Academics, policymakers, and practitioners do not agree on whether the United States has too much or too little roadway infrastructure. Nevertheless, federal, state, and local governments spend hundreds of billions of dollars annually building, expanding, rebuilding, and maintaining roads. Beyond their disagreements on the costs and benefits of roadway infrastructure, researchers and policymakers do not even know how much land is dedicated to roadways, where it is located, or how much it is worth. Without better accounting, it is difficult to assess whether there is too much or too little roadway and whether outcomes such as commute times, wealth, health, or employment vary with the amount of roadway.

For all the benefits roadways provide motorists, they take land that could otherwise be used for homes, businesses, shops, and open spaces. A better understanding of roadway space and value is essential for assessing state and federal transportation policy, conducting cost–benefit analyses, and helping local officials understand how their city, town, and metropolitan areas compare with others.

Our research uses a novel methodology by combining publicly available datasets to generate predictions of roadway widths and areas across US metropolitan areas. We matched these predictions to estimated land values for all 316 primary metropolitan statistical areas (PMSAs), where approximately 80 percent of the US population resides.

Our findings indicate that roadways account for a fifth to a quarter of all urbanized land in PMSAs—about 22,000 square miles—which is equivalent to the total land area of West Virginia. This roadway was worth approximately $4.1 trillion in 2016, with large, wealthy PMSAs such as New York City, Los Angeles, and Chicago representing a substantial share of this value. Suburban neighborhoods within PMSAs have generally dedicated more land area—but less valuable land—to roadways.

Additionally, we conducted a rough cost–benefit analysis of US roadway investments. Land is an important component of the cost of providing roadways and worth more per year than the amount governments spend on building and maintaining the roadway system. Our analysis reveals that dedicating more land to roadways would likely lead to net losses in social well-being. Specifically, the costs of widening roadways by 10 percent would, on average, exceed the benefits of time savings for drivers and truckers by a factor of three after accounting for the value of land. After accounting for the social harms of driving, such as greenhouse gas emissions, our findings suggest that the costs would be four to five times higher than the benefits.

Prior research indicates that widening roadways by 10 percent increases driving speeds by up to 1 percent. However, after accounting for land values, our analysis finds that speeds would have to increase by 3 percent for a 10 percent expansion in the widths of urban roadways to have economic benefits that exceed costs. When we also consider the social harms of driving, our analysis suggests that the increase in speeds would need to be closer to 50 percent to justify this roadway expansion.

Dedicating more land to housing, offices, and other such uses instead of roadways would likely increase net social benefits on average. Our research finds that a 10 percent reduction in urban roadways—accomplished by removing, narrowing, or downgrading roadways—would result in $27.8 billion of net economic benefits per year. At some point, reductions in roadways would result in economic harm, but across US urban areas today, reducing the amount of space dedicated to urban roadways could generate substantial gains and reduce pollution, greenhouse gas emissions, and traffic fatalities.

Note
This research brief is based on Erick Guerra, Gilles Duranton, and Xinyu Ma, “Urban Roadway in America: The Amount, Extent, and Value,” Journal of the American Planning Association (August 2024): 1–15.