AGENT–BASED KEYNESIAN MACROECONOMICS - An Evolutionary Model Embedded in an Agent–Based Computer Simulation

Subject of the present study is the agent-based computer simulation ofAgent Island. Agent Island is a macroeconomic model, which belongs to thefield of monetary theory. Agent-based modeling is an innovative tool thatmade much progress in other scientific fields like medicine or logistics.In economics this tool is quite new, and in monetary theory to this datevirtual no agent-based simulation model has been developed. It istherefore the topic of this study to close this gap to some extend. Hence,the model integrates in a straightforward way next to the common privatesectors (i.e. households, consumer goods firms and capital goods firms)and as an innovation a banking system, a central bank and a monetarycircuit. Thereby, the central bank controls the business cycle via aninterest rate policy; the according mechanism builds on the seminal ideaof Knut Wicksell (natural rate of interest vs. money rate of interest). Inaddition, the model contains also many Keynesian features and aflow-of-funds accounting system in the tradition of Wolfgang Stützel.Importantly, one objective of the study is the validation of Agent Island,which means that the individual agents (i.e. their rules, variables andparameters) are adjusted in such a way that on the aggregate level certainphenomena emerge. The crucial aspect of the modeling and the validation istherefore the relation between the micro and macro level: Every phenomenonon the aggregate level (e.g. some stylized facts of the business cycle,the monetary transmission mechanism, the Phillips curve relationship, theKeynesian paradox of thrift or the course of the business cycle) emergesout of individual actions and interactions of the many thousand agents onAgent Island. In contrast to models comprising a representative agent, wedo not apply a modeling on the aggregate level; and in contrast toorthodox GE models, true interaction between heterogeneous agents takesplace (e.g. by face-to-face-trading).