Caught in the Web -- Finance & Development September 2012

ANDREW Haldane, Executive Director for Financial Stability of the Bank of England, once compared the collapse of the investment banking firm Lehman Brothers in late 2008 to a virus outbreak. Fear that Lehman’s demise would spread caused widespread panic that led to a freeze in borrowing and lending in a number of financial markets—much like the fear of a virus might lead people to avoid contact with one another. The effects of Lehman’s fall were difficult to isolate. In fact, the reach and macroeconomic costs associated with Lehman’s demise were much larger than one could have anticipated based on the size of the initial event alone (Haldane, 2009). It is not clear what enables relatively small shocks, such as Lehman’s demise, to reverberate through the larger economy and why such shocks are often difficult to isolate. The answer seems to lie with the complexity of the financial connections among agents, institutions, and countries. When unexpected bad events occur (negative shocks, in economic parlance), uncertainty sets in and disrupts economic activity. The failure of Lehman, which many believe sparked the global financial crisis, suggests that some institutions are so central to the financial system that their failure can be catastrophic. Similarly, some countries are so integrated into the global financial system that a negative shock there can reverberate through the entire global economy. Commentators coined the term “too interconnected to fail” to identify players (financial institutions and countries) that are so caught in the web that they pose a systemic risk— that is, their failure would threaten the stability of the entire financial system. Several years into the crisis, it has become clear that to safeguard financial stability, it is important to understand how interconnectedness can be defined and measured and how it relates to the concept of systemic risk. A methodology that can help do this is network analysis, which is used in epidemiology to investigate the spread of diseases. Network analysis looks at the financial system as a set of players connected with one another through financial contracts (see box). For example, the interbank market is a network of banks that lend to each other on a short-term, often overnight, basis. At every point in time, the banks in the network have exposure to each other represented by the total claims or liabilities accumulated through lending and borrowing. There are a number of ways that a bank becomes more important, or central, in the network. It can have borrowing and lending relationships with a large number of banks. It can have claims on other important banks. It can network analysis looks at financial connections in a globalized world and the risks they pose