An Agent-Based Computational Model for Bank Formation and Interbank Networks

Abstract

The aim of this thesis is to study the role of banking in society and the effect of the interbank market on the performance of the banking system. It starts by reviewing several studies conducted on empirical banking networks and highlighting their salient features in the context of modern network theory. A simulated network resembling the characteristics documented in the empirical studies is then built and its resilience is analyzed with a particular emphasis in documenting the crucial role played by highly interconnected banks.It is our belief that the study of systemic risk and contagion in a banking system is an integral part to the study of the economic role of banks themselves. Thus the current work focuses on the fundamentals of banking and aims at identifying the necessary drivers for a dynamical setup of the interbank market.Through an agent-based model, we address the issues of bank formation, bank runs and the emergence of an interbank market. Starting with heterogeneous individuals, bank formation is viewed as an emergent phenomenon arising to meet the needs for investment opportunities in face of uncertain liquidity preferences. When banks work in isolation (no interbank market), in the long run and through a long experience with bank failures, banking turns into a monopoly or a market with few players.By equipping banks with their own learning tools and allowing an interbank market to develop, fewer bank failures and a less concentrated banking system are witnessed.In addition, through a scenario analysis, it is demonstrated that allowing banks to interact does not weaken the banking system in almost all the cases, and improves the performance on multiple occasions.The work is concluded by studying the effects of a banking system on individuals and the economy in what is called social measures. We establish that the effects of banking on social measures such as consumption level, consumption inequality between individuals, long term investment and economic waste, varies significantly based on the structure of the society.iii

Modeling Emergence of the Interbank Networks

“…Castiglionesi and Lavarro (2011) presented a model with endogenous network formation in a setting with micro-founded banking behaviour. 3 These advances notwithstanding, owing to the complexity of the equilibrium-based studies of network formation, agent-based modeling of financial networks is one promising avenue that can be followed (Markose, 2012;Grasselli, 2013).…”