Curbs In: The Coming Octabox for the Masses

When the DJIA fell over the last couple of months from 14000 into the 7000s, one thing you didn’t see on CNBC was the “Curbs In” graphic, an indicator that NYSE had halted program trading temporarily due to excessive volatility. Why didn’t you see it?

Two reasons. First, in November 2007, the NYSE did away with the NYSE Collar (Rule 80A), the curb on program trading that kicked in when the NYSE Composite Index rose or fell 2% from the previous close. The NYSE cited the curb as ineffective in stemming volatility. Most of the time when you saw “curbs in” on CNBC, Rule 80A triggered the curb. As the curb applied to moves both up and down, you were most likely seeing it more often due to rallies than due to declines over the past several years.

The second reason you didn’t see the curbs in graphic is that the NYSE circuit breaker (Rule 80B) is far more difficult to trigger. In the event of a 10%, 20% or 30% decline in the DJIA, the rule goes into effect, halting trading based on the magnitude of the decline and the time of day the index crosses a threshold. If, for instance, a 10% decline occurs late in the day, trading continues. Other triggers can close the market entirely. And unlike Rule 80A, there are no curbs on the upside.

The rub is that the 10%, 20% and 30% declines aren’t dependent on the previous close—they’re dependent on the average closing value of the DJIA for the month prior to the beginning of the quarter. For Q4 2008, that was 11000, meaning that curbs won’t kick in this quarter unless we hit losses of 1100, 2200 or 3350 points.

So when the DJIA was scraping 7550, the market could have declined 14% and still not have triggered the curb.

Looking forward, let’s say that over the course of December, the average close of the DJIA ends up at 7500 due to more bad news and more de-leveraging. When the curbs are set for Q1 2009, the 10% curb will be at 750. Most of the huge daily drops over the last month have been in the range of 7% to 8%. If the DJIA gets to its 7500 average by lurking for most of December in the low 7000s and finishes strong at the end of the year at say, 8500, then rises a little more on New Year’s optimism to 9500, we could be in a situation where the same volatility we’ve been seeing (8-11% positive or negative days) could make it more likely to see curbs kicking into effect early next year, particularly if there are the kind of inter-week swings like we were seeing in October.

What will be the psychological effect of seeing a 750 point drop combined with a one-hour market halt to the average idiot retail investor?

UpsideTrader and Howard Lindzon see the CNBC Octabox respectively as a joke and a contrarian signal. The idiot retail investor, after a few months of routine 300-700 point changes sees a 750 point drop as old news. But trading halted, combined with a Deca-box of CNBC tools opining, AND a Curbs In graphic?

Chaos.

What I haven’t found a good source for is an estimate of the amount of de-leveraging that funds still have to accomplish, and whether or not the bulk of that is already done (or will be done) before the end of the year. From what I’ve read on the topic, it used to be common for program trading to take into account the effect of curbs going into effect, and adjust strategy accordingly; a number of times this fall, we seemed headed for 10% drops only to pull back at the last minute. Luck? Or could it be that no one really wanted a “markets halted” headline when they were sitting on piles of assets they couldn’t unload?

Regardless, most scenarios I can envision involve a negative feedback loop that involves continued high volatility and rallies that can’t be sustained. For the market to unwind in an orderly fashion, a buffer is needed. Since I’m a self-proclaimed idiot retail investor, I can make ridiculous predictions and look back at them and laugh. So my predictions:

• DJIA curbs for December based on 8900 average (890 points)
• Major January 2009 and February 2009 sell-off continues, finishing February at 7400
• Minor March rally back into the 8000s, with a Q2 set of curbs set at around 790

If the conventional wisdom is that you wait until hundreds of stocks are hitting 52-week highs before getting back in for the next bull rally (figuring you can afford to miss the first 6-9 months of a true bull market and save yourself getting burned by these false recoveries in the meantime), then I think the earliest one can safely jump back in will be sometime in November of 2009, and very likely it could be February 2010 if the scenarios above play out. For instance, AAPL hasn’t closed above $100 since November 3; for most of September it was above $120, and barring any rallies this spring, it’s 52-week high will be $107 on October 1, 2009, but only $91.86 come November 15, 2009. And it will very likely be a leader in the rally.

Is this just a retail investor’s capitulation, and the rally’s already begun? Possibly. Meanwhile, I’ll be making quick in-and-out hits and keeping most of my powder dry until I see some real strength.