Abstract for: Modeling Sex Trafficking and Commercial Sex Work Policy in Massachusetts

Sex trafficking affects many individuals in the United States, yet severe data scarcity hampers evidence-based policymaking. Commercial sex work (CSW) policy is tightly intertwined with trafficking outcomes, and current policies of criminalization disproportionately harm sellers who are often trafficking victims themselves. This paper investigates how alternative CSW policy regimes affect trafficking in Massachusetts using a system dynamics model. System dynamics modeling (SDM) is a type of computational simulation that allows us to capture the causal structure of the system even with scarce data. We developed an SDM of the Massachusetts CSW industry through iterative subject matter expert elicitation and calibrated it against available state-level data estimates. Sensitivity analyses identify the most impactful and policy-relevant parameters. We compare the outcomes of four potential policy regimes against the default, full criminalization of CSW. We found that full decriminalization increases trafficking, buyers and demand by removing buyer risk. Partial decriminalization and an extension that also increases seller services reduce total and trafficked sex sellers, though both modestly increase trafficker counts. Expungement of seller criminal records produces consistent protective effects across outcomes. The model demonstrates that partial decriminalization consistently outperforms full decriminalization on trafficking-related outcomes which is a counterintuitive finding that, nonetheless, has support from prior empirical and simulation research. Criminal record reform represents a high-leverage, low-disruption policy option. We also demonstrate how the model could be used to investigate non-legislative interventions and future work should prioritize data collection to narrow uncertainty bounds for decision-making. AI used to help build analysis scripts and, sparingly, to edit poorly flowing paragraphs in the paper