FFeature:eature: TTradingrading SStrategiestrategies

Charles Dow’s Theory Still Valid for the 21st Century By Jack Schannep

Article Highlights • Dow Theory uses the Dow Jones industrial average and the Dow Jones transportation average to judge the market’s trend. • The movement of one price average must be confi rmed by the other before reliable inferences may be drawn. • Followers of the theory must accept that day-to-day manipulation is possible, the averages discount everything and the theory is not infallible.

Every day we hear about the activity. He recognized that if the stocks Dow rising or falling, but we may in his averages were going up, that pointed not stop to think who Dow was, to future business being good, and vice what the Dow averages are all versa. On January 31, 1901, Charles Dow compared the stock market to the tides about and what the implications of the ocean when he wrote in The Wall might be. Street Journal: “A person watching the tide I’ll start with who Charles Dow was coming in and who wishes to know the and then how his theory, which has served exact spot which marks the high tide, sets so well for over 100 years, can still be used a stick in the sand at the points reached by in the 21st century as a guide to timing the stock market and the incoming waves until the stick reaches a position where making money from it. the waves do not come up to it, and fi nally recede enough to show that the tide has turned. This method holds good in Dow Theory Origins and Background watching and determining the fl ood tide of the stock market.” If you think of the Dow Jones industrial average as being Charles Henry Dow was born in November 1851 on the measure of the tide on one part of the beach and the a farm in Sterling, Connecticut. At the early age of 18, he Dow Jones transportation average as the measure on another began his career as a reporter and developed an early interest part of the beach, and you use both to determine that the in business. Subsequently, he and his friend Edward Jones tide is indeed coming in or going out all along the seashore, formed Dow Jones & Company in 1882. They published the you will understand what Dow was getting at: Confi rmation handwritten Customer’s Afternoon Letter, precursor of The by both averages is an integral part of the Dow Theory. Wall Street Journal, with Dow as its fi rst editor. It was not The amazing thing about Dow’s theory was that it until May 26, 1896, that the Dow Jones industrial average was developed with only fi ve years’ worth of data on the was born with 12 “smokestack” companies. A year later, a two averages from which to form the basis. Unfortunately, separate average began to keep track of the railroad stocks, Charles Dow had little time to write about and expound on which were the primary transportation mode of the day. Both his theory. He never wrote down a complete description of averages have been expanded over time, with 30 stocks now his theory, never dedicated a complete editorial to it, never in the industrial average, and the rail average (renamed the gave it a name and, most unfortunately, never wrote a book. transportation average) expanded to include airline, air freight, By 1902, Dow was in failing health and sold the company. delivery service, marine transport and trucking companies. He died shortly thereafter, on December 4, 1902. Most of Dow saw the stock market and his idea, yet to be named what we know of the theory came from a series of editori- by others as the “Dow Theory,” as a barometer of business als in The Wall Street Journal written by Dow’s successor as

September 2012 7 Figure 1. The Classic Buy Signal averages, all of which may be in progress at one and the same time. Dow Jones Industrials Industrial and Transport The Primary Trend patterns are The fi rst, and most important, is Transports interchangeable the primary trend: The broad upward or downward movements known as bull or bear markets, which may be of several years duration. The correct determina- tion of the direction of this movement 4) Higher is the most important factor in successful Bounce 2) Bounce speculation. There is no known method Breakout of forecasting the extent or duration of Level a primary movement. The primary trend, (Buy signal) once in place, is assumed to continue in place until defi nitely proven otherwise. 1) Lows 3) –3% Pullback to This is an offshoot of Isaac Newton’s Higher Lows law, which states a body in motion tends to stay in motion unless compelled to The classic buy signal is outlined as follows: change its state. A primary bear market is a long 1. Market lows (“the sticks in the sand”) 3. Pullback (hold above the lows) downward movement interrupted 2. Bounce 4. Breakout (above the bounce high) by important rallies. There are three principal phases of a bear market: The fi rst represents the abandonment editor, William Peter Hamilton, during ary reactions are subject to such an of the hopes upon which stocks were the period from 1902 until his death in infl uence to a more limited degree, purchased at infl ated prices; the second 1929. Hamilton did write a book about but the primary trend can never be refl ects selling due to decreased business Dow’s theory, “The Stock Market Ba- manipulated. and earnings; and the third is caused rometer,” in 1922 (reprinted in 1998 by 2. The Averages Discount Every- by distress selling of sound securities, John Wiley & Sons). thing: The fl uctuations of the regardless of their value, by those who The most organized and thorough daily closing prices of the Dow must fi nd a cash market for at least a description of the Dow Theory as we Jones rail and industrial averages portion of their assets. knew it in the early 20th century came afford a composite view of all These phases go from complacency from the book of that name, written by the hopes, disappointments and to concern and, fi nally, to capitulation. Robert Rhea in 1932. Rhea, who was knowledge of everyone who knows “Capitulation” was a term never used bedridden, had the time and inclination anything of fi nancial matters. For by Dow, Hamilton or Rhea, but it was to analyze the 35 years of data available that reason, the effects of coming alluded to by the description of the to him to further refi ne the work of events (excluding acts of God) are third phase of a bear market, and the Dow and Hamilton into what I consider always properly anticipated in their further statement: “After a market has the defi nitive work on the original Dow movement. The averages quickly drastically declined…and then goes into Theory. The book was reissued in 1993 appraise such calamities as fi res a semi-panic collapse, it is wise to cover by Fraser Publishing Company. Portions and earthquakes. short positions and even perhaps make are reprinted here with permission. 3. The Theory Is Not Infallible: commitments for long accounts.” In Robert Rhea, after many years of The Dow Theory is not an infallible my interpretation of the Dow Theory studying the writings of both Dow and system for beating the market. Its for the 21st century, I do indeed use Hamilton, set out a “few hypotheses,” successful use as an aid in specula- capitulation as the time to start buying. which he said must be accepted “without tion requires serious study, and the A primary bull market is a broad reservation whatsoever” if one is to use summing up of evidence must be upward movement, interrupted by sec- the theory successfully in order to know impartial. The wish must never be ondary reactions, and averaging longer when to buy and sell and to make money allowed to father the thought. than two years. There are three phases in the stock market. They are: of a bull period: The fi rst is represented 1. Manipulation: Manipulation is The Three Movements by reviving confi dence in the future of possible in the day-to-day move- business, the second is the response of ments of the averages, and second- There are three movements of the stock prices to known improvement in

8 AAII Journal FFeature:eature: TTradingrading SStrategiestrategies

corporate earnings, and the third is the Figure 2. The Classic Sell Signal period when speculation is rampant and infl ation is apparent—a period Dow Jones when stocks are advanced on hopes Industrials Industrial and Transport and expectations. patterns are While the primary trend has not Transports interchangeable been specifi cally defi ned, my own re- 3) +3% Bounce to search shows that a bull market primary 1) Highs Lower Highs trend will have advanced in excess of 19% on the Dow Jones averages and the S&P 500 index. A bear market primary Breakdown trend will have declined in excess of 16% Levels on both. These percentages are recipro- (Sell signal) cal numbers, unlike the traditionally used 2) Pullback 4) Lower Lows +20% and –20%, but there are more important reasons to use them, which space does not allow me to expand upon. The classic sell signal is outlined as follows: Secondary Reaction 1. Market highs (“sticks in the sand”) 3. Bounce (to below the highs) The second, and most deceptive, 2. Pullback 4. Breakdown (below pullback) movement is the secondary reaction: This is an important decline in a primary bull market or a rally in a primary bear fi gures are not written into the theory. reliable inferences may be drawn. Con- market. For the purpose of this discus- What is precisely defi ned is the extent clusions based upon the movement of sion, a secondary reaction is considered of the “return move,” the pullback after one average, unconfi rmed by the other, to be an important decline in a bull a bounce up from a bear market bottom are almost certain to prove misleading. market or advance in a bear market, or the bounce after a pullback from a A common complaint is that the usually lasting from three weeks to as bull market top, and that it shall exceed railroads (transports) are of inconse- many months, during which interval the 3% on either of the averages. quential importance these days, which price movement generally retraces from After a secondary reaction, the makes the theory ‘out of date.’ I would 33% to 66% of the primary price change primary trend is reaffi rmed when both remind the reader that the transportation since the termination of the last preced- the industrials and transports return average is actually made up of 20 stocks ing secondary reaction. These reactions to extend that trend, but confi rmation representing at least six transportation- are frequently erroneously assumed to need not occur on the same day. In a related industries. The stocks in the represent a change of the primary trend, bull market, such a move to new highs average deliver raw materials and com- because obviously the fi rst stage of a is often described as being “in the clear” ponents to industry and then distribute bull market must always coincide with and is sometimes labeled as a new buy the manufactured products to the world. a movement which might have proved signal, which is incorrect. The buy signal Therefore, their business fortunes are to have been merely a secondary reac- dates to the original signal; this move still intertwined. tion in a bear market; the contra being merely affi rms that signal. There is a modern-day misconcep- true after the peak has been attained tion that says both the industrial and in a bull market. The time frame of Daily Fluctuation transports must make new all-time highs “usually lasting from three weeks to The third, and usually unimportant, for a bull market to be in force. But a as many months” has been taken by movement is the daily fl uctuation. Infer- bear market changes to a bull market many Dow theorists as being cast in ences drawn from one day’s movement at the low point, not after it gets to a stone. Actually, at one time (January of the averages are almost certain to be higher point than the last bull market. 4, 1902), Dow himself wrote, “The misleading and are of little value. Granted, the new bull market is not secondary movement covers a period immediately determinable at that low ranging from 10 days to 60 days,” and Determining the Trend point, but after a time it can be seen as that is the defi nition I use. A review of having been the start. the Dow Theory signals implies that a The movements of both the The levels at which I consider that secondary trend will usually extend at railroad and industrial stock averages a market attains bull- or bear-market least 4% on both the industrials and should always be considered together. status are +19% and –16% as men- transportation averages, and usually The movement of one price average tioned previously. I use those levels as one or both will exceed 7%, but these must be confi rmed by the other before (continued on page 13)

September 2012 9 SStocktock SStrategiestrategies

1.06 for the median stock. Morningstar or low brings out some of the same Figure 3. Morningstar’s Equity doesn’t have enough small-cap stocks companies we’ve uncovered already. Style Box under coverage to create price/fair value Vulcan Materials fi ts the bill, as does ratios for the small-cap style-box squares, Cisco Systems. A few additional compa- Equity Style Box but based on the aforementioned pat- nies fi t our criteria, however, including terns, it’s a reasonable guess that a) Johnson & Johnson (JNJ), Medtronic Large 0.88 0.92 0.99 small-cap stocks look more expensive Inc. (MDT), Microsoft, Exelon Corp. than mid- and large-cap names and b) (EXC) and Pfi zer Inc. (PFE). growth stocks look more expensive than Medium 0.89 0.92 1.06 value stocks. Broad Takeaways That’s not to say that there’s no gas left in the tank for growth stocks: Com- Although we’ve viewed the equity panies that can increase their earnings universe through a variety of differ- Small N/A N/A N/A despite a slack macroeconomic environ- ent lenses, there are some overarching ment are still apt to be rewarded. But our themes that emerge. Value Blend Growth data do suggest that if you’re rebalancing Value stocks look relatively attrac- your portfolio or deploying new money, tive to our equity analysts relative to value-oriented stocks or funds are a good growth names right now. That’s espe- like real estate and utilities, meanwhile, place to focus your efforts. cially true if investors are willing to delve look relatively expensive and could be Screening for large- and mid-size into value-oriented companies that are good sale candidates if investors are value stocks with wide moat ratings, more cyclical in nature, such as those in trimming long-held winners from their price/fair value ratios of 0.90 or less and the energy and basic materials sectors. taxable portfolios in anticipation of fair-value uncertainty ratings of medium Many stocks in dividend-rich sectors higher dividend and capital gains taxes. 

Christine Benz is director of personal fi nance for Morningstar and senior columnist for Morningstar.com. She is a frequent speaker at AAII Investor Conferences and Local Chapter meetings. Find out more at www.aaii.com/authors/christine-benz.

FFeature:eature: TTradingrading SStrategiestrategies

(continued from page 9) a buy signal for the developing bull the next decline as measured by both the a ‘stop-point’ for completing the Dow market. industrial and transportation averages, Theory for 21st century signals, in case Figure 1 represents how the Dow a sell signal is generated, indicating a they were not completed prior to those Jones industrial average and the trans- bear market. thresholds being attained. portation averages might look. Figure 2 on page 9 illustrates a clas- More than one bounce can occur sic sell signal. Buy Signal within the confi nes of the bounce highs The classic buy signal is developed and the lows. A common variation is as follows: After the low point of a when one of the averages makes newer Performance primary downtrend in a bear market is lows but then rejoins the other in mak- established (point 1 in Figure 1 on page ing newer highs. Historically, the traditional Dow 8), a secondary uptrend bounce will Theory has outperformed a buy-and- occur (point 2). After that, a pullback Sell Signal hold approach by 1% to 2% a year, a (point 3) on one of the averages must The classic sell signal is determined seemingly small percentage until one exceed 3%, according to Robert Rhea. in much the same way, but opposite to considers the compounding effect over That setback must, ideally, hold above a buy signal. When a bull market tops a number of years, where the difference the prior lows on both the industrial and sets back, with a subsequent rally is substantial. and the transportation averages. Finally, that goes back up (again, over 3%) but Improvements to the Dow Theory a breakout above the previous rally high falls short of reaching the previous high for the 21st century would appear to in- (point 4) by both averages constitutes and then penetrates the recent lows on crease that advantage going forward. 

Jack Schannep is editor of TheDowTheory.com and author of “Dow Theory for the 21st Century: Technical Indicators for Improving Your Investment Results” (John Wiley & Sons, 2008). Find out more at www.aaii.com/authors/jack-schannep.

September 2012 13