Roads and transportation are costly. Building ten miles of a four-lane rural interstate highway costs about $200 million. For an eight-lane urban interstate, it is closer to $1 billion. Roads and transportation also come at a high environmental cost. The transportation sector accounts for about one sixth of world carbon emissions and the absence of roads is central to the process of taming (or spoiling) wild places. On the other hand, roads and transportation are the bedrock of the modern world. They facilitate the trade and specialization of tasks that is fundamental to economic growth since the industrial revolution. Devising transportation policy to balance the economic benefits of roads against their economic and environmental cost is an important and formidable problem.
Over the past several years, my coauthors and I have been conducting research to quantify the impact of road construction on various aspects of our modern world. A past issue of PERC Reports reported on research establishing that the total miles driven in a city each year increases in lockstep with the extent of a city’s road network. Other research done at PERC examines the reasons why driving is faster in some cities than others and estimates the benefits that might accrue to society if we were to adopt a simple scheme for pricing access to roads (as we price access to other goods and services, like candy bars or housing). In a paper forthcoming in the Review of Economic Studies, also supported by PERC, we examine the effect of interstate highway connections between cities and of interstate highways within cities on patterns of trade.
More concretely, we analyze data describing trade-flows between a sample of large US cities and the interstate highway network. The central statistical problem is that of “reverse causation.” If we observe that cities with extensive highway networks trade a lot, we do not know whether the highways are built to accommodate already high trade flows, or if the construction of highways causes trade. Much of our research effort is devoted to developing statistical procedures to solve this problem so that we can be sure that we are estimating the amount of trade that highways actually cause.
We find a 10% increase in the length of the highway network within a city causes about a 5% increase in the weight of the city’s annual exports but no measurable increase in its value. That is, improving a city’s highway network causes a city to become more specialized in the production of things that weight more per dollar of value. We confirm this pattern using 50 years of employment data. Cities with more extensive highway networks employ a higher share of their residents in industries that make heavier goods. It is not obvious that this is a desirable outcome.
We know from other research that an increase in the extent of a city’s highway network has a small effect on the city’s employment: a 10% increase in highway miles causes about a 1.5% increase in employment over 20 years. Taken together these results suggest the construction of new highways in US cities, like the one proposed for Nashville, are unlikely to have the large beneficial effects on economic activity that proponents claim, and that these benefits will not obviously be larger than environmental and construction costs.
Taken together, the available research on the effect of the interstate highway system, and transportation infrastructure generally, suggests that the “roads as economic development” policy deserves careful scrutiny if not outright skepticism (although it does not address the value of road maintenance, which will likely fare better in a cost-benefit analysis). Among possible investments of public money, new transportation infrastructure (and highways in particular) appears to have far lower and less certain return than alternatives such as education or health care for the young. Although most of the research currently available is based on data from the United States, on the basis of this research the call for developing county infrastructure put forth in the UN’s millennium development goals also deserves careful examination.
Matthew A. Turner is a 2013 Lone Mountain Fellow and a professor in the Department of Economics at the University of Toronto. His current research focuses on the economics of land use and transportation.