« Know Your Products | Main | HIV vs. Cancer »

A Bad Idea, But A Real Question

Image Courtesy DOE/NRELThe recently-appointed head of the California Department of Motor Vehicles, Joan Borucki, advocates replacing the 18 cents-per-gallon gasoline tax in California with a "tax by the mile" system charging a fee for distance traveled in California, as determined by a Global Positioning System device installed into every vehicle, and collected at the gas pump. The reason? California car buyers have put so many hybrids and other high-mileage vehicles on the road that the amount collected in California in gas taxes since 1998 has declined by 8% (inflation adjusted), even while miles traveled has increased by 16% in the same period.

This would mean that someone considering switching from a 15 mile-per-gallon truck to a 45 mile-per-gallon hybrid sedan for a 30 mile commute would pay the same amount of tax for the trip.

This system is currently being considered in Oregon, under the "Road User Fee" label, and a test of 250 vehicles in Eugene, OR, is set to begin later this year. Other states are watching the results of the Oregon tests and the California discussion. As proposed, however, the "tax by the mile" system won't fly. There are too many problems with the plan's design. Nevertheless, the longer-term issue of declining gas taxes still has to be dealt with. Read on for more.

There are, in my view, a number of problems with the "tax by the mile" proposal, some more obvious than others. It reduces to some degree the financial incentive to move to a more fuel-efficient vehicle, and even if the per-mile tax is relatively small, it will be perceived as a serious disincentive. It also doesn't accurately reflect the different damage inflicted on roads by vehicles of different weights -- the aforementioned 15 mile-per-gallon truck may weigh two to three times as much as the hybrid, and cause significantly more damage per mile. It's subtly regressive, as many people (in California, at least) live far from their employment because periphery housing is more affordable on smaller incomes. And the idea of using a GPS device as a tracking system is just silly on a number of fronts: it's appalling from a civil liberties perspective, it will cost tens or hundreds of millions of dollars to get the devices into cars already on the road, and it wouldn't work for cars brought in from out of state (either by tourists or people seeking to evade the law). In short, the likelihood of the "tax by the mile" proposal actually coming about in this way is virtually nil.

But the appropriate push-back against this particular proposal masks a larger question: as advocates of higher-mileage, newer-technology vehicles become ever more successful, how can states continue to pay for transportation services? Gas taxes are often (if not always) devoted to paying for transport maintenance and development, and not just road work. Alternative and mass transit expenses are often (depending upon the state) paid in part by gas taxes. If gas consumption declines (as we would like to see happen), gas tax revenues go down, meaning that services supported by gas taxes lose funding.

Ultimately, a road use fee will probably be necessary. It should, however, be a function of miles traveled and vehicle weight. Furthermore, a GPS device is not necessary for this to work. Most recent vehicles have electronic odometers; it would not be difficult to install an RFID-type chip a microprocessor with a short-range transceiver to watch accumulated miles and keep track of when the car was fueled up, and to communicate that figure (modified by a hard-coded vehicle weight factor) with appropriately-enabled gas pumps. The per-gallon gas tax would remain for vehicles without such chips, and would be the default add-on when the gas pump can't communicate with the vehicle. There would still be ways to game the system (and gas stations just over the state line would do booming business as people gassed up before returning home), but a system such as this would mitigate the disincentive for fuel efficiency by including weight as a factor, and be both less-intrusive and less-costly than GPS.

It's not a very adventurous idea, though. Let's think big: assume that, over the next decade, the combination of people demanding more efficient cars and carmakers actually supplying them pushes the average miles-per-gallon way up for vehicles on the highway; as a result, transportation funds from per-gallon gas taxes drop precipitously. What innovative funding for transit would you suggest to replace gas taxes? It should be fair, roughly comparable in money generated to per-gallon gas taxes, and not include perverse incentives to misbehave. Got any ideas?

Comments (21)


IMO, if taxes are to rise as more efficient cars are on the road, the incentive to get an efficient vehicle will also rise.

A win-win situation. More efficient cars, and the same revenue (due to tax increases.) Cleaner air, better roads (efficient cars are generally lighter and less damaging to the road surface.) Those who want to drive gas guzzlers will have to pay more, but it costs more to allow them on the roads in higher maintenance costs, and their size means fewer actual people can fit on the roads.


Political difficulties aside, why shouldn't we just raise the per-gallon gas tax?

Also, I would think it'd be much cheaper to measure miles traveled when people renew their vehicle registrations, either at the DMV or some licensed facility like AAA or a smog check station. Vehicles with untrustworthy odometers could pay a simple weight-based registration tax on top of the license fee.

This is another one of those ideas that radical right-wingers are pushing along with things like "HOT Lanes" - co-opting notions like "congestion pricing" and "pay as you go" schemes. Unfortunately, a lot of well-intentioned people get duped by these things and support them without clearly seeing the consequences and reasoning behind them.

If a state wanted to solve a revenue shortfall from higher mileage vehicles (which is a specious argument to begin with, since VMTs are up and the average fuel economy for new cars sold is either stagnant or down), then the state should simply raise the gas tax rate, as Yami suggested.

But seeing as people can write off a $100,000 vehicle that isn't even subject to emission or fuel economy standards because it's too big, and the most that someone can write off for a hybrid (federally) is $2,000 (which is being phased out), then we know where policy priorities are with government these days.

What I also find telling is that the same people who propose shifting to a mileage-based gas tax also oppose pay-at-the-pump auto insurance.

I just got done participating on the Minnesota Citizens League Transportation Study Committee, and we wrestled with all these issues. Unfortunately, the Committee findings end up tacitly supporting HOT lanes and mileage-based gas taxes, simply because they *seem* to adhere to congestion pricing and road-use approaches to pricing.

The problem with gas taxes is that its a wedge issue for progressives - half of the people support them because it's better for the environment, while the other half oppose them because they are a form of regressive taxation which hurts the poor and middle class. But we end up with $2/gallon gasoline anyway, except the price premium is used to make a whole host of anti-democratic nations flush with cash, instead of providing a tax revenue stream to invest in a transportation system which weans us off of our heavy dependence on automobiles and petroleum.

The more we can do to create options which shift automobile expenses from fixed to variable costs (like carsharing and pay-at-the-pump auto insurance), the more likely we'll see positive change. But as long as most auto costs are fixed, people have a strong incentive to drive to amortize those costs.

If anyone's interested in the details of these things, here's a spreadsheet showing the "Automoble Corporate Average Fuel Economy (CAFE) Standards versus Sales-Weighted Fuel Economy Estimates, 1978–2003", which basically means comparing the CAFE standards to the actual fuel mileage of the new auto fleet:


As you can see, the national fuel economy for new vehicles for 2003 is the same as it was in 1992 - obviously because of the explosion in the sales of light trucks (ie, SUVs) during the 1990s. This other chart:


shows the CAFE standards for light trucks, which are about 7 mpg less than for passenger cars.

Meanwhile, vehicle miles traveled (VMT) per household keeps increasing:

while obviously the number of households has increased as well.

Meanwhile, petroleum consumption in the transportation sector is up 20% since 1990:

So, I really don't see where they can make an argument that they're seeing revenue shortfalls in gas taxes - certainly not at the federal level.

Jamais Cascio:

Reasonable points, folks, so here are some hopefully reasonable respones:

Simply increasing the gas tax over time has the advantage of being straightforward, but the disadvantage of being terribly politically unpalatable. It's going to be hard to convince legislators to take that course directly. However, it may end up being the default solution to the gas tax revenue problem when the DMV chair's awful plan is formally proposed and eventually rejected.

Yami, the drawback to odometer checks at registration time is that it doesn't account for miles traveled outside of the state. People who drive cross-country on a summer vacation would be hit as hard as someone putting on the miles doing long commutes within the state for the same total distance. As this is a state-level revenue and service issue, this would generate a lot of opposition.

Joseph, while vehicle miles traveled are certainly up (by 16% in CA since 1998), the LA Times article I linked to in the piece states that revenues from gasoline taxes declined by 8%, inflation adjusted, in that same time period. The issue of reduced gas tax income is apparently a real one, at least in California.

Even if you want to argue that the decline in revenues while VMTs are up has a different source -- and I'm perfectly willing to accept such an argument -- the larger point that higher efficiency vehicles will lead to lower fuel tax revenues, assuming there are no changes to the system, still stands.

Jamais Cascio:

And thanks for the links, Joseph. Useful information, if a bit discouraging.

As you note, those are datasets for the federal level, not California.

Assuming that the LA Times' sources are correct, and gas tax revenue in CA declined over the last 7 years while VMTs increased, what (other than higher overall efficiency) could have caused that?

" the LA Times article I linked to in the piece states that revenues from gasoline taxes declined by 8%, inflation adjusted, in that same time period."

That's part of the problem, too. Gas taxes are set in absolute dollar amounts - not percentages like most other taxes. Therefore, over time, if everything else is constant, the real dollar amount of the tax revenue will go down. Cumulative inflation since 1998 is 15.9%, for example, so the revenue shortfall in California is simply because the tax doesn't adjust for inflation itself. The current dollar revenues to the state are obvioulsy up during that period.

When people buy gasoline, almost no one microanalyzes the cost components. Ask anyone on the street what their state and federal gas taxes are, and I guarantee less than 2% of the people will get it right. They also won't know that federal gas taxes increased more during both the Reagan and Bush presidencies than during the Clinton presidency. So, I don't understand how upping the gas tax, at least by indexing it to inflation by making it a rate instead of a fixed amount, would be politically unpalatable, especially considering the alternative involves tracking your vehicle with a GPS unit.

Jamais, here's some information on historical gas tax data by state:


which I found here (the California DOE website):

As you can see, the last increase in the federal gas tax was in 1994 (a 3.1 cent increase). I remember Clinton first proposed a 50 cent hike and members of his own party were the ones who beat that down to 3.1 pennies.

Meanwhile, average state taxes since 1994 have only increased 2.8 cents. Consequently, as you can also see, the inflation-adjusted value of those taxes are 5.6 cents lower than in 1995.

As of 2001, Monthly Gross Taxable Gasoline Sales in California are 2 billion gallons per year higher than in 1992.


The most current monthly data I could find was for September 2004, here:


The amount of gasoline consumed in California was 2.2% higher than in September 2002, despite the fact that gas prices were about 60 cents higher in 2004 than in 2002.

According to this:

the last time the state excise tax was raised in California was in 1994 (seemingly part of a phased-in total increase of 9 cents, starting with a 5 cent per gallon increase in August 1990).


What about business vehicles and other large vehicles that drive through the state on a regular basis. How do we charge for a fully loaded rig going from Seattle to Arizona?

The RFID at the gas pump idea is really bad from a privacy point of view. Do we really want private corporations and the goverment maintaining databases of everywhere we refuel, when we refuel, and how much we bought?

Should my employer be able to find out where I drive when I'm on my own time? Should the state be able to track every citizen's movement so easily?

Jamais Cascio:

How do we charge for a fully loaded rig going from Seattle to Arizona?

In the "tax by mile" proposal now floated, that remains to be answered. In my suggested variation, the big rig would pay the same kind of diesel tax it pays now.

The RFID at the gas pump idea is really bad from a privacy point of view.

I should have been more detailed -- the radio chip version wouldn't need to include any identifying information at all, simply a "this many miles have been traveled since last refueling, and this vehicle is rated at X thousand pounds" value sent to the gas pump, which then would tack on the appropriate miles~weight gas tax in place of the default per-gallon tax. I used "RFID-type" in a broad and inaccurate way, to refer to a small processor/transceiver combo chip, not to mean an actual ID tag.

Should my employer be able to find out where I drive when I'm on my own time? Should the state be able to track every citizen's movement so easily?

No and no. Those are exactly the kinds of questions which will doom the DMV chief's GPS-based proposal.

Jamais Cascio:

[I've modified the text of that paragraph to make the above point more clear -- I don't mean RFID identification tags, but a simple chip & transceiver setup.]


wouldn't it be just as easy for the car to tell the pump how many miles per gallon its getting (most car 'brains' notice that nowadays), and have a variable tax rate based on that, to promote more efficient cars? (Oh I forgot, the job of the government isn't to lead in the right direction, but to put more money in the pockets of the corporations that bought it.) I can't be the only person to have thought of this, surely. This wouldn't hurt the poor and middle class any more than having an emission standard included in vehicle inspection which, what, a handful of state do already, since it would only apply to brand new automobiles and would thus have to be phased in. (would it?)


I don't think putting a GPS in every car is going to be more "politically palatable" than a sensible tax increase plan (stepped over time). I mean, we know that some people are going to reject the GPS for personal/political/privacy reasons ... what are they going to do, send people to jail?

Anyway, I drive in California and the problem is not the few Prius, it is that they are still outnumbered by new huge SUVs.

The best plan would be to announce a stepped program, of increasing fuel taxes over five years, and let people plan accordingly. That will have the immediate affect of shifting purchases, but that's pretty much what we need.


What they are trying to do is very simple and frankly the only thing they can do. Increase the tax base without being canned in the process;/

They know they have to change the way they fund road work because soon hydrogen and made at home fueling systems will be taking over and no manner of at the pump system will work as there might not even be a pump anymore.

We have new batteries comming in within 5-10 years that could explode the level of plug ins used.

We have reforulation stations that will turn any natural gas line into a fueling station.

And we have an old taxation method that cant cope with any of it and the basic need to fund roads. Nope fuel taxes wont cut it they will in fact be forced to tax road use.

Mars Saxman:

What on earth is "pay-at-the-pump auto insurance"?

Wintermane, if there's no pump anymore, how are they going to implement a system which collects a mileage tax at the pump?

I find these "hydrogen is coming, let's tax cars by the mile" argument disingenuous, to be kind about it. Hopefully that's not the argument you're making.

Mars, it's also referred to as "Pay As You Drive (PAYD) Insurance. Here is a page that discusses the concept.

The basic idea is that insurance rates don't vary much based on miles driven, so that the less you drive, the higher your per mile cost of auto insurance. Instead, the basic proposal is to pay insurance premiums as part of gasoline purchases.

In that way, it runs into the same problems that these mileage-based gas tax proposals have, which is how to calculate the cost. Personally, I favor a smart card or bluetooth system, such that the car can identify itself and indicate it's fuel mileage, either using standard EPA ratings or actual mileage as calculated by the car itself. Then, when you buy a certain amount of gasoline, it knows how many miles that represents and charges accordingly. Risk adjustments could be made to the premium based on a driver's record or even based on the automobile self-monitoring actual driving behavior.

In any case, the system gets rid of the problem of uninsured drivers, creates a common risk pool for auto insurance, and eliminates a lot of the auto insurance bureaucracy in the process. As you can imagine, the insurance industry isn't exactly happy with the concept.

In terms of ecological benefit, this shifts what was a fixed cost to a variable cost, such that driving a car becomes more "pay as you go" instead of sinking all the costs upfront then amortizing them by driving. As someone who works at home and doesn't use his car too often, though need to own one because of the lack of public transportation and other car alternatives where I live, I pay an absurd portion of my yearly auto costs to insurance. I'm penalized for not driving much.

A mileage-based gas tax is also pay-as-you-go, but so are existing excise taxes on gasoline. The problem, as many here have stated, is that it disincentivises fuel efficiency.

Somehow I didn't get the link to work in that last post. Here it is:


Mars Saxman:

Ahh, that's a very interesting idea, though fancy solutions involving complex electronic gadgets always make me suspicious. A simple odometer report seems perfectly adequate to me; here in Washington, you already have to take your car into a state-authorized emissions test facility every two years anyway, so the infrastructure necessary for periodic validation of the odometer readings already exists.

Actually, come to think of it, why not just make this proposed mileage tax part of the yearly license plate renewal process instead of using fancy, failure-prone wireless technology that would have to be retrofitted into all existing cars? Every time you renewed your tabs, the DMV would simply subtract your current odometer reading from your last reading and charge you some per-mile rate based on the result. The DMV could certify odometer inspection facilities like the state EPA currently certifies emissions test facilities.


It isnt just hydrogen that is causing the issue to pop up its also methane and reformed methane as well as direct electricity.

In the end the only way they will do it is to track who goes where. Or find a totaly different funding model that doesnt deal with the fuel at all.

Rose Bittle:

Taxing miles driven instead of gallons used will be devastating to truckers and those who have to live miles from their home to work. My daughter lives in Stockton, but travels to Martinez to her job. She wants to move to Martinez, because of her work and because she loves it there, but she cannot afford a home or rent in that area. So charging her (and those like her) for mileage would be detrimental. The price of gas is astronomical for them now, but charging miles would be more so. I pray that this is rethought, and decided against. I'm retired so it doesn't bother me, but my heart goes out to those still working. Thanks.


This page contains a single entry from the blog posted on February 15, 2005 10:52 AM.

The previous post in this blog was Know Your Products.

The next post in this blog is HIV vs. Cancer.

Many more can be found on the main index page or by looking through the archives.

Powered by
Movable Type 3.34