# An applied spatial agent-based model of administrative boundaries using SEAL

## TL;DR

This paper presents SEAL, an empirical spatial agent-based model for metropolitan regions in Brazil, enabling policy analysis by simulating markets and municipal governance impacts using official data.

## Contribution

It adapts an abstract model into an empirical framework for Brazil, integrating spatial boundaries and official data for policy experimentation.

## Key findings

- Single metropolitan government may benefit citizens.
- Model suggests potential improvements in quality of life.
- Preliminary results support policy integration benefits.

## Abstract

This paper extends and adapts an existing abstract model into an empirical metropolitan region in Brazil. The model - named SEAL: a Spatial Economic Agent-based Lab - comprehends a framework to enable public policy ex-ante analysis. The aim of the model is to use official data and municipalities spatial boundaries to allow for policy experimentation. The current version considers three markets: housing, labor and goods. Families' members age, consume, join the labor market and trade houses. A single consumption tax is collected by municipalities that invest back into quality of life improvements. We test whether a single metropolitan government - which is an aggregation of municipalities - would be in the best interest of its citizens. Preliminary results for 20 simulation runs indicate that it may be the case. Future developments include improving performance to enable running of higher percentage of the population and a number of runs that make the model more robust.