A couple of days ago, the Council for the city of Jacksonville voted on extending a six-cent gasoline tax for another twenty years. While this might seem like a relatively minor occurrence, it got me thinking about the merits of whether an excise tax should be levied on the sale of fuel, particularly on the state and federal levels. Typically, the gasoline tax acts as a de facto imperfect user fee to help fund transportation-related projects, e.g., highways and other infrastructure, which is important because the infrastructure system is the lifeline of the American transportation system that provides transport, job, economic growth, and mobility. One of the issues with the gasoline tax is that overall, inflation-adjusted revenues have been on the decline.
The reasons for decreased revenue have to do with more fuel-efficient cars, inflation eating into purchasing power, as well as more people taking public transit, all of which make the gasoline tax less effective. We should be worried about a deteriorating infrastructure because with it comes higher costs, more congestion, increased accidents, and reduced mobility. Infrastructure is a metric of a country's economic wealth, which it makes me wonder if we should stick with the gasoline tax or if we should try something else. Aside from its unpopularity, the gasoline tax comes with its issues, including the regressive nature of the gasoline tax, inelastic demand (Bento et al., 2009; Li et al., 2013), improper reallocation of federal funds, increased prices of other goods, inability to disincetivize driving during peak hours and decrease congestion (Congressional Budget Office [CBO], 2011, p. 10), noticeable market distortions in secondary markets (Taylor and Van Doren, 2007), and an unfair burden on rural and remote areas [and the poor] (Ferguson, 2007). Stricter fuel-economy standards might seem like a good idea, but a gas tax is up to six times more effective than stricter fuel-economy standards (Karplus et al., 2013). Perhaps we can index the gasoline tax to inflation. Using the Consumer Price Index wouldn't be the best idea because it overstates inflation and has a substitution bias. Alternatively, the Institute on Taxation and Economic Policy shows that the tax can be adjusted to the inflation of construction costs to more accurately reflect the costs of infrastructure projects. Regardless, it could help recover some lost funds. Even so, indexing it to inflation is no guarantee of the quality of the roads (Hartgen et al., 2010).
There were some other ideas that I came across while doing research. One is to install toll roads. Although that would act more like a user fee than the gasoline tax, the upfront costs to replace gasoline taxes with toll roads would be astronomical. The Rand Corporation suggests that instead of implementing a tax on gasoline, we should implement one on oil (Crane et al., 2011), which has the advantage of making the tax more diffuse. There is the possibility of implementing a private-public partnership, but looking at other public-private partnerships, it may or may not work.
The idea that seems to best work is to tax the miles in what is known as a milage-based user fee. The Government Accountability Office conducted a study in 2012 showing that a milage-based user fee is more efficient than a gasoline tax. It's equitable (CBO, p. 15) and it's fiscally sustainable (O'Toole, 2012), not to mention that it cuts back on financial crisis issues. However, it comes with its issues. One has to do with tracking the miles. If one uses the odometer, that leaves it open to odometer tampering, i.e., fraud and tax evasion. A GPS might solve that problem, but it would considered an invasion of privacy because the GPS would track where and when people would drive. There have been implementations in which the tracking system only collects certain data, but with what the NSA did with regards to acquiring metadata, it's difficult not to be skeptical. Alternatively, there is the possibility of using your own GPS, smartphone, or other system to wirelessly track your milage without the government invading privacy. Furthermore, the tax would be collected at the gas pump, which means that electric cars would be exempt. Implementation would eat up a good chunk of revenues because adding in trackers into each car would be costly, especially with tracking and assessing a user fee (CBO, p. 18).
On the whole, I would say that we should relegate funding of the roads to the states because the federal government muddies things up. At least that way, there would be more competition and innovation of ideas. To make the infrastructure system more market-based, and thereby more efficient, the solution is going to be a portfolio of the aforementioned policy alternatives. For example, start the milage-based user fee with trucks only. Although they make up four percent of automobiles, they are responsible for twenty-five percent of milage travelled (CBO, p. ix). Increase the petroleum tax while decreasing the gasoline tax. States can install more toll roads while decreasing the gasoline tax. Whatever the mixture of policies looks like, one thing is for sure: the gasoline tax is on its way out.
June 22, 2014 Addendum: Brookings Institution senior fellow Clifford Winston wrote a paper on how the private sector can improve our infrastructure system. Worth the read.
July 11, 2014 Addendum: The Left-leaning Center for American Progress even realized the wonders of the mileage-based user fee in their latest report.