Impact of Referral Ties on Worker Earnings in the Gig Economy: Evidence from an On-Demand Delivery Platform in India

With digital platform “gig” work expanding globally and providing livelihoods to millions of low-income workers in the Global South, understanding the drivers of these workers’ earnings has become increasingly important. Whereas existing research emphasizes platform design and worker characteristics as drivers of gig worker earnings, the authors examine another potential driver that is understudied: their social connections with more experienced workers through referral ties. The authors argue that such referral ties can boost gig workers’ earnings through information provision by the referrer. Analyzing proprietary data from an on-demand delivery platform in India, they find that referred workers earn 14.2% more per hour than a matched sample of non-referred workers. Using difference-indifferences estimation for better econometric identification, they find that following the referrer’s exit, the referred worker’s hourly earnings decline by 12.5%, erasing most of the referral premium and demonstrating that the above earnings effect depends on the referrer’s continued presence on the platform. Additional analyses further support an information-exchange mechanism: (i) when the referrer is concurrently online, the referred workers earn more, complete more jobs per hour, and are able to position themselves closer to high-demand areas; and (ii) upon the referrer’s exit, the referred workers experience most prominent earnings declines for the more informationsensitive earnings components, where the referrer’s guidance would have helped the most. Their study thus demonstrates how the mechanism of information exchange through interpersonal networks might be important even in a gig setting, at least in the Global South, where formal channels of information are often underdeveloped.