Do ridesharing apps decrease drunk driving? It's more complicated than you might think
There was a time, not so long ago, when people getting ready for a night out used to choose one person from among their group to be the designated driver, the person whose job it was to drive a group of friends to and from the bars and would thus forgo drinking for the evening.
At least that’s what was supposed to happen. Too often people just drove drunk, resulting in tragedy. Today, with the ability to hire a ride on demand, both the designated driver and impaired driving may thankfully be going extinct.
Ride-hailing services like Uber and Lyft have claimed for years that they reduce DUIs by providing an easy alternative to drinking and driving for would-be drunk drivers. But do the stats support the claim?
It turns out that answering that question is trickier than you might think. Past studies that cross-reference the number of DUIs or traffic fatalities in a community with the introduction of ride hailing have yielded inconsistent results. Contradictory studies have found a decrease in DUIs following the introduction of ride hailing, no relationship between the two variables, or even an increase in traffic fatalities associated with Uber. The problem is that the mere introduction of ride-hailing services in a city doesn’t necessarily mean that people are using them in large numbers, and doesn’t account for the unique variables of each city.
Finally, thanks to a pair of just-released studies, the research is in, and the answer to the question of whether Uber et al decreases DUIs is a resounding “Yes, but….”
In the paper “Uber and Alcohol-Related Traffic Fatalities,” published in July 2021 by the National Bureau of Economic Research, study authors find “robust, large, and statistically significant negative impacts on alcohol-related traffic fatalities”—a 6.1% decrease, to be precise, associated with the widespread adoption of Uber in any given community. While that may seem like a small number, 6.1% nationwide corresponds with thousands of lives that would otherwise have been lost due to impaired driving.
Unlike previous research, this study, authors say, is the first to use proprietary rider data from Uber itself, meaning the study isn’t dependent on drawing a tenuous connection between the date ride-hailing is introduced in a city and its actual widespread adoption by residents.
Here’s the but: as study authors note, these findings conform with a widespread, decades-long, nationwide trend of decreasing DUIs—drunk driving fatalities in the U.S. have fallen by half since 1982. That is a steep and substantial decline that is attributable to a whole host of factors, including a decrease in rural living (where impaired driving tends to happen with the greatest frequency) and even lower alcohol consumption in the United States in general. But the granular proprietary Uber data allows researchers to map observable decreases in DUI fatalities with real time usage of ride-hailing in lieu of drunk driving.
Another study published in JAMA Surgery in June 2021, looked strictly at ride-sharing data in Houston, Texas, to find a notable decrease in impaired driving convictions associated with widespread use of Uber. The effect was most pronounced on Fridays and Saturdays, which saw decreases in convictions for impaired driving of 18% and 15%, respectively. The Uber effect on impaired driving was especially significant for drivers under 30 years old (the cohort most likely to adopt new ridesharing technologies). Motor vehicle crashes are a leading cause of death for drivers in this age range, meaning the impact of ridesharing on lives saved is especially strong for younger people.
Thanks to these new studies, we can say with confidence that ride-hailing services do indeed lead to a drop in driving under the influence. While the decreasing trend in impaired driving overall accounts for a significant drop all by itself, the new data help illustrate how ride-hailing technologies make a meaningful contribution to keeping our roads and loved ones safe.
Notwithstanding the good news on the drunk driving front, a more recent development complicates these trends: the ubiquitous decriminalization of marijuana. Because THC—the compound that makes pot smokers high—is detectable in the body for long after its effect on the user wears off, measuring the effect of marijuana use on impaired driving is much more difficult than with alcohol. Still, numerous studies have shown that marijuana use can sharply increase the odds of being involved in an automobile collision—in some cases by up to twice as much. While we rightly celebrate the decrease in drinking and driving in recent years, we should remember that drunk driving is not the only kind of impaired driving that technology can help us eliminate.
And in that effort, if you believe the science, ride-hailing services like Uber, Lyft, and others, certainly have an important role to play.