Systemic risk in the financial system is risk tied not to one specific firm, but to the interactions between firms – for example, through possible avalanches of spreading financial distress. One important form of systemic risk arises from indirect links between financial institutions, created when financial institutions invest in the same assets. Such “overlapping portfolios” can help the consequences of financial shocks propagate through the system, especially through the devaluation of common assets. Devaluations trigger asset sales, which lead to further devaluations in an amplifying feedback.
During “the Great Moderation,” for example, a period starting in the mid-1980s until 2007, portfolios of financial institutions became increasingly similar as banks around the world held mortgage-backed securities and collateralized debt obligations in their portfolios. During the U.S. sub-prime mortgage crisis, financial contagion linked to overlapping portfolios led to losses of more than $500 billion. Due to the lack of empirical data, however, very little research has explored the dynamics of financial distress in networks of institutions with overlapping portfolios.
In recent work, LML Fellow Fabio Caccioli and colleagues develop a novel method to quantify systemic risk from overlapping portfolios, extending the notion of systemic importance in financial networks to bipartite networks of financial institutions and securities using a unique data set for Mexican banks over the period 2004-2013. Building on earlier work, they go on to define a novel risk measure to quantify the expected loss due to systemic risk while taking cascading effects into account through explicit use of financial network topologies. They finally also use this risk measure to quantify the marginal contributions of individual direct and indirect exposures to the overall systemic risk.
The paper is available at https://arxiv.org/pdf/1802.00311.pdf