Economists vs. Realists on Mileage-Based User Fees

Finding the best way to replace gas taxes with mileage-based user fees

For nearly 25 years I’ve been an advocate of road pricing, for all the usual reasons: it’s a far better user fee than the blunt instrument of fuel taxes, it’s the most effective solution for freeway congestion, and it could even include charges for negative externalities.

The last decade has seen growing support, both among economists and within metropolitan planning organizations and state departments of transportation, for the goal of shifting highway funding from fuel taxes to some kind of mileage-based user fees. Simultaneously, there is growing evidence from the increasing number of high-occupany toll (HOT) lanes projects that congestion pricing actually works the way economists said it would, significantly reducing congestion in priced lanes, spreading out peak loads, and encouraging shifts to alternative modes for peak-period travel.

Yet as the movement toward implementing some kind of mileage-based user fee (MBUF) system gains momentum, I’m struck by what seems to be a huge disconnect between the systems being modeled by academic economists and the proposals emerging from field tests and demonstrations involving actual motorists. This disparity was very much in evidence during the sessions at the 2012 Annual Meeting of the Transportation Research Board.

On the one hand, there were presentations by economists on the results of complex modeling exercises. One presentation explained how a proposed per-mile charge was constructed by separately estimating the cost of a list of negative externalities (conventional tailpipe emissions, greenhouse gases, congestion, noise, etc.) plus (almost as an afterthought) a small charge for the cost of maintaining the highway infrastructure. The total came to 6.4 to 9.6 cents per mile, with only 0.3 cents of that being the infrastructure cost-all the rest was externalities. The researcher also noted that, alas, the impact of this charging structure would be regressive, unless there was significant redistribution of the revenues.

In stark contrast were presentations from researchers involved with focus groups and field experiments in which ordinary motorists were asked to consider (or simulate) operating a vehicle under various forms of per-mile charging systems instead of today’s gas taxes. The general thrust of these presentations was on simplicity and on paying for the costs of building and maintaining the infrastructure.

At the conclusion of one such study, the researchers suggested implementing a three-tiered system aimed at gaining public acceptance:

  • The base tier would be a flat annual fee based on an assumed level of miles driven. This would presumably be set somewhat higher than what many people drive, but has the advantage of requiring no technology on the car at all.
  • The second tier, which people could opt into, would require an annual odometer reading, done during the annual vehicle inspection. This would be attractive to those who drive less than the tier one assumption. It also requires no new technology in the car.
  • The third tier would be aimed at those who drive partly out of state (and who would be over-charged if the fee was based solely on their annual odometer reading). They could opt for on-board mileage recording using technology that could distinguish between in-state and out-of-state miles driven, administered by a third-party vendor (for privacy protection reasons).

To be sure, I deliberately selected these two examples as polar opposites, but they serve to illustrate my point that much of the work being done on MBUFs in academia is almost entirely disconnected from what is going on in the real world where transportation researchers are trying to figure out realistic implementation strategies for making this historic shift from unsustainable fuel taxes to a sustainable replacement based on miles driven.

In the real world, making such a fundamental change in a highway funding system that has endured for over 90 years will be very difficult. Much of the public is highly skeptical of government, especially when it comes to “taxation,” what they see as Big Brother-type invasions of privacy, and “social engineering.” Hence, the more functions that we expect a mileage-based user fee system to support, the more complex it will be and the greater the odds of political rejection.

Consequently, I think it’s incumbent on those of us who favor road pricing to do some hard thinking about the trade-offs involved. We can certainly imagine a federal law mandating a costly GPS box in every car that distinguishes not only which road each mile is driven on but which lane is used, and that also knows the vehicle’s propulsion source, weight, and even the number of passengers on board, so that the charge can be fine-tuned to account for every possible impact. But my considered judgment is that this type of road user charging system has close to zero chance of being implemented.

So let’s back off a bit and reconsider our objectives. The number one problem facing our roadway system is the non-sustainability (and inefficiency) of a highway funding system based on the gallons of petroleum-based fuel consumed. Unless we can transition to a more sustainable funding source, our hugely valuable, multi-modal highway infrastructure is very much at risk.

In my view, urban traffic congestion (especially freeway congestion) is the second major problem we need to address. The 2011 Texas Transportation Institute Urban Mobility Report puts the annual cost of just the lost time and wasted fuel at $101 billion. When you add to that the lost productivity in urban areas, the true cost is likely double that amount.

There is, of course, a set of negative externalities from the use of motor vehicles-smog and particulates, CO2 and other greenhouse gases, noise, and runoff. My first suggestion for simplifying the mileage-based user fee transition is to leave those problems to be solved by other tools. Catalytic converters and corporate average fuel economy (CAFÉ) regulations have already led to major reductions in conventional and CO2 emissions, and the latest CAFÉ standards will cut them in half again over the next several decades. Noise and runoff are increasingly being addressed by design standards required for environmental mitigation. Hence, one huge simplification would be to eliminate this set of externalities from needing to be part of the system requirements for MBUFs.

Freeway congestion is easily addressed with current technology-low-cost transponders and license-plate imaging. This kind of charging could be phased in a lane at a time, starting with individual express or HOT lanes and building those into a seamless network of lanes with variable pricing, aimed at trips with the highest time value. Once such networks are in operation, it would then be time to pursue modest peak-period pricing of the rest of the freeway system, to extend the benefits of pricing to the majority of far less time-sensitive trips. Again, no costly mandated GPS box is required for any of this.

Transponder and license-plate tolling can also be extended to limited-access inter-city highways, such as the Interstates. Here the rationale would be the growing need to reconstruct this most important component of our highway system as it begins wearing out, estimated to cost $2-3 trillion. Tolling would be added to an Interstate corridor only when that corridor needs significant rebuilding and modernization. I have called this approach “value-added tolling.” It is being proposed now for a major reconstruction of I-70 across Missouri, Illinois, Indiana, and Ohio, in a project that would include the addition of truck-only lanes.

If we can address the funding and pricing of limited-access highways and freeways via low-cost all-electronic tolling, that leaves only the need for a baseline MBUF to handle all other roadways. And because what I’ve proposed above would address both congestion pricing and externalities, what is left could readily be handled by the kind of several-tier basic system outlined earlier.

There is still an enormous amount to be done in figuring out how to transition from fuel taxes to mileage-based user fees. What I urge transportation researchers to do is to reconsider the prevailing view that one high-tech system must be defined and imposed, to solve all highway funding, congestion, and externality problems. That path most likely leads to political failure. But by separating the problem into its component parts, I think we can address each of them in ways that are both cost-effective and politically do-able.

Robert Poole is director of transportation policy and Searle Freedom Trust Transportation Fellow at Reason Foundation.