Monday, June 4, 2012

Systemic Risk and Sovereign Defaults

An interesting 2011 paper from Andrew Ang of Columbia Business School and Francis Longstaff of UCLA Anderson Business School, develops useful findings which apply to current discussions about sovereign risk.

The authors take the U.S. and the ten largest states (CA, TX, NY...) and compare them to the Eurozone countries, using CDS spreads as a measure of changes in sovereign risk.  Their goal is to get a feel for how much of sovereign risk is truly systemic. An underlying principle of the statements of most international public officials is that global systemic risk is the bugaboo behind real risks to the system. Ang and Longstaff use CDS spreads as their market-based measure for pricing risks and changes in risk.  They then partition the overall risk into systemic risk and idiosyncratic risk. 

In the U.S., they find that California as a sovereign issuer has 5x the average risks of the other 9 U.S. states in the sample.  New York, by contrast, has surprisingly almost no systemic risk: risk associated with its securities are almost all idiosyncratic.  So, New York's securities could be held in a diversified portfolio of U.S. sovereign securities, reducing the portfolio risk.  Applying the same reasoning to California paper wouldn't yield the same risk-lowering benefit. 

For all ten of the largest U.S. states in the sample, the average systematic risk percentage is only 12.2%.  California's systematic risk percentage is 36.8%, certainly well above average.  However, in somewhat of a counter intuitive conclusion, systemic risk for U.S. state issuers is significantly lower than idiosyncratic risk specific to the issuer in question.

Europe by contrast shows results which suggest that the nature of sovereign systemic risk is quite different between Europe and America.  The average systemic risk percentage for all Eurozone nations is 30.9%, 2.5x higher than the average for American issuers.  The country with the highest proportion of systemic risk in total risk is France at 53.2%. 

Even as macroeconomic discussions today focus on the relatively small size of the Greek economy in the Eurozone, the authors find that the Greek systemic risk component is 3x that of Portugal, Spain and Belgium, the other weak sisters in the Eurozone. All of them, however, carry much more systemic risk than does German paper. 

What factors best explain changes in systemic risk?  What might constitute a transmission mechanism?  Changes in VIX and variations in overall equity market returns are the major factors explaining systemic risk in the EU and in American sovereign issuers.

Think about the findings here and ask yourself again, "Why would Germany be sanguine about moving towards fiscal union?"

No comments: