Reducing Interference Bias in Online Marketplace Experiments Using Cluster Randomization: Evidence from a Pricing Meta-experiment on Airbnb

The authors are grateful to Lanbo Zhang, Minyong Lee, and Sharan Srinivasan for their assistance with the design and analysis of the experiments in this paper. The authors thank numerous other Airbnb employees who have assisted this project and are grateful to Iris Fung, Jonathan Niles-Weed, and Peter Coles for their assistance. The authors also appreciate the helpful feedback they received from Dean Eckles, Andrey Fradkin, Hannah Li, Alex Moehring, Hong Yi Tu Ye, the Berkeley Haas Management of Organizations (MORS) Macro Research Lunch, the Massachusetts Institute of Technology (MIT) Sloan Social Analytics Laboratory, and attendees of the 2019 Winter Conference on Business Analytics and the Harvard Business School Digital Doctoral Workshop. This experiment was classified as exempt by the MIT Committee on the Use of Humans as Experimental Subjects under Protocol 1807452488. David Holtz was previously an employee of Airbnb between 2014 and 2016 and was an unpaid contractor while working on this paper. He currently holds a material financial interest in Airbnb, Inc. Ruben Lobel and Inessa Liskovich are currently full-time employees of Airbnb. Felipe Lobel was previously a paid intern at Airbnb and was an unpaid contractor while working on this paper. Airbnb had the right to review this paper subject to the following stipulations: (1) delaying disclosure or publication until any intellectual property rights have been adequately protected; (2) removing Airbnb confidential information from the work product; and (3) adding Airbnb as an author, as appropriate, of the work product and/or adding an appropriate disclosure concerning Airbnb’s support of the project.