Joos Akkerman attending two Italian summerschools on macroeconomic agent-based modeling

by Joos Akkerman

This summer I was able to visit two summer schools in Italy. Both focused on agent-based modeling (ABM) in macroeconomics, and covered two of the most important paradigms in macro ABM: the K+S model and the CATS model.

In July, I attended the ‘Agent-based Macroeconomics’ PhD course at the Sant’Anna School of Advanced Studies in Pisa taught by Prof Andrea Roventini. Sant’Anna is the birthplace of the ‘Keynes meeting Schumpeter’ (K+S) model, one of the most prominent macro-economic ABM frameworks. This model combines long-term economic growth through technological development with endogenous short-term business cycle fluctuations. I was already quite familiar with this framework, and it was great to meet the people that developed it! This model also forms the basis of the CRAB model, which was developed by the SC3 team (Alessandro Taberna, Tatiana Filatova and Liz Verbeek) in cooperation with the scholars at Sant’Anna. I was able to discuss my own plans with the CRAB model and get some great insights. And I also got to meet the nice students in the PhD cohort, who were very welcoming and made my stay in Pisa very memorable!

In September, I attended the summer school on stock-flow-consistent agent-based modeling at the Universita Politecnia Delle Marche in Ancona. This provided a thorough introduction into the stock-flow consistent paradigm, and how this can be used in ABMs. Applying stock-flow consistency to ABMs is a way to ‘discipline’ them by not allowing for any unaccounted sources of money in and out of your model. This adds a layer of realism to your model. Ancona (together with Milan) is also home of the `Complex Adaptive Trivial System’ (CATS) macro-ABM model, another prominent macro-economic ABM framework. Among others, this model has been extensively applied to macro-financial interactions, which is very relevant for my study of the economic and financial effects of natural disasters. The summer school was closed by a keynote lecture by Sebastian Poledna on forecasting using agent-based models. The recent paper he coauthored, which uses a stock-flow consistent ABM to forecasting macroeconomic variables, truly shows the large potential of the ABM-paradigm. It was also very stimulating to meet so many other PhD candidates, working on a highly diverse set of topics. Their approaches ranged from complexity, neoclassical, post-Keynesian and historical, but all found use for the ABM methodology.

I am excited to employ this knowledge to develop further my PhD project on revealing economic and financial tipping points in transformational adaptation to climate change under the umbrella of the corresponding VIDI project.