Approaches to Accounting for the Financial Sector in Agent-Based Modeling

Abstract

The article discusses approaches to introducing the financial sector into agent-based models, as well as various options for modeling the behavior of agents (firms and households) directly interacting with the financial system. Agent-based models can explain how some important macroeconomic phenomena can be generated by evolving networks of interactions between boundedly rational agents in economies, where the underlying rules of interaction can evolve endogenously over time. The importance of accounting for financial activity in macroeconomic models for reproducing business cycles and predicting financial crises is emphasized. Consideration of financial blocks in macroeconomic agent-based models includes an analysis of approaches to modeling the choice between household savings and consumption, making decisions about the amount of investment by private sector firms, as well as the rules for setting interest rates and regulatory limits by banks. Possibilities for improving approaches in terms of taking into account interest rates when modeling household savings are proposed, as well as long-term planning based on the calculation of discounted cash flows with an assessment of the impact of changes in output on market prices when forming firms’ investment plans.