Richard Florida is a co-founder and editor at large of CityLab and a senior editor at The Atlantic. He is a University Professor and Director of Cities at the University of Toronto’s Martin Prosperity Institute, and a Distinguished Fellow at New York University’s Schack Institute of Real Estate.
The U.S. is a nation divided not just by how people get around, but by how fast they drive.
America is famously divided along many lines: red and blue states, high-income and low-income areas, educated and uneducated cities—the list goes on. But, startlingly, America’s political and economic divide extends to fatality rates as well, and specifically to where people are killed in their cars.
The map below shows America’s Car Death Belt, made up of two distinct parts: the Deep South and the Great Plains states.
The map is from a new study by Michael Sivak and Brandon Schoettle at the University of Michigan Transportation Research Institute. The study used data from the National Highway Traffic Safety Administration and the Center for Disease Control to map the death rate from road crashes on a per capita basis in 2013. There were a reported 32,719 traffic fatalities that year—a very concerning number, especially given that these deaths are largely preventable. Keeping them from happening requires knowledge of where they occur the most and why.
States With the Highest and Lowest Road Fatality Rates
|States With the Highest Road Death Rates||Fatalities per 100,000 pop.|
|States With the Lowest Road Death Rates||Fatalities per 100,000 pop.|
|District of Columbia||3.1|
Montana tops the list of road deaths with a rate of 22.6 road fatalities per 100,000 people—more than twice the national rate—followed by Mississippi, North Dakota, West Virginia, and Alabama. On the other end of the spectrum, the states with the lowest rate of road crash deaths are Washington, D.C. (which is of course entirely urban), followed by Massachusetts, New Jersey, New York, and Rhode Island. These states are all located along the Boston-NY-Washington Corridor on the East Coast, where mass transit is much more available and widely used.
Road crashes also make up a much higher percentage of all deaths in some states than in others, as the map above shows. Again we see the same Car Death Belt spanning the Deep South and the Great Plains.
States with the Highest and Lowest Relative Road Fatalities
|States With the Highest Relative Road Fatalities||Road Share of All Fatalities|
|States With the Lowest Relative Road Fatalities||Road Share of All Fatalities|
|District of Columbia||0.4%|
Nationally, road crashes make up 1.3 percent of fatalities from all causes. But car deaths make up the highest percentage of overall fatalities in Montana and North Dakota (2.4 percent), Mississippi (2.0 percent), and Wyoming, South Dakota, and Texas (1.9 percent). The lowest percentages are again located in Washington D.C., Massachusetts, Rhode Island, New Jersey, and New York, all of which had a road fatality share of less than one percent.
What factors might be related to fewer or greater car deaths? To get at this question, my Martin Prosperity Institute colleague Charlotta Mellander ran a basic correlation analysis to determine associations between road fatalities and certain key demographic, social, and economic factors. As usual, it’s important to keep in mind that these correlations do not imply causation, but instead point to associations between variables.
For one, car deaths are clearly lower in denser, more urbanized states. The rate of road deaths (per 100,000 people) is negatively correlated with state-level density (-0.51) and state-level urbanization (-0.63). This is not surprising since people drive more and faster in less populous states and less and slower in denser, more urbanized, and more populated states. The states with the highest rate of road deaths all have top speed limits of at least 70 mph and some have top limits of 80 mph. Conversely, many of the states with lower rates of road deaths have lower speed limits. Only three of the states with the lowest car death rates have speed limits of 70 mph or more and the majority have top speed limits of 65 mph or less.
Car deaths are also higher in poorer states. The road death rate is negatively associated with GDP per capita (-.47) and even more so with per capita income (-.66).
People in states with more high-tech knowledge-based economies are far less likely to die in their cars. Our analysis found negative associations between the rate of road deaths and the share of adults who are college grads (-.69) and the share of workers who are members of the creative class (-.60). Conversely, car deaths were higher in states where greater shares of workers are members of the blue-collar working class (with a positive correlation of .53).
Car deaths were also higher in states with higher poverty rates (.57). There are a couple of likely reasons for this link. For one, people in less advantaged states tend to drive older, less expensive cars. These cars are less likely to offer good crash protection and be equipped with the appropriate safety technology. In contrast, more affluent and knowledge-based states have smaller percentages of people who drive, and even those who do are often forced—either by law or other factors like congestion—to drive slower.
Car deaths also track political ideology. They are higher in conservative states and lower in liberal ones. Car deaths are positively correlated to the share of voters who voted for the Republican candidate Mitt Romney in the last presidential election (.66), and negatively correlated to the share who voted for Obama (-.66). Of course, more conservative states are often the more sprawling and car-oriented ones, while liberal ones are denser and more transit-oriented.
At the end of the day, America is a nation divided not just by how people get around, but by how fast they drive. What’s more, this dependence on cars is influenced by certain socioeconomic factors like low income or GDP, as well as by the sheer density of a location. To prevent more road fatalities from occurring, future policy will have to bear this in mind.
UPDATE 5:45 p.m.: Jed Kolko tweets with another chart to bolster these findings: