Policy challenges of VMT in the USA

Dr. Travis Dunn of D’Artagnan Consulting* has prepared an interesting presentation (that he gave at the recent IBTTA Symposium on Mileage-Based  User Fees and Transportation Finance) on some of the challenges US policy makers when considering a shift from fuel excise taxes to distance based charging for road use (commonly referred to as VMT or Vehicle Mileage Tax**) .

He makes some fascinating points (sources in his presentation): 

- From 1994 to 2010, the real value of the Federal gas tax has declined by over 40% compared to the Highway Purchasing Price Index (the cost of paying to maintain and build roads). It has not increased in nominal terms since 1994. It is US$0.04/litre (US$0.184/gallon);

- At the state level, 18 states have also not increased their State gas taxes in nominal terms since 1994, 2 have reduced them (Connecticut and New Mexico) others have increased them. The average across all states is an increase of 2.8c/gallon since 1994. This roughly equates to a decline of around 25% in real terms, but obviously some states have seen far higher declines, and a few (notably North Carolina and Washington) have kept up pace with inflation and some. 8 states index their own gas taxes to inflation;

- Vehicle fuel efficiency has improved by around 15% in terms of gallons per mile since 1994, although it had declined between 1994 and 2004 by around 7%, this has been more than reversed from 2004 onwards. 

From this, he concludes that the timing for VMT to replace gas tax has been since that rise in efficiency, as it is a combination of inertia in indexing gas tax to inflation and the increased fuel efficiency of the fleet that is putting pressure on state and federal budgets dependent on these hypothecated taxes. 

He concludes there are a few more fundamental policy points: 

- Whilst VMT protects revenue from erosion due to reductions in fuel consumption per vehicle mile, it does not protect it from inflation. VMT will still need to be increased to offset that;

- Federal gas tax already means a complex set of cross subsidies between states. Four states get less than 90% of what they contribute in Federal gas tax (Texas, South Carolina, Utah and New Jersey), whilst 17 states get more than 110% of what they contribute (DC gets nearly five times what it pays, Alaska gets nearly four times). VMT may mean states are likely to want to keep the revenue they generate, and perhaps it means rates might vary from state to state;

Yet if VMT was introduced to replace gas tax, the picture would change, assuming funding stayed the same. Texas would gain by paying $411 million less than it does now (reflecting its fuel consumption is high relative to distance travelled). Florida would lose by paying $510 million more than it does now (reflecting the opposite). 

Obviously there are other dimensions that Travis doesn’t take into account, which could help mitigate this, primarily because discussion of VMT has been as a straight replacement for gas tax.   However, as a tool it can be used to significantly improve pricing, after all vehicles already pay differently according to fuel consumption, which reflects weight, topography, road network and demand conditions.

I'd argue some of the minimum components of VMT should be:

-    VMT shouldn’t be a flat rate for all vehicles, because gas tax isn’t. It would be highly preferable to match VMT to also reflect weight, so that whilst cars pay a common rate, trucks and buses would pay more to reflect the greater wear and tear they impose on roads, and the capital requirements for highway structures to be built to cope with such weights.   Other taxes (such as tyre taxes and registration fees) that may be proxies to reflect this would need to be taken into account, and adjusted downwards if desirable;

-   VMT should reflect geographic variations in cost.  Even if there is a federal VMT, it seems reasonable to be cost oriented by having VMT rates that could vary from state to state because costs vary.  As long as this is oriented towards cost recovery, it should be less controversial, but it ought to also mean revenue raised in one state is spent in that state (or on routes approaching it). 

-  VMT at the interstate level can be varied to reflect recovery of the costs of major new capital works, replacing conventional tolls.

His whole presentation is here (PDF), it's worth a look through.  If you have any questions, his contact details are at the end.

Obviously it is a long way to go before VMT will be rolled out on a wide scale replacing existing taxes, but the challenges driving the concept are here now and unless some of the key dimensions of the current system and what VMT will bring are understood, it will difficult to move forward and the current problems will simply get worse.   Travis's presentation raises just some of those issues, which highlight the obvious but often ignored point that the existing system produces "winners and losers", another system will change who they are - the bigger challenge is making sure whatever comes ahead, that VMT can be seen as being fair.

* Disclosure: I have known Travis for about five years and have done work for D'Artagnan Consulting.

** There is ample debate about whether it should be a tax or a fee. Advocates of the word “tax” are coming from the view that it replaced existing taxes by being a new way to raise revenue for government, which will probably (but not necessarily) spend the money on transport. Advocates of the word “fee” see it as changing the characterisation of highway usage and payment for highways toward buying road use as a service payment, which is more commercial, and could be to a private provider. In this case, I’m neutral as the usage of the term will depend on the jurisdiction and the policies adopted.