Politics and infrastructure policy

Aerial view of roads and equation

In their Barcelona GSE Working Paper (No. 1057) “The Political Economy of Transportation Investment,” Edward L. Glaeser and Giacomo A.M. Ponzetto develop a model of the political economy of transportation investments to explain theoretically the historical transition of infrastructure policy that the United States experienced over the course of the twentieth century.

As Altshuler and Luberoff (2003) describe, America has gone from a period of overbuilding to a period of underbuilding and over-abatement of the downsides of construction. Their book Mega-Projects: The Changing Politics of Urban Public Investment highlights three different phases in America’s 20th century urban infrastructure building. In the first phase, which lasted until the 1960s, massive projects such as New York’s Lincoln Center were constructed in the center of metropolitan areas. These projects were mostly funded with federal support and there was very little concern about local opposition. In the second phase, community opposition became organized and activists succeeded in blocking most large-scale projects. The third phase saw mega-projects return, but with far more attention to community concerns, so construction costs ballooned to pay for compensating benefits for local residents.

In order to explain this three-phase transition, Glaeser and Ponzetto present a political economy model of transportation investments that highlights the role of inattention and nuisance. They analyze how the inattention of voters to federally-funded projects in other states, and the nuisance that local residents experience because of a project being undertaken—in the form of noise, air pollution or congestion—affect the equilibrium amount of investment.

The model

The paper considers a spatial model that includes three different geographic units: a city, its suburbs and the rest of the country. The main focus is on investment in a transportation system in the city that benefits city and suburbs and has the possibility of generating positive spillovers for the rest of the country.

Each location has a homogenous population. Each agent makes a constant number of trips using the transportation system, and transportation investment reduces the time required for each trip in the city, thereby increasing the welfare of agents in all locations. On the other hand, investing in infrastructure creates an inconvenience cost for city residents. The investment is partially financed through national taxes, and partially paid locally by the metropolitan area.

In this environment, two opposing political parties compete in an election and choose their policy proposals for transportation investment to maximize the probability of obtaining a majority of votes. Voters have policy preferences but also idiosyncratic tastes for the two parties. Each resident of a certain location observes the parties’ policy proposals with a given probability and decides accordingly whom to vote for.

Theoretical results

Each region’s desired investment increases with the cost of congestion and aggregate income. If congestion is a more serious problem, more infrastructure investment is needed to relieve it. In addition, as the opportunity cost of time spent travelling increases with income, higher income means the benefit of infrastructure investment is also higher. The city’s desired investment naturally decreases with the inconvenience created.

Moreover, if local financing of the project increases, then the investment desired by local taxpayers decreases whereas the investment desired by outsiders increase. This shows how the source of income matters for local support of spending on infrastructure. This result can explain why there were many huge projects after World War II. As these projects were mainly funded by residents of other states, local leaders were enjoying the benefits from the projects while free-riding on their costs.

The equilibrium level of investment increases with the cost of congestion and with aggregate income, whereas it decreases with the inconvenience it causes to local residents. Because politics do not correctly internalize all social gains and losses from infrastructure spending, in equilibrium investment may turn out to be too high or too low compared to its efficient level. In particular, this depends on how well informed local residents in the city are.

Since the late 1960s, information spreads quickly in the city center through the efforts of community organizations, and city dwellers have become politically more effective. Their information is more likely to reduce investment when inconvenience and the income of city residents are higher. Rising income leads to higher nuisance costs and when local residents are sufficiently well informed, it leads to less investment. These results match the experience of post-war urban America.

The model also predicts that when decisions over where and how to invest in infrastructure are made by the national government rather than by local governments, spending increases in cities that receive many visits from outsiders and decreases elsewhere. This result is consistent with a comparison of the US and UK. In the United Kingdom transportation policy is largely set by the central government, and infrastructure spending is skewed in favor of London, by far the main magnet for domestic visitors. Conversely, decision-making in the United States is highly decentralized and transportation spending favors low-density regions.

This paper is now published as “The Political Economy of Transportation Investment” in Economics of Transportation 13: 4-26 (2018).