Measurement of Systemic Risk and its Drivers

Abstract

The recent global financial crisis has taught us a tough and important lesson- there is a pressing need for containing systemic risk in the financial system. However, prior to Containing this risk and form the necessary regulations we need to measure this risk properly and study its sources. Systemic risk is a vaguely and broadly defined term that has changed significantly in the course of the recent 2007-2009 financial crisis.

This research is exploratory in nature and has three objectives. First is the measurement of systemic risk in the financial system and its evolution over time on a daily basis. The second objective is to study the linkages between systemic risk and the macroeconomic environment. Third is the role of firm specific characteristics in explaining systemic risk contributions.

Having been severely hit by the financial crisis in 2007-09, the US financial system provides the perfect setting for studying the evolution of systemic risk and its drivers if any. This research has collected the data on daily equity returns and quarterly balance sheets of 30 large US financial firms from 2001 to 2011. The sample firms have been divided into four groups 1) Depository institutions, 2) Broker-Dealers, 3) Insurance and 4) Other financial firms to study the group differentials in the evolution of systemic risk. The Systemic expected shortfall (SES) has been taken as the systemic risk contribution measure of a financial firm. The total risk in the system is the sum of the systemic risk contributions of all the members of the system.

The SES for a bank or a financial institution is defined as the expected shortfall of a bank's equity value below the target level conditional on a systemic crisis. The concept of this SES measure was given by Acharya et al. (2010) on the basis of a two period theoretical model of the economy having banks. The advantage of this model is that it is based on a theoretical economic model and also possesses the characteristics of a reduced form model.

In the present study, Systemic Expected Shortfall (SES) estimation results showed that there are indications of continuous increase in the systemic risk levels of the system before the Lehman Brothers' failure but probably this ex-ante risk was neither measured nor taken very seriously by the regulators. Systemic risk was at its peak during the crisis among all the groups though the level of the peak was different for different groups. Depository institutions and broker-dealer firms seem to contribute more towards systemic risk and though the level of risk came down after the crisis, it was still high towards the end of 2011.

Unemployment rate, industrial production growth (proxy for economic growth at monthly level), credit spread and realized volatility along with the estimated systemic risk were studied to examine the inter-linkages between risk and the macro-financial environment. We found that systemic risk is not affected by macro-economic shocks but the systemic risk Granger causes the macro-economic environment.

Role of firm specific characteristics, that is, which characteristic plays more prominent role in contributing towards the systemic risk has also been examined by employing a panel data analysis. Results have shown that there is a dynamic relationship between systemic risk and firm specific variables.

The research contribution of the present study to the systemic risk literature is at several levels. First, it seems to be the first study estimating the time series of Systemic Expected Shortfall on daily horizon. Second, firm specific share in the total systemic risk contribution of the firm is estimated in the study which gives new insights into studying systemic risk. Third, macroeconomic linkages with Systemic Expected Shortfall are studied for the first time, to the best of my knowledge.