Some Mildly Positive Signs, and One Bad One
Monday, May 24th, 2004 7:50pm EST
A couple of administrative notes…due to a large number of requests, we have added total assets in the Rydex bull funds and total assets in the Rydex bear funds, as well as total assets in the leveraged bull and bear funds to the Rydex Sector section of the site. Also, there was great interest in the SPY and QQQ volume deviations from their underlying stocks, on which I commented last week. I have added those two charts as well, called the SPY and QQQ Liquidity Premiums, under the Breadth section of the site. You can click on the headings to get a more detailed discussion of what they represent.
Small Traders Still Not Scrambling
Bottom Line: The ROBO put/call ratio continues to show a lack of the type of scramble for put protection that we saw at other major lows. While it is high compared to other recent readings, it is not high enough to declare that we are seeing excessive pessimism.
With all of the recent excitement over the historic put/call ratios that we’ve been seeing, I have been asked to comment on the ROBO put/call ratioTM that is posted to the site. Recall that the ROBO ratio takes care of the theoretical problems associated with the normal p/c ratios that come from the Chicago Board Options Exchange (CBOE). The data from the CBOE does not tells us who is trading these options (many small traders, or a few large ones?) or what kinds of strategies are making up the volume (are they buying these options to open, or selling them?). These deficiencies could make a very large difference in how the p/c ratios should be interpreted.
Fortunately there is a way to correct for those deficiencies, but unfortunately the data is released only weekly, and includes all option activity, not just that from the CBOE. The ROBO p/c ratio considers only those trades for 10 contracts or less, so that answers the first question about who is making up the volume. The average transaction size typically ranges from $200 to $2000, so we know that we’re talking about the smallest of options traders. We also only consider those transactions that are bought to open, so that answers the second question about the strategies being employed. Since these are such small traders, it is highly unlikely that they would be involved in complex options strategies. Most likely, if they are buying a put as an opening transaction, then they are betting on falling prices, and if they are buying a call, then they are betting on rising prices.
So the ROBO p/c ratio is a much more “pure” read on what should be contrary sentiment than the traditional put/call ratios. Since options traders tend to be more speculative than others, and since we are isolating the smallest of traders, this should be a very good contrary guide. We saw the ROBO ratio spike to 1.0 or more, meaning these small traders were buying more puts than calls, at each of the major lows since 2000, while they were aggressively buying nearly two calls for every put near most of the peaks. The ratio was no good at all last fall, as these traders were betting on rising prices and that’s just what they got (what’s that saying about a broken clock?). In January, the ratio reached a low not seen since 2000, and that time it was a good heads-up that speculation was overdone.
Currently, the ratio is up to 0.60, its highest level since the week ended August 15th, 2003. While it is encouraging to see these traders betting on the downside more than they have been in seven months, we are not seeing anywhere near the panic we saw at the other major lows over the past couple of years. Of course, at those points we were in the midst of a bear market and cascading prices.
Put/call ratios can become extreme for three reasons – either call volume is low, put volume is high or some combination of both. Over the past week, it was the latter, as call volume dropped off from what we had been seeing while put volume rose significantly.
The chart below highlights this phenomenon, as it shows the difference between the number of put contracts large traders were closing that they had previously bought and the number of call contracts that small traders were buying to open. If large traders are closing put contracts they had previously bought, then that is a sign they may be more optimistic about rising prices – after all, why would they cover contracts that would profit from a falling market? On the other hand, when small traders buy calls to open, that is a pretty clear bet on a rising market. So when the indicator below (the red line) is high, it shows that large traders are closing their puts to a large degree, while small traders are showing some hesitancy by backing off new call purchases. Conversely, a low reading from the indicator would mean that large traders are not covering their puts (since they think the market will fall) while small traders are aggressively buying calls.
1 In March 2000, small traders were buying calls to open with abandon. At the same time, large traders were not closing out their puts – hoping to hold on for the downtrend they thought was imminent. Turns out they were right and small traders were very wrong.
2 In the week after 9/11, small traders held off on buying calls to open, waiting to see what would happen next, while large traders were aggressively closing puts that they had bought earlier. A vicious rally ensued.
3 Once again, small traders were buying calls aggressively. For the first time since the year 2000, they were buying to open more than 1,000,000 calls every week. Meanwhile, large traders were not doing much with the puts they had bought earlier, preferring to hold on for a possible downtrend. They got it.
4 Last week, we saw a significant drop-off in small traders buying calls to open, while large traders were being quite aggressive in closing their put contracts. In fact, last week saw them closing their previous put purchases at a rate that ranks 4th in the past four years. This has pushed the indicator to a level unseen except for September 2001.
The fact that large traders were covering puts last week while small traders backed off buying too many calls is a good sign – it is the type of divergence between large-money and small-money traders that often precedes a low. While this data supports a rally from these levels, it is not a reason in and of itself to expect one. If we were able to see small traders truly scrambling for puts while shunning calls, then that would be a different story. So far, they are not seeing the need to be aggressive in protecting themselves, at least not to the degree seen previously.
The Cash is There – How Will They Spend It?
Bottom Line: Cash balances at NYSE clearing firms are still quite high. While this can be a positive, we need to see a willingness of these investors to put the money to work in the stock market.
I’ve gone over the topic of margin before, but as a quick refresher it is a loan that one takes out from their brokerage firm. Similar to any other loan, you pay interest on the balance you borrow and can use the proceeds for whatever purpose you wish – a new car, a nice vacation, or (most likely) to purchase additional stock in your brokerage account. As long as the equity in your account doesn’t drop too low, your account is in good standing and you simply keep paying interest on your debt.
When you open a margin account, brokerage firms segregate your funds behind the scenes. They create different “types” of accounts – there is a type 1 account, which is simply a cash account where everything needs to be paid for in full; there is a margin account, where you may borrow against the securities currently held in your account; and there is a type 3 account, where short sales are kept. Not every firm follows these procedures, but it is the typical setup.
Each month, the NYSE and NASD report the amount of total borrowings brokerage firms have lent to their customers (total margin debt). They also report the amount of cash held in both cash accounts and margin accounts. For the month of April, customers with accounts at NYSE-designated clearing firms were holding a total of just under $85 Billion cash in their cash accounts. This is significant because it is the 2nd-highest total in history (who says there’s no cash lying around?). The highest total was actually in March 2000, as these customers were holding a little more cash than they are now. You may logically point out that March 2000 was near the top of the bubble, but there are two distinct differences this time around. In March 2000, total margin debt was a whopping $278 Billion – it was “only” $181 Billion last month, a difference of nearly $100 Billion in debt. Also, total cash sitting in margin (type 2) accounts was only $65 Billion in March 2000, but it was nearly $104 Billion last month. So in March 2000 these customers had $278 Billion in debt compared to $150 Billion in cash. Last month, they had $181 Billion in debt compared to $188 Billion in cash. Quite a difference.
These figures make up the “Available Cash” indicators that are posted to the site. The indicator is calculated simply by subtracting the liabilities (margin debt) from the assets (cash balances) to come up with the amount of cash these customers could theoretically spend. Historically this is a negative figure, because many brokerage firms do not pay you interest on your cash balances, unless you are a very good customer. So most often we see high margin balances, but small cash balances. In fact, it is so rare to actually see a positive “available cash” figure that other than the recent time period, we would have to go back to 1950 to see it.
The chart below shows this available cash figure for NYSE accounts over the past six years:
We can see that in 2000, customers were borrowing themselves silly, pushing the net worth of their brokerage accounts further and further into a record hole. As the bear market took hold, things began to reverse, and the accounts hit an opposite modern-day record in the fall of 2002, just as the market was searching for a bottom. As the bull market of 2003 progressed, customers began to borrow more and more on margin, further fueling the rise.
As of April, this figure is still positive, and is the largest we’ve seen since September of 2003. It is obvious from the chart that there is plenty of firepower left for customers to spend on more stock purchases if they so desire. The thing that should be watched is not only how much cash is left on hand, but also the trend of margin borrowing. Bulls should want to see customers borrowing on margin, as it shows a willingness to accept risk. That is not a bad thing, so long as it doesn’t get out of hand.
The Rally That Wasn’t
Bottom Line: Some breadth measures have cycled back to overbought territory despite a flaccid market. Historically, that has lead to additional short-term declines.
In the last comment, I noted that the longer we hung around the lows without getting some upside going, the more wary I would become. Unfortunately, that is exactly what has happened, and I want to put some numbers to my feelings of uneasiness.
Two weeks ago, a 10-day moving average of the up issues ratio hit an extremely oversold level of 35%. That means that over the prior two weeks, issues that closed positively on the day accounted for a mere 35% of total changed issues on average. That was a level of oversold normally seen only at intermediate-term market lows. However, since that time the market has basically done nothing but go sideways, and tonight that same up issues ratio is at 59% - an overbought reading. I went over the past 40 years of breadth data to see if we have ever seen such a flat market following such oversold readings. What I checked was every instance of the up issues ratio dropping to 35% or below, then looked at where the market was by the time it rebounded back up to at least 55%. In other words, I wanted to check how much previous markets had rallied from the time they cycled from deeply oversold to overbought.
Not surprisingly, our current situation is one of the flattest on record. On average, the S&P 500 was 4.5% higher once it reached overbought after becoming extremely oversold. Only twice before, out of 33 total instances, has the S&P changed less than it has this time. Both preceded additional weakness – 20 days later, the S&P was lower both times, for an average loss of 1.8%. However, after 90 days both were nicely positive, with an average gain of 9.1%.
If we look at the 10 occurrences with the smallest changes after this oversold/overbought cycle ran its course, a clear pattern of future weakness emerges. 20 days into the future, the S&P was higher only 3 out of the 10 times, and its average return was -1.7%. Six months later, 6 of the 10 had returned to positive territory, but the average return was an anemic 0.5%. Even up to one year later, the average return was essentially unchanged, at 0.7%, while 6 of the 10 became positive.
On the other hand, the 10 occurrences with the largest changes after the oversold/overbought cycle for the most part showed additional strength in the future. After 20 days, the S&P was higher 6 times, with an average gain of 0.5%. But six months later, 7 out of the 10 were positive, and the average return was a respectable 6.5%. One year later, 8 of them were positive, and the average return had climbed to 10.7%.
These findings help confirm my fears from the last comment – if the market was not able to muster some type of rally from our deeply oversold conditions, then it increased substantially the risk that we would see more weakness ahead. From the study above, it is clear that those instances with the smallest changes after the market went from oversold to overbought underperformed up to 20 days later. Even longer-term, the S&P in general was not able to generate much in the way of returns. This is disappointing, as it suggests that we have further work to do before a sizable rally is likely, unless we can get going NOW.
Conclusion
All of the facts that I have pointed out over the past week and a half remain just that. Extended runs of positive markets simply do not roll over into new bear markets, or at least they haven’t in the past 100+ years. We haven’t yet suffered the average correction after such runs before new upswings kick in, but we’re close. And the true historic extremes we saw over the past week or so lent a lot of credence to the likelihood of higher prices over the next 60-90 days and beyond. But I am very troubled by the inability of the broader market to rally in the face of these readings. As we saw above, in the past when the market cycled from deeply oversold to overbought, on average it was rewarded with a 4.5% gain for the effort. This time around we got squat, and historically that type of limp performance has lead to further weakness over the next few weeks. I suspect a break of last week’s lows would squeeze many of the buyers last week out of their positions, and I imagine we would see a rather quick decline. I continue to believe that such an instance should be used as an opportunity to add more long exposure, not as an excuse to sell, as the chances of higher prices over the intermediate-term are high.
In the short-term, more of our shorter-term indicators are negative than positive for the first time since May 5th. The STEM.MR model continues to hover around a low level, and whatever oversold readings we have been getting result in nothing but one-day wonders. None of these are positive signs, but I don’t believe they are enough to warrant establishing short positions as long as we’re mired in this range. Probably the only edge that will appear is when everyone else recognizes the same thing – a break above 1108 or below 1077 on the S&P.
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.