Less drunken driving and Uber may have correlation
Some studies find correlation between ride-hailing app and lower rates of alcohol-related accidents
It seems like common sense: People will be less tempted to drink and drive if they have an easier, safer way to get home after a night out. Uber has plenty of incentive to make this case and has reported on declines in drunken-driving incidents in several major cities after the company rolled out its ride-hailing services, beginning in San Francisco in 2010.
A recent independent study backs this up. It found that in four boroughs of New York City, excluding Staten Island, there has been a 25 percent to 35 percent reduction in alcohol-related car accidents since Uber came to town in 2011, as compared to other places where ride-hailing company doesn’t operate.
That’s a significant reduction, amounting to about 40 fewer collisions per month in those boroughs. And it’s good news for Uber, which could use some positive attention after months of hurtling from one public relations crisis to the next.
“We need more evidence, but the trend seems to be pointing toward ride-sharing reducing drunk driving incidents,” said Jessica Lynn Peck, a doctoral candidate at the City University of New York Graduate Center who wrote the study, a working paper that was published in January.
But not all studies have reached the same conclusion. One report, published in the American Journal of Epidemiology last year, looked at 100 densely populated counties across the United States and found no correlation between the rollout of Uber services and the number of traffic fatalities.
Noli Brazil, a postdoctoral research associate with the University of Southern California who wrote that paper with David S. Kirk, an associate professor at Oxford University, said commonsense arguments — that ride-hailing apps should prevent drunken driving — made some sense on an individual level.
But he warned against making broader assumptions, especially since those who would drive drunk are not necessarily rational decisionmakers. Researchers on the subject, he added, had to deal with a dizzying array of variables, including state laws, time frames and communities’ access to public transportation. All of these can affect a study’s conclusions, and there is plenty of opportunity to cherry-pick data.
“In order to explain our results, we pointed out the fact that the proportion of individuals who use Uber is quite small relative to the number of drivers in a given county,” Brazil said.
Most studies, like Peck’s, have noted a correlation between Uber services and lower rates of alcohol-related accidents. A 2015 report from Temple University found that Uber was associated with a decrease in motor vehicle homicides in California. A report this year from researchers at West Carolina University also found that Uber service rollouts led to declines in fatal accident rates across the country.
But none of these reports has been as unequivocal as the one that Uber itself released in 2015. It stated that in several major cities, Uber ridership peaked at times when drunken-driving accidents tend to happen.
It also found that in Seattle, the introduction of services was associated with a 10 percent decrease in arrests for driving under the influence. And in places where Uber was launched in California, it reported, the number of alcoholrelated crashes every month decreased by 6.5 percent among young drivers.
“Several independent studies have shown Uber’s presence in cities can help reduce drunk driving,” a company spokeswoman said. “We’re glad to provide an alternative to drunk driving that helps people make safer, more responsible choices.”
Peck, whose research used collisions data from the New York Department of Motor Vehicles and the state’s Department of Transportation from 2007-13, agreed that the growing body of research suggests ride-hailing services lead to fewer alcohol-related car accidents. But she noted that analyzing data sets and teasing out causation from correlation is slow and tricky work.
“I think anyone who does statistics for a living is going to be really careful about saying they are sure,” she said. “Because we are scientists, and we are never sure.”
While scientists are known for being cautious, Uber is known for being brash. The young company has disrupted the taxi industry since its foundation in 2009 and is reportedly valued at close to $70 billion.
As a workplace, Uber came under fire in February amid reports that the company had an aggressive internal culture that allowed some infractions to go unpunished.
“Uber has been in the news a lot in the last year, in ways that I think all of the authors on his subject find a little cringe-worthy,” Peck said.
She added that the point of her work was not to evaluate the company itself but to provide public health data that would be useful to city and state legislators.
Brazil said that as Uber becomes more popular across the United States, a growing body of research could indeed show that the app leads to lower drunken driving rates — but it’s too early to say for sure.