Monday, March 16, 2009

Rethinking the Gas Tax.... Badly

Fred Hiatt suggests that gas taxes should be supplemented by, and probably ultimately replaced by, a tax based upon miles driven. This would entail installing a GPS device in every vehicle on the road, that would be tracked and monitored by tax authorities.
It would keep track of where you drove your car, and when, but the data would not be shared beyond the vehicle so privacy would be protected.
That's reassuring.

The current gas tax, as everybody knows, works like this: You buy gas, and a percentage of the purchase price goes to taxes. If you drive a more fuel efficient vehicle, you pay less per mile in taxes. If you do not, you pay more per mile. If the gas tax produces a revenue shortfall, it's possible to increase the tax. Now let's look at Hiatt's proposal:
Whereas 20 years ago user fees - gas taxes, tolls - covered as much as 75 percent of the wear and tear on roads and other direct costs, today the ratio is down to 60 percent, Atkinson says. People are being subsidized to drive - and that's before you even count the indirect costs of noise, traffic, wasted time and pollution.
So here, nothing that couldn't be addressed by a simple increase in the existing gas tax, rather than creating a parallel system of taxation.
It could be set to charge more per mile driven for Hummers than for Civics....
Something that happens by default under the present system, as Civics are vastly more fuel efficient than Hummers. So far, we're just adding technical and bureaucratic complexity to the current system. So what's this really about?
It could be set to charge ... more during rush hour than in the middle of the night; more for driving on congested bridges than on empty roads.
So what Hiatt is actually proposing is a congestion tax. He draws a parallel to another existing means of addressing road congestion - toll roads. He doesn't - and probably can't - explain why it's superior to add GPS tracking to every vehicle in the nation, and creating a new tax bill for every vehicle owner in the nation, rather than adding a few new toll roads in congested areas.

Here's where it gets ugly. Auto manufacturers, truck drivers and trucking companies, and similar groups lobby against higher gas taxes on inefficient vehicles; other groups, or maybe even the same auto manufacturers, will argue for tax holidays or subsidies for certain "fuel efficient vehicles". Cities lobby against suburbs, claiming a higher cost of road maintenance; suburbs lobby against cities, claiming that their taxpayers are being asked to subsidize urban roads. People who "have no choice" but to drive during rush hour or on congested roads will respond to peak pricing by protesting that they're being unfairly taxed, and perhaps also taxed beyond their means. And so it goes....

There are areas of this country that would benefit from some form of congestion pricing to discourage vehicle traffic, particularly during peak hours. There are even more areas of this country that could benefit from improved mass transportation so that people have a reasonable alternative to driving to work during rush hour. I think it would be better to address those issues directly, rather than creating the costly, cumbersome system Hiatt proposes.

Also, what are the chances that privacy protection would survive any sort of major public incident. What are the odds that concerns about driver privacy would have trumped the analysis of driving records to find the freeway snipers? To find out if suspects in the Unabomber case or anthrax mailings were in the vicinity of a post office where a mailing occurred? Does Hiatt believe that the agency holding the tracking information would reply, "Sorry, that's private"? (Even assuming that they would make that response, backed up by legislation, when confronted with a subpoena from somebody who was tracking suspected infidelity of a spouse in a divorce case.) Is this where we get the usual response - a shrug, and the declaration, "Well, I have nothing to hide"?


  1. One of the blogs you link to had an adulatory post up about John Stossel, with a series of clips from a recent special he did. I say that with some degree of amazement that you would link to somebody who shows that level of Stossel worship, and also because the second clip (the last I could bear to watch) ties into this subject.

    Stossel was essentially arguing that people's commutes would be easier if there were more toll roads - and specifically if more public roads were sold to private companies to manage as toll roads. Yet quite obviously unless more roads were built, that would not be the case. Putting a public road into private hands as a toll road may lighten traffic for those who pay to use the road, but with the obvious consequence of increasing the number of people using the roads that remain public. If the toll were so low as to deter no one from using the road, congestion would not be affected.

    Stossel also suggests that these highway deals generate huge amounts of revenue, which is true to some degree. The state sells a highway that could produce $80 million per year in toll revenue to a private company, putting $400 million into the general fund. The state then spends the $400 million, with future budgets being short the actual or potential $80 million in revenues that now go to the private company. It's a great fake way to balance a budget at the expense of the future, but it's not the miracle cure Stossel pretends - it may be an atrocious long-term business decision.

    Sell is, of course, a misnomer. These are more typically lease deals, 50, 75, 90-year leases. The company can depreciate the entire cost of the purchase over the first fifteen years or so of the lease, so as to minimize any tax payments. The company may also be able to walk away from the lease at any time without penalty (or with no meaningful penalty - if the lease is the corporation's only asset, it may be impossible to collect a penalty), potentially leaving the public with the cost of repairing a neglected toll road (which can then be leased to the successor company, and the cycle starts over). Governor Edward G. Rendell of Pennsylvania engineered a lease of the turnpike that allows the private company to default, consequence free, at any time. Only the taxpayer is on the hook.

    You know what would be a better solution in almost all cases? Simply raising the toll. Something the private company is going to do anyway.

  2. As I'm sure you know, the fact that I link to a blog doesn't mean I agree with any particular post, or any of its posts at all.


Note: Only a member of this blog may post a comment.