Where Would They Have Been Without Platform Work? A Matching-Based Analysis of Post-Platform Exit Careers

• In Event: Organizations, Occupations, and Work Section Roundtables
  In Refereed Roundtable (90 minute): Table 7: Platform Work

Abstract

Periods of labor market disruption, from recessions to the COVID-19 pandemic, have pushed many displaced and underemployed workers into the platform economy. With minimal entry barriers and rapid onboarding, app-based platforms work as an income buffer and a last-resort safety net for workers facing limited options in the formal labor market. A growing body of research documents the challenges of exiting platform work, ranging from stigma around platform experience, employer penalties in hiring, and the absence of transferable skills. Yet we still lack evidence on the effect of platform work on careers for post-exit mobility, specifically relative to traditional low-wage employment.
This paper asks how post-platform wage and skill trajectories differ between low-skilled platform workers and occupationally similar workers who never engage in platform work, and what mechanisms drive those differences. I draw on labor process theory, human capital theory, and social capital theory to conceptualize how deskilling and social isolation influence low-skilled ridesharing and food delivery platform workers' career trajectories. I use labor process theory to highlight how algorithmic control and task fragmentation can intensify deskilling in ride-hailing and food-delivery work compared to traditional driving and delivery jobs Human capital theory helps explain why these deskilled workers may struggle to develop transferable skills or accumulate skills to support upward mobility while social capital theory helps to explain platform workers' fewer opportunities to build and mobilize job-relevant ties, recommendations, and referrals, though ride-hailing and food delivery may differ in this respect.
Using a quasi-experimental design, I analyze individual-level career history data from Revelio Labs, which tracks millions of online professional profiles. I employ Coarsened Exact Matching (CEM) to compare self-reported ride-hailing and food-delivery workers to traditional counterparts (e.g., taxi drivers and couriers), balancing for pre-treatment covariates including education level, career stage, and prior occupational history.

Author

• Frantony Jeremy Lewis, University of North Carolina at Charlotte