Countering urban segregation in Brazilian cities: policy-oriented explorations using agent-based simulation

In this study we use agent-based simulations to explore the impact of social-mix policies on the segregation dynamics of São José dos Campos, a medium-sized Brazilian city. We use the model MASUS, Multi-Agent Simulator for Urban Segregation, to test two policy strategies: one based on the spatial dispersal of poverty, and the other on the spatial dispersal of wealth. The experiments indicated that these strategies reveal varying shortcomings and complementary benefits in cities such as São José dos Campos. While poverty dispersal provides immediate results on segregation levels and direct benefits for the assisted families, wealth dispersal can produce long-term outcomes and promote a positive change in the overall levels and patterns of segregation in the city. <br> <b>Keywords:</b> urban segregation, social mix, urban policies, social simulation, agent-based model, Brazil