Working Paper Explores Literature on Financial Networks and Interconnectedness
Published: October 20, 2015
The OFR released a working paper today that reviews the rapidly expanding research on network models of the financial system.
Since the financial crisis of 2007-09, policymakers have highlighted interconnectedness between financial firms as a potential amplifier of financial shocks. In 2011, the international Basel Committee on Banking Supervision adopted interconnectedness as one of five categories of indicators to identify global systemically important banks (see the recent OFR briefs on this topic).
Network models are tools to analyze interconnectedness. The working paper “Contagion in Financial Networks” focuses on networks of payment obligations between financial firms created through lending, derivatives transactions, and other financial contracts. These connections can allow beneficial transfer of risk and capital allocation. But the same connections also create channels for shocks to spread from one part of the financial system to others.
The paper reviews research analyzing the balance between these conflicting effects and finds that the potential for contagion depends on network structure and factors such as the size, leverage, and asset quality of individual financial institutions.
Connections between the balance sheets of financial firms create two mechanisms for spreading shocks through the financial system: default cascades and funding runs.
In a default cascade, the failure of one firm to meet its payment obligations causes other firms to fail. Concerns about this type of scenario led to the 2008 rescue of American International Group, or AIG, an insurance giant that was unable to meet obligations to counterparties in credit default swaps and other securities.
A funding run works in the opposite direction, when lenders withdraw credit or increase collateral demands from borrowers who are then forced to liquidate their investments. This mechanism describes the response of the short-term funding market after Lehman Brothers collapsed.
Both channels have been studied through network models, sometimes in combination. The OFR paper describes examples from past crises, details the mechanics of default cascades and funding runs, and surveys empirical research on features of interbank lending networks around the world.
Although market participants and supervisors have limited information to precisely measure interconnections among financial institutions, the paper shows that it is possible to bound the extent to which interconnectedness amplifies shocks. The analysis leads to simple indicators of interconnectedness that can be used in practice for monitoring purposes.
OFR research has taken several approaches to network analysis. We have mapped portions of the financial system for example, to explain liquidity and funding in securities financing transactions (see A Map of Funding Durability and Risk). Another recent paper used an agent-based model to analyze the vulnerability of markets to fire sales and runs (see An Agent-Based Model for Financial Vulnerability). An earlier paper estimated the extent to which interconnections increase expected losses and defaults under a wide range of shock distributions (see How Likely is Contagion in Financial Networks).
We’re also interested in the potential risks posed by central clearing. One paper showed that large clearing members can pose increasing concentration risks to a central counterparty over time (see Systemic Risk: The Dynamics under Central Clearing). Another showed that margin charges collectively create incentives for swap dealers to split their positions among multiple central counterparties, effectively “hiding” potential liquidation costs (see Hidden Illiquidity with Multiple Central Counterparties).
The new working paper serves as a valuable tool for researchers by providing a comprehensive review of work in this area. It discusses the challenge of understanding how and why networks form in the first place and reviews initial progress in addressing these questions.
Greg Feldberg is Acting Deputy Director for Research and Analysis at the Office of Financial Research