Monday, May 21, 2012

21/5/2012: Quick note on US Markets' Crash Indices

The risk-off thingy is starting to bite - with a few frantic calls over the weekend from across the Atlantic. People are shifting strategies like feet in Swan Lake's pas de deux. Here's an nice set of charts that shows we are in a precarious starting point to the risk-off market indeed.

The Yale University Crash Index - latest data takes us only through April, shows that the base off which we have entered May markets is already loaded with high risk:



April 2012 Institutional Index came in at 26.94 reading, which compares unfavorably to historical average of 36.86 and to crisis period average of 31.27. Jittery markets mean that 2011-present average is 29.88 - worse than crisis period average and that April 2012 was even worse than that. Meanwhile, individual investors index showed usual lags, with lower pessimism in April at 28.47, which is a better reading than 26.57 for crisis period average and better than 24.76 for 2011-present average. Still, individual investors are more risk conscious than historical average of 33.70.

One interesting bit - disregarding the issue of lags, historical correlation between two indices is 0.76 while crisis period correlation is 0.82, which suggests that May reading should come down like a hammer for individual investors. The same is confirmed by looking at changes in indices volatility. Standard errors for Institutional investors responses have compressed from historical 3.82 average to crisis period 2.99 average to 2.85 average for the period since January 2011. Similarly, for individual investors, historical average standard error is 3.36, declining to 2.731 for crisis period and 2.724 average since January 2011.

Note that per charts above, since the beginning of the crisis in mid-2007 (data shows clear break in data at June 2007), Individual investors index has been flat trending (volatile along trend), while Institutional investors index has been trending down (with loads of volatility, too).

No comments:

Post a Comment