Platform Leakage and Disintermediation Prevention
(A2023-114520)

AUTHORS

Pradeep K. Chintagunta, Booth School of Business, University of Chicago; Liqiang Huang, Zhejiang University; Wei Miao, University College London; Wanqing Zhang, Bayes Business School, City, University of London

ABSTRACT

Disintermediation has threatened the platform business model. Despite a prevalent issue, disintermediation is understudied due to its unobserved nature. In this study, we overcome this challenge by analyzing a randomized field experiment in which independent contractors were hired as undercover buyers to reach out to sellers under different experimental conditions. We investigate how willing sellers are to engage in buyer-driven disintermediation. The experiment varied (1) if the communication could be monitored by the platform and (2) the magnitude of commission savings from disintermediation. OLS estimates indicate that communications being unmonitored significantly increased the level of disintermediation by up to 21.7 percentage points. Meanwhile, the average treatment effect (ATE) of higher commission savings is statistically insignificant. Next, we apply the causal forest algorithm to estimate the conditional average treatment effects (CATE) for each individual observation. Our second-stage analyses reveal salient treatment heterogeneity across sellers and provide important managerial implications for online platforms.