In researching questions relating to economic geography, I rely on agent-based models for at least four reasons:
1) Agent-based models are inherently micro-founded, every decision is taken by an individual agent in the system. This lends well to a “scale-agnostic” view of the world.
2) The models are easily able to handle the messy world of increasing returns. While incorporating increasing returns with other modeling techniques is difficult at best, agent-based models thrive under such specifications. Non-equilibrium results are not avoided, they are embraced.
3) The field of agent-based economic geography is in its infancy. Despite the ability of agent-based models to incorporate sociological forces that economic geographers highlight and increasing returns that give geographical economists difficulty, there is little work in economic geography that utilizes the power of agent-based modeling.
4) Finally, and most importantly, agent-based economic geography is exhilarating. The models are “alive”! The modeler gets to witness the evolution of the outcome, not just the outcome itself. Examining the underlying processes and forces pushing the model along are both insightful and inspiring.
Keith Waters 