Manufacturing Productivity with Worker Turnover
68 Pages Posted: 25 Sep 2018 Last revised: 2 Sep 2021
Date Written: September 1, 2021
Today's most sophisticated manufacturers produce immense quantities of complex goods in rapid response to fast-evolving near-term demand. To maintain high productivity under such circumstances, they schedule and staff large bodies of workers to efficiently dispatch highly variable workloads. Their success depends on managing the assignment of experience and skills to specialized tasks and effectively coordinating work on production lines. Worker turnover may disrupt such productivity, and we study the manufacturer's problem of controlling turnover and its effects. We collect data regarding production schedules, staffing, productivity, and pay from within the supply chain of a major consumer electronics manufacturer. Studying 52,214 China-based workers manufacturing millions of devices weekly, we find that worker turnover adversely affects the ability to coordinate by weakening knowledge-sharing and relationships among assembly line co-workers. Using publicly available cost information, we estimate a cost impact of $206-274M in units failing stringent quality control alone. We solve the manufacturer's problem of production planning which embeds flexibility in scheduling and staffing and structurally models workers' turnover decisions in dynamic equilibrium. Our counterfactual analyses predict significant gains from a less turnover-prone, hence more productive, workforce. Increasing wages by 10% reduces the manufacturer's variable production costs by an estimated 4.5%, or $928M for the product we study. Intuition suggests that by building inventories to buffer against variability in workforce productivity, firms may ignore turnover. We instead find that using inventories in production planning can keep workforce utilization high, hence raising the marginal returns to productivity and motivating self-interested firms to retain workers.
Keywords: Data-driven workforce planning, Empirical operations management, Employee turnover, Experience-Based Equilibrium, Production planning, Productivity, Stochastic optimization, Structural estimation
Suggested Citation: Suggested Citation