Editorials

Uber and Lyft: Customer Reviews and the Right-to-Control

Benjamin Sachs

Benjamin Sachs is the Kestnbaum Professor of Labor and Industry at Harvard Law School and a leading expert in the field of labor law and labor relations. He is also faculty director of the Center for Labor and a Just Economy. Professor Sachs teaches courses in labor law, employment law, and law and social change, and his writing focuses on union organizing and unions in American politics. Prior to joining the Harvard faculty in 2008, Professor Sachs was the Joseph Goldstein Fellow at Yale Law School.  From 2002-2006, he served as Assistant General Counsel of the Service Employees International Union (SEIU) in Washington, D.C.  Professor Sachs graduated from Yale Law School in 1998, and served as a judicial law clerk to the Honorable Stephen Reinhardt of the United States Court of Appeals for the Ninth Circuit. His writing has appeared in the Harvard Law Review, the Yale Law Journal, the Columbia Law Review, the New York Times and elsewhere.  Professor Sachs received the Yale Law School teaching award in 2007 and in 2013 received the Sacks-Freund Award for Teaching Excellence at Harvard Law School.  He can be reached at [email protected].

Yesterday, I moderated an ABA “webinar” on Uber, Lyft and the sharing economy with three great panelists: Shannon Liss-Riordan, Richard Reibstein, and Evan Spelfogel.  One of the most productive exchanges came during the discussion of how to determine whether Uber and Lyft (and similar “sharing economy” or “gig economy” firms) have sufficient control over the work carried out by their drivers to be deemed employers under the relevant employment laws.

In particular, we discussed the relevance of customer reviews to the “right-to-control” question; the question that is at the heart of essentially all tests of employment status. As is well known, both Uber and Lyft rely on customer ratings when deciding to terminate drivers. Both firms allow customers to rate their drivers with a 1-5 star system, and both firms terminate drivers – denying them future access to the relevant platforms – when the driver’s average star rating falls below a certain level.  For example, Judge Chen’s order states: “Uber may terminate any driver whose star rating ‘falls below the applicable minimum star-rating,’ and a significant amount of evidence in the record indicates that Uber does, in fact, terminate drivers whose star ratings fall below a certain threshold determined by Uber.”  It also appears that both companies may use the star ratings as a means of enforcing specific work rules: for example, rules regarding cleanliness, type of music to be played in the car, where and how to pick up passengers, etc…

My view – reinforced by yesterday’s discussion – is that the star rating system, as currently practiced, does suggest that Uber and Lyft are exercising employer-like control over termination decisions.  I think Judge Chen and Judge Chhabria have this right.  Yesterday’s discussion also suggested that this conclusion depends on the specific role that Uber and Lyft are themselves playing in the customer review systems.  That is, the firms are (1) soliciting customer feedback, (2) setting relevant performance levels, and then (3) making termination decisions when the customer feedback reveals that drivers are not meeting the performance levels set by the firms.  This is what employers do.

The control that flows from the current structure of customer reviews can be seen even more clearly if we contrast the current structure with an alternative performance evaluation system that the firms might adopt.  In brief, if Uber and Lyft moved from the current system (in which the firm sets the relevant performance standards) to a system in which customers decide for themselves what they want their performance standard to be, the firms’ right-to-control would dissipate. Thus, in the variation I have in mind, customers would rate drivers with the same type of five-star system now in place.  But, and here’s the change, Uber and Lyft would not terminate any drivers based on the star ratings they received.  Instead, the firms would allow customers to include a star-rating requirement along with their ride requests.  So, when requesting a ride, the app would tell the customer that the average driver in her area gets 4.3 stars and the top ten percent of drivers get 4.8 stars (I’m making up the numbers).  The passenger could then indicate that she only wants a driver with an average star rating of 4.8, or 4.3, or 4.5, or whatever.  The effect of this system would be to shift work to the best performing drivers – indeed, it might ultimately have the same practical effect as termination on drivers with low star ratings.  But the system would be entirely and genuinely customer driven and thus not, in my view, indicative of control by the firms.

Now, of course, this system would require more of customers then the current structure, and for this reason it might be less desirable from the firms’ perspective.  But I suppose that is the point: firms like Uber and Lyft must choose between maintaining control, and thus being subject to the responsibilities of employment status, or shifting control to customers and bearing the costs that come with such delegation.

Finally, to be clear, there are many factors that go into the “right-to-control” analysis.  And, there are many elements other than right-to-control that go into the ultimate determination of employee status.  But on this one piece of the right-to-control question, it appears that Uber and Lyft are currently acting like employers.  Seeing the alternatives available to the firms helps reinforce this conclusion.

Enjoy OnLabor’s fresh takes on the day’s labor news, right in your inbox.