Tag Archives: VMT

LA Land Use Patterns Help Reduce VMT

Sometimes you find things in the darnedest places. While reading Randal O’Toole’s testimony on Washington’s Growth Management Act (spoiler: he’s opposed), I see he references work by David Brownstone down at UC Irvine:

As University of California (Irvine) economist David Brownstone concluded after thoroughly studying this issue, the link between land uses and driving is ‘too small to be useful’ in attempting to save energy or reduce emissions.

Hmm, as someone they tell me was a Great Communicator used to say: trust, but verify. So let’s see what Brownstone has to say in his most recent paper:

The estimation results indicate that residential density has a statistically significant but economically modest influence on vehicle usage, which is similar to that in previous studies. However, the joint effect of the contextual density measure (density in the context of its surrounding area) and residential density on vehicle usage is quantitatively larger than the sole effect of residential density. Moving a household from a suburban to an urban area reduces household annual mileage by 18%.

I’ll leave you to speculate as to why O’Toole would cite authoritative sounding sources that, on closer review, clearly do not say what he would like you to think.

Nevertheless, the result of the Brownstone paper is very important: density on the census block level has a relatively small impact on vehicle miles traveled (VMT). Regional effects dominate. In other words, density is much more important on the regional scale than the local scale. If you want to decrease VMT, you need to increase regional density, not just build TOD projects at transit stations.

This study lends support to things we’ve explored from an intuitive perspective before (and data is almost always better than intuition). It explains how places like LA and Orange County can show up in lists of lowest household gasoline use* – even if you have to drive, you never have to drive very far. And it also shows a possible way forward for a region that shows up on lists of highest household gasoline use – the IE. Rather than focus on building TOD projects near transit stations, officials in the IE should upzone everywhere. They should allow things like Palms-style apartments and redevelopment of Cudahy-style lots the way they’ve been redeveloped in their namesake city. Because while the IE will probably never be able to emulate New York City’s travel patterns, it could certainly emulate LA’s.

*Note: 7 of the 10 worst gas guzzling cities are in the South (excluding Texas), which also makes sense in the context of my post on suburb types.

The Right Argument for Tolls

Aaron M. Renn, aka The Urbanophile, has an article out today about the need for tolling in Rhode Island. There’s no doubt that things in the Ocean State need some work; I was on the 195 recently and saw some terrifying levels of spalling concrete like that seen in the picture of the Warren Ave overpass. Renn pitches tolls for facilities like the Sakonnet River Bridge, and VMTs, which he calls de facto tolls.

I’ve written before about how VMTs do not make sense and do not actually solve any of the stated problems, so I don’t have anything to add there.

However, the issue of traditional tolling is another good one to discuss, and I think Renn’s article runs into trouble there. First, while tolls solve the problem of inadequate funding, they do not solve the other major issues that Renn identifies: corruption and incompetence. Toll money can be wasted just as easily as gas tax money; it does not force the state to stop deferring maintenance. There is no substitute for building competent governing institutions, and evidence would suggest Rhode Island has a ways to go with that.

Beyond that, though, Renn uses the same faulty user fee logic that I wrote about in the context of VMTs. Renn says that “it’s intuitively fair for those who use something to pay for it”, an argument frequently heard from progressive writers who don’t care for cars. As I said, this is a disastrous line of reasoning for progressive causes, since transit users don’t pay for the entire cost of transit either. We don’t provide roads “for free” any more than we provide bus service or public education for free.

There is a logical framework with which you can make an internally consistent argument that highway users should pay the entire cost of highways, but transit users shouldn’t pay the entire cost of transit. If you think that highways are not public goods, but transit is a public good, then you’re good to go. But if roads are not public goods, then there’s no reason for the public sector to supply roads at all, and the solution would be to just privatize roadways. That argument seems plausible regarding limited-access highways, but certainly not with respect to local roads and bridges.

If you think that roads are public goods, there are still arguments in favor of tolling. I think there are two, one solid and one marginal. The solid argument for tolling is congestion charging as a way of capacity management. Congestion is a negative externality of driving, and drivers should pay for it. Unlike VMTs, congestion charges account for a negative externality that cannot be properly captured by the gas tax.

The marginal argument for tolling is that users of expensive facilities, like bridges and tunnels, should pay more. This is a marginal argument because it partly relies on the faulty user fee framework. People who have children that require special education use more expensive school facilities, but we do not make them pay additional school tax. The strength of tolls for expensive facilities is that it forces local users to pay for the facility, instead of allowing them to fleece distant taxpayers who may not be paying attention, which provides an incentive to control costs. Note that this is also partly a substitute for competent governing institutions.

So again, like the gas tax, there are good reasons for tolling, but you should always think twice about the user fee approach.

Gas Taxes Are Not User Fees

I wasn’t planning on making the first post on this blog about gas taxes, but transpo funding is in the news a lot lately, and I’m getting sick of writing the same comment over and over again on Streetsblog, The Atlantic Cities, etc.

America’s infrastructure could use some sprucing up. If you live in the Northeast or Midwest, your bridges are rusting away from years of road salt. If you ride one of America’s big legacy transit systems, every day you’re putting your safety in the hands of components so old they can’t even be replaced, because no one makes them anymore.

So, where should we get the money to address this problem? That’s easy. Currently, the federal government can borrow money at negative real interest rates. Considering that any tax increase is going to have a negative effect on aggregate demand, which is the last thing the economy needs right now, it’s an open and shut case.

However, as is usually the case in America, the obvious policy solution is blocked by political intransigence, so lately the debate has been about how to raise more revenue. With a hike in the gas tax off the table, other ideas like a vehicle-miles tax (VMT) have been floated. The basic idea of the VMT is that everyone pays a fixed rate for each mile they travel, with miles traveled being recorded through low-tech means like odometer readings or high-tech means like GPS. For reasons I fail to comprehend, the VMT is getting a lot of traction in the progressive community, capped off by a recent GAO report that declared the VMT more “efficient and equitable”. But in fact, the VMT is a bad idea, and by supporting it, progressives are doing a huge disservice to the cause of better pedestrian, bicycle, and transit facilities.

The supposed benefit of the VMT is that it charges people based on how much they use roads – in effect it’s a user fee. Later, we’ll get to why that’s a bad idea in and of itself, but for now, note that the gas tax can be considered a de facto user fee. The more you drive, the more gas you use, and the more you pay. In addition, the gas tax has the benefit of making people who drive inefficient vehicles pay more. A VMT rewards people who drive Escalades at the expense of people who drive Volts. How is that more “equitable” than the gas tax? Sure, you could make the VMT higher for vehicles that are less efficient, but why go to all that trouble?

Which brings us to the second reason the VMT is a bad idea – it is complicated to implement. Any system, from low-tech odometer readings to high-tech GPS, would be subject to rampant fraud. High-tech systems like GPS would require every vehicle in the country to be outfitted with a VMT unit, at considerable expense. The GAO found that depending on the system, up to 33% of the revenue would be lost to implementation costs. Compared to a gas tax hike, which would lose 0% of revenue to implementation costs, how is that more “efficient”?

But wait, didn’t we say that a gas tax hike was off the table? Ah, now we’re getting somewhere. Why is a gas tax hike off the table? Because politicians decided it was off the table. So my question to progressives is this: do you really think that GOP pols will reflexively block any increase in the gas tax, but are perfectly willing to go tell the Agenda 21 crowd that they all have to install government-monitored GPS units in their cars? It’s just ridiculous. No Republican politician takes the VMT seriously. It’s a red herring, an issue that can be used to distract from the need to raise the gas tax. They’ll call for studies of the VMT, and then cite privacy concerns and implementation difficulties as reasons it won’t work.

Beyond all of this, though, if progressives engage in the debate using the “user fee” framework, we’ve already lost, no matter what the outcome. A lot of people have jumped on the “drivers should pay the full cost of roads” bandwagon just because they don’t like cars. But the obvious next step in the argument is that transit riders should pay the full cost of transit, bicyclists should pay the full cost of bike improvements, and so on. The likes of Cato and Reason, who love highway user fee concepts, already make that argument. If you look at the gas tax or VMT as a user fee, you’ve set yourself up for Randall O’Toole to make a convincing argument that gas taxes and VMTs should only fund highways.

Okay, so if the gas tax isn’t a user fee, then why do we have it? Simple. It’s a tax on the negative externalities, which include pollution, climate change, noise, and so on.

Here’s the right way to think about it. On one side, we have public goods that cost money – things like schools, roads, transit, etc. Some projects are better than others and more deserving of funding. On the other side, we need to raise revenue, and there are better and worse ways of raising revenue. The gas tax is a great way to raise revenue because it imposes a cost for generating negative externalities in direct proportion to the rate at which a vehicle produces those externalities, and because it already exists. A tax on bicycling would be a bad idea, because bikes generate positive externalities that we would like to increase. Similarly, a VMT punishes nascent alternative technologies that we ought to be encouraging. Note that by this logic, it is perfectly reasonable to have property or land taxes contribute to funding transportation – land is worthless without access to transportation, just like it is worthless without police to stop crime. We don’t demand that user fees pay for police service either.

Note that this framework also allows you to build a coherent argument for why the gas tax should be increased but transit fares should not. It obviates the “I pay for my roads so you pay for your bikes and trains” talking point – cycling and riding transit have positive externalities, so they should be promoted. It’s also the right way to argue in favor of HOT lanes and congestion pricing, which are taxes on the negative externalities of traffic congestion. The user free framework, on the other hand, is an argument you can’t win. It’s a trap.


As for the supposed fatal flaws of the gas tax, they are all weak arguments that fail to hold up to scrutiny:

  • Inflation: the easiest to dispense with. The argument goes that inflation has eroded the purchasing power of the gas tax. The obvious solution is to index the gas tax to inflation or construction costs or what have you, or periodically raise the gas tax. You know, like we did for 60 years.
  • Fuel efficiency: the argument here is that increasing fuel efficiency has made it impossible to raise enough money from the gas tax. Note that this is logically inconsistent with the idea that enough money can be raised from a VMT – as long as we are driving gasoline powered vehicles, a gas tax and a VMT can be seen as functionally equivalent, with the difference being that a gas tax is a tax on inputs and a VMT is a tax on outputs. Electric and natural gas vehicles may be an issue someday, but right now they are an inconsequential portion of the vehicle fleet. And at any rate, the data shows that in historical context, current increases in fuel efficiency are not out of scale with what happened in the 1970s.


  • Not popular with drivers: and the VMT won’t be either. Shockingly, people prefer not paying to paying.
  • Politically impossible: the same kind of thinking that leads people to argue for spending billions of dollars on trans-Hudson rail tunnels before maximizing existing capacity by through-routing.

At the end of the day, there is simply no compelling technical or economic reason to switch from gas taxes to VMTs. On top of that, a VMT would be a major victory for the kind of people who want to user-fee everything in the public realm. The VMT is counterproductive for progressive causes and we need to treat it as such. The answer is simple. Raise the gas tax because it is a charge on negative externalities.