Digital Mapping of Financial Contagion : Introduction


A major premise of this project is that the comprehensive market and regulatory failure of 2007 which has led to the global economic crisis, can only have arisen from deep doctrinal flaws of the dominant economic and financial theories, deficits in the knowledge base and also a lack of appropriate quantitative modelling tools. The origins of the financial contagion from the sub-prime crisis in the US can be traced back to the development of financial products such as Residential Mortgage Backed Securities (RMBS), Collateralized Mortgage/Debt  Obligations (CM/DOs) and Credit Default Swaps (CDS) which were subjected to little or no regulatory scrutiny.  Over the period of the last 15 years or so when financial innovations were progressing at a rapid rate, there has been a marked underdevelopment of a modelling framework to articulate the massive interrelationships in the financial system implied by the workings of these new financial products.  Academic economists, policy makers and regulators were and continue to be restricted in their analysis of the crisis by a woefully inadequate set of modelling tools.

 

The COMISEF project is concerned with developing a multi-agent based computational economics (ACE) framework that can articulate and demonstrate the interrelationships of the financial contagion with a view to aid policy analysis.  The provenance of ACE has been attributed to a new paradigm based on markets as complex adaptive systems (see, Markose (2002, 2005, 2006) and Markose et. al  (2007) which include three  recent Special Issues) which subscribes to the view of computational incompleteness, self-reflexive decision problems which lead to heterogeneity in strategies, ‘surprises’ or innovation based structure changing dynamics and network interconnectivity of agent relationships which are prone to non-linear and extreme power law dynamics. These implications of the ACE framework stand in marked contrast to the extant representative agent analytical or econometric models with Gaussian and completeness assumptions of perfect rationality which can not include the type of outcomes recently observed as being likely.  Neither can the traditional framework provide appropriate quantitative modelling tools for complex market systems.  A large scale multi-agent model of the financial sector, which is capable of accessing fine grained data bases of the different financial players involved and also mapping the links and feedback loops between sectors, is being developed. Databases include the spatial and dynamic data on sub-prime mortgagees in the US, on and off balance sheet data of banks and other financial entities, data feeds from stock markets and so on.  It is intended that updated reports are given periodically to reflect the developments and experiments from a multi-agent based simulator of the financial contagion built to demonstrate the unfolding of many different scenarios of the crisis. In this report analysis is restricted primarily to the US commercial banking system. 

 

Under the rubric of how ACE works to give quantitative assessments of the ongoing financial crisis, it will be contrasted with extant macro- econometric modelling.  The major objective of the analysis is to assess the fragility of the US financial system and the extent to which the capacity of the US commercial banks to lend has been impaired.  On route a number of critical aspects of the crisis which include the CDS insurance relationships, the valuation of toxic mortgage backed securities with ongoing defaults of US mortgagees, the role of ‘bear raids’ which emanate from stock market mark downs are explicitly developed in the ACE model.

 

Financial Contagion

 

The diagram below maps the financial contagion, click on each section to browse through the related page.