Putting the pieces together...

Minor Update

The stock markets are barely off last week's highs, but sentiment has gone quite bearish again. The 2cs of bearish sentiment has risen from  88 a week ago to 111 tonight!

We seem to get occasional spikes up of 25-50 SPX handles and then "fake a top" and move down a bit for a few days. What a great strategy!

The political-economic news is dreadful day after day, and it is driving public opinion or vice-versa. I read somewhere today that pro/con presidential sentiment is the major driver of the four year cycle (not a new idea) and therefore the markets. More on that another time, but it's imperative to try to put the news aside and look at market sentiment and other breadth data  without emotion. Mark Hulbert looks at subscription newsletters for clues, instead of market data as in the 2cs, and this is his conclusion:

http://www.marketwatch.com/story/contrarian-analysis-of-stock-market-2009-09-29

2CS Update

As of tonight the 2cs has dropped to 88, rounded. It hasn't been that low since two days after the high (close) of SPX in May 2008. You may recall that SPX hit the 400 dma on the very day of the high on May 19 2008 at 1425.35 (close). Today the 400 dma is @ 1059.90. I sure would like to see a couple of days up to put the 2cs under 80..it was 72.31 on May 19 2008. Four days before that it had been 89.14.

I'd like to see a Dow clx indicator confirmation, but it's possible we had it last week.  It was a week early in May 2008.

 

I did not get bullish immediately in early March 2009. I'm investing in retirement, so I'm not a rabid high risk trader as I once was. But I started getting long in equity substitutes like corporate bonds and beaten-down closed-end funds in April and May, and I'm up 12.6% on the year to date which is nearly double my goal.

 

Despite all that, the 2CS has kept me long through all the uncertainty since March/April. It's possible we will slow down here again and extend the timing out further. I am certainly not hoping for negatives. But we are set up for a possible sell if the market gains more this week. On any real strength this week I am going to be selling the parts of accounts that are most likely to be influenced by an equity decline.

 

http://screencast.com/t/AbQ8ZHnNwP2N

2CS Makes All-time Top 5

The simple 2CS sentimeter takes the daily CBOE VXO and daily CBOE combined Put/Call ratio, multiplies them, and sums the most recent five days' results.  At the post 9/11 stock low 2CS got to 262 and at the 2002-2003 lows 228, 246, and 165.  1998 was 237.  Today 218.  Ms. Market is in good company for a price low at this time.

 

Also do look at Helge Sundar Loekke's website which you can click to under "Blogs I Like" on the lefthand column. I have followed his work and opinions for 6-7 years and value his work highly. He has a longer term stock market cycle low due in this current two week period.

Silver

I'd been lightening up considerably on the commodity side this year, particularly so in mid July. Although I'm a long term bull on commodities, I'm still a short to intermediate term bear. I have expected a plateau type correction with a lot of volatility for at least a year from March 2008. But we could be getting to the floor or base of that consolidation range rather soon in some commodities.

 

The silver chart shows the great bull market range from $1.29 per ounce in 1967 to the $40.5o per ounce high (nearby futures) in 1980, and it shows the divisions of that range in classic Gann or Fibonacci measurement. Just as crude oil's 2008 high was almost exactly four times the 1991 high, the 2008 high for silver was almost exactly at one-half the 1980 high. As W. D. Gann wrote and spoke extensivley, nowhere better than in his 1942 book, "How to Make Profits in Commodities", these mathematical occurences are not mere coincidences and in fact are quite common. (The 1951 edition of Gann's book is still in print.)

 

Note that as of today (the chart is through yesterday) silver is less than a dollar from its 1987 high which is also the Fibonacci  0.786 level of the entire 1967 to 1980 level. This could be a good place to cover shorts and/or take first long positions for long term investors or short term traders. This is not a recommendation, which I'm not qualified to make, just my thinking and planning for myself. I don't expect new silver highs this year.

 

Although gold broke its 80 week moving average this week, a bearish sign, silver could still make an earlier bottom as it did in the 1990's.

 

Gold On Long Term Support

I am writing this on Friday afternoon just before the regular COMEX floor close for gold. Since the high in middle March, gold has given back over 25% of its bull market since 1999. At this time constant 2 month forward New York gold futures are resting on the 80 week moving average which has as excellent a record for gold as it does for the S&P500 index. Gold has not closed a week under the 80 week moving average since late 2001, and "seems not to want to" now.  A break here or next week would convince me that this is a bear market and not simply a "healthy" correction.

 

2Cs Sentimeter Update

The red numbers are the extreme daily readings of 2CS for short to intermediate term SPX moves.

During 2007  there was marked bullish sentiment divergence from January to October. A bearish outcome was confirmed by the extreme bearish sentiment at the summer low (198) which was the most bearish market opinion since 2002. After the rally to October still showed declining bullish opinion compared to January 2007, the bear decline continued.

 

Rally tops after the October high could only generate bearish readings of 112, 124 and 125 followed by increasingly bearish readings on successive new lows in SPX. Technically the May 2008 rally top was a good place to short on moving averages and retracement percentages. But the 2CS reading of 72 at the May SPX top, far more bullish than any prior rally top  after October, suggested that the next new low might be the last.

 

True to that promise, the SPX low in mid August 2008 sported a 2CS of 161, showing bullish sentiment divergence. So far neither the action in SPX nor in 2CS has been very bullish, and that in turn suggests another revisit toward the lows for a re-test of bearish strength.

 

The 2CS Revisited

The 2CS Revisited

The 2CS Sentimeter is my favorite gauge of the extremes of investor complacency and panic.  The 2CS is quite low at stock market price highs and quite high at important market lows. It is very simply calculated daily by multiplying the daily CBOE combined put/call ratio by the daily CBOE VXO. (You could use the CBOE VIX instead, but I prefer VXO as it allows me to compare with charts and data I have back to 1996.) Add up the daily P/C*VXO of the past five days and you have the 2CS. (Click on chart for larger version.)

The first monthly chart of the SPX shows the extreme 2CS value in red numerals at important highs and lows from 1996 to 2007. 2CS has been at or above 200 at most of the major crash lows since 1996, with the greatest bearish extreme of sentiment coming right after 9/11/01. The market low after the Russian default and LTCM crash of 1998 also had an extreme bearish reading as did the double lows of July and October 2002. At the re-test of the market low in March 2003, 2CS was 165.

The second and weekly chart of SPX since June 2007 shows up 2CS as an accelerated version of the 1998 to 2002 market. After August 2007 the SPX made new all time highs, but 2CS did not go as low as it had in February and March 2007. I call this "sentiment divergence" which was also seen in 2000. WE also sometimes see sentiment divergence on daily charts during smaller period turns. And, of course, the same can happen at lows.

As the market fell from the early October 2007 SPX high, two 2CS readings near 200 were seen at the January and March 2008 SPX lows.

Another 2CS historical finding, not all of which is shown on these charts, is what happens on the rallies between two successive >200 2CS market lows, such as 1998, 2002-2003, and 2008. In all cases the 2CS reached levels between 120-130 on the rallies between the lows, most recently 123 on April 7, 2008.

My judgment is that SPX may have seen "the" low but that another retest may have started this past week which will probably make a higher market low and smaller 2CS reading than at the previous lows of January and March 2008.

2003-2008: One Opinion

2003-2008: One Opinion

Excepting July 2006 to February 2007, the SPX and other broadly-based US stock indexes had a "heavy" or corrective feel and profile since March 2004. Chart analysis and sentiment analysis confirm that. The movement up from March 2003 to March 2004 was definitely A bullish impulse wave which was not continued thereafter.

I have borrowed my Elliott Wave approach from Elliott, Frost and Prechter, and from Neely. Each of the successors to Elliott has stressed a different aspect of Elliott's work: Frost and Prechter favor the simple 12345-abc with basic waves, and Neely favors complex waves mostly consisting of triangles. I use some of each, all of which is seen in Elliott's own works.

From March 2004 to July 2007 the modest rise consisted of back and forth movements suggesting an ascending triangle. For a long time I thought that the triangle started from the red 1 in March 2004 and ended in July 2006. That is still a viable pattern with the red 1 changed to red A and the black d to red B. That would make A to B into a running B wave triangle. However, a running B wave implies a lot more followup power than we saw thereafter. SPX went up ~374 points from the March 2003 low to March 2004, but from July 2006 low to July 2007 high SPX only gained ~357 points. So the power projection failed to validate that move as a second impulse wave,
or red wave C.

Given my assumptions and method, I think it makes more sense to consider that last move from black d to black e as a diagonal or terminal triangle which ended at the November 2007 highs. This is not a Glenn Neely count, but I see the whole move from August 2004 to November 2007 as a "bow tie-shaped" triangle irregular B wave whose end was above the end of red wave 1.

From the November high to the March 2008 low was clearly an impulse wave down with the five waves being easily discernible even on a weekly chart. Thus a-b-c equals the wave 2 correction of wave 1 of 2003-2004. In nominal terms wave 2 ended ~20% higher than wave 1.

But if you multiply the 2003 low and the 2007 low by the US Dollar Index prices of those two low price dates, wave 2 last week was only 10% above the 2003 low. At the 2007 high, the same "real" increase was 60% from March 2003. The annualized "real gain" to March 2008 was only 2.16% as compared to >16% at the 2007 high! October 2007 to March 2008: that's a bear market indeed.

The black numbers in parentheses are the 2CS bearish sentimeter readings at the chart label points. 2CS is the five day running total of the daily CBOE VXO times the daily CBOE P/C ratio. Bearishness was very low at the red wave 1 high, lower than at any time during the bear market from 2000 to 2003. In fact bearishness was lower than at any point in the bull market after 1996. For me this confirmed that it had been a bullish impulse move in 2003-2004.

The 2CS bearish sentiment at the 2007 high points was greater than in 2004 and similar to high index price points in the late 1990's. At the November 2007 high, bearishness was already over 100 and went to 201 last week: more bearishness than at the 2003 low. For me Elliott Wave labels, the annualized real return of just 2.36% since the 2003 low, and the 2CS of bearishness all support my impression that wave 2 may be over. Nevertheless, in the pursuit of a conservative approach to retirement investing, which I live on, I will make no significant changes until I see if my "guess" is correct.

I'll be away for about ten days, so I'll say a few words on gold and commodities. Just as in 1987, gold stocks started getting sold when "generic" stocks started getting sold in October 2007. If you look at a chart of the XAU gold stocks index divided by gold bullion, you'll see that it peaked in October while bullion itself didn't peak until last week. This is due to margin calls in stocks which forces many people to sell everything, the good and the bad stocks. In 1987 gold held up for six weeks after the stocks market and gold stocks crashed.

Gold and commodities normally consolidate for quite a while before making their next move up. I believe that we are in a long term inflationary wave with ten to fifteen more years to run, as opposed to 1987 when we were not even half way through a long bear market in commodities. If I am correct, the stock market will start up again now. Commodities and gold will lag until the economy shows signs of re-heating, as it will do.

As I've mentioned, I had lightened up on metals stocks last year but I did NOT sell any gold bullion then or now. My commodity funds are 100% backed by US Treasury TIP's or Tbills to generate income, and are unleveraged. The natural gas stocks are owned exclusively for their high dividend yields. Natural gas futures were actually up on Thursday, but margin call selling knocked the trusts down. I added a bit and hope to add more later. Natural gas was a laggard so far in the commodity bull market, but it should outperform as new gas pipelines and imported LNG push US gas prices up to world prices over the next 1-5 years.

Even though the "real" US dollar-adjusted SPX lost 31% of its value from October to March, a diverse portfolio of cash, bonds, stocks, gold and commodities is ahead 2.2% in the same period, and it is still positive year to date. Not a brilliant return, but for a retirement program during great market turbulence, I'll take it!

Chart Doodles & Sentiment

A most notable quote of Ed Seykota, pioneer trend follower and specialist in trade control, comes from Jack Schwager's "Market Wizards". Schwager asked Seykota:

"What are your thoughts about using fundamental analysis as an input in trading?"

Seykota made this now famous reply:

"Fundamentals that you read about are typically useless as the market has already discounted the price, and I call them "funnymentals." However, if you catch on early, before others believe, then you might have valuable "surprise-a-mentals."

My understanding of this was then and is now that the daily news or even weekend reviews  that we are inundated with  is totally worthless. If reporters know and are telling you something, it's been known for a while and already acted upon. I rarely watch TV except for sports and weather which can be "real " news. And I never watch TV market programs. Kill the TV used to be the motto of thinking people, and it's a good one. Keep your mind clear and not distracted or disturbed by bad news which is 95% of what's reported.

Seykota's second sentence gets less attention and respect, but is of equal importance: gaining some insight into future moves ahead of the crowd. If we have some general idea about where markets are going over the next few years, and WHY, we might have a valuable "surprise-a-mental".

In the late 1990's when I was buying gold, chat site cyber-friends (and enemies) used to tease or scold me for being so stupid.  The "surprise-a-mental" I had was knowledge of the Economic Long Wave Cycle of approximately 50 years which was close in time to making a low. Obviously I couldn't tell anyone or myself  the day, month or even precisely the year, but I was pretty sure it would be  between 1998 and 2003 on the schedule it had been on since the late 18th century, and likely long before. So buying gold gradually on weakness and ahead of the crowd was a good idea.

Also buying things exposed to China was a good idea since they had a lot of people to "demand supply". I bought my first China fund in 1996 and a few nndividual Hong Kong stocks. That turned out to be way too early, and 1997 was not kind to them. But I wasn't overloaded or leveraged in either category, so I could hold and wait.

If one uses technical analysis, including chart analysis and rocket science indicators, and sentiment, with the Long Wave as a "surprise-a-mental" backdrop, and then ignore the noise of news and exercise patience, one can make a lot of money like Ed Seykota who doesn't even, or didn't then at least, have a qu0te screen on his desk.

So I decided last night to try to forget the news, which is uniformly horrid, and just look at a few very simple technical lines and numbers on charts of the SP500. I looked to see if there was any connection between the big  SP500 crashes of the past 2o years: 1987, 200-2003, and 2007-2008.

I can imagine some readers are thinking, "Tom, I can see 1987 and 2000-2003 being on the crash chart, but 2007-2008 doesn't make the cut even though it took 20% off many stock indexes". Good point, but let me show a sentiment chart that makes a good case for 2007-2008 as "crash-eligible". this comes from cyber-friend, Kirk Lindstrom (click on chart for larger image):

The ten week AAII bulls minus bears indicator is lower than it has been at any time since 1991 during the Savings & Loan crisis, which this one resembles in so many ways. Investors are more bearish than they were at the 2002-2003 lows! So I do think the drop into January, and even since, qualifies as a crash, at least psychologically.

Here is a log chart of SP500 futures  which draws a line from the 1987 low through the 2002-2003 lows, and what do you know? SP500 touched it on January 22, 2008. Granted that this might not convince a judge in a law court, but it is is quite suggestive when both  a sentiment chart shows massive bearishness at a low and a percent line on a log chart of stocks connects the three biggest sentiment crashes of the past 20 years.

Last I'll show something more tenuous but which could indicate what has been going on since the market made a low on January 22. This is daily chart of SP500 futures again with the numbers and alphabet letters us in Elliott Wave counting. It fits (but isn't proved) with the moves from January 22 being a bullish first wave up and a corrective wave down. No proof. Just suggestive. But if you forget the news, it might just be that the SP500 is going up again. Stay tuned to the chart, not the news.

 

Sentiment Failure in 2007

Earlier today I posted this below at Bill Luby's superb Vix and More blog. Excellent sentiment analysis is always rare, as one would  hope. It requires art and even perhaps the talents of a forensic psychiatrist. Bill's site is the best. I have seen nearly all sentiment sites available to private traders.

The thread at Bill's shop today concerned sentiment failure, and the learning experience it provides for skilled practitioners. I re-edited this post a bit to correct some minor errors:

I think that sentiment "failure" showed itself at the July top and then was confirmed at the first low. The 2CS, which I define as the five day running total of daily VXO and the CBOE combined P/C, had been giving multiple deliriously bullish readings (<50) in late 1996 and then at the February 2007 highs and once only in April as the indices resumed their climb out of the March 2007 lows. But as the market marched up to July, the 2CS began to rise. It was 71 at the July high. This was a major change of sentiment character. This was a reversion to 2CS range extremes seen at rally highs during the bear market to 2003.

I mentioned my monitor label, "sentiment divergence happens", but that is a short term happening. When divergences are as great as that from February to the July highs, there is definite range shifting going on, and that means that the markets are "in the process of" making a longer term top.

From the July highs to the end of July the 2CS was already nearly 150, not seen since the first half of 2003. This was the confirmation of probable longer term trend change. The delirious bullish high range had shifted but also the rather placid low range had now become somewhat delirious. Bearish delirium reached 198 on July 17.

By then it was evident that a broadening distribution top, called different things by different people--reverse point waves, expanding triangle, etc.--was also forming in many indices. Since distribution in a top takes a long time, it was reasonable that it should last longer that three months from May to August.

On the rally to October new highs 2CS only got as low as 63. No more delirious 50's or 40's as in 2006 and up to February 2007. 63 is almost exactly the number attained on the rally into May 2002 and on other bear market rally tops. The top was in.

Personally I did not run out that day and shout in the streets or liquidate everything and load up on puts. I had been trimming into strength all year.

Sentiment does diverge and it exhibits range-shifting longer term, as it also did in 2002/2003 at the lows. To grasp it requires a mental shift from the notion that the majority is always wrong at the wrong times, as you clearly see. Range shifting is the sentiment equivalent of Dow Theory price waves. I think it's even better.

Chart 1 shows monthly and longer term 2CS sentiment (in red and, notably, inverted to price) at intermediate term highs and lows from 1997-2007. After the 1998 lows  it was reasonable on many technical grounds to expect a large distribution top. Despite the internet and junk bubble extending gains, sentiment began to cool decidedly by March 2000 and the April low.

The same was true for the 2001-2003 lows.  In both cases it took about 18 months to play out from the first disjunction or divergence point and about six months from the divergence.  Although post is primarily about 2007, the early warning signs were already present in 2006 at the May high and June/July lows. Delirium reached it's peak in May 2006 but at the summer lows bearish sentiment greater than any since 2003 was seen. At the time this was seen, correctly, as being quite bullish, but it also was an early warning.

Chart 2 is a daily close-up of the 2007 action, already discussed, with a summary of my interpretation. (Click on chart images for a larger view.)

It's 1998 all over again...

This is just a quick note due to being busy, busy today. Nevertheless I wanted to share my analysis of what I think has just happened in the markets.

This is a repeat of 1998. There is no yet proven economic cause for a bear market at this point, except for the putative anticipation of a recession in the future.  A panic collapse of a greedy financial idea caused each market collapse: in 1998 it was hedge fund stupidity and the Russian default; this time it is ignorant and/or fraudulent mortgage generation and repackaging with poor credit oversight at every step.

Both market collapses took parts of four months, and both shaved over 20% off the market. I am basing that percentage shave in the current case on the overnight SPX on some data vendors, including mine, which hit a low of 1249.94 overnight, making the drop off the October top  >20%. Some indices were down >20% within NYSE market hours.

In both 1998 and now, sentiment and some breadth measures reached their lowest points in years, and the 400 day moving averages were cracked. Consumer sentiment  was not unduly affected either then or now.  Few consumers, on a percentage basis, own stocks, so the financial panic did and does not extend to them if it is over quickly. (Even from 2000-2002 it didn't discourage the consumer except for the mildest recession and quite possibly the shortest one in history.)

Two further reminders: Greenspan intervened strongly near the end of the collapse in 1998, and the market went to new highs within a few months. Obviously I can't guarantee that will happen, and we could at least retest the overnight lows in a few weeks. But it may not be as bad as it has looked this past week or so.

As for differences between 1998 and now: interest rates were higher in 1998, the dollar was stronger, and general price levels were falling rather than rising, and there was no presidential election looming. But those are what Ed Seykota called "funnymentals" not technical or historical market comparisons. Fundamentals are harder to use in prediction. The next recession will  not be certain until it's nearly over. Not every slowdown leads to recession.

Julian Robertson and the Long/Short Hedge Strategy

Julian Robertson lost a large bundle from 1998 to 2000, in the same way that Warren Buffet did in an unleveraged manner, and as John Hussman (See "True Confessions" below) has been doing on a much smaller and unleveraged scale for the last few years: Robertson was buying "quality" stocks and shorting "crap" stocks against them.

Robertson's problem was that everyone else was doing the exact opposite from 1998 to 2000: they were selling the good stuff and buying internet and tech crap. After massive losses in 1999, Robertson liquidated Tiger and his other hedge funds right into the 2000 top. To do that, Robertson had to buy back his shorted and now bloated crap stocks and sell his good stocks on their lows and exactly when everyone else was suddenly doing the opposite. O my!

Had Robertson not been leveraged and been able to hold on another six months in 2000, he would have triumphed and prospered as "value" trumped junk in the new century. Not being leveraged, at least in the usual sense, Buffet survived to win again. Hussman's timing was perfect for value versus junk from 2000 through 2003.

Here is a brief history: http://neumann.hec.ca/~p283/gpmba/Tiger.htm

Also look at a monthly chart of Warren Buffet's Berkshire Hathaway BRK.A from 1998 to March 2003. Many forget how much he lost from 1998-2003.

Sentiment and Personal Psychology

I was always interested in market sentiment, but I typically used anecdotal comments or polled sentiment surveys (II, AAII, etc.) until late in 1996 when the VIX and CBOE Put/Call ratios became available to retail traders on a timely daily basis. I looked at both data series, and it occurred to me that a combination of the two made sense. I imagined in my devious sentiment mind that "they" could fool us some days with one measure and some days with the other; so if I combined them I got a better "fingerprint" of current market sentiment as reflected only in what people actually were doing rather than in what they were saying to pollsters.

Other people must surely have thought of this same very simple indicator and kept their mouths shut. I started doing a daily multiple or product of the CBOE combined P/C Ratio times the VIX (now VXO). I called it the 2CS, and I have tabulated it daily since mid 1996. To smooth it a bit, but still keep it close to daily action, I decided early on to use a five day running total of each day's product of VXO times P/C. There is no hidden or proprietary stuff here. I've written a lot about it at this blog and elsewhere.

Sentiment dogma states that the great majority of all investors and speculators is always wrong. And that's a safe bet most of the time. The majority gets most bullish at tops and most bearish at bottoms. And that's true for short and intermediate term high and low swings within a longer term trend. But I learned later on that toward the ends of longer term trends, some part of the majority "gets it right" and sees ahead. So at extreme highs and lows we often get what I call "sentiment divergence" or "range shifting".

2CS range shifting is something I had begun to worry about since last summer (2007). We made new highs several times into October, but the 2CS (with VXO and P/C ratios inverted) didn't make new lows as compared to February 2007.

The 2/20/07 2CS was 44.46 (SPX close 1459.7). 7/9/07 2CS 63.06 (SPX close 1531.85). 10/09/07 2CS 63.11 (SPX close 1565.15). Price was making new highs, but bullishness was lagging and  making new lows at those price highs. When we start to see this counter-intuitive divergence or price and bullisness, I believe that we have to consider that the longer term trend is in the process of changing.

At the other end of the price and sentiment range we suddenly got a very great increase in bearishness at the market low in August 2007 when 2CS ran higher than at the 2003 market  low! I commented on it here and elsewhere, but I don't want to pretend that I totally understood the implications of it myself immediately or communicated them crisply ....at least consciously.

However, all year in 2007 I was feeling increasingly nervous that I was going to lose a lot of money if I didn't cut back on stocks. I was doing the 2CS and seeing those numbers, and my gut was understanding them. But i wasn't verbalizing or accepting it at a rational level. But I did cut back anyway, and drastically all year, although I felt like a "wimp" while doing it. I tend to be optimistic, and I like to be bullish. This may be a fault, but it's what I am.

Whether this makes any sense to other investors I can't say. I couldn't accept or communicate my market feelings, BUT I was  acting on them. And it has saved me a bundle so far this year. I told all this to my wife the other day, and she said there's hope for me after all. :) Women know all this stuff.

I'll post a chart of 2CS another time. See if you can follow the argument first.

"Survival of the Foulest"

The markets have obliged me by roughly following the schematic I outlined in July in "Heading for the Hills"and in "Second Half". My thinking was based upon the annual cycle or "seasonal" pattern of US stock market prices as well as the "decennial cycle" in which the year ending in "7" in each decade is typically more muted in price amplitude. The market did peak in mid July as anticipated or "hoped" and made significant corrections into lows in August and October.

Perhaps I'd be wiser to rest on my laurels and not push my luck. But November and December are the most reliable times of the year to be long. The Dow Jones seasonal runs from 1928 to present and the SPX seasonal from 1950 to present (see below).

I have to confess--since you've already read it--that part of my willingness to accept a falling market for this past summer, and early this  fall, was due to "concerns" about the economy in the US. However, recent data suggest that the economy is continuing to run fairly well. I am not an economist, but the few economists that I follow regularly seem to be agreed that things are not that bad. It's hard not to hear news wherever you go, and as usual it's all bad. But that's normal for, as Barabara Tuchman the British narrative historian said,

"A ...hazard, built into the very nature of recorded history, is overload of the negative: the disproportionate survival of the bad side--of evil, misery, contention, and harm. In history this is exactly the same as in the daily newspaper. The normal does not make news. History is made by the documents that survive, and these lean leavily on crisis, calamity, crime, and misbehavior, because such things are the subject matter of the documentary process--of lawsuits, treaties, moralists' denunciations, literary satire, papal Bulls......."

Normal or happy events are rarely heard from, but most of the unhappy events are heard and recorded. This is what Nassim Taleb, author of my current favorite, "The Black Swan", calls the "fallacy of silent evidence". Namely, we humans tend to make value judgments and form beliefs based on what is presented to us as definitive ancient and current outcomes to events. But in fact far more outcomes are lost to the news and to history than are saved to it, and so we often base many of our opinions upon what I would call the Darwinian "survival of the foulest" outcomes.

When we hear this concept we almost all will think, "of course", but we don't change our methods of establishing opinions. However, one surprisingly good outcome of this defect of human nature is that it gives rise to the profitable technique of sentiment analysis. In speculation and investing  we don't need to collect millions of personal anecdotes or news items to find out how gloomy market participants currently are. We can measure the degree of ambient gloom and doom by seeing what the two major classes of market participants--the "retail" investors and the knowledgeable professional insiders-- are actually doing with their money. And we can do this each day and to a limited extent within each day.

All that said, my lean or bias toward accepting the historical record of the annual cycle or seasonal pattern for the Dow and SPX for the second half of 2007 was at least partly formed on the basis of economic hearsay and my own non-professional  evaluation of the economy. I let the gloom and doom of late June's economic hearsay influence my investment approach, and it was wrong, as the economy did not weaken until now as hearsay predicted.

However, the seasonal pattern worked anyway!  So I have learned a lesson, which I will doubtless forget and then allow myself again to become prey to news and unbalanced current opinions. As the late Peter Cook's character,  Sir Arthur Streeb-Greebling, exclaimed in Good Evening: "Oh, yes, I've learned from my mistakes and I'm sure I can repeat them exactly." Read Nassim Taleb to learn more about how to improve upon one's human nature by means of sceptical empirical testing and prove Sir Arthur wrong, if you can.

The US markets, and many around the world, fell nearly back again to the  October 22 lows late this past week. The historic odds favor a rise to the end of the year, and sentiment of the statistical type, based on what market players are actually doing, supports that idea despite, or, perhaps "because of", current preoccupations with gloom and doom stories based on the credit issue.

   

Tuesday, September 18th, 2007

The first paragraph is by a poster at WW, the rest mine with some help from Richard Russell's great Dow Jones book:

Tuesday was reportedly only the 7th day since 1950 with more than 25 stocks up in volume versus every 1 down. The average market gain 3 months later was 9.4%. Each signal day was followed by quite a rally. In the case of 1982's signal, well it ran all the way up into year 2000. The dates are 12/27/50, 1/11/51, 10/23/57, 11/1/78, 8/17/82, 8/20/82, and 9/18/2007.

The 1950 example was the example of 25/1 up/down volume most like now. It came about 18 months after the great stock crash low and commodity low (Kondratieff Wave low) of June 1949, Dow 161. There had been a good rally for a year, but then the market sputtered with the terrible geo-political gloom and doom (USSR, China, Korea) and made a higher low at 228 right after Christmas.

The 1957 example was out of the October crash low of 425. Severe recession. Political gloom and doom. Communists pressing their luck in the US and throughout the world.

The 1978 example was out of October crash low of 790. President Carter abandons the Shah of Iran and gives away the Panama Canal; FED Chairman Miller newly installed (worst ever); dollar and bonds collapsing; gold breaks above 240 not long after.

1982's example was out of the bitter August crash lows of 772. Auto production was the lowest in 35 years, factory capacity utilization 69%, 10.1% unemployment.

In every case of a 25/1 up/down volume day the possibility of a big rally was totally dismissed since the economy, and the world could obviously only get much worse.

2CS Sentimeter 1997-2007

In 1996 when CBOE's VIX became more readily available to small private investors on the internet,  I got interested in sentiment measuring. I started calculating the daily product of the two indicators: the CBOE VIX and the CBOE combined Put/Call ratio. I had noticed that sometimes one or the other of the two was the leader or more indicative of current sentiment, and it seemed logical to combine them. I soon decided on a five day running total of the product of the two numbers as being sufficient short both to center near price highs and lows and to register sentiment extremes fairly well. I still do the 2CS by hand each day as I still draw some graphics  charts by hand,  being a survivor from before personal computing.

The monthly chart of the SP500 since 1997 shows only the extremes of the 2CS Sentimeter (red) near those price extremes. We have had two bull markets since 1996, one nasty bear market and several crashes in both bull and bear markets. So I have developed a feel for this indicator.

I'll let the chart speak largely for itself except to point out the obvious. There have only been three greater extremes of bearish sentiment than our current reading indicated by the 2CS since 1996: the two bear market panic lows of 2002, the post 9/11/2001 panic low and the 1998 Russian default and LTCM panic low. Therefore I think this current episode qualifies as a panic even if it's only a 10% decline in most US averages. If one looks at the pattern of changing levels of the 2CS at successive highs and lows, I think there are some hints perhaps for the next few years. But I also think we can be pretty certain of making a good low and sustaining a good or excellent rally in the markets rather soon. It's possible we've already put in a low, but it's too soon to say on the basis of this indicator alone.

Back in the Saddle

After my longest vacation in decades, I am somewhat at a loss for words. We saw many old friends, talked and listened for hours, experienced much good music and art, and ate too well and too much: in short a successful vacation! The earlier time this summer in Hawaii was more truly "R & R" time: rest and recuperation. This was fun for all the senses.

I've grown tired of lugging a laptop computer around and took only my "nano-computer", a T-Mobile Dash (HTC) which runs the Windows Mobile 6 OS. This very slim machine runs internet, messaging, and email off either TMobile's GPRS "EDGE" network or any nearby wi-fi output. T Mobile has greatly extended their capability since I bought my first Sidekick unit over three years ago, both geographically and in network speed. I tested the network speed several times through http://performance.toast.net/ and got a rough average of 160 Kbps or about double the effective speed of a 56 Kbps dial-up service. This is about like ISDN was years ago, and is quite acceptable.  It worked well enough to keep track of the basics once or twice a day. After all a vacation is for "taking leave of" and I didn't want to be an "inter-nut" and spoil the fun for myself and others.

Before I tell you how my portfolios did, let me tell you what they consisted of with all of them under my management consolidated. The "pie" does not include personal checking accounts, stored gold, collector items, personal real estate (home) and other family and personal possessions. It includes only investible and invested funds. 49% was/is in "cash" in the form of Vanguard's Prime money market fund @ 5.13%. The rest is approximately equally divided amongst nine funds: BEARX, DODBX, FAIRX, HSGFX, PRWCX, SGENX, VWINX, VWEHX, and "MIX". The last one is  a collection of much smaller pieces of funds and ETF's including energy, mining, real estate,  and "emerging markets" exposures. VWEHX is Vanguard's "high yield" municipal bond fund, but a very conservative one.

DODBX, FAIRX, PRWCX, SGENX, AND VWINX are all considered by me to be "balanced funds" with either a traditional stock/bond/convertible mix or who are primarily stock investors but who almost always keep >20% in cash for buying value after pullbacks. They (in total) are estimated by me to be holding about 70% in stocks and 30% in bonds or cash at this time.

BEARX is both a long/short fund emphasizing the short, but is also a gold fund. HSGFX is a mid cap long fund with 100% short hedging via Russell 2000 and SP500 futures or options. "MIX" is the inflation kicker or spice to improve inflation protection but not enough to torpedo the plan. That plan is volatility reduction which is important to me now partly since I am retired and live off our funds and because I felt increasingly that the markets were due for a larger drawdown.

Now for the comparisons: from the peak of the stock markets on about July 18 (and using daily closes only) the Dow 30 was -5.89% to its lowest close to date and the SP500 was -6.58% to its lowest close. The long US Treasury bond was up about 2.5% over the same period and the US Dollar Index and gold were virtually unchanged over that same period.

Drum roll please........ the low volatility fund described here was down 1.17% over the same period to its lowest daily closing. Thus it had less than 1/5 the drawdown percentage of the SP500. As of Friday's close, the SP500 fund at Vanguard is up 3.56% on the year, and my low volatility fund is up 3.16%.  Clearly there is an opportunity cost to a low volatility program, but the much lower "ulcer index" is worth it to me at my age and given the current technical situation of the market.

I haven't done all my work on the markets since I got  back a few days ago, so I don't want to say too much. But I think that based on historical considerations the market will probably remain under pressure this summer. How it responds to this currently grossly oversold condition in the next week or ten days will  tell the tale. Several very different but reputable intermediate term predictors are turning positive right now. They both turned negative just before I went away in mid July.

 

Heading for the Hills

Additional evidence is accumulating for a pullback starting this month. In "Second Half", posted here on July 1, 2007, I discussed some economic and historic data favoring a decline of more than the usual 2-3% we have grown accustomed to since March 2003. Shorter term data are now beginning to appear as well.

My own short and intermediate term sentiment measures are finally showing some frothiness of a degree nearly adequate for a decline. Also several excellent neural network folks I respect are reporting that their forward projections are weakening.

Within each month US stock markets tend to bottom in the middle of the week before options expiration and peak just after expiration. Don Wolanchuk has long and correctly trumpeted this very real phenomenon. http://wolanchuk.com/

And I want to return to the historic "7 year" of the decade and the annual tendency for a July top and an October low. (Remember or refer back to 1957 and 1987.) There is also cycle evidence of several types consistent with a possible decline this month.

Sentiment, or the "animal spirits" of Adam Smith, can always shove the statistical or historical evidence off the stage and into the wings without shame or blinking, so don't throw everything out at once. All investment or speculation is a judgement call. But I am lightening up again on the long side. I have done this four or five times since 2002, and several times it has cost me some gains opportunity for a while. But that is my preference at this stage of my life: not to maximize risk and profit but safety and income instead.

If I am guessing right, we will have a turn down of some meaningfulness next week into August and possibly October.

It's summer and very hot, so I am heading for the hills for a few weeks or more in cooler northern New Mexico USA.

Overbought?

30 years of the US 30 year Treasury bond