An Uber driver once told me he keeps a gun under his seat because “Uber doesn’t have my back”. Another was told by a customer that he should “go back to his own country”. Yet another was punched in the back of his head while driving.
Anyone who has ever worked a customer service job knows that customer abuse against workers is rampant; one study found that, on average, call center workers interact with abusive customers more than ten times a day. The gig economy is no exception.
Yet if platform workers are “their own boss”, why are they subjected to constant customer abuse? And why does this issue seem to be getting worse over time?
In my recent ILR Review article, I explore how the way that platform companies maintain high service standards leaves workers defenseless against abusive and harassing customer behavior.
Many service organizations rely on customer-feedback systems to manage workers. If you have ordered a pizza or called a customer service line recently, there’s a good chance you were asked to rate the service interaction on a 1 to 5 scale.
One critical difference between platform companies, such as Uber or Instacart, and more traditional service companies is that platform companies give customers direct control over workers’ performance evaluation. The most prominent consequence for a poor customer rating is what these companies euphemistically refer to as “deactivation” (i.e., termination). For Uber and Lyft drivers, they are required to maintain roughly a 4.7 out of 5 “star” rating in order to avoid “deactivation”. As one driver told me, a 4-star rating just means “fire me slowly.”
There are other possible consequences for poor ratings. For example, Instacart only allows workers with the highest ratings (roughly 4.95/5) to work the most lucrative “batches” (orders). This means that one blemish on their record can cost a worker hundreds of dollars while they “work off” their rating.
Giving customers direct control over workers makes workers choose between confronting a passenger and potentially losing their income. As a result, workers are placed in a situation where they must conform to customers’ demands – including illegal (“It’s cool for me to drink this beer while you drive, right?”), dangerous (“drive faster, I’m late for work!”), or harassing (“Do you have a boyfriend?”) demands – in order to continue working on the service.
Yet given the shortage of gig workers and increasing competition in key platform industries, why have platform companies not attempted to address customer abuse on their services? In my article, I found that platform companies “launder control” – empowering customers to punish workers – in order to shape worker behavior without using the traditional tools of management. Doing so allows platform companies to maintain high service standards without triggering a formal employment relationship.
At the same time, workers find clever ways of punishing platform companies for designing these systems. During the time of my study, Uber and Lyft offered workers important differences in terms of driver support. Lyft, at a competitive disadvantage compared to Uber, attempted to attract drivers by branding itself as the “pro-driver” service. A key component of this was in-app tipping, phone support, and a more forgiving customer evaluation system (i.e., giving them less power over workers).
As Lyft offered workers slightly more protection from customer abuse (and the ability to contact a person through the customer service line), I found that drivers would spend more time on Lyft in order to ensure that customers who used the platform would receive reliable service.
Beyond subtle methods of supporting Lyft, I found that drivers would also actively poach Uber passengers for Lyft by passing out Lyft referral cards to them. Drivers realized that passenger distribution prevented them from using the service that better protected them and actively attempted to rearrange their local market.
This study helps us understand the nexus between control and customer abuse in the platform economy: In order to avoid a formal employment relationship with workers, platform companies elevate customers to a position of power. It is long said that in customer service, the “customer is king”. While service organizations may keep this mantra, platform companies formally ordain it.
Yet in elevating the customer to a position of power, it enables them to force workers to engage in illegal or dangerous behaviors in order to keep their jobs. As a result, this convenient workaround to an employment relationship has left workers defenseless against abusive customers. Hence why one driver told me that they – contrary to local regulations – carried a concealed weapon while driving.
This study also illustrates that high-road employment practices are possible in platform work. Workers recognized that, during the time of the study, Lyft provided them with slightly more favorable labor conditions. This protection was valuable to workers, and workers responded by trying to help Lyft gain market share.
This finding helps explain why Uber and Lyft are now playing a “shameless game of copycat” and have largely converged in their treatment of workers: When there is a disparity in treatment across competitors, workers will attempt to reshape the market in subtle (and sometimes not-so-subtle) ways. Accordingly, in order to keep workers from defecting to a competitor, both Uber and Lyft rapidly mimic each other’s innovations.
Does this mean that labor market competition will naturally raise labor standards in the platform economy? This seems unlikely because labor standard elevations – such as in-app tipping – can be easily copied by a competitor, improving workers’ experiences yet leaving companies (potentially) worse off. Yet if workers could band together and bargain collectively, they may be able to bend the trajectory of the gig economy toward a worker-centered one.
Michael Maffie. “The Perils of Laundering Control through Customers: A Study of Control and Resistance in the Ride-hail Industry”. ILR Review 2022.
image: Ivan Radic via flickr (CC BY 2.0)