2004 All Over Again? Hedging a Falling Market

|
 |  Includes: IWV, SH
by: Todd Feldman

Based on anecdotal evidence and behavioral finance modeling, we suggest that the US stock market in 2011 will be similar to 2004.

Let's take a look at the calculated return of the S&P 500 based on data from Yahoo Finance from 2000-2002, 2003, 2008, and 2009-2010.

Table 1: S&P 500 Return Comparison

2000-2002 2008 2003* 2009-2010*
Return -39.54% -40.09% 27.26% 43.85%
Peak to Trough Return, Trough to Peak -48.92% -56.65% 49.05% 98.13%
Click to enlarge

*2003 and and 2009-2010 period returns include first two months of 2004 and 2011.

We classify the periods of 2000-2002 and 2008 as crisis periods and 2003 and 2009-2010 as post crisis recovery periods. The returns are comparable between the two classified periods, except the period 2009-2010 did post a a much greater return compared to 2003.

These comparisons leads to my prediction that 2011 may be similar to 2004. However, the results above are anecdotal at best.

Let us take a more scientific approach. We will use behavioral modeling to better understand whether these periods look similar. To do so we have created two propritiery measures that are akin to sentiment measures. These measures are similar to the VIX in that a higher measure captures more fear in the market. However, we find these measures produce more predictive power (Feldman 2010, Journal of Behavioral Finance). The measures are based on portfolio manager behavior. We have created a long and short measure. The difference between the two is the memory rate. The long measure puts more weight on past data where as the short measure puts most of the weight on the most recent data.

The chart below overlays the two measures for both the periods November 2003 to March 2004 and November 2003 to February 2004.

Figure 1: US Behavioral Measures for 2004 and 2011 Periods

Propritiery Behavioral Measures

Take a look at the two measures beginning in November of 2003 and November of 2010. We notice a similar pattern in the movement. The long measure is exactly the same for both periods. The measure is downward sloping, which is a bullish signal, however moving to a contrarian position. The short measure in 2004 is moving in the same manner to the 2011 short measure. The short measure in 2011 is lower, implying greater short-term bullishness than in 2004.

What does this mean? Well, if you take a look at the March 2004 short and long measures, it can been seen that they increased significantly at the beginning of March. During that time the S&P 500 fell almost 5%. If the 2011 behavioral indicators are to continue following the 2004 indicators, we predict a pullback this week as the 2011 short-term measure is one week ahead of the 2004 short-measure. However, since the 2011 short-term measure is below the 2004 short-term measure, we believe the pullback will be more substantial.

As I write this piece the markets already fell for the week. We predict markets will continue to fall. Therefore, it is best to protect your portfolio. The easiest way for investors to protect their portfolio is invest in a short ETF such as SH. The amount an investor should hedge is based on the beta of their portfolio relative to the S&P 500. If the portfolio worth is $10,000 and your beta is 2, than you should buy $20,000 of SH. Wait for a 5-10% correction and then remove the hedge.

Disclosure: I have no positions in any stocks mentioned, but may initiate a short position in SH over the next 72 hours.