Discriminating Tastes: Uber's Customer Ratings as Vehicles for Workplace Discrimination

Consumer-sourced rating systems are a dominant method of worker evaluation in platform-based work. These systems facilitate the semi-automated management of large, disaggregated workforces, and the rapid growth of service platforms — but may also represent a potential backdoor to employment discrimination. Our paper analyzes the Uber platform as a case study to explore how bias may creep into evaluations of drivers through consumer-sourced rating systems. A good deal of social science research suggests that aggregated consumer ratings are likely to be inflected with biases against members of legally protected groups. While companies are legally prohibited from making employment decisions based on protected characteristics of workers, their reliance on potentially biased consumer ratings to make material determinations may nonetheless lead to disparate impact in employment outcomes. Hence, the mediating role of the rating system opens the door to employment discrimination.We analyze the limitations of current civil rights law to address this issue, and outline a number of operational, legal, and design-based interventions that might assist in so doing. The analysis highlights how innovative work structures challenge traditional legal frameworks, and require creative design, development, operation, and regulation to ensure that they do not facilitate discriminatory outcomes against historically disadvantaged groups.

[1]  A. Krueger,et al.  An Analysis of the Labor Market for Uber’s Driver-Partners in the United States , 2016 .

[2]  Alex Rosenblat,et al.  Algorithmic Labor and Information Asymmetries: A Case Study of Uber’s Drivers , 2016 .

[3]  Paul Dourish,et al.  Standing Out from the Crowd: Emotional Labor, Body Labor, and Temporal Labor in Ridesharing , 2016, CSCW.

[4]  Mitchell H. Rubinstein.  Employees, Employers, and Quasi-Employers: An Analysis of Employees and Employers Who Operate in the Borderland between an Employer-and-Employee Relationship , 2012 .

[5]  M. Banaji,et al.  Race Effects on Ebay , 2011 .

[6]  Emilio J. Castilla.  Bringing Managers Back In , 2011 .

[7]  Devin G. Pope,et al.  What’s in a Picture? , 2011, The Journal of Human Resources.

[8]  Margaret S. Collins.  Growing and Growing Up , 2011 .

[9]  Jennifer L. Doleac,et al.  The Visible Hand: Race and Online Market Outcomes , 2010 .

[10]  Paul A. Pavlou,et al.  Overcoming the J-shaped distribution of product reviews , 2009, CACM.

[11]  Noah D. Zatz.  Managing the Macaw: Third-Party Harassers, Accommodation, and the Disaggregation of Discriminatory Intent , 2009 .

[12]  Emilio J. Castilla,et al.  Gender, Race, and Meritocracy in Organizational Careers1 , 2008, American Journal of Sociology.

[13]  M. Buckley,et al.  The existence and nature of racial bias in supervisory ratings. , 2005, The Journal of applied psychology.

[14]  A. Jøsang,et al.  Filtering Out Unfair Ratings in Bayesian Reputation Systems , 2004 .

[15]  R. Town,et al.  The Effects of Race and Worker Productivity on Performance Evaluations , 2001 .

[16]  Chrysanthos Dellarocas,et al.  Immunizing online reputation reporting systems against unfair ratings and discriminatory behavior , 2000, EC '00.

[17]  Vicki Smith,et al.  Consumers' Reports: Management by Customers in a Changing Economy , 1991 .

[18]  G. Rutherglen.  Disparate Impact under Title VII: An Objective Theory of Discrimination , 1987 .

[19]  Kurt Kraiger,et al.  A meta-analysis of ratee race effects in performance ratings. , 1985 .

[20]  A. Hochschild,et al.  The Managed Heart: Commercialization of Human Feeling. , 1985 .

[21]  William H. Mobley,et al.  Supervisor and Employee Race and Sex Effects on Performance Appraisals: A Field Study of Adverse Impact and Generalizability. , 1982 .