Center for Financial Studies (CFS)
Refine
Document Type
- Report (2)
- Working Paper (1)
Language
- English (3)
Has Fulltext
- yes (3)
Is part of the Bibliography
- no (3) (remove)
Keywords
- networks (3) (remove)
Institute
We develop a dynamic network model with heterogenous banks which undertake optimizing portfolio decisions subject to liquidity and capital constraints and trade in the interbank market whose equilibrium is governed by a tatonnement process. Due to the micro-funded structure of the decisional process as well as the iterative dynamic adjustment taking place in the market, the links in the network structures are endogenous and evolve dynamically. We use the model to assess the diffusion of systemic risk, the contribution of each bank to it as well as the evolution of the network in response to financial shocks and across different prudential policy regimes.
We develop a dynamic network model with heterogenous banks which undertake optimizing portfolio decisions subject to liquidity and capital constraints and trade in the interbank market whose equilibrium is governed by a tatonnement process. Due to the micro-funded structure of the decisional process as well as the iterative dynamic adjustment taking place in the market, the links in the network structures are endogenous and evolve dynamically. We use the model to assess the diffusion of systemic risk (measured as default probability), the contribution of each bank to it as well as the evolution of the network in response to financial shocks and across different prudential policy regimes.
The analyses of intersectoral linkages of Leontief (1941) and Hirschman (1958) provide a natural way to study the transmission of risk among interconnected banks and to measure their systemic importance. In this paper we show how classic input-output analysis can be applied to banking and how to derive six indicators that capture different aspects of systemic importance, using a simple numerical example for illustration. We also discuss the relationship with other approaches, most notably network centrality measures, both formally and by means of a simulated network.