Last week we mentioned the concept of measuring market turbulence as an additional risk indicator. One approach for assessing the level of turbulence is to look at the difference in the tail (possibility for extreme events) as measured by the Normal VaR (or ETL) and the Cognity Fat-tailed VaR (or ETL) model. In the case of “normal” markets (i.e. no excess probability of extreme events), the two approaches would coincide. Widening of that spread shows increasing market turbulence, and - associated with that - increasing probability of extreme events.
To analyze the current market environment, we typically look at different markets and factors comparing the current Fat-tailed versus Normal VaR spread for each one of the segments to the average values observed during various important historical periods.
Today we present such an analysis on several MSCI regional indices:
All Countries World Index
For the MSCI AC World Index™ we observe that the 2016 Turbulence indicator (the violet line which is the average 2016 -VaR spread) is the same as the Average Turbulence post-crisis (market in blue). It is close to one-third of the turbulence during the crisis (indicated in pink on the charts - Sept-Dec 2008). We see that the pre-crisis period of 2006-Q22007 turbulence (in green) was at zero.
For the US Market, the picture tells a similar story with the exception being that in absolute terms the Crisis-period turbulence was about twice as high. Still, at present we can say that the 2016 turbulence is at average values for the post - 2008 market. One interesting observation, however, is that the beginning of 2016 is characterized by more pronounced spread levels, which is indicative of the increasing nervousness around the state of the US economy.
The Canadian market is at present above the average post-crisis turbulence, most probably a reflection of the dependence on oil- market and related industries. It is still less than one-third of the 2008 crash turbulence level.
Europe: UK, France, Germany, Greece and Russia
Turning to European markets: We see current turbulence in Germany and France at the average post-crisis levels. UK exhibits more pronounced turbulence since the beginning of the year reaching almost 50% of the 2008 crisis levels. Although this might seem worrying, it’s worth mentioning that the UK crisis turbulence was lowest in absolute levels among the analyzed markets.
Taking Greece as another interesting case: During 2015 Greece reached tremendously high levels of turbulence forcing the average post-crisis VaR spread almost to that of 2008, exceeded only by the 2016 spread.
Russia’s turbulence indicator is at present similar to its pre- and post-crisis averages. This market was never characterized by normality but in terms of through-time comparison, 2016 is a more or less steady period for the Russian market.
APAC: China, Japan and Korea
Turning to Asia: For China, we of course spot recent uptick in the turbulence, with current levels above average and approaching 40% of the 2008-crisis levels. Japan is at almost peak levels of turbulence, while pre-crisis it was a market characterized by normality. Korea is showing stable behavior.
Volatility of Volatility
Of course turbulence is not volatility - it is more a volatility of volatility, so absolute level comparison between markets should be done with care.
Obviously such characterization of risk adds an important layer of information. Having this type of analysis on a global equity portfolio in terms of risk contributors across various regions can guide important reallocation and hedging decisions should turbulence start to accumulate in a certain market.
What is of essence is that increasing volatility of volatility (turbulence) can be spotted before the absolute level of volatility raises. It can be used, for example, for hedge construction, when those are still relatively cheap.
The FinAnalytica Cognity system, now part of BISAM’s suite of portfolio analytics, provides similar analysis across asset-classes. Each factor of the Cognity multi-asset class global model is applied to depict specifically which factors could drive increased probability of extreme events. By monitoring the spread between the normal and Fat-tailed VaR within the system, our current Cognity clients are armed with early warning indicators of future hot spots for potential market extremes. Moreover, even in case of steady markets, such deeper understanding of the risk characterizations across markets brings more reliable assessments of the true risk contributors and diversifiers.
Robust risk analytics are essential for understanding and driving performance of investment portfolios. Cognity®, now a BISAM solution, is a unique, comprehensive, multi-asset class platform for market risk, portfolio construction and investment decision analytics. Visit www.bisam.com/risk to request a demo, or send an email to [email protected].