Using Agent-Based Models to Inform Post-pandemic Return-to-Work Policy Decisions

In response to shifting workplace norms and expectations brought about by the COVID-19 pandemic, many employers are deciding what level of workplace flexibility to implement as society moves out of the pandemic. The influx of remote and telework during the pandemic resulted in increased favorability towards flexible work, leaving workers resistant to return to pre-pandemic in-office norms. These new policy decisions are multi-faceted, balancing business need with employee sentiments. The recency of this phenomenon limits the amount of available data on post-pandemic employee sentiments and behaviors, and thus presents a challenge to making data-driven policy decisions. Here we demonstrate the use of an agent-based model to explore the effects of various remote work policy implementations on the workforce population of a government agency (~19,000 employees). We found that an agent-based model could successfully identify trends in employee behavior that could then be used to inform policy. Namely, we observed that workplace flexibility has temporary effects on attrition and that addressing hiring practices may be a better long-term strategy for combatting attrition. The implementation of our model as a tool for informing remote work-related policies provides a case for the application of agent-based models within business and human resource settings.