Wednesday, June 22, 2011

The CISS Indicator (fg)

There is a new indicator for measuring the systemic risk component of financial risk generated by ECB staff and recently published in the latest ECB Financial Stability Review; it's called the “composite indicator of systemic stress", or simpliy CISS.

The aim of CISS is to monitor the current level of systemic risk in financial markets by aggregating individual risk measures of various market segments including the bank and non-bank financial intermediaries sector, money markets, equity and bond markets, and foreign exchange markets. According to Hollo/Kremer/Duca (2011), each sub-index is calculated as the simple mean of the transformed values of three individual stress measures for each market segment. The five sub-indices are then aggregated on the basis of their time-varying cross-correlation structure in the same way as the overall risk of an asset portfolio is calculated from the risk characteristics of its individual assets. The novel feature of CISS is that the systemic risk measure relatively increases in times of high correlation of the sub-indices where stress prevails in several market segments at the same time. Such a situation implies that for financial investors, it becomes difficult to diversify and hedge their portfolios since market segments are nearly perfectly correlated.

Meanwhile the indicator nicely features the massive soaring of systemic risk starting in 2007 and it quite solidly depicts the implications of the sovereign debt crisis in the euro area, there is one decisive weakness in this measure: it is a measure designed to give a current picture of financial stress. However, it is incapable of mirroring the build-up of financial instability in the run-up to the financial crisis.

For monetary policy and macroprudential regulation, what is essentiel is the reliance on an ex-ante measure for the degree of financial instability and the likelihood of a near-time costly bust in economic terms. In this respect, the CISS is, like any other currently existing indicator, not able to quantitively sketch the build-up of systemic risk in the background of the financial system, where, at the same time, most current risk measures do NOT indicate future financial stress (see, for instance, the low measured corporate risk spreads in 2004-07 which did not indicate financial instability despite evolving financial imbalances due to massive lenghtening of financial intermediation chains and increased network activity). Consequently, it would turn out to be productive, if the CISS would be applied in order to address the issue of evolving financial imbalances ex-ante.