What Will Replace the Gas Tax?

As many have argued in this blog, the gas tax has its flaws as a vehicle for raising money to fund ongoing roadway maintenance and construction. The flaws will become increasingly evident as Americans shift to vehicles with better gas mileage — or to electric automobiles that don’t consume any gasoline at all. Some, including myself, have argued that Virginia should move to a system that charges automobiles on the basis of Vehicle Miles Driven: The more you drive, the more you pay.

The issue has been this: Does the technology exist to track Vehicle Miles Driven cheaply and unobtrusively? We have an answer: Almost. A Canadian company is developing technology, which it expects to roll out in a year, that will transform the way people think about transportation.

Toronto-based Applied Location Corporation was profiled by Business 2.0 magazine in a list of 11 disruptive new technologies. Founder Bern Grush has created a satellite-based system, Skymeter, for toll collection, traffic congestion management, and pay-as-you-drive insurance. Writes Business 2.0:

He says he’s created an algorithm that corrects for the noise and missed signals to which GPS is prone, especially in urban canyons, so Skymeter can reliably pinpoint a car’s location to within 1.5 meters. Once that becomes possible, cities can start to manage traffic better by charging more for driving during peak times. (London already does this, but with an expensive system of cameras.) “The problem is congestion,” Grush says. “We need to send pricing signals to motorists to drive at alternate times.”

Skymeter is disruptive more for the new economic models it might spur than for what it might displace. For instance, in addition to congestion tolls, it could be used to charge for parking anywhere in a city, even where there are no meters. Or the technology could enable insurance companies to offer better rates to safer drivers, since it records speed, time of day, and driving route. …

Grush thinks his product could ultimately help to supplement or even replace the fuel tax now used to pay for the upkeep of our roads. “The need to fix road financing and congestion is so acute,” he argues, “that a dramatic shift will necessarily occur.”

Virginia is running out of money to maintain its road network. We need to envision a future that abolishes the gasoline tax and replaces it with a tax based on Vehicle Miles Driven, adjusted perhaps for the weight of the vehicle. This tax would cover the maintenance of the road network. As maintenance costs increase (or conceivably decrease, if we outsource maintenance), the tax would adjust automatically each year.

Funds for new construction would come from tolls — ideally, tolls that varied the price according to congestion levels. With such a two-track system, Virginia could ensure that (a) all maintenance costs were covered, (b) that tolls encouraged people to drive during off-peak periods or to shift to carpooling, buses, rail or telecommuting, and (c) that the money for new construction came directly from the people who used the new facilities. We could then turn the conversation from how much money we need, and where to get it, to a more productive dialogue: how best to align our transportation systems with our human settlement patterns.

As lawmakers discuss Virginia’s transportation funding in the soon-to-convene special session, they need to be thinking about the larger context: emerging technologies and new, user-pays transportation-funding models that are sweeping the world.

Share this article


(comments below)


(comments below)


22 responses to “What Will Replace the Gas Tax?”

  1. Anonymous Avatar

    There are a lot of unknowns with a system that would toll people based on congestion levels.

    Case in point:

    Mr. Commuter, who lives in Frederick County, VA., drives every day to his 9-5 job in Fairfax County. In order to get there, he drives through a minimum of 4 localities. In this case it’s Frederick, Clarke, Loudoun, and Fairfax. He must work 9-5…no exceptions.

    The traffic is steady the entire time but it becomes extremely bad once you arrive in Loudoun and continue through to Fairfax. On the way home the traffic is extremely bad from Clarke until he arrives home in Frederick County.

    Based on this, how is VDOT going to determine which locality can access a toll on his commute? Why would Loudoun and Fairfax be able to do it while Clarke and Frederick not be allowed to do it? Is there some definition of what congestion is?

    How in the world could all 4 localities make him pay a toll and still expect him to make a living?

    The battle over who could/could not toll a car in their locality would be never ending.

  2. Anonymous Avatar

    As you propose such things, make sure you are consistent in your approaches to sacrificing privacy.

  3. The gas tax already charges you more the more you drive, and it charges you more if you drive a heavier vehicle, or one that is less efficient. If it turns out that higher fuel prices are effective at doing just that, which seems to be your complaint, then we need to adjust the fuel tax upward to match the costs of maintaining the roads to the current technology.

    While you are at it, extend the fuel tax to all fuels, not just gas. That will encourage people not to live in fuel guzzling homes, and if they charge their electric car at home, it still gets taxed.

    Are you seriously suggesting that a guy who drives his Moped to work should get taxed on VMT the same as the guy with a giant SUV? I’m sorry, but that is nuts, and for everybody else the same argument holds but as a matter of degree.

    Sure, lets create a whole new business, based on satellites that cost $4000 dolars in fuel per pound to launch, in order to save on fuel. And then let’s use it to charge for parking where we don’t bother to now because there is plenty of space. And you don’t think that is a new tax?

    Boeing just won a major new border control contract, and did it rely on satellites or unmanned aerial vehicles? Nope, they proposed to put up a bunch of low tech towers. This is Boeing mind you.

    We had toll roads once, and traffic light controlled grid street systems, and electric trolleys, and roundabouts, and we mostly got rid of them for rasons that are still valid today. Each of them was new technology at the time.

    Just because it is new technology doesn’t mean it’s a good idea. This one ranks at the bottom of the list of the very worst. How am I going to get charged for the weight of my dump truck, empty or full? My gas gauge knows the difference automatically – no satellite required.

    Think of what this will mean to those that use the least congested roads. If the people who use the roads have to pay for all the maintenance, many rural roads will shut down. At the same time those that use the most valuable and expensive roads will get charged not only for the maintenance but for congestion charges on top of that. But that’s OK because they can shift to even more inefficient systems like buses, rail, or even car pooling.

    We already know how to align our transportation system with human settlement patterns: supply more transportation where people choose to live and choose to work.

    This is going to be the worst of the worst. People who take Larry’s attitude that they don’t need to pay as much because rural roads cost less, will soon find out there are fewer people to support them, too. On the other hand, they may soon find company from all those who flee the double whammy imposed in the city.

    As a result our settlement patterns will be changing so fast that transportation can never catch up.

    I just can’t see anything good in this idea, unless your goal is to control everything and everybody down to 1.5 meters.

  4. Interesting idea. If the technology were available, would you want to also base the cost per mile driven on number passengers transported? For example, someone carpooling to work with 3 other people in the car would pay less than someone driving just themselves. It may even encourage public transportation or mass private transportation. Maybe not.

    I’m not sure what do on the commerical traffic, and obviously the technology would need to be developed.

  5. James Atticus Bowden Avatar
    James Atticus Bowden

    I agree with Ray. The gas tax is a good user tax on consumption.

    We have a ways to go on diminishing returns (better engines lower the amount of gas used for mor miles) to starve the maintenance of roads and highways. (I know the alarming estimates of projected costs, but compare that to estimates of revenue.) If and when the time comes, you can raise the gas tax or add tolls to pay for the maintenance.

  6. Larry Gross Avatar
    Larry Gross

    Some of the issues pointed out… like efficiency of the vehicle and number of passengers… convince me also that we’re still a ways away from charging on a per vehicle basis. (I’ll throw one more into the pot… SLUG pooling.. ad hoc carpooling).

    I think also… there exists, some confusion as to the PURPOSE of whatever methods/technology is chosen.

    There are two issues:

    1. – congestion/transportation network efficiency/”shape” rush hour by depressing the “peaks”

    2. – conservation – less fuel use

    I don’t think these two things are directly connnected but others do.

    Look at HOV lanes and ask folks what their primary purpose is.. and you’ll not only get these two answers but folks will vehemently disagree as to which one is the primary purpose and which one is secondary.

    Electronic TOLLs by themselves do not address the environment per se. It’s all about how many cars are on the road at a given time – a vehicle is a vehicle regardless of it’s other characteristics.

    Electronic TOLLs combined with a dedicated restriction lane for high occupancy DOES take that into account.

    Note there is actually a bill in the GA that seeks to designate all HOV lanes as HOT lanes. This represents one level of thinking at least by the sponsor of it.

    The quickest and easiest way to deploy electronic tolls would be on existing infrastructure…. First… while any plans to add lanes .. proceed..on a much slower basis.

    For instance, any law that enables HOT Lane TOLLING (congestion pricing) could go into effect immediately wherever EZ-Pass and Smart TAG already exist..- basically just made a software change in the billing algorithms…to vary t he toll charge per hour.

    Expanding HOT lanes on existing infrastructure designated as HOV but not currently tolled would only require installing high-speed TOLL gantries… and connection to the existing EZ-Pass computer network to enable congestion pricing.

    … and this might be a method to generate to then back-fit new HOT lanes…

    How these various scenarios play out might actually make this special session a bit more exciting that previous sessions.. especially if the House is on this electronic TOLLING track and the Senate is on a new money track.

    But who is more likely to get thrown out of office…? the guy who votes for a tax increase.. or the guy who votes for electronic tolling?

  7. We already have a system that provides feedback to not drive during peak times — it’s called congestion.

    I think the gas tax could be set by law to provide a certain revenue, and adjusted once per year based on the same rules as escrow is set on your mortgage.

    But if you want to add some level of charge that isn’t based on fuel efficiency, why not add a tax to the property tax collection, and require people to report their mileage every year.

    We have state inspections, so you could instead take the mileage at that time as a check — in fact they DO take your mileage.

    Raise the penalties for messing with your odometer, if you have to.

    The only down side is that it doesn’t allow time-based metering, but again I think we get that from congestion already. I know I never drive during those hours unless I’ve had a bout of temporary insanity.

  8. Virginia Centrist Avatar
    Virginia Centrist

    one problem with this is that over the last 20 years, people haven’t been switching to higher mileage vehicles – they’ve been going to minivans, then suvs.

    sure, we’ve seen a blip on the radar as gas prices climbed, but i doubt it will last. the latest fad car will appear and people will go back to gas guzzlers.

  9. Anonymous Avatar

    Check this out;


    Thought someone here would like to take a look at it.

  10. I think Larry is right, as usual. There are two issues and they are not completely related.

    The congestion tax is a special issue for special places.

    The gas tax is more generally applicable and more generally fair. I know for certain that if the GA comes up with a plan that increases registration fees and insurance fees rather than actual usage, then three of my four farm trucks are going to the dump, and the biggest, most useful, most wasteful one will remain.

    That makes no sense to me. It’s bad enough as it is. I ought to be able to insure me as the driver with my most expensive vehicle, and have it transfer automatically whenever I’m driving one of the others. I’m only using one at a time, why charge me more for being efficient?

  11. I don’t think so VC.

    SUV are an affront to our sensibilites, so maybe that makes them stick out. (Not to mention aggressive behavior that seems to accompany these behemoths.)



    Shows clearly that fuel consumption per vehicle dropped off sharply after 1975 and has barely increased since. It also shows that we have taken advantage of the increased efficiency in greater miles driven.

    In that respect Jim gets credit for his usual comments on VMT. But, it also shows that we have run a thirty year experiment that shows higher prices lead to more efficiency. What we haven’t done is raise fuel taxes sufficiently to account for the greater efficiencies.

    How we use that efficiency may leave something to be desired, or maybe not.


    Shows clearly that the ratio of energy use for transportation compared to other energy uses has remained nearly constant since 1950. So maybe we travel as much as we do in order to accomplish what we need to do.

    The parent site is
    http://www.mnforsustain.org/table_of_contents.htm I have not looked at much of this, but it appears to be an excellent source, being mindful of the usual amount of agenda driven drivel.

  12. Larry Gross Avatar
    Larry Gross



    awesome map! thanks for posting.

    I encourage others to take a look – in one graphic, one can easily see and understand the jobs/traffic condundrum in No.Va.

  13. Larry Gross Avatar
    Larry Gross

    The are at least 3 problems with raising the gas tax:

    1. – the higher it goes, the less gasoline will be purchased – and most analysts believe..that at the end of the day – there will be a net decrease in revenues.

    This is exactly why most states, including Va legislators are looking for alternate paths.

    2. – raising the price of gasoline will not alter congestion levels. Folks will continue to drive at the busiest, most crowded time-periods and cut out other trips.

    3. – whether one agrees or disagrees with 1. and 2. above – the reality is that an increase in gasoline taxes will not happen in the GA.

    It’s not in the cards. What is Plan B even if you have to hold your nose?

    …or those who prefer a gas tax .. is plan B essentially to stand pat and wait until a majority of legislators will approve a gas tax increase?

    I’m just asking the question… folks.. cuz my impression is that the special session is going to produce “something” and it probably won’t be a higher gas tax – though who knows… strange and unpredictable outcomes are not uncommon at Va’s GA sessions.


  14. Larry Gross Avatar
    Larry Gross

    There’s news in this mornings papers with respect to the U.S 460 issue.

    Specifically, bids were received from the companies interested in building and operating U.S. 460.

    * – Bids are 1 Billion or more – 2-3 times VDOT’s estimate (deja vu all over again per Yogi – VDOT has NO credibility).

    * – depending on one’s point of view, the fact that all of the companies are asking for public funds and help with financing could be a sign that the project is not economically viable as a PPTA.

    * – estimated tolls are from $4 to $13 bucks over the 55 miles.

    * – the cost per mile is about 20 million ( 1 billion / 55 miles).

    * – back of the envelope analysis

    One billion spread out over 50 years would require a payback of about $60,000 a day.

    Assume 20,000 cars a day = $3 toll
    or 10K cars = $6 toll.

    This does NOT include financing costs NOR maintenance costs both of which could add substantially to the toll.

    So the TOLL itself could double or triple…

    and THAT is the rub because the higher the toll – the less folks will use US 460 and that part is the 64 dollar question for private investors.

    PPTA Tolls on high volume roads like I-95 or I-81 are feasible because the base load traffic is very high but on a road like US 460 – the base load traffic may not be as high AND it would compete with I-64 which probably would siphon drivers during lower volume periods.

    This is exactly why PPTA is not going to be the answer to every “perceived” road need.

    A good question is that if a road is found to be not feasible as a PPTA – does that mean the road is really not needed?

  15. Toomanytaxes Avatar

    Perhaps, all of the facts suggest that transportation cannot be fixed in parts of Virginia (and in many other places) at a reasonable cost. Maybe it is not feasible to move this many people back and forth from home, to work, and return every working day. This seems to be true, whether we drive or take mass transit.

    Today’s technology simply cannot handle the job.

    At the same time, I don’t think Ed Risse has made his case that by moving everyone into apartments in a few locations, we solve the problem — unless we recreate company towns. If you live in Metrowest, you can work for one of five companies. If you leave their employment, you also need to move.

    I’m not arguing that we stop building transportation facilities. But we need to encourage more good-paying jobs in outlying areas and extensive telecommuting for many “knowledge” and “information” jobs.

    I’ve hauled my standard up the flagpole. Have at it!

  16. Larry Gross Avatar
    Larry Gross

    Here’s a shocking statistic:

    “Between July 2005 to July 2006, the region (NoVa) added nearly 75,000 jobs”


    Does this mean .. that NoVa is ALSO adding 75,000 more cars a year to it’s transportation grid?

    good gawd!!

  17. Jim Wamsley Avatar
    Jim Wamsley

    I see one flag on Toomanytaxes flag pole. I need two for a solution.

    Create an abundance of Condo’s and Apartments, served by good schools, rail transit, and lively neighborhoods in a few locations. This should bring the cost of housing and taxes down and meet the needs of about a third of the population.

    Encourage home centers and more good paying jobs in outlying areas. This also should meet the needs of a significant part of the population and reduce taxes.

    The big problem is these two solutions require changing land use and transportation thinking. They have to be approached as synergistic solutions. A “lets chose one and reject the other” approach will lead to failure.

  18. It is probably more than 75,000 cars, based on the current jobs to vehicle ratios.

    The graphs I posted show that your argument number one isn’t true. Higher gas prices led to more efficient vehicles, but we drove more miles and consumed more gas, even in the face of higher prices.

    Looking at the first graph would lead you to believe that we could raise the gas tax quite a bit, reduce VMT a little and still be moeny ahead.

    But then, looking at the second graph leads you to wonder what reducing transportation energy use would do to the rest of the economy.

    However, your point number two and three are right. Except that we apparently haven’t raised the price enough to reduce driving, so we don’t know what the effect on congestion would be.

    You only have to reduce a little traffic to have a big effect on congestion. Likewise, you only need a little more capacity in the right places to have a similar effect on congestion. (Although that situation may change over time.) Or, as TMT points out, you only need to move a few jobs to move the congestion someplace else.

    It looks like what the GA is going to come up with is new fees that are not taxes and are not related to driving inorder to “solve” the problem.

    As Larry said, good gawd!

  19. Toomanytaxes Avatar

    JW – a couple of data points for the discussion. I reviewed the developer’s website for Metrowest near the Vienna station. The proposal would add 2248 residential units, including 1600 high rise units. Plans call for 100,000 square feet of commercial/retail and 300,000 square feet of office. Fairfax County approved 5288 new parking spaces. The website did not post any target prices for the housing.

    I’ll add that the Orange Line is already near its capacity, but WMATA does plan to add more rail cars as a result of the 2004 county bond issue.

    Does this help or make things worse?

    It strikes me that adding 5288 more vehicles does not make things better.

  20. Toomanytaxes Avatar

    JW – My understanding is that all (or at least most) of the new parking spaces will be built underground and probably won’t be accessable for commuter parking.

  21. Larry Gross Avatar
    Larry Gross

    re: What we don’t know.
    How much parking the commercial/retail needs and how much of this need can be met by sharing residential spaces during the day and office spaces during the evening.”

    Traffic generation for most conventional (well defined) development is documented in the Traffic Generation manual of the Institute of Transportation Engineers.

    I’m not sure exactly what the relationship is between Traffic Generation per se and parking – perhaps Jim W does but the manual will state for a particular use – how many auto trips a day it will generate and I assume this also means that parking is part of the planning equation. For instance a 500 unit apartment complex might generate 10 trips per residence but I’m not sure how the calculation for number of spaces is done.

    Many places already only allow one space per unit with a spillover area and I presume that the number of spaces that go along with a unit – are market-driven. For instance, a married couple with two jobs is going to likely need two autos in a conventional (non TOD) housing complex so they would not buy/rent where only one car was allowed and, if necessary may have to make economic and/or locational decisions to meet their perceived needs.

    I’m not sure where the .8 vehicle per residence comes from for TOD… is it in the traffic generation manual?

    Also .. is the part about double-duty of parking spaces in that manual or is it basically an assertion by TOD advocates?

    I think it sounds like a good concept but wouldn’t it basically “work” by charging for the spaces and/or allocating spaces .. that can be used but have to be vacated… in the evenings? I don’t think this will “work” on an honor system… you’d have to have rules and enforcement and penalties for non-comformance.

    Is the point here to do something that DISCOURAGES auto use in a TOD or is the approach to INCENTIVIZE transit use – and it’s not necessarily and either/or path.

  22. The traffic generation manual of the Institute of Traffic Engineers is not the Gospel. It has come under fire for usiong gross generalizations in specific instances where it does not apply. It has also come under fire for using statistics to four decimals that imply a precision that is unfounded.

    Lary is right. It isn’t an either/or path. It could be both and they could be mutually interdependent. But in the end there is only one most cost effective solution, and artificially restricting parking spaces guarantees that you won’t find that solution.

Leave a Reply