There’s been a bunchofcommentintheblogosphere today about hiking gas taxes—with the rough consensus that it’s ok environmental policy, tough on the poor, and politically risky (though perhaps not quite as unthinkable as it once was).

So it’s interesting to note that Oregon—often considered a policy innovator among US states—is in the middle of an experiment that could eventually lead to a repeal of the state gas tax.

Oregon’s transportation department is recruiting volunteers to test a system that would charge people based on how far they drive, not on how much gas they use. The trial will test two rate structures—some participants will pay a flat rate of 1.2 cents per mile, while others will pay a variable rate depending on whether they’re driving during rush hour; a control group would continue to pay normal gas taxes. (See here for details, or if you’re interested in volunteering.)

  • Our work is made possible by the generosity of people like you!

    Thanks to Deborah Clark for supporting a sustainable Northwest.

  • The state is interested in this sort of approach for a bunch of reasons, but if I read things correctly, they’re mostly worried that gas tax revenues are poised to fall, perhaps significantly, over the upcoming years.

    Here’s the issue: if gas prices remain relatively high, or keep rising over time, economists project a gradual per-capita decline in gas consumption, as people replace their cars with more efficient ones. And if cars keep getting more and more fuel efficient, then total gas revenues could actually fall, even as the demands on the road network increase. Ultimately, states may be forced to choose between continual gas tax hikes and persistent funding shortfalls for roads—both of which are bound to make lots of people unhappy. Under a pay-by-the-mile system, however, transportation funding would be keyed to how much people actually drive—so the state would still keep up its funding no matter how efficient the cars get.

    Obviously, shifting from gas taxes to mileage-based taxes has some signficant downsides. First, the gas tax does create a slight incentive for fuel-efficient cars; getting rid of it could put Hummers on a more even footing with hybrids. And second, it’s not entirely clear that a falloff in highway funding would be a bad thing. Obviously, most drivers want the roads maintained in good working order; but, in my mind at least, a lot of highway spending seems like it’s pretty wasteful.

    That said, we’ve beenprettyinterested in the pay by the mile idea, despite the potential downsides. Fully developed, the technology could facilitate two innovations that could be far more powerful at promoting fuel conservation than the exisiting gas tax. First, the same technology used to track mileage for taxing purposes could also be used for a more comprehensive congestion pricing system—which could simultaneously clear up congestion, reduce driving, and promote bus ridership by keeping streets and highways flowing. And second, it could pave the way for Pay-As-You-Drive car insurance, which would have the same effect on driving overall as roughly doubling the cost of gas.

    If you want to get really fancy, you could even fine tune the pay-by-the-mile taxes, to increase fees for cars with the worst pollution or CO2 emissions, the most road space (SUVs require longer stopping distances, and tend to use up a little more space on streets and highways), or the worst safety records.

    Of course, pay-by-the-mile taxes suffer from one huge drawback: they’re much more complicated than gas taxes, not just because they require special technology but also because a properly “fine tuned” system—one that accounts for all the different externalities of driving, ranging from pollution to congestion—could be pretty incomprehensible to the average driver. And mileage-based taxes would be vulnerable to all sorts of political shenanigans, as car manufacturers would jockey for special exemptions or rates for certain kinds of vehicles. All of which means that, even though I’m very excited by the tests, I’m not yet ready to support the idea—and certainly not until we see how drivers really react to the system.