In Defense of a VMT Tax

Matt writes:

I’m a supporter of higher taxes, so if I were a Senator and someone was bringing a vehicle miles traveled tax to the floor I suppose I’d be prepared to support it. But the sporadic bouts of enthusiasm for this idea are really baffling. As public policy, a VMT has no advantages whatsoever over higher gasoline taxes. Raising a given quantity of funds through a new VMT rather than through higher gas taxes creates (a) more administrative headaches, (b) more civil liberties concerns, (c) fewer environmental benefits and in exchange you get nothing. They say higher gas taxes is a hard sell politically (though given that elites from both parties privately acknowledge that it’s a good idea you could always just not sell it) but a VMT would be just as hard a sell.

As Andrew Samwick says “The appropriate tax instrument to make up for declining or inadequate gas tax revenues is…a higher gas tax rate.” If at some future date vehicles have become so incredibly fuel efficient that no gas tax rates could possible raise the revenue we need, then that would be an excellent problem to have. And it’s certainly not a problem we have right now.

I think he makes a number of good points, but I don’t think the case against a VMT is that foolish. The problem with a gas tax is that its purpose is unclear. It’s not really a user fee, but it’s meant to fund transportation. It’s kind of a tax on negative pollution externalities, but its heart isn’t really in it since its value has been falling over time. I also think it’s kind of a problem that revenue to fund transportation varies inversely with oil prices, and I think the civil liberties problem is significantly overblown, since pretty soon most Americans will be walking around with GPS in their pockets anyway.

The biggest advantage to a VMT is that because its implementation has to be “smarter”, it offers you the flexibility to do some very useful things. In particular, GPS-based VMT taxes can be used to implement variable congestion tolls. When you get in your car in the morning, your NAV system could plot several routes, the rates for which have been set based on congestion levels, and it could tell you which way is cheapest. The ability to price roads individually, on a varying basis, and without erecting toll booths or traffic cameras all over the place is a huge advantage.

Obviously, that kind of thing would be very difficult to pass at the federal level. On the other hand, we haven’t raised the gas tax in nearly 20 years, so it’s not clear to me that attempting to try something new is a totally silly idea.


  1. Alex B. says:

    It’s not just that a VMT tax would be difficult to pass politically, it requires a massive new set of infrastructure just to collect it. The gas tax does not, and that’s one of the gas tax’s primary advantages.

    What’s frustrating is the stubborn refusal to even consider raising the gas tax in the interim and look at VMT as the future – say, 5-10 years down the road. Any proposal to raise the gas tax is dead on arrival. It’s going to have to be a part of any real plan to move forward simply because it can be implemented right away.

  2. Rob says:

    This is an issue whereI feel like Yglesias has been all over the board. In the past he’s argued that a motor fuel tax is no better than a broader tax on energy. The case you lay out might be an easier political sell where a system of tolls already exists, because it could be argued as a better and more efficient means of collecting those tolls.

  3. Nick J says:

    Do not forget about the ease of collecting gasoline taxes. In California, the state and federal gasoline excise taxes are paid at one of the over fifty distribution terminals. When you see a tanker truck (jobber) that fuel has already been taxed and paid for before it arrives at a retail station. Seems much more feasible and less costly to collect a higher gasoline tax from 50 locations than collect a VMT tax from every car.

  4. Tom says:

    The “gas tax is not a user fee” argument seems overly technical to me. I mean, yes: it doesn’t track road consumption (though highway trust fund dollars don’t really cover maintenance to the extent that this argument implies). Some of the money goes to transit. Various external factors affect collections, as you point out.

    But I still think it makes sense to think of it as a user fee. That’s what it’s *trying* to do — make highway users pay for highways (or at least to carry more of the burden than others). I don’t think there can be too much doubt about that.

    I still think that the GPS proposals are unrealistic. Now that I see the congestion-pricing hope it holds for you, I understand your enthusiasm. But I just don’t see how it could be technically workable. The capital costs are *way* higher than, say, EZ-Pass — not just for the GPS and data logger, but for the associated payment infrastructure. The technology is brittle and could be easily gamed by users. And this is to say nothing of the civil liberty concerns, which I think are enough to doom this idea on their own.

    You might have some luck (and save more money) making odometers more fraud-proof and levying the VMT as part of the vehicle registration process. But I don’t think that would offer enough benefits to merit a switch from the simplicity of the gas tax. Certainly it wouldn’t bring any of the congestion-pricing benefits you envision.

    I think a Snow Crash-style system of P3 highways with dynamic rates is much more likely to be the way that congestion pricing comes to America. People don’t seem to mind abandoning their privacy when it’s to a corporation rather than the government.

  5. David says:

    I think I agree with Matt here: the technological, political, and civil liberties concerns present a barrier that makes discussion of a VMT tax more academic abstract than practical.

    Further, while imperfect, in my opinion the gas tax presents consumers with incentives to change behavior on two dimensions: decreasing vehicle miles traveled, and increasing the fuel efficiency of the vehicle they drive.

    The VMT would only incentive decreasing the number of miles traveled.

  6. BruceMcF says:

    If the advantage of the VMT is that it is supposed to make congestion charging easier – that is absurd. Far less disruptive to only require the congestion charging equipment on those benefiting from the congestion charge with less congested streets.

    The fact that someone who knows that Yglesias supports congestion charges has whispered this reason in his ear suggested that there is something else going on, since the argument presented is so flimsy.

    Oh, wait, … the beauty of the gas tax is that it “feels” like a user fee, while being a net cross-subsidy from urban motorists to rural and suburban motorists.

    That is, of course, also its flaw: as part of a system to encourage suburbanization, it is part of the substantial increase in the share of cross-subsidy recipients and reduces the share of cross-subsidy providers.

    Of course, the Federal VMT would retain the cross subsidy from urban motorists to rural and suburban motorists. Indeed, maintaining the subsidy by urban motorists of rural and suburban motorists seems to be the point.

  7. BruceMcF says:

    Oh frack, shouldn’t write after midnight. Matt’s right, someone’s whispered in Ryan’s ear.

  8. sponson says:

    The existing method of taxing per gallon of fuel perfectly fits the priorities and takes care of such contingencies as people burning fuel in generators, compressors and other devices that never move down the road. It also provides precise and proportional (also known as “fair”) incentive to conserve and reduce usage. I can see no possible outcome of a VMT as implemented by our corporate-directed Congress that won’t end up being more unfair than the current system.

  9. SeanDC says:

    Considering a VMT as a replacement for a simple fuel tax is absolutely nutty. It would have vast inefficiencies and implementation costs, and it does not incentivize efficiency or conservation at all.

    It’s like if the power company charged you for having lights on, rather than for your actual electricity consumption. How many energy-efficient compact flourescent light bulbs would be installed if we moved to this pricing structure?

    The gas tax should be increased significantly (think $1.50 per gallon eventually) in a well-telegraphed amount over the next 5-10 years, with *all of the incremental tax revenue* going back to taxpayers in the form of reduced payroll taxes. Taxpayers would not see an overall impact on their finances, but the pain at the pump would cause significant behavioral shifts and provide a real boost for efficient vehicles, efficient driving habits, etc.

  10. JC says:

    Robin Chase: it’s not necessary or desirable to have only one or two sources of funding for transportation.

    So by that way of thinking, our choice need not be either/or. If the fuel tax and VMT tax both create desired (and different) behavioral incentives, why not implement both policies?