Where turnover contagion tips: Coworker network structure, job embeddedness, and retention targeting in an agent-based model
Year: 2026
Type: preprint
Abstract: Employee turnover spreads among coworkers, but the organizational consequences of that contagion are poorly understood. We build an agent-based model in which employees embedded in a coworker network start searching for a new job when shocks and the share of searching or vacant coworkers exceed an embeddedness-dependent threshold. Across about 63,000 simulation runs, contagion strength produced a sharp tipping point between modest amplification of turnover and a self-sustaining regime. Clustered networks tipped earlier than random ones, whereas hubs alone did not lower the tipping point. Higher mean job embeddedness delayed tipping strongly. Under a fixed budget, retention practices that reached hubs or brokers and were attached to positions rather than individuals raised the tipping point most. The model is stylized and uncalibrated, so the findings are hypotheses for empirical testing.
Source: Research Square
Institution: Lincoln University College
Language: English
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W7234249747
