.
These two findings suggest that, if the effect of neighborhood-level land use patterns is small, the effect of metropolitan-level land-use policy is potentially profound. Find a way to fully supply the latent market demand for compact neighborhoods near downtown (which, as Arthur C. Nelson notes in his response to Stevens’ paper, is huge), limit exurban sprawl, and you are a good chunk of the way there.
The policy strategies that can contribute to achieving these goals are many, and, as we wrote in our report, A New Way Forward, they vary depending on the city. In fast-growing coastal cities, the priority might be on liberalizing zoning to allow more compact development near downtown. In the Rust Belt, it might mean encouraging neighborhood renovation and solving long-standing fiscal woes that degrade public services like schools and push people to leave cities who really might rather prefer to stay. In the West, it might mean building more intensely around new transit nodes. And in the Sunbelt, it might start with simply not building any more darned highways to limit exurban sprawl.
Strategy: To Plan or To Price? Or Both?
In his response to Stevens’ paper, Michael Manville raised the most direct challenge, asking why the planning profession is so obsessed with the effect of development patterns on VMT when large amounts of public space are given away, for free, to people who drive. He writes, “When communities come to us concerned about excess driving, we can talk to them about promoting dense, transit-friendly development. But first we should remind them that their roads are free, that their parking spaces are free, and that their zoning for off-street parking feeds their problems with on-street driving. Travel’s influence on the built environment is a function of the built environment designed for travel. We should stop pretending otherwise.”
On one level, I could not agree more. We have written often of the various ways that driving is subsidized in the United States. Removing those subsidies is an important step toward getting the design of our communities right and reducing the damage caused by car dependence.
But, we also need to recognize that, when doing battle against global warming or the countless other impacts of car dependence, you go into battle with the electorate you have, not the electorate you might want. And the electorate, by and large, consists of people skittish about their own economic situation who are highly dependent on cars.
It should come as no surprise under those circumstances that public opinion polling on pricing strategies is positively brutal. I think advocates, public officials and planners need to tell the electorate hard truths about subsidization of driving, even when it rubs people the wrong way. Eliminating subsidies where we can, as Joe Cortright writes, is a powerful strategy. We just should not expect it to be a winning strategy in the short run.
The best route to solving the subsidization problem in the long run may very well be to change the politics in the short run. And the best way to do that – in addition to educating the public and decision-makers about the real economics of transportation, as Manville does and as we try to do in much of our work – is to increase the number of people with less day-to-day stake in the price of gas or roads, and reduce the number of people for whom those issues are legitimate day-to-day worries. Everything we can do at the margins, therefore, to give people additional transportation choices and satisfy latent demand for less car-dependent forms of living helps to bring the day when we can have a productive discussion about pricing ever closer. The two strategies are not competitive with one another, they are complementary.
Vision: Saying Goodbye to the 20th Century
There is a big question lurking at the center of Stevens’ paper, one that is touched upon lightly in the responses but deserves to be confronted directly. It is the question of what, in 21st century America, constitutes “normal”?
Stevens implies that we should consider late 20th-century suburban development patterns as the norm, and compact, traditional forms of development as the exceptions. He writes, “the burden of proof is arguably upon planning researchers and practitioners to demonstrate that the benefits of planning interventions (such as compact development) outweigh the costs.” Writing in the same issue of JAPA, Susan Handy pushes back, noting that all planning is inherently an “intervention” and that “[c]urrent planning practice … inhibits the market from pursuing compact development; a loosening of planning restrictions would produce more compact development.” (emphasis mine)
The debate between Stevens and Handy is emblematic of a divide – maybe the most important divide – at the core of American politics in 2017. It is the divide between those who believe that the conditions of consistent, rapid economic growth that prevailed in the U.S. in the late 20th century (and their attendant trappings, including suburbanization) represent the natural, attainable and desirable order of things, and those who don’t.
This is a divide that transcends left and right. Regardless of whether your vision of what was good in the 20th century was go-go consumerism and rampant economic growth, redistributive public policy and a growing welfare state, or (if you’re Steve Bannon) patriotically inspired displays of industrial might, if you think that vision is achievable in the same way it was 50 years ago, you are very likely to wind up with a different analysis of the 21st century than if you believe, as I do, that conditions and priorities in the world have fundamentally changed in ways that we may not fully understand but that we have to try to comprehend and confront.
The most laughable example of this in the current discourse is the debate about whether the Millennials will “move to the suburbs” as they age, just as their parents and grandparents did. Would Millennials like to move to a brand-new, affordable, cozy, government-subsidized house with a small yard in a neat suburb a short, uncongested drive from downtown – like their grandparents did? Boy howdy, would they! But that is not what is on offer – and it is something the planning profession and public officials are powerless to deliver, at scale, ever again.
In my neck of the woods (Boston), here’s what is on offer: a) that same bucolic suburb at many times the price your grandparents paid to live there; b) a faded version of the same suburb where the residents are all retiring, retail is struggling, the tax base is going to seed, pension and infrastructure maintenance expenses are kicking in, and your kid has to shell out a fortune just to try out for the football team, or c) a semi-bucolic new-build exurb that’s so far away from the city center that you might as well kiss your kids goodbye at the age of two and say “daddy’s home” when you walk in the door at 18, having barely seen them in the meantime.
Promising a lifestyle that is impossible to deliver – and that in many ways was never all it was cracked up to be – is how you burn public trust. It’s how you wind up with a housing bubble (and a nascent auto bubble). It’s how the balance sheets of transportation agencies and other government entities wind up a smoking wreckage, even as we heap on more long-term liabilities. It is how you get Donald Trump … and whatever emerges in reaction to Donald Trump.
It is also a perspective that causes us to overlook opportunities that are right in front of our noses – to treat things like the legacy urban fabric of Rust Belt cities as liabilities, rather than as underutilized assets. It leads us, like the dog in the old fable, to drop the bone we currently have in favor of a better-looking one that turns out only to be a mirage.
Planners, of all people, should be acutely aware of all this, and be imbued with the sense and sensitivity to help point the way out of the abyss. I’ve known planners over the years with this kind of courage and insight. They need resources to help them not just understand the effects of micro-level changes in urban form but also to understand and tell the macro-level story of how America is changing and must change, so that they can help their communities find a way forward.
There are, as we argued in our report A New Way Forward, a vast array of tools available to policy-makers and the public to reduce VMT and address global warming – as well as a nearly infinite assortment of ways to combine those tools to best address the particular needs of particular places in a changing 21st century world. Studies like Stevens’ can provide important tactical insights, and for that, I am grateful. But the real and meaningful debates about the future development and design of our communities are much deeper, and the solutions are most likely to be found elsewhere.