Short-term Looking Dubious, if No Relief Buying
Sunday, May 30th, 2004 11:50am EST
Another One Bites the Dust
Bottom Line: An effective indicator over the past six decades has succumbed to the effects of program trading, as the proliferation of these computerized orders has transformed the Specialist Short Ratio into an unreliable gauge.
One of the most popular – and effective – indicators years ago was the Specialist Short Ratio. Recall that this ratio reflects the percentage of short sales on the NYSE that were done by specialists on that exchange. Specialists are market makers, buying and selling the stocks they are responsible for to match customer orders and trade from their own account if necessary to keep an orderly market. As we saw with the controversy several months ago, specialists yield a lot of power over the stocks they control, and have an unmatched ability to know what kind of demand there is (or is not) for their stocks. It is for that reason that watching how much stock specialists are buying or selling each week was an excellent indicator for many decades, and in fact had been somewhat effective up until only recently. When specialists made up a large percentage of total short sales, say 50% or more, then it was a good sign that the public was demanding so much stock that specialists had to short it to them, and that normally coincided with market peaks. On the other hand, a low Specialist Short Ratio, say 30%, meant that specialists were large buyers of stock and that type of activity was normally seen near market lows.
Over the past year and a half, the Specialist Short Ratio has been on a steady downtrend. It is one reason that some have given for the rise in stocks, as it indicated that the public was shorting so much stock it was a sign of excessive pessimism. I, too, have noted the extremely low ratio and in fact last year showed that on a long-term basis, it was the lowest in history. Other periods of prolonged public shorting invariably lead to positive markets. One thing that I noted consistently, though, was the large amount of convertible debt issuance last year. A very popular strategy among hedge funds is “convertible arbitrage” where they buy convertible bonds then turn around and sell short the underlying stock, thereby hedging their bonds in the case of a price decline. Since so much debt was issued, it created more need for short sales, and thus likely depressed the Specialist Short Ratio. My best guess has been that this strategy alone accounted for at least 20% of all short sales.
For the latest data released from the NYSE, covering the week ended May 14th, the Specialist Short Ratio dropped to 21%. This is extraordinarily low historically – in fact, it is the third-lowest reading in the past 62 years, exceeded only by the week after 9/11 and the week ended 02/06/04. If we had seen some terrible market declines over the past couple of months, the recent readings may be understandable. But the way it is, they just do not make sense. The probable culprit? Program trading.
Here is how the NYSE defines program trading:
“Program trading encompasses a wide range of portfolio-trading strategies involving the purchase or sale of a basket of at least 15 stocks with a total value of $1 million or more. Program trading is calculated as the sum of the shares bought, sold and sold short in program trades. The total of these shares is divided by total reported volume.”
For the week ended May 14th, program trading accounted for 50.5% of all NYSE volume - for every 100 shares traded on the exchange, more than 50 of them were triggered by a computer program. Of that volume, 44% was attributed to principal transactions, and 54% to agent transactions. “Principal” transactions are those that NYSE members do for their own account, so it would be, say, Lehman Brothers wading into the exchange to buy or sell for their own trading capital. “Agent” transactions are those that broker/dealers do to accommodate their customers. If you send a buy order to Lehman, and they forward your order to the floor, then they are acting as your agent.
So nearly 51% of all NYSE volume was program trading, and 54% of program trading was done by non-members, so we can conclude that 28% of all volume was program trading from the public (according to the NYSE, “public” includes hedge funds). The largest single strategy accounting for program trading was index arbitrage, at 14% of all program volume. Index arbitrage is when funds try to profit by discrepancies between underlying stocks and derivative instruments like futures contracts. The other 86% of program volume was not broken out by the NYSE.
The correlation between program trading and the Specialist Short Ratio is -.43 from 1989 – 2004. Using weekly data since 1999, the correlation climbs to -.79, meaning that movement in one measure could theoretically explain about 62% of the movement in the other. This is a very strong correlation, and given the number of data points, the odds that it is due to chance alone are basically zero. The chart below shows the relationship between program trading and the Specialist Short Ratio.
In 1999, the Specialist Short Ratio chopped around its average range while program trading remained muted at around 20% of total volume. However, beginning in 2000, when programs began to account for a larger and larger share of volume, the short ratio began falling in kind. Over the past three years, as noted by the correlation above, the short ratio has fallen almost in lockstep as program trading has gotten to the point where it regularly accounts for over half of all NYSE volume.
Similar to the impact of the e-mini contract on the Commitments of Traders data, I think program trading has essentially destroyed the Specialist Short Ratio as a consistent sentiment indicator. While de-trending it helps, as is posted to the site, I believe we’ve seen a shift in market dynamics that has made it much more difficult to give much weight to these readings.
Rydexers Feeling Good, and That’s Bad
Bottom Line: These wrong-way traders have forsaken low-beta funds for riskier ones, to an extent that has normally preceded at least a short-term pullback.
While it is perfectly normal for breadth to become overbought very soon after the severe oversold conditions of most intermediate-term lows, when investor sentiment becomes excessively optimistic, we normally see at least a short-term correction. We’re seeing such a thing now with Rydex traders.
Our Rydex Beta Chase index is calculated by comparing asset flows into and out of low-beta funds (or “safe” funds that do not move as much as the market) to high-beta funds (or “risky” funds that often move at least 2 times as much as the market). For example, a fund with a beta of 1.0 would gain around 1% if the S&P gained 1%; a fund with a beta of 2.0 on the other hand, should gain around 2% (of course, that works on the downside, too). By watching the money moves among these various types of funds, we can get an idea of how risk-averse or risk-seeking these traders are. Since this works as a contrary indicator with this group of traders, when they are actively seeking out risk, the market usually pulls back, and when they are extremely risk-averse, the market typically rallies.
The chart below shows a snapshot of the Beta Chase index over the past year.
When the index hits 10, it essentially means that these traders are 10 times more willing to invest in risky funds than safe ones. Even during the strong uptrend from March of last year, such displays of overconfidence lead to short-term consolidations or declines with a high degree of consistency.
Another way to monitor this type of activity is via the Rydex Enthusiasm indicator. This one looks only at the index funds which track the S&P 500 and Nasdaq 100, and compares asset flows to the performance of the underlying indexes. If the S&P rises by a couple of points one day, but an inordinate amount of money pours out of the bearish funds and into the bullish ones, then the index will be high. If the indicator is low, then it means that traders are either betting against a rallying market or they are not completely buying into it – a bullish sign.
The chart below shows a snapshot of the Enthusiasm index over the past year.
While the overheated instances match the ones from the Beta Chase index pretty closely, there are a couple of differences. It is disconcerting to me that the Enthusiasm index hit its highest point since April 2003 on Thursday, and backed off only slightly on Friday. Once again, we can see that other peaks in the indicator (highlighted in red) have coincided quite well with short-term consolidations or declines. Conversely, very low readings in the indicator, showing those times Rydex traders were extremely pessimistic about the market’s prospects given how it had performed, were invariably excellent buying opportunities.
Conclusion
Most of the breadth measures that became historically oversold around May 10th or so have reversed course and are now between somewhat and extremely overbought. This is entirely normal if we are coming off an intermediate-term low, but it seems as though sentiment is turning more bullish very quickly. At times like this, I like to go back to my daily notebook to see what kinds of environment we were seeing at various times. The following are entries directly from my notes around the time the market was making its lows a couple of weeks ago:
5/10…Over wknd, Chechyn president killed in bomb explosion, Nikkei -5% overnight, rumors of forced selling by Andor Capital, nobody recommend buys here that I read – NOBODY – all say technical damage too great, risk too high, stay in cash.
5/13…Schaeffer’s, Lowry’s out w/intermediate-term sell signals, getting lots of press. So many wondering if yesterday reversal was short-term low, nobody I read calling it major low despite incredible breadth readings. Keep hearing about seasonally weak time of year, horrible Berg beheading, Iraqi prisoner abuse, oil >$40, gas keeps pushing new highs, mkt not responding to good EPS or econ #, bonds keep falling.
5/17…Head of Iraqi governing council assassinated, India -17% trading halted, Japan -3%, Taiwan -5%. Numerous calls for “Black Monday” on message boards, a few on respectable news services. TheStreet.com continually pushing its “Short Advisor” service on site & email to “profit from such volatile times”.
Looking over the environment now, many of those dire signs in the posts above have disappeared. While I still see a significant number of opinions such as “this is just a bounce from oversold conditions”, it appears as though the action last week has put an end to many of the concerns from earlier in May (it’s amazing how a market rally tends to do that).
I mentioned last time that last Memorial Day, when terrorism fears where as strong as they are now, the market was mixed heading into the holiday, but when the weekend passed without major incident, the markets rocketed higher the next day. That may be a fact of life now, and concerns over terrorism may change previously reliable seasonal biases around holidays – particularly major ones with national or religious overtones. While we’re still quite short-term overbought, if the holiday passes without incident and we begin to see some relief buying pressure, I don’t really think overbought readings will matter – the broader market could easily rally right through them.
I continue to find little reason to believe the action in May is anything other than an intermediate-term low. However, in the short-term the signs are not especially encouraging and it appears as though we are likely about to test buyers’ resolve with a pullback. The largest question mark is the amount of relief buying that may come in if we see a safe Memorial Day, but after that it may become increasingly difficult to maintain higher prices.
Jason Goepfert
President and CEO
Sundial Capital Research, Inc.
Disclosure: no positions
This disclosure is not intended as trading advice in any form. It is meant as a note to subscribers that the author may have a position directly affected by the market outlook reflected in the commentary. Although the author takes great pains to remain objective in any commentaries, it is only fair that readers should know that the author may have taken positions in accordance with his market outlook. Positions can and do change at any time, without notice to the reader.
© 2004 Sundial Capital Research, Inc. All Rights Reserved.