© Universität Bielefeld
Agent-based Modeling for Economic Policy Analysis
Herbert Dawid, Philipp Harting, Jasper Hepp, Dirk Kohlweyer, Tomasz Makarewicz and Mariya Mitkova
In recent years it has been widely acknowledged by economic scholars that the explanatory power of standard representative agent models is in many instances limited. This has lead to a surging interest in the empirical exploration of bounded rationality of behavior in economic decision problems, mainly by means of laboratory experiments, and attempts to incorporate heterogeneity in endowments or behavior into economic models. A particularly natural and promising approach to account for economic phenomena that result from the (bounded) rational interaction of heterogeneous economic agents is the use of agent-based computer simulation models. Phenomena of such type are abundant (the avalanche-like dynamics in the network of connected commercial banks inducing the current economic crisis is just one prominent example in that respect), and a large amount of insightful agent-based research has addressed a wide range of relevant economic issues (see e.g. the Handbook of Computational Economics Volume II edited by Tesfatsion and Judd (2006) and Volume IV edited by Hommes and LeBaron (2018) for overviews).
A main research topic at ETACE is the development of micro-founded macroeconomic heterogeneous agent-based models that can be used as a unified framework for policy analysis in different economic policy areas. Based on work carried out in the EU-funded Eurace Project (2006-2009), the Eurace@Unibi Model model has been developed and used as a tool for the analysis of various economic policy questions related to issues of technological change and economic growth, labor market policies, social cohesion and convergence, and to study banking and credit market regulations. The Eurace@Unibi model is among the most sophisticated and well-documented models in this domain of economic research. It has strong empirical microfoundations, reproduces a large set of empirical stylized facts and has been the basis for work in the EU-funded projects ISIGrowth (2016-2018) and GROWINPRO (2019-2022). The Eurace@Unibi model is continually adapted and extended to suitably address concrete research questions.
In addition to work using the Eurace@Unibi model, the ETACE group keeps developing and applying other simulation models for the analysis of important economic issues. These include structural change in light of technological progress and automation, economic and epidemic effects of lock-down policies to contain a pandemic or the role of expectation formation processes. Also, members of ETACE develop and apply statistical methods and concepts (e.g., spline functions) for model calibration as well as systematic and rigorous analysis of policy experiments using agent-based simulation models.
Publications by ETACE members in this Research Area: