Funding Roads and Bridges to Perdition
Gasoline taxes, road tolls and highway infrastructure spending are issues at the forefront of a lot of heated debates in state legislatures across the country. I am going to write about the issue a couple of times this week. Some lawmakers want to raise sales or other taxes to pay for infrastructure and others want to increase gasoline taxes and other “user fees” to pay for it. The highway infrastructure spending and revenue issue can illustrate classic principles of sound fiscal and economic policy so it is too bad that the debates have generally taken the “low road” by framing the issue almost entirely as either one of “who wants to raise taxes and who doesn’t,” or “who wants to makes roads and bridges safe and who doesn’t”.
User fees are a good thing and it is sound fiscal policy to have the users of roads pay for them via gasoline taxes, road tolls, and other fees that reflect an individual’s usage of roads and bridges. When general revenues are used to pay for roads and bridges people who don’t necessarily use them wind-up paying for a portion of highways and subsidize the usage of roads of those who travel them a lot. When you subsidize something you can bet you are going to get more of it than you would have gotten without the subsidy and in this case that means more travel on roads which, of course, means there will be more need for roads and spending on roads and that means more subsidy and that approach is surely a road to perdition.
It was nice to see New Hampshire rank high in a recent report (issue brief) by the Tax Foundation on the percentage of highway spending that is funded by user fees like gasoline taxes, tolls and other fees. Unfortunately, in making good points about user fees, the Foundation draws the wrong conclusion about the data it uses to make them. That happens a lot when you use bivariate analysis to draw conclusions in a multivariate world. Instead, using multivariate (regression) analysis on the data, it becomes clear that it is less the use of good principles of fiscal policy that results in states paying for a higher percentage of the costs of highways with user fees, than it is a function of the volume of federal government grants they receive. So a cursory look at the Tax Foundation’s report can give NH a sense of superiority in fiscal policy over many states (while I generally think that is true about NH it is not so much in this case), and especially over Vermont because that state funds just under 20% of its highway spending with user fees compared to NH’s 42%. The real reason those percentages are what they are is that Vermont receives about 64% more federal highway funds per capita than does NH ($220 to $134 in 2010). The chart below shows the simple relationship between the percentage of highway spending in a state that is funded by gas taxes and user fees and the amount of federal highway funding per capita in each state.
States like NH that fund a higher percentage of highway expenditures with user fees do generally receives lower amounts of highway funds from the feds (the data point slope downward to the right). There are even more intervening variables, like the amount of federal highways (by mile) and as a percentage of all highways that are in a state but still, by far, the amount of federal highway funding per capita is the best predictor of the volume of highway spending per capita in each state. The amount of motor vehicle-related user fees per capita were a distant second but still significantly related to highway spending.
Almost everyone agrees that NH’s (and every other state’s) roads and bridges are in need but I don’t think the debate is ever going to be about the wisdom of user fees versus general revenues in paying for highway infrastructure. It is too bad because if it were we just might reduce the need for more spending in the future.