RELATIVE VOLATILITY OF STOCKS & BONDS

This post concerns the volatility in the US Bond Market relative to the US stock market. In particular, why is it that volatility is currently higher in bonds than in stocks? Does this mean anything? To be honest, I am not sure, but I can lay out some possibilities.

First, the “flight to quality” as economic armageddon approached was significant, unique, and had never been seen before. Worldwide money flows initially went into US bonds and dollars, but I still think the level of uncertainty is high regarding credit conditions in the US. Given all the money that was pumped into the system the fear of higher inflation lingers. Others feel that anemic economic conditions will dampen inflation and inflationary expectations. My opinion is that most people do not understand or appreciate the magnitude and reality of what has really happened over the past few years. The bond market is also trying to figure everything out, thus the higher volatility.

Second, very little to nothing has been done. America just borrowed more money, threw it at the problem, and continued as if nothing has happened. America may have trouble financing its problems into the future. The bond market has noticed this and I believe it is another reason for its higher relative volatility.

Third, America needs to get its fiscal house in order. How we are perceived by the rest of the world is reflected in our markets. How we handle our economic problems matters. Markets are affected by many variables, but two of the most significant factors are confidence and perceptions. These are human emotions driven by many things, nonetheless, they can easily change. Lower confidence is a result of higher uncertainty. In my opinion, lower confidence and negative perceptions of the US have also contributed to the higher relative volatility of bonds to stocks.

RELATIVE VOLATILITY 10YR NOTE TO SP500

In some earlier posts I discussed dissipative systems and entropy. For purposes of this post I will use the interaction of two open and dissipative systems, the US bond market (10 year notes) and the US stock market (SPY). In addition, I am making some assumptions. The first is that I like to think of entropy in the markets as equivalent to the amount of information and uncertainty that exists. The second is that higher standard deviation is equivalent to higher entropy. A higher standard deviation means prices are more variable and erratic. This is a result of more uncertainty, and consequently, more entropy in the market. In addition, this can also be thought of in reverse--less risk leads to less uncertainty, and consequently, less entropy.

In general, standard deviation is used as a measure of volatility in a market. Markets become more volatile as the level of uncertainty grows. I have created two graphs which show the standard deviation of both markets since the beginning of 2007. One graph shows that if the relative volatility is greater than 1, then stock prices have a higher level of entropy compared to bond prices. If less than 1, then bond prices have more entropy than stock prices. The second graph shows the actual standard deviation of each market. Two quick conclusions can be made from these charts. The first is that volatility has significantly declined in both markets since the height of the financial crises, and second, the volatility in bonds is higher relative to stocks. More next time…

OUR PROPRIETARY VOLATILITY INDICATOR

We have tested this statistic going back to the 1950's on the SP500 Cash Market. Above I have placed 3 graphs of varying time lengths. As can be seen by the graphs, when our volatility statistic is above .95%, the market has "high volatility," and below .95 it has "low volatility." The higher the number, the more volatile the market, the lower the number, the less volatile the market. Over the long term this statistic is quite accurate in defining the underlying volatility of the market. I have never had much success with any other indicators or calculations of volatility. The way we use this figure is that we change our stops when the volatility goes above or below the .95% threshold. We will publish our figure each Friday in the Volatility Statistic page above.

VOLATILITY CLUSTERS AND PERSISTS

This is a graph I produced of the daily range of the SP500 Emini futures from July 2002 to the end of last week. The lesson is that volatility clusters and persists. A comparison of 2004-2006 ranges to 2007-2008 ranges clearly shows how the ranges when low, stayed generally low for a couple of years. Likewise, when the ranges began to expand at the end of 2007 through 2008 they tended to continue to stay large for a couple of years. Large ranges and high volatility are a double-edged sword. They can bring large gains but also large losses. The key is to try to understand when the underlying volatility conditions of the market are changing and to modify your trading plan. For example, using different stops or adjusting position sizes.