Systemic risk is a classic example of an emergent phenomenon. It occurs when, through their interactions with one another, individuals acting for their own purposes generate extreme behavior at a system-wide level. Such behavior is often difficult to anticipate based on the low-level behavior of the individuals. Ironically, in many cases each individual seeks to minimize his or her own risk, and the actions taken to do this collectively generate large risks at a systemic level.
Recent events in financial markets provide a good illustration on many different levels. A combination of new, supposedly risk-reducing derivative products, excessive use of credit, and a classic speculative bubble, led to the deepest recession since the depression. Among underlying determinants of systemic risk, liquidity risk has been recognized to be among the most prevalent and pernicious risks facing financial markets and institutions in terms of their function and stability. Furthermore, liquidity risk, both fundamentally and in its manifestation, is an extremely complex phenomenon, which manifests itself in spread risk, counterparty risk and settlement risk, and thus has implications for credit risk and operational risk. The current crisis was further exacerbated by a tangled web of complex overlapping agreements and a general lack of transparency, which meant that no individual actor could easily collect the data to understand the potential systemic risks, much less understand them.
The SFI systemic risk initiative will seek to bring complex systems ideas to bear on the problem of systemic risk. While this initiative will focus on financial markets, it will also bring insights from many other disciplines, including ecology, epidemiology, and network theory. It will be comprised of academics, practitioners, and policymakers. The initiative will seek to apply concepts, principles, and methods from complexity sciences and an interdisciplinary approach to develop a useful framework for understanding systemic risk. On a large scale, the workshop will focus on the following areas and questions:
- Connectivity. What is the role of the increased complexity of the financial network (in terms of connectivity, nonlinear interactions and feedbacks, etc.) in determining the robustness of the system to exogenous or endogenous shocks? Which insight can we gain for other disciplines (ecology, epidemiology, archaeology) where networked agents are subjected to stress?
- Adaptation. Financial system is a complex adaptive system with bounded rational and partly informed agents. Feedbacks and behavioral biases strongly affect agent's response to systemic events. What is the role of adaptation in systemic risk?
- Homogeneity. What is the role of the increased homogeneity of the financial system (in terms, for example, of risk hedging strategies) in determining the fragility of the system. When is risk minimization at a local level beneficial to the risk minimization at the global (system) level? When is increased connectivity beneficial in dispersing risks?
- Uncertainty, metrics, and data availability. What empirical knowledge of the financial system at a global level do we have? Which type of data should be made available to scholars and regulators to improve the understanding of the robustness and fragility of the financial system? Which information should be made available to market participants to improve their response to stress? Can we imagine a way of mapping the financial system? What are the proper metrics to quantify systemic risk?
- All the above areas will be extremely relevant for understanding the role of regulators and policy makers in designing system rules able to minimize the susceptibility of the financial system to systemic crises.
- A complexity approach to systemic risk in financial markets would address issues of relevance for traditional notions of financial fragility applied to a system with increasing intermediation by large and/or complex non-bank financial institutions like hedge funds, broker-dealers, and government-sponsored enterprises. Specific practical implications related to the study of complexity issues related to systemic risk underlying financial markets would include:
- How does interconnectivity and coordination influence price dynamics via network externalities?
- What is the role of large and complex non-bank financial intermediaries in modern financial markets, including their interactions with banks (and between banks) and the effect of these interactions on systemic risks?
- What is the appropriate role for institutional non-market mechanisms and how can market and non-market mechanisms be harmonized to address systemic risk?
- What is the ability of investors and regulators to observe and quantify risk?
- What is the potential role for regulation, public policies, market mechanisms, and financial innovation to reduce (or increase) systemic risks?
- How do leverage, funding, and asset liquidity relate to systemic risk?
- What are the determinants of liquidity and the role of liquidity in maintaining financial system stability?
- What are the systemic implications of trends in industry practice, such as clearing and settlement systems and market-risk management techniques?
- What is the structure and mechanics of debt and equity securities markets, including related derivatives markets?
- What is the relationship between real options and systemic risk, and what can and do market mechanisms serve?
How does one measure “too big to fail” versus “too interconnected to fail”?
The purpose of the initiative is to establish, explore, and promote a common framework and approach for analyzing and addressing issues of systemic risk within the real economy and the financial markets, based upon complexity science approaches to dynamic and adaptive economic systems; in particular, the study of economic and market phenomena related to externalities, state- and path- dependence, feedback loops, nonlinearities, interdependence which manifest themselves as observed episodic liquidity and contagion effects, information cascades, runs, bubbles, and crashes. Of particular interest are methods that include (but are not limited to) social network analysis, agent-based computational modeling, and (extensive-form) evolutionary game theory to institutional structure/dynamics and market microstructure.
- Morten L. Bech, Senior Economist, Money and Payments Studies Function of the Research and Statistics Department, Federal Reserve Bank of New York
- D. Wilson Ervin, Chief Risk Officer, Credit Suisse
- J. Doyne Farmer, SFI Professor
- John Geanakoplos, SFI External Professor, Science Board member, Chair of Science Steering Committee, and Ex Officio Trustee, and Professor of Economics, Yale University
- Andrew W. Lo, Director, Laboratory for Financial Engineering, MIT
- David K.A. Mordecai, Senior Advisor, Compass Lexecon, as discussant
- George Sugihara, Professor, Scripps Institution of Oceanography, UCSD
- Duncan Watts, SFI External Professor, Research Scientist, Yahoo! Research and Adjunct Senior Research Fellow, Columbia University
If you do not see videos and you are logged in, please refresh the page.