Showing posts with label Eijffinger. Show all posts
Showing posts with label Eijffinger. Show all posts

Dec 11, 2009

Systemic Risk

In an excellent paper called 'Defining and Measuring Systemic Risk', professor Sylvester Eijffinger of the Tilburg University discusses actual developments around one of the most interesting risk topics of this moment: systemic risk (not to be confused with systematic risk).

Just a short warming up to actually download and read this excellent article:

Main target of the 2010 launch of the European Systematic Risk Board (ESRB) is trying to identify and avoid future financial crises before they start. This implies that ESRB's main issue is 'how to detect systemic risks '. All this -of course - under the lead of the European Central Bank (ECB).

First of all the ECB does not have a clear concept of systemic risk, nor in the academia there exists a generally accepted definition. However, the G10 definition provides a good starting point:

Systemic risk
Systemic risk is the risk that an event will trigger a loss of economic value or confidence in, and attendant increases in uncertainty about, a substantial portion of the financial system that is serious enough to quite probably have significant adverse effects on the real economy

This still sounds pretty complex, and it is.
To get the right feeling, take a look at the next diagram illustrating a network of Credit Default Swaps (CDS) contracts:

In his blog 'complexity is our enemy' Steve Hsu, Professor of physics at the University of Oregon, explains in short and in simple words the principles and problems of the Credit Default Swap Market.

Hsu perfectly illustrates why some financial institutions are 'too connected to fail', as opposed to 'too BIG to fail'. Systemic risk is all about complexity.

New early warning models

There are several new models that can predict a financial crisis. Key challenge is to find a model with an indicator that predicts a potential crisis (just in time) with high probability, while at the same time minimizing errors of type I errors (missing crises) and type II false alarm).

One indicator can be qualified as the best current performing indicator: 'The global private credit gap', by Alessi and Detken (2009). This method predicts 82% of the crises correctly and has a 32% share of false alarms. 95% of the crises (price boom/bust cycles) are signaled in at least one of the 6 preceding quarters and the difference in the conditional and unconditional probability of a boom following a signal is 16%

Individual Institutions’ Contribution to Systemic Risk
For measuring risks of individual banks, a measure called CoVaR was developed by Adrian and Brunnermeier. The CoVar model measures the marginal expected shortfall (MES) as used in Value at Risk (VaR) as well as the systemic expected shortfall (SES).
Eijffinger's Conclusion
Finding new early warning instruments that are effective, easy to use, and independent of the interest-rate instrument seems to be an impossible task. And yet there is a solution according to Sylvester Eijffinger: "Central banks should give the growth of (broad) money supply more prominence in their monetary policy strategies."

The ECB with its often criticized monetary pillar may have a head start. Important central banks, such as the Bank of England and the United States Federal Reserve, kept their key interest rates too low for too long leading to a long period of double-digit growth in money supply.

The ECB was more cautious. To be sure, the fall of he risk premium on financial markets, the development of all kinds of exotic derivatives, and these derivatives’ subsequent misuse sowed the seeds for this crisis, but those factors could not have caused the crisis without the plentiful rainfall that allowed those seeds to grow.

What can pension funds and insurers learn from this?
The answer is simple:
  • Make Risk Management top priority nr. 1
  • Develop and implement in advance - cross financial institutions - early warning models.
  • Insist upon regulators to create a world wide central registration data base that registers and reports all possible derivate transactions in the financial market. Every financial institution has to report every transaction in a preformatted form.
  • New financial products are subject to approval ('no objection') by the regulator before market launch.
This way regulators will have a complete transparent view cross financial institutions. Systemic problem solved.

- Eijffinger:Defining and Measuring Systemic Risk
- The global private credit gap
- CoVaR modelM
- Steve Hsu: complexity is our enemy