Off the Rails
July 10, 2017
The GPS Portfolio program tries to select ETFs at the start of an upward swing after a pullback, so with the market moving up one day and down the next it has found little to buy. SPY and XLY (Consumer Discretionary) are attractive on their merits, but the market is unattractive on general principles.
This is probably because
so many individual indicators remain negative even though the Dow Theory buy
signal has been reconfirmed by the Dow Transportation Average.
A list of current and closed trades appears with the table above at the regular link (Discretionary GPS Portfolio), which you can also find among the links at the beginning of every report.
and S&P 500 finished the week N++S+, while the Dow Transportation
and Industrial indexes finished T++I+. Both are in their most
frequently seen configurations, and both have records that are roughly equivalent
to randomly selected sessions.
the combination has a record for gains the next day of 54%, very little different
from the 53% profitability for days selected at random since 1971.
The most reliable combinations are those in bold type. The very best performers are colored green, while the very worst are in pink. The tables will be updated each week so we can follow the results going forward. Additional explanatory material was posted here, here and here.
General Market Comment
Our guess in the last report, that the results of trades opened at the close the Friday before the 4th of July and sold at the close Thursday would produce losses, was accurate.
Similar trades have been profitable in 11 of the previous 15 years (73%), with the second session after the holiday traditionally the most profitable. However, the nearly 1% loss last Thursday reversed earlier gains and resulted in an overall loss of 10 S&P 500 points.
After toying with us for a week, the Dow Transportation Average finally closed at a new all-time high on Monday. A review of the Dow Theory revealed this to be reconfirmation of the last buy signal, rather than a new one.
All stock market indicators are subject to changes in interpretation as new data is accumulated, so with the Dow Theory now well over 100 years old, it is not surprising to find many differing interpretations.
From the chart below one can support a claim that the last major buy signal occurred in 2016, in April, July or December (solid black arrows). The dashed arrow represents the last lows by both the Industrials and Transports, so a buy signal can be dated from any of the three pairs of new highs that followed.
Many would declare the start of the bull market at the first solid arrow in April, some would delay the declaration until at least one average made a new all-time high (July arrow) and others might wait for new all-time highs by both averages (December arrow).
However, since there has not been a new confirmed pair of lows since the buy signal occurred, all would have to agree that the latest pair of all-time highs is confirmation of the already established buy signal, rather than a new one. Nevertheless, the recent highs by both averages are positive for the market going forward.
More good news came Wednesday, in the form of a 20-day buy signal from the indicator based on the final hour of trading.
If the S&P 500 closes above 2432.54 on Wednesday, August 2, it will become the 23rd profitable buy signal of the 26 given in the last four years. This will make it our most accurate 20-day indicator, with a success ratio of 92%!
In contrast to the good news reported above, we must note that even after doubling Friday's percent gain by the NYSE, the NASDAQ weekly net advances-declines has not regained the lead it enjoyed until June. The difference between the cumulative net advances turned higher again, but it is still 5 advances short of a new high.
The loss of the lead by the NASDAQ is quite apparent from a comparison of its chart to that of the S&P 500. The S&P is holding above its breakout, and above its 50-day moving average, while the NASDAQ is trading below both. This is of concern because the market has traditionally been most profitable when the NASDAQ has been in the lead.
Another negative is the recent sharp downturn by Semiconductor stocks (SOX). This chart plots the ratio of SOX/S&P 500 against the S&P. The red vertical lines are drawn from highs in the ratio to show the S&P gains and losses after the semiconductors turned lower.
In the last seven years there were 15 semiconductor stock highs before the most recent one. The S&P turned lower in the days following 12 of them (80%).
The S&P 500 MACD sell signal continues to hold the market back.
The sell signal is likely to continue, based on the low number of stocks with positive MACDs. There was a spurt of strong stocks to 223 on Monday, but the number slumped back to 186 on Thursday and Friday.
Just like the technical indicators, economic indicators were also mixed. To skip to the technical, please click here.
The 3-month average margin debt made another new all-time high in May. However, the total margin debt for that month did not make a new high. Instead, it receded by $10 billion from the all-time high it reached the month before at $539 billion. Currently it stands just $3 billion above its March level.
The monthly rate of growth of margin debt fell in May for the first time this year, so a decrease in the 3-month average may have already occurred.
It is important to follow margin debt carefully. The last time there was a decline after an all-time high, back in June 2015, the S&P 500 fell 11% before we knew that margin debt had peaked.
Margin debt also peaked in March 2000 (five months before the S&P 500 made its high) and in July 2007 (three months ahead of the market high). This suggests a possible market high this fall.
The closely watched (by the Fed) Labor Market Conditions Index (LMCI) (click here for definition) was revised sharply higher in March, to a level not seen in two years. It was also revised higher for January and February.
April was just revised to a new 2-year high of 3.7, but it slumped in May (the most recent figures available) to 2.3.
The LMCI is a reliable indicator of the health of the labor market since it cannot be as easily manipulated as the unemployment rate, which ignores shrinkage in the labor pool.
It was not all bad news last week. Help wanted ads rebounded in May and June, and while they are still growing more slowly than a year ago, the situation has improved by 1/3 from its low.
Also good news, as we anticipated from the sizable increases in the Chicago Business Index and the Richmond Manufacturing Index reported last week, the ISM Manufacturing Index recovered sharply after two slow months. It was reported at 57.8% for June, a 2-1/2 year high.
The VIX Index had its ups and downs last week but ended where it started. Once again it is correctly priced to reflect market value, after jumping 25% during the turbulent trading on Thursday.
The VIX Volatility Oscillator also changed very little from Friday to Friday, but at 83.4 it is very close to a buy signal on the VIX Index, which will occur if it reaches 86.
The S&P 500 Short Trend Indicator (STI) fell from positive to neutral on Thursday. It has been positive since May 5, so it has not been possible for it to give a combined sell signal with the Swing Indicator (SWI). However, it no longer has that protection.
The SWI has fallen to -180, so it will be some time before it will give a combined signal in either direction, but at least it is now possible.
The Dow STI is positive and the NASDAQ STI is negative. This limits combined sell signals to the Dow and combined buys to the NASDAQ, but neither is likely in the next couple of weeks. However, the Dow has more of a chance to give a sell signal than the NASDAQ has to give a buy.
The Quick Summary Index sell signal persists, but it continues to be a rather weak and ineffective signal. The Summary Index remained steady at 9.20 last week, but it will fall to 8.30 this week if the indicators remain unchanged at 12 negative, 13 neutral and 4 positive.
The S&P 500 200-day moving average (green line in the chart above) gained 5.66 points to finish the week at 2299.63, while the 50-day moving average (blue line above) added 3.95 points to 2414.29. The S&P fell below its 50-day average on Thursday but recovered the next day.
The Trend Indicator (TI) (pink line below) gained 0.06 points to 13.78. It has been confined between 13.19 and 13.99 since March.
The S&P 500 14-day Relative Strength Indicator (RSI) remained neutral at 47.5%. It will drift lower to 44.2% this week if the S&P remains where it is.
The S&P 500 Price/Volume Chart continued to drift in a narrow counterclockwise spiral. However, it has an opportunity to bullishly exit the pattern with a rally on higher volume along the green arrow.
The SPDR Gold Trust (GLD) spent the last month between 118.18 and 119.82 before taking leave of its narrow range to the downside three days ago.
GLD now stands at a four-month low, its Short Trend Indicator (STI) is at zero and its Swing Indicator (SWI) is sufficiently negative to make a reversal impossible for at least two weeks. So much for a positive breakout from its declining monthly trendline by the end of July.
The Average Signature still has one positive chart, but it is unlikely to produce a buy signal in the next few weeks.
Fred Goodman, CFP, is a fee-only Certified Financial Planner based in Los Angeles. To send Fred your questions or comments, click here: Fred@MarketMonograph.com. E-mail sent to Fred may be edited for clarity and brevity and published on this web site, and may include your name unless you request anonymity or specify not for publication. The charts and commentary represent what Fred thinks about the market and what he is thinking of doing for his own account and for accounts he manages at the time of writing. Fred, his clients, or his family may have positions or may make trades in securities mentioned in these commentaries. There is no guarantee that you will profit from trading as discussed herein. You may lose money and Fred assumes no responsibility for what you do or do not do with this information. Copyright©2001-2017 Fred Goodman. All rights reserved. For information purposes only, offered as a periodical of general circulation; not to be deemed to be recommendations for buying or selling specific securities or to constitute personalized investment advice. Derived from sources believed to be reliable, but no warranty is made as to accuracy.