Since November 2009 I have been noting that the S&P500 (represented here by SPY) has been closely retracing the path that it followed in 2003 and 2004. Now that 2010 is drawing to a close, I took a look back at 2005 to get one vision (hallucination?) of what lies ahead for 2011. My updated graph showing SPY, normalized SPY volume and VIX over those 3 years is shown below:
A couple things jumped out to me, looking at the 2005 results:
- 2005 was a sideways year, with very little movement. The low for the year was around 114, the high 128—boring.
- Volume really picked up later in the year. Did the retail investors, burned by the tech crash finally come back to the market?
- Volatility in 2005, as measured by the VIX was quite low. People are remarking about our recent lows of 17, but 2005 averaged 12.8.
Although the 2009—2010 price path has crossed over the 2003-2004 path at least every couple of months, the 2009-2010 volatility has been consistently much higher than the numbers 6 years ago. Is this due to lingering fear from the truly scary days of 2009, or has something structurally changed in the market? There is no question that the inherent speed of the market has multiplied (Nanex reports that the Flash Crash propagated from Chicago Futures to the New York exchanges in 20 milliseconds). For reference light takes about 4 milliseconds to cover that distance. I’m pretty sure the market linkages were at least a factor of 10 slower in 2004. Whether this directly leads to increased volatility is open for debate, but I don’t see how it could inherently reduce volatility. I’m pretty confident that 2011’s volatility will on average stay higher than 2005’s results.
I don’t believe that price pattern coincidences are necessarily predictive. I think there are macro-economic similarities between the two periods, but there are also a lot of differences (e.g., increasing correlation between asset classes, on-going intervention by the Fed, very low inflation, etc.). My gut feel is that 2011 will be better than 2006, but my crystal ball is not particularly reliable…