Investing

Return to Article

Following the Flight to Low Quality

James Altucher

Part of this is due simply to the laws of physics -- large objects require more energy to move. In other words, a company making $100 billion in annual revenue is less likely to double than a company making $10 million in revenue and just beginning to gain market share.

However, because of this move in the Russell, we are getting the usual pundits claiming that we are seeing a "flight to low quality" and that this usually precedes a crash of some sort.

Let's find out if this is true.

To verify this, I took the Russell 2000 and the S&P 500 indices and mapped the average difference, or spread, between the two indices. I then calculated out when the spread for any given one-month period exceeded the average spread by more than 1.5 standard deviations in either direction. In other words, if the Russell 2000 ran up for a month with the S&P 500 largely flat, then this condition would be triggered. Or, if the Russell 2000 moved down sharply, or more aggressively than the S&P 500 moved down in a one-month period, then the condition would trigger.

If the Russell 2000 moved up too fast, the simulation would short the Russell 2000 ETF (IWM Quote - Cramer on IWM - Stock Picks). If the Russell moved down too fast, the simulation would trigger a long. The trade would exit when the spread would go down to 0.5 standard deviations over the average spread.

In other words, when the spread got to the point at which the pundits were yelling about it on TV, then a trade would result. When the spread was back to normal, the system would exit the trade.

First off, what happens when the Russell 2000 moves up much faster than the S&P 500?

Many analysts think this "flight to low quality" is a bearish sign that "the bull is getting tired" and one should start short-selling.

This would be a quick way to lose money.

Since July 6, 2000 (the IWM's inception date), there have been 51 occurrences of this short signal, and you would have made money on the trade just 19 times. You would've lost money on each other instance and the average return across all 51 occurrences is a negative 0.46%. What's more, the trade would have cost money on eight of the last eight occurrences. Going long the S&P 500 at these points results in an average return of +0.16%, but that result is not statistically significant.

Currently, this simulation has an open trade that started on March 24. If you had shorted then, you would be -2.28% down.

So what if the Russell 2000 goes down much more quickly than the S&P 500?

Going long at these points (52 occurrences) would've resulted in a successful trade 71% of the time for an average return of 0.77% across all 52 trades. The average holding period is six days.

An example is spelled out in the chart below. The top panel shows the number of standard deviations the spread of the Russell 2000 and the S&P 500 is from its norm. The middle panel is the ratio of Russell 2000 over SPY and the bottom panel is the Russell 2000, including the pink circle indicating a short trade was entered on March 24, 2006.

Work the Long Side The long trade works better

Click here for larger image.

Source: Formula Capital