Only 1 percent of Uber customers always tip, while nearly 60 percent never do, according to a new study from the National Bureau of Economic Research (NBER). The average tip is 50 cents a ride, but for those who do tip, the average is more like $3. Men are more likely to tip than women, but female drivers get tipped more than male ones.

Tipping has long been a source of fierce debate, especially as it relates to ride-hailing companies like Uber. For years, the company rejected efforts to add a tipping option to its app, arguing it would overly complicate the seamlessness of the experience. But Uber eventually caved and now drivers earn hundreds of millions of dollars in tips every year. The NBER study on tipping with Uber is sure to throw more fuel on the fire.

The paper was authored by Stanford University’s Bharat Chandar and University of California-San Diego’s Uri Gneezy, as well as John List, former chief economist at Uber who is now at Lyft, and Ian Muir, current head of economics at Lyft. The researchers were uniquely positioned: in addition to combining big data analyzation with field experimentation, the team actually helped Uber implement its in-app tipping option, which rolled out in June 2017. As such, they were able to develop data from more than 40 million trips.

What they found was not a whole lot of tipping. Roughly 16 percent of Uber rides are tipped. Yet, most riders (60 percent) never tipped over the research team’s four weeks of data collection. Of those who do tip, very few (1 percent) tip on every trip. The remainder of people only tip on about 25 percent of trips.

This is likely because of the unique aspects of app-based ride-hailing. All payments take place in the app and riders simply hop out of the vehicle when the ride is over. It’s only after the fact — sometimes long after the fact, depending on when you open the Uber app next — when riders are asked whether they want to tip. As such, the question of tipping is removed from the experience of the ride, Gneezy explained.

“I think Uber drivers are tipped less than taxi drivers because tipping happens after the ride is over and not face to face,” Gneezy said in an email. “In a sense, I think that this is the right way. Riders don’t tip automatically, but only if they are happy with the service. Hence, tips provide incentives to drivers.”

In an email, Chandar cautioned that the data was from 2017 and could be considered “stale.” That said, he didn’t believe that Uber should be seeking to encourage people to tip more. “It is not obvious to me that getting people to tip more, on rideshare or otherwise, should be the goal,” Chandar said. “As we show in the paper, while tipping has some relationship with trip quality, it is also associated with other factors not evidently related to quality. Effects on earnings can also be ambiguous, as we show in our other paper released today.”

There were other interesting findings, such as riders who have a five-star rating tip more than twice as often as those with a 4.75 rating, and when they do tip they do it nearly 14 percent more. The team also noticed important correlations between gender and tipping:

Male riders tip 23% more than female riders, a result largely driven by the fact that men are more likely to tip than women (approximately 19% more often). Further, female drivers are tipped more than male drivers—a fact that is true regardless of rider gender: men (women) tip female drivers nearly 12% (11%) more than they tip male drivers.

Uber has a fraught history with tipping. It was reportedly Travis Kalanick, the former CEO and co-founder of Uber, that most fiercely rejected calls to add tipping. And it was only when Kalanick was about to be ousted from his job during Uber’s scandal-plagued 2017 that the company eventually caved and added the option.

Moreover, Uber sought to undermine the very notion of tipping. Uber published a post on Medium in April 2016 that went into greater detail about the company’s position on tipping. “Whether consciously or unconsciously, we tend to tip certain types of people better than others,” Uber said. “This means two people providing the same level of service get paid different amounts. With Uber, drivers know that they earn the same for doing the same trip, no matter who they are or where they’re from.”

The company cited a 2008 Cornell University study that found that “consumers of both races discriminate against black service providers by tipping them less than white service providers.” But it’s worth noting that the study predated the gig economy by several years. And now Uber has trained consumers to leave their cash at home, and that ratings, not dollars, translate into better service.

Uber has recently changed its tune on tipping. A year after making in-app tipping possible, the company touted the fact that drivers had earned more than $600 million in tips in the US and Canada. That number has risen considerably since then: according to a spokesperson, in the past two years Uber drivers and Uber Eats couriers have collected nearly $2 billion in tips. “We’re committed to developing and improving features that help detect and mitigate bias on our platform,” a spokesperson said.

Update October 21st, 5:22PM ET: The story was updated to include a comment from Bharat Chandar and a statement from an Uber spokesperson. The headline also was amended to include the source of the information.

https://www.theverge.com/2019/10/21/20925109/uber-tipping-riders-drivers-percentage-gender-nber-study