5-STARS

01.27.15 10:55 AM ET

The Mysterious Way Uber Bans Drivers

Ever wonder how and why Uber drivers get banned? If it’s not for criminal behavior, the waters get murky.

What do Uber and Harvard have in common? Both only use the top 10 percent or so of their rating scales. For a Harvard student, the median grade is an A- and for an Uber driver, anything below a 4.7 out of 5 average rating from passengers spells trouble. Or, as one driver memorably puts it on Uber People, a popular web forum for rideshare drivers: “4.6 is [a] death zone.”

If you’re an Uber driver or a seasoned Uber rider, you probably already know that anything less than a perfect rating is seen as a failing grade. Forbes writer Jeff Bercovici observes that Uber’s high rating requirement puts financial pressure on drivers to seek a five-star review and social pressure on passengers to give them out. As he writes of a recent, less-than-optimal Uber driver, “I gave him five stars, of course. What do you think I am, a psychopath?”

But if, like me, you only take Uber when you’re in a pinch in a strange city, you might find yourself wondering why your driver apologizes so profusely for a wrong turn. The answer: your driver could be one angry customer away from losing a job.

The penalty for falling below a certain rating and into the “death zone” is something that’s appeared in virtually every news story about Uber for the past year: deactivation, a little understood but common practice for the ridesharing giant. Uber drivers are not employees but independent contractors and, as such, they rely on the Uber mobile application in order to work. But Uber can cut off access to this app at any moment for any number of reasons, reasons that haven’t been made entirely clear.

I interviewed an Uber spokesperson to try to gain more clarity about the circumstances in which the company decides to lock its drivers out of the service. Uber PR seems to be a bit of a walled garden at the moment—understandably so, after the company’s reputation-shattering 2014—and so I had to submit my questions via e-mail and await a response. To be frank, I might be even more confused after our interview than I was before it happened.

Many cases of deactivation seem relatively clear-cut. When a Washington D.C. Uber driver allegedly slapped a passenger for burping, he was deactivated immediately. Last January, an Uber driver who was then suspected of—and has since been charged with—vehicular manslaughter had his account “immediately deactivated.” Last October, an UberX drive in San Francisco who was cited for battery of a female passenger was also suspended by Uber as well as Lyft. And earlier this month, when a gay couple claimed that they had been thrown out of an Uber for kissing in the backseat, Uber suspended the driver in question right away.

It is in these cases that Uber is clearest about its deactivation policy. Uber spokesperson Molly Spaeth wrote in an e-mail to The Daily Beast: “It is our policy to immediately and permanently remove anyone—rider or driver—who behaves inappropriately on the system. For any serious matter, we immediately suspend the account or accounts in question as we gather the facts and investigate the situation.”

But the deactivation waters only get murkier from there. Uber has a history for deactivating drivers for reasons that have nothing to do with criminal behavior. Last October, UberX driver Christopher Ortiz discovered that he had been banned from the service after tweeting an article about Uber driver safety with the comment: “Driving for Uber, not much safer than driving a taxi.” He was told in an e-mail from Uber manager John Hamby that his account had been “permanently deactivated due to hateful statements regarding Uber through Social Media.” After the media picked up Ortiz’s story, Uber reactivated his account and dismissed it as “an error by the local team.”

In response to my question about Ortiz, Spaeth re-emphasized Uber’s official response that “[his] account never should have been deactivated.” I had also asked whether or not Uber drivers, as a general rule, were allowed to be critical of Uber on social media without fearing suspension. That question, as well as a question about the existence of precautionary measures to keep hypervigilant regional managers in check, both go unanswered.

When Uber drivers on the Uber People forum first caught wind of Ortiz’s suspension, they were immediately concerned with their own social media presences, making comments like “I better calm down,” “I deleted a few tweets as soon as I saw that,” or, sarcastically, “I don’t think I have mentioned lately how great working for Uber is.” Ortiz might have been reactivated but the incident has still left Uber drivers nervous about the extent of the latitude they may or may not have as independent contractors. It’s clear that Uber wants to distance itself from the Ortiz incident but it’s not apparent what steps they are taking to prevent similar incidents in the future.

As for the so-called “death zone” beyond which drivers risk deactivation for poor performance, Uber cannot say precisely where that lies. Each driver seems to have had different experiences skirting that line. One driver on the corporate review website Glassdoor reportedly dipped as low as a 4.56 rating before getting the ax. Another claimed this week that he was deactivated for logging a “2 week rating below 4.4” even though he had an overall rating of 4.7. In a thread this month on the Uber People forum, the mysterious cutoff point seems to be something of an oral tradition: “From what I have seen it's around the 4.6 or lower that they get concerned.” Jacobin claims that, “drivers with an average below 4.7 can be deactivated.” So which is it—4.5? 4.6? 4.7?

The Uber spokesperson I corresponded with could not tell me the secret number because there might not be one at all. She wrote: “There is no single threshold below which a driver or rider is deactivated due to the wide range of trips that riders and drivers have taken on the platform, and due to the wide range of reasons that riders and drivers may receive suboptimal ratings.” On one hand, then, Uber may have more sophisticated and individualized algorithms to track driver performance but, on the other hand, the lack of a clear guideline leaves many Uber drivers nervous about crossing the line. One new driver on the Uber People forum, for example, reportedly spent his first six weeks on the job obsessing over a drop from a 4.75 rating to a 4.66 rating.

“I think most people rated me at 4-stars thinking this is Amazon or Yelp,” he said.

But this lack of clarity around the deactivation threshold is more than just confusing, it could also be grounds for an argument that Uber drivers are improperly classified as independent contractors and should be considered employees instead. As Ellen Huet argued in a Forbes editorial last fall: “The more Uber fires its drivers, the less it has the right to, legally speaking.” To make this case, Huet points to recent court opinions that contend that being able to “discharge [a] worker without cause” means that a company cannot have an employer-contractor relationship with that worker. But so far, Uber’s classification of drivers as independent contractors, or “driver partners,” still stands.

In the meantime, the Uber deactivation mysteries are only getting deeper. Earlier this week, Buzzfeed broke the news that Uber allegedly suspended at least a dozen California drivers for complying with state laws that require they register their cars as commercial vehicles. When asked for comment, Uber did not comment on the alleged suspension e-mails and simply said: “It is not our policy to deactivate driver partners who wish to register their vehicles as commercial.”

Why do Uber drivers get deactivated? Who can Uber suspend with cause and why? No one seems to know for sure, not even the drivers who keep the service running.