MMR-511 Jul 16

450 INSIDE 445 N STRONG BREADTH DATA P. 1 440 LA THE CLEL GOLD & YEN CORRELATION P. 4 435 c KET PRESIDENTIAL CYCLE P. 5 AR TIMING MODELS P. 6 430 T DJU VS. DJIA P. 7 425 OR REP SUNSPOTS AND PROTESTS P. 8

420

415

Prepared After the Market Close, July 26, 2016 Report #511, July 27, 2016

3500 DJIA - Log Scale 4.3 High Summation 4.25 3000 QE3 Begins 4.2 Index QE1 QE2 Ends 4.15 2500 Ends High on our list of true mar- 4.1 ket maxims is this: “Gobs of 4.05 2000 Final price high Final price high breadth is a good thing.” Strong made here, 4 Final price high made here, at a breadth can only happen when 1500 peak at +283 3.95 made here, peak below +500 liquidity is strong, and money is +500 3.9 peak at +500 failure plentiful enough to lift the major- 1000 3.85 ity of stocks. That is not a condi- 3.8 tion which typically goes away 500 3.75 3.7 rapidly. 0 The strong breadth shows up 3.65 3.6 in the form of a new all-time high -500 3.55 for the NYSE’s A-D Line. The -1000 3.5 major tops of the market’s history 3.45 Ratio-Adjusted Summation Index (RASI) typically announce themselves -1500 3.4 with a divergent condition, with 01/04/10 01/03/11 01/03/12 01/02/13 01/02/14 01/02/15 01/04/16 prices making higher highs and level. That has two simultaneous mes- updated in the lower chart. We wanted the A-D Line making lower highs. We sages. First, it says that the market has an answer to the question of how much do not have that condition now. gotten a bit stretched, and so a normal risk there is if we have a new high in The strong upward slope of the A-D corrective period is in order. Second, it the A-D Line. So the lower portion of Line helps to push the McClellan A-D says that this is not the end of the the chart shows the worst 3-month for- Summation Index up to really high lev- uptrend. Big price tops usually arrive ward drawdowns seen after a new els. The first chart shows the Ratio- with the RASI below +500. If the 3-year high for the A-D Line. Adjusted Summation Index (RASI), RASI had failed to make it up above Generally, risk is limited to about which factors out the changing number +500, that would be really worrisome, 10%, and this “rule” has persisted for of issues traded on the NYSE, and thus but that’s not the condition we face decades. But there are a few historical it is better for long term analyses. right now. exceptions to that rule. The ending of The RASI is now up at a really high Years ago we did a study, which is QE1 and QE2 brought larger than nor- BOTTOM LINE 20% 3.4 In an election year with 2 new chal- 3.3 SP500 - Log Scale 3.2

lengers running, the market tends to do 15% 3.1 worse than with an incumbent running. 3 That worse performance is usually 10% 2.9 2.8

focused into September and early Octo- 2.7

ber, but there is preliminary choppy 5% 3-Mo Forward Max SP500 Drawdown 2.6 work to do. Look for a sharp drop into After a 3-year high in the NYSE A-D Line 2.5 2.4 a bottom due Aug. 3-5, and another bot- 0% 2.3

tom due Aug. 22-23 before a late- 2.2 August pop. That pop should set up for -5% 2.1 the September decline. Gold prices and 2 1.9 -10% T-Bond prices should also fall, just as 1.8 they previously were all rising together End of QE1 1.7 with stocks. After the October low, we -15% 1.6 expect a robust rally for both stocks and 1.5 End of QE2 1.4

gold into early 2017. Bonds, we’re not -20% 1.3 so sure. 01/04/88 12/30/91 12/26/95 12/23/99 12/31/03 01/04/08 01/05/12 01/07/16

mal drawdowns. There was also a rare 19000 4/28/15. A-D Line 12 13% drawdown in late 1978 (not shown topped 4/24/15 in that chart). But most of the time, risk DJIA 1/18/16 over the 3 months after a new high in 11 the A-D Line is really minimal. 18000

That does not mean a normal 10 garden-variety correction cannot hap- 7/22/16 pen, and indeed we expect to see one 17000 9 between now and October. 4/17/16 The chart here on page 2 is an update of one shown in our last Report, 7/13 to 8 in a discussion of our still-experimental 7/21 BC Indicator. This is BC’s Big Brother. 16000 It’s big turns have a pretty amazing 5/13/16 7 track record at coinciding with major 3/30/16 BC's Big Brother turns for stock prices. Less impressive 9/4 15000 6 is its consistency at declaring whether 04/03/14 07/03/14 10/02/14 01/01/15 04/02/15 07/02/15 10/01/15 12/31/15 03/31/16 06/30/16 09/29/16 those turns should be tops or bottoms. In other words, it inverts sometimes. tion Index has managed to move spent some time above zero in the That does not seem to be the case slightly above its April 28 high of recent action following the deep nega- right now. Big Brother showed a top +4805.2 with a highest posting of tive posting associated with the Brexit due July 22, and the DJIA’s actual clos- +4865.2 on July 22. This area between low. But it has not shown the initiation ing high (so far) was on July 20. The +4700 and +5300 is prime reversal ter- height that it demonstrated subsequent SP500 made its high on July 22. So ritory for the Summation Index. It is to the September and February price we’ll call this one “pretty good”. Com- not surprising to see this indicator stall lows. It will need to bounce up higher ing up, Big Brother shows a drop that out where it is now. When the Summa- to maintain its postings above zero or resembles one from last summer. tion Index is at such a high level, there drop below zero to complete the current Bottom Line: Having really strong A-D is the need to relieve the overbought positive complex pattern. numbers tells us that liquidity is plenti- condition. But the price moves associ- Chart 7: The DJIA Price Oscillator was ful, and so the risk of a major decline is ated with a correction that begins at pushed below zero by the Brexit results. extremely small. But at the same time, such a high Summation level tend to be By dropping below zero the presump- we are seeing an overbought condition very limited. There tends to be a larger tion of the DJIA making a higher high which merits a corrective period. price correction following the next was removed. As it has turned out, the Summation high that follows the initial DJIA has moved to a higher high any- Page 3 Charts drop. way. The dip below zero does not rule Chart 1: The A-D Line has had an Chart 4: The Volume Summation out a new higher price high, it just exceptional run up from the February Index is not anywhere close to equaling makes it more iffy. The Price Oscillator 2016 low. That run has taken it to a or exceeding its April high. This is in turned down on July 26, the day before new all time high. This action shouts keeping with the under performance of the Fed announcement. Other Price out that there is huge liquidity sloshing the Daily Volume Line. But unlike the Oscillators failed to confirm that rever- around that is available to be put to use A-D Summation Index the Volume sal. Either they will confirm by turning in the stock market or economy. This Summation Index has not yet turned down after the Fed does its thing or the run up has taken the A-D Line as far down to confirm a top being in. Per- DJIA Price Oscillator will have more above its 1% Trend as it has historically haps the Fed announcement on upside work to do. As it stands now been able to get. So, now it will be up Wednesday July 27 will cause the top- there is a significant divergence with a to the 10% and 5% Trends to see if they ping action to be completed or further higher price high on the DJIA and a can contain any correction. the breakout move by moving the Sum- lower Price Oscillator high. This indi- Chart 2: The Daily Volume Line chart mation higher. cator topped at +207.2 and there are doesn’t show the highs that it made last Chart 5: The McClellan Oscillator many many top reversals near this level. year. So, it isn’t apparent that the Daily topped out July 12 at +193.36 before Chart 8: The CBOE Volatility Index Volume Line has not yet been able dipping below zero for only one day on (VIX) has had a couple of closes below make a new high. But it is very close to July 25. The complex Oscillator struc- 12 and five intraday lows below 12 here last year’s highs after the latest bounce ture above zero made it clear that the in late July. These are very low levels up off the June low. It is because of the bulls were in control. The one day dip for the VIX. There have only been a reluctance of this indicator to make new below zero said the bears weren’t able couple of instances which were widely highs that we see the disparity between to take back control. If a simple up and separated in time where postings below the DJIA and SP500 making breakout back down below zero shows up now it 10 were seen in the last 25 years. If the moves above last year’s price highs, would be clear that the bulls were not stock market is to continue trending while the broader NYSE Comp. has able to regain control. It is necessary higher, it will be incumbent on the VIX only been able to retrace back up to the for the Oscillator to remain below zero to remain near these recent low levels. breakdown point just before the August for the Summation Index to correct its But if there is a correction to start soon, mini crash. current very high level. the VIX should acknowledge that by Chart 3: The McClellan A-D Summa- Chart 6: The Volume Oscillator has moving back up to higher levels.

NYSE CUMULATIVE ADVANCE-DECLINE LINE DAILY CUMULATIVE UP-DOWN VOLUME LINE 375 185 370 1 180 2 175 365 170 360 CUM. A-D 5% TREND 165 10% TREND 1% TREND 355 160

350 155 Millions Thousands 345 150 145 340 140 CUM. UV-DV 5% TREND 335 135 10% TREND 1% TREND 330 130

325 125 04-Jun-15 04-Aug-15 02-Oct-15 02-Dec-15 03-Feb-16 05-Apr-16 03-Jun-16 03-Aug-16 04-Jun-15 04-Aug-15 02-Oct-15 02-Dec-15 03-Feb-16 05-Apr-16 03-Jun-16 03-Aug-16

McCLELLAN SUMMATION INDEX: ADV-DECL McCLELLAN SUMMATION INDEX: VOLUME 9000 1130011000 11300 10000 8000 NYSE COMP 3 109009000 NYSE COMP 4 10900 7000 8000 10500 105007000 6000 6000 10100 101005000 *** 5000 ***** ** **** ***** *** ** * *** * ** 4000 ****** *** *** * 9700 *** ** ** * **** ** **** ***** ** *** ** * ** *** 4000 ****** ***** ** *** * 97003000 ** * * * ** ** * ** * * ** ** * ** ** * *** * * **** * ** ** ********* 2000 * * ** * * 9300 * ******** *** * *** ***** ******* ** 3000 * **** * *** * *** ** 93001000 ***** *** ****

Thousands ** **** * *** * * * ** * * ** * 8900 ****** ** 0 ** * **** ** * 2000 ** * ** * ********* * ******** * ** * ***** * *** ** ** * * ****** * 8900-1000 ** * ** * * ** * *** *** ** ** ** * ***** *** * ** ****** * 8500 1000 * * * **** * -2000 ** ******** ******* * ***** ** * * * ** * * * **** * ** ** *** * * * * * * * * * ********** **** *** * * * 8500-3000 * ****** * * * ** ** ** **** * *** * * * ** * * * 8100 0 **** ** ************** * ***** * * * * ***** *** * ** *** * ** * * * * -4000 ******** * * * **** * * * ***** * ****** * * * * *** ******* * ****** * 8100-5000 ** 7700 -1000 ** ** *** * * * * ***** ****** -6000 * -2000 7700-7000 7300 04-Jun-15 04-Aug-15 02-Oct-15 02-Dec-15 03-Feb-16 05-Apr-16 03-Jun-16 03-Aug-16 04-Jun-15 04-Aug-15 02-Oct-15 02-Dec-15 03-Feb-16 05-Apr-16 03-Jun-16 03-Aug-16

McCLELLAN OSCILLATOR (ADVANCE-DECLINE): NYSE McCLELLAN OSCILLATOR (VOLUME): NYSE 400 700 600 6 300 5 500

400 200 300

200 100 100

0 0

Thousands -100

-100 -200

-300 -200 -400

-500 -300 -600

-400 -700 04-Jun-15 04-Aug-15 02-Oc t-15 02-Dec -15 03-Feb-16 05-Apr-16 03-Jun-16 03-Aug-16 04-Jun-15 04-Aug-15 02-Oc t-15 02-Dec-15 03-Feb-16 05-Apr-16 03-Jun-16 03-Aug-16

DJIA CLOSE & PRICE OSC CBOE VOLATILITY INDEX (VIX) 18800 9002200 65 2150 18400 800 SP500 60 2100 7 700 18000 2050 55 600 2000 17600 50 500 1950 8 17200 4001900 45 16800 * ** * 3001850 ****** ***************** 40 * * **** ** ** 16400 ** ** * ** **** 2001800 * *********** * * ** ** *** ** ** * * * * *** ** 35 * ** ** *** * 1001750 16000 * * * * *** * ** **** ** ** ******* *** * 50MA, 1-Sigma BBands * * * ** * ***** ******* *** 1700 ** ** ** * ***** ** * *** 0 30 15600 ************* ***** ** * ** *** ***** ** ***** * * * 1650 **** **** * * * ********* * ** -100 25 15200 * * * * 1600 * * * ** * * * -200 * * ** * 1550 20 14800 ********** * * ** ******** -300 * * * *** 1500 **** * * * ******* ****** 15 14400 * -4001450

14000 -5001400 10 04-Jun-15 04-Aug-15 02-Oct-15 02-Dec-15 03-Feb-16 05-Apr-16 03-Jun-16 03-Aug-16 04-Jun-15 04-Aug-15 02-Oct-15 02-Dec-15 03-Feb-16 05-Apr-16 03-Jun-16 03-Aug-16

2300 1.35 Gold Matching 2200 Start of 1.3 JAPANESE YEN 2100 Abenomics 1.25 Yen’s Movements 2000 1.2 Gold conspiracy buffs are a unique 1900 1.15 Abenomics II, 1.1 class of people. There are not as many 1800 the sequel? 1.05 of them these days as there were in the 1700 1 1980s and 1990s, the years which fol- 1600 0.95 1500 lowed the huge bubble top in 1980. 0.9 1400 They are dying off, and not being 0.85 1300 replaced at the same rate. And yet they 0.8 1200 still exist, and still say funny things. 0.75 1100 0.7 One such conspiracy theorist 1000 NEAR MONTH GOLD recently asserted to us that gold prices 0.65 900 0.6 “are controlled by the COMEX”. If 800 0.55 that’s true, then the correlation evident 700 0.5 in the first chart says that the COMEX 600 0.45 must also be controlling the yen. The 500 0.4 U.S. does nearly $200 billion of trade 01/03/06 01/04/08 01/05/10 01/04/12 01/03/14 01/05/16 per year with Japan, and somehow the and speculative actions can overwhelm The implication of this correlation is COMEX is controlling the exchange the actions of any small group, some- that the immediate future path of gold rates of the two currencies? times including central banks. If any- prices is inextricably tied to the future The problem with conspiracies is one believes that the COMEX has the of the yen. And PM Abe just got ree- that they are difficult to pull off, and power to determine the paths of both lected with a mandate to do more to even more difficult to keep secret. On gold prices and the yen, then that person help Japan’s economy, which has tradi- top of that, the massive market forces needs to do a bit more research. tionally meant driving the yen lower. All of this is unfolding as 3.3 2.2 gold prices are making their final Oct 3.1 CASH GOLD - LOG SCALE 2008 plunge toward a major price low

2 as forecasted by 2 separate and 2.9 Gold finally trades independent cycles. The 8-year freely on U.S. cash Feb Feb 2.7 1985 1993 cycle is depicted in the middle market, JAN 1975 Mar 1.8 2001 chart. Based on when these 2.5 Aug cycle lows have arrived in the 1976 Collapse2.3 of 1.6 past, the next one is due some- Bretton Woods time between August 2016 and Agreement,2.1 15 AUG 1971 March 2017. 1.9 1.4 Our thesis is that we are

1.7 going to see this next 8-year 1.2 cycle bottom arrive coincident 1.5 with the 13-1/2 month cycle low 1.3 due later this year, as shown in 1 the bottom chart. Ideally that 1.1 Idealized 8-Year Cycle cycle low is due in October, but

0.9 0.8 it can arrive plus or minus a 1968 1972 1976 1980 1984 1988 1992 1996 2000 2004 2008 2012 2016 month and still be considered

3.3 “on time”. One other message from this cycle is that coming out of that 3.2 Near Month Gold - Log Scale next cycle low, we ought to see a big rally to a higher price high. That message comes from the 3.1 Mid- fact that the current cycle has Cycle seen “right translation”, meaning Low Mid- 3 Mid- Cycle that the high on the right side of Cycle Low the midpoint is above that on the Low In 2012 it split in left side. That is a bullish con- Another 2.9 two 255 figuration, promising us better split 250 245 things in the next cycle. 240 Bottom Line: Gold is joined at 235 the hip right now with the yen. 13-1/2 Month Cycle 230 225 Both should head lower into a 220 major cycle low due later this 01/04/10 01/03/11 01/03/12 01/02/13 01/02/14 01/02/15 01/04/16 01/02/17 year.

1.44 SP500 Current Period 2200 Presidential Cycle 2150 1.4 Shows Decline 2100 2050 Coming 1.36 Election years are different from Way off 2000 other types of years, in terms of the track here 1.32 July 11 1950 way that the market follows its nor- Aug 31 1900 mal seasonal pattern. And among 1.28 election years, there is a further dif- 1850 ference depending on whether an 1.24 1800 incumbent president is running for Back on track starting 1750 reelection, or if instead there is the 1.2 condition we have now, with two here 1700 1650 new challengers contesting for that 1.16 Presidential Cycle Pattern job. Second Term Presidents 1600 Under that condition, the market 1.12 1550 does not do as well in election 01/02/15 04/06/15 07/06/15 10/02/15 01/04/16 04/05/16 07/05/16 10/03/16 years. And following an early July top, the market typically transitions into arrive July 11. The actual highest clos- with the exact day of the tops. But in a bearish mode as worries about the ing high for the SP500 on the recent up every case where there is a reading of election outcome start to occupy inves- move was on July 22, but the SP500 14 or 15, it is associated with an impor- tors’ thoughts. rose up into this little plateau on July tant top. We just may have to wait a Ideally, this version of the Presiden- 14, very close to the ideal date. couple of days for that top to finish tial Cycle Pattern says that a top should There is one more top shown in the being formed. pattern due Aug. 31, ahead of the more This indicator reached a reading of

450 concerted drop into October. A low in 15 on July 18. The final price high (for 445 N October also happens to fit with the the moment anyway) was on July 20. 440 LA THE CLEL leading indication from the NINO4 The last spike high in this indicator 435 c T ARKE 430 T weather data featured back in MMR brought a reading of 16 on March 29, 425 R REPO #509. which was followed a few days later 420 We can worry about identifying the with a DJIA top on April 1 (and a low point in October when it gets slightly higher high on April 20). The Sherman McClellan Publisher closer. For now, we have an over- key point is that if the bulls can get eve- bought condition appearing roughly on ryone marching in formation together Tom McClellan schedule to deal with. One indication for a protracted period, they end up con- Editor of that overbought condition is the sim- suming all of their energy after a while, The McClellan Market Report, ple indicator in the lower chart. It and cannot keep the band together. A ISSN 1094-8163, is published 24 counts how many times the DJIA has corrective period is the natural remedy, times a year by email & web access. closed higher over the past 20 trading and one is due now. Annual subscription rate: $195 days. Bottom Line: We are now out of the Readings above 14 or below 6 show bullish phase of the Presidential Cycle, Visa, Mastercard, and Discover cards accepted, or mail payment to: extended conditions. The really low and getting ready for a meaningful McClellan Financial Publications readings are more reliably associated decline starting in a few weeks, and bot- P.O. Box 39779 with bottoms than the high ones are toming in October. Lakewood, WA 98496-3779 Phone (253) 581-4889 fax:584-8194 Subscriptions: (800) 872-3737 4.3 26 or at www.mcoscillator.com DJIA, Log Scale 24 email: [email protected] 4.25 ©2016. All rights reserved. Any reproduction 22 of the words, graphs, tables, or analysis, in 20 whole or in part, without the express consent of 4.2 McClellan Financial Publications is strictly prohibited. Analysis is derived from data Up Closes in Last 20 Days 18 believed to be accurate, but such accuracy or 4.15 completeness cannot be guaranteed. It should 16 not be assumed that such analysis, past or 14 future, will be profitable or will equal past per- 4.1 formance or guarantee future performance or 12 trends. Information is provided to assist readers in forming their own understanding of market 4.05 10 structure and no statements made herein should be construed as recommending any specific 8 course of action. All trading and investment 4 decisions are the sole responsibility of the 6 reader. From time to time the principals of McClellan Financial Publications may have 3.95 4 open positions in the markets covered. 01/03/12 01/02/13 01/02/14 01/02/15 01/04/16

TIMING MODELS Stock Indices (DJIA, SPX, Nasdaq, NYSE Comp., etc.) Treasury Bond Prices

SIGNAL SOURCE PREDICTED ACTUAL SIGNAL SOURCE PREDICTED ACTUAL Top DJIA ST Price Osc July 7 July 6 Top T-Bond Stochastic July 13 July 11 Top DJIA Price Osc July 8 July 6 Bottom T-Bond Close/Sum July 14 July 18 Bottom DJIA Price Osc July 8 July 7 Top T-Bond ST Price Osc July 26 Bottom DJIA Close/Sum July 11 July 7 Top T-Bond Up-Dn Osc July 27 Bottom Uncommon A-D Osc July 27 Bottom T-Bond Up-Dn Osc Aug 12 A Bottom NYSE Vol. Summ Aug 3 A Bottom DJIA ST Price Osc Aug 3 Gold and Precious Metals Stocks A Bottom SP500 Price Osc Aug 4-5 SIGNAL SOURCE PREDICTED ACTUAL Bottom Uncommon A-D Osc Aug 15 Top Gold ST Price Osc July 7 July 6 Bottom SP500 Price Osc Aug 22-23 Top Gold Price Osc July 25 July 21 Top Nasdaq Up-Dn Osc Aug 26 Top GDM ST Price Osc July 27 Bottom NYSE A-D Osc Aug 29 Bottom Gold Price Osc Aug 17 Bottom SP500 Close/Sum Sep 22 Bottom GDM Price Osc Aug 22 Experimental Indicator, “BC” Top Gold ST Price Osc Aug 23 Predicted Signal How It Turned Out Top Gold Up-Down Osc Aug 24 Implied Top July 4 Top July 1 business-unfriendly Implied Top* July 22 Top July 20 tax hikes and mini- What To Expect Stock prices drop after the July 27 *This signal from BC’s big brother, see p. 2 mum wage hikes. FOMC meeting announcement. We A Volatility Event Aug 3 And the FOMC just started a 2-day don’t know what they are actually going Implied Bottom Sep 14 to say, just that the market is set up to Implied Top Oct 2 meeting on July 26, with the post- not appreciate whatever it is. Volatility meeting announcement due out in the should rise into a climax due Aug. 3-5. The Signals afternoon of July 27. That announce- Another important looking bottom is The big news is that the top signal ment could have a weight on how these due Aug. 22-23, so far coming from from BC’s Big Brother that we wrote signals play out. only one signal and not yet confirmed about in our last Report seems to have An action on rates by the FOMC, or by other ones. But gold does show a come in right on schedule. That top even an adverse comment, could help to bottom due then as well. was curiously not identified by our nudge the stock market out of the very T-Bonds have been moving side- regular Timing Model signals. narrow range we have seen over the ways for the past week, attempting to Those signals did correctly mark the past 2 weeks, and could help fulfill that hold support. A top due July 26-27 is top we saw at the beginning of July, and Aug. 3 “volatility event”. likely going to be related to the FOMC they do offer confirmation of a special And that volatility event from our meeting announcement, and should be type of signal for the BC Indicator experimental BC Indicator coincides more of a top to go down out of rather known as a “volatility event”. We call with a cluster of bottom signals due than up into. it that because when there are those lit- Aug. 3-5 from our conventional signals. Gold shows important looking bot- tle wiggles in the BC Indicator (see The bold-letter designation on one of tom signals due Aug. 17 and 22. MMR 510), there tends to be a jump in those indicates that the nature of that Those could be referring to the same volatility which is otherwise hard to signal is similar to other big ones in the bottoming event, as they are actually characterize. past that have had greater importance only 3 trading days apart. But that is As we look at these signals, it is for stock prices. So get ready for the not likely to be the final price low for worth keeping in mind that the Demo- market to get more exciting! gold’s decline into the 8-year and 13- cratic National Convention is underway 1/2 month cycle lows due later this as we write this, with renewed calls for year. HOW THEY WORK These models do not catch every market turn, but the signals usually These timing models are based on our proprietary calculation method. show some effect in the market action. It is important to understand that the This technique involves a computationally complex comparison of two or market does not have to go up from a bottom; it may just stop going down. It more carefully selected indicator values. This yields the date and direction of does not have to go down from a top, it may just stop going up. Some bot- a projected future turning point. Making several such comparisons can help toms turn out to be just a flat spot before a continuation up. paint a picture, one reversal point at a time, of the future structure. The BC indicator is an experimental new tool, not related in method to Once generated, signals remain in effect, though the result can have the other signals. greater or lesser significance based on what the market is doing when the date "Actual" dates listed for NYSE Indices are for the NYSE Comp/Dow arrives. Certain indicators are slightly less accurate in pinpointing the exact Jones Industrial Average. Letter groups (A, B, C, etc.) denote clusters of sig- date, so we may print a range of dates. Price Oscillators and Summation nals. ST Prc Osc means “Short Term Price Oscillator.” Index signals are usually more important, though sometimes not as precise in Past performance of these mathematically generated turning point pro- time. Uncommon A-D refers to an oscillator derived from NYSE stocks that jections in no way guarantees future results. These dates may be useful in are not part of the Common Only list in Barron’s. Dates in bold denote sig- planning for the future, or giving greater confidence at turning points. We nals of greater potential strength according to our research. would not, however, attempt to trade any of the markets based solely on these models.

Utilities Divergence 740 22000 DOW JONES UTILITY AVERAGE 21500 700 Is Bad News For This higher high 21000 Stocks 660 clearly blew it as a 20500 bullish message 20000 Normally the movements of the 620 DJIA and the Dow Jones Utility Aver- 19500 age are pretty well correlated. That 580 19000 makes sense, because after all, each is 18500 540 made up of stocks. It is when they dis- 18000 agree with each other that we can get 500 useful information. 17500 Usually when there is a disagree- 460 17000 ment, it is the DJU that ends up being 16500 420 right about where both are headed. 16000

That is important because the DJU 380 DOW JONES INDUSTRIALS AVERAGE 15500 made its top back on July 6. The DJIA 10/02/14 01/02/15 04/06/15 07/06/15 10/02/15 01/04/16 04/05/16 07/05/16 10/03/16 continued higher and its all-time high doing the opposite of each other, but we recently moved to a new all-time high (so far) was on July 20, creating the foresee a likelihood of both types of net short position in T-Bond futures, bearish divergence we see now. assets falling in price over the next 3 evidently betting on a big drop for bond That news is of immediate concern months or so. prices. to stock market investors. But it could One big reason is the data in the The commercials are often early in also mean a big decline for utilities Commitment of Traders (COT) Report, getting to a big skewed position like stocks, as the “chase for yield” suddenly the highlights of which we feature every this. But they are nearly always proven becomes not quite so fashionable. Friday in our Daily Edition. The “com- right when they do (eventually). So the And that leads us to a similar mercial” traders are the big money, and message to take from this is that bond expectation for bond prices. We have usually also the smart money. They prices really have no business being up all gotten used to stock and bond prices 40% 180 this high, and are going to have to do Near Month T-Bonds something to remedy that. 170 There is one fly in the ointment for 30% 160 that hypothesis, though. The bottom chart shows a comparison of the current 20% 150 plot of the Treasury Yield Index (TYX) NET SHORT 140 versus that same index from a period in 10% 130 the 1990s. There is an impressive pattern correlation between the two. 120 0% If that pattern correlation is going 110 to persist into the future, then that

-10% would mean the TYX should fall further NET LONG 100 from here. The TYX measures the cur- 90 rent yield-to-maturity of the most -20% COT: COMMERCIALS' NET SHORT POSITION 80 recently issued 30-year Treasury bond. AS PERCENTAGE OF TOTAL OPEN INTEREST So for the TYX to go lower as indi- -30% 70 01/03/07 01/03/08 01/02/09 01/04/10 01/03/11 01/03/12 01/02/13 01/02/14 01/02/15 01/04/16 cated, then bond yields would have to fall, and bond prices rise. 4 9.4 That is in conflict with the message 3.8 9.2 from the COT data, and so we have a Treasury Yield Index (TYX) 9 3.6 quandary. The two cannot both be 8.8 3.4 right. Our preference is to follow the 8.6 COT data, because pattern analogs like 3.2 8.4 3 8.2 this can and do break correlation, and usually a the moment when one is most 2.8 8 7.8 fervently counting on it to continue 2.6 7.6 working. 2.4 7.4 Perhaps if the world’s central banks 2.2 7.2 realize that ZIRP is killing Deutsche- 2 7 bank and other major banks, they’ll let 1.8 6.8 up, and allow rates to normalize again. 6.6 1.6 6.4 Bottom Line: Commercial traders are 1.4 making a huge bet now on lower bond TYX 1994-97 6.2 1.2 6 prices (higher yields). They are usually 1 5.8 right. 03/05/13 07/03/13 10/31/13 03/05/14 07/03/14 10/31/14 03/05/15 07/06/15 11/02/15 03/04/16 07/05/16 11/01/16

2016 Protests Not 450 1946 - post-WWII 400 Likely to Have Last- rail, coal, steel 1934 - "Long strikes led to 350 1925 - 35,000 KKK March". Taft-Hartley Act ing Significance in 1947 1904 - 1917 - protesting members march 1935 - Dec. 9 2016 has seen a lot of protests in 300 General suffragettes on Washington protest the news, or so it seems. The two strikes in arrested (led to DC; May 30th movement in Italy, 19th amendment movement strikes China; Relief major parties’ presidential nominating 250 Russia, in 1918); Russian in China Camp strikes conventions have seen some protests, Japan revolution in Canada which is pretty typical. There are also 200 the protests over police shootings by groups like “Black Lives Matter”, and 150 counter-protests in support of police. 100

But these protests are not likely to Monthly Avg Sunspot Number have lasting historical significance. We 50 say that because the really significant protest movements in history, move- 0 ments which led to actual change in 1900 1905 1910 1915 1920 1925 1930 1935 1940 1945 1950 policy or practice, have tended to 450 1965 Watts 1989 - Fall of appear during the ascending phase of Riots 1979 - Iranian Berlin Wall; the 11-year sunspot cycle. This is a 400 student protests, 1967 Vietnam Tienanmen overthow Shah 1998 Indone- thesis first put forth by Russian war protests Square sian protests, 3501955 - Rosa researcher A. L. Tchijevsky who corre- Massacre fall of Suharto; Parks sparks 1968 - General 1999 WTO riots lated instances of wars, revolutions, Montgomery 300 strike in France; in Seattle riots, etc. with the sunspot cycle, going bus boycott MLK murdered; DNC convention 250 back over 2000 years. He wound up protests 2011 - Arab spending a long time in a Russian gulag Spring; because his views about what sparked 200 "Occupy Wall Street" the 1917 Russian Revolution conflicted with those of Josef Stalin. 150

The first two charts on page 8 100

show the monthly sunspot number, and Monthly Avg Sunspot Number the 9-14 year cycle period is fairly evi- 50 dent. Typically we see the major protest movements on the ascending phase 0 of each cycle, and the Arab Spring in 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015 2011 was a great example. problems finally come to a head. forward by the average 11-year period This is not to say that there are not Curiously, the rising phase of the length gets us another stock market low protests at other times in the sunspot sunspot cycle also tends to coincide around 2020. If we use the 9-14 year cycle. But the point is that protests aris- with rising economic prosperity, and a range of values for the period between ing at other times do not typically rising phase for the stock market. And sunspot minima, then we get 2018- arouse the same degree of mass support the sunspot minima tend to mark impor- 2023. from the public. Something about tant lows for the DJIA, plus or minus a One other point worth noting about increasing solar activity seems to work year. this relationship is that there is often a on human brains, increasing our collec- That’s important because the last major market dip halfway through the tive uppitiness such that long-festering solar minimum was in 2009. Counting sunspot cycle, roughly coinciding with the peak in the monthly sunspot num- 550 4.3 4.1 bers. We did not see a mid-cycle low in 500 3.9 the 1920s, as the roaring boom/bubble 3.7 then painted over it. And we arguably 450 DJIA - Log Scale 3.5 3.3 did not see one in the current sunspot 400 3.1 cycle, with the Fed’s QE programs Sometimes the 2.9 350 keeping the market aloft. sunspot cycle peak 2.7 sees a mid-cycle dip 2.5 Once we get past the next solar 300 for DJIA 2.3 minimum in 2018-2023, we can expect 2.1 250 another major protest movement to 1.9 erupt sometime in the mid-2020s. And 200 1.7 1.5 we can also expect a big boom for stock 150 1.3 1.1 market prices at the same time. But as 100 0.9 we slide into the next solar minimum, Monthly Avg Sunspot Number 0.7 the expectations of higher long term 50 0.5 0.3 stock prices (absent more QE) should 0 0.1 be reduced. 1890 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 2010