CHART 21.14 DJIA versus DJ Utilities 1980–2013
The relationship between the Utilities and the Industrials is often overlooked because they usually give their loudest message when other market activity is at its most exciting. It is normal behavior at market tops for the Utility Average to quietly decline while investors, analysts, and the media are excited about huge price advances yet to be seen. Chart 21.14 shows a classic example in 1987. In August, the Industrials were at an all-time high, but the Utility Average was already in a well-established bear market. At market bottoms, fear, depression, and sometimes panic reign while the Utility Average is very quietly in the process of turning up.
The Unweighted Indexes
An unweighted index is calculated by adding the prices of a universe of stocks and dividing the total by that number. The resulting average is then weighted by price rather than capitalization. The most widely followed is the Value Line Arithmetic.
Unweighted indexes are useful because they closely represent the price of the “average” stock often found in individual portfolios, as opposed to the blue chips, to which institutional investment is more oriented. Unweighted indexes are also helpful in gaining an understanding of the market’s technical structure since they have a tendency to lead the market (i.e., the DJIA) at market tops. When a persistent divergence of this nature between the DJIA and the Value Line develops, it almost always results in the Dow being dragged down as well. Once a divergence starts, a cautious approach should be maintained until both the DJIA and the Value Line break out from price patterns or declining trendlines, etc.
A show of good RS by the unweighted indexes at a time of sustained weakness in the major averages often indicates that a significant rally will follow when the decline is over. This occurred in 1978, when the Value Line Composite Index made its low in late 1977, several months ahead of the DJIA.
Chart 21.15 shows the Value Line Arithmetic against the S&P Composite between 1984 and 1990. In late 1985, the Value Line made a lower low than it did at the beginning of the year, but the S&P made a higher low. This out-of-gear situation was a negative sign but was never confirmed by the S&P violating its 40-week EMA. We see a similar type of situation in 1986, but again, this potentially negative discrepancy was not confirmed by a negative S&P MA crossover. The situation in 1990 was different because the S&P not only penetrated its MA but violated a major up trendline as well. This reaffirms an important principle—that of confirmation. There are countless situations where we can compare two indicators or averages and observe disagreements. However, just as divergences in oscillators should be confirmed by price, so these disagreements, whether they are positive or negative in nature, must be confirmed before we can come to a conclusion that the trend has reversed.
CHART 21.15 Value Line Arithmetic versus S&P Composite, 1984–1990
The NASDAQ
The technology boom of the 1990s brought the NASDAQ Composite into a kind of prominence that it had never experienced before. This capitalization-weighted index is dominated by large technology companies and has become a proxy for the technology sector. The NASDAQ has no consistent leading characteristics like the Utilities, probably because several technology sectors, such as semiconductors, have lagging tendencies. However, it can be used with relative strength analysis. Chart 21.16 features the NASDAQ together with its RS line against the S&P Composite. Note how a joint trendline break in 1991 signaled a major rally. Later on, another down trendline break in the RS line was confirmed. This time, it was a solid break above a resistance trendline, the violation of which resulted in an acceleration in the speed of the bull market.
CHART 21.16 NASDAQ Composite, 1983–2000 versus NASDAQ RS
Also worthy of note is the fact that the RS line diverged positively with the NASDAQ Composite Index at the 2009 low. This was quite different from the previous bear market where the NASDAQ Composite outperformed on the downside. While there was a nice breakout above a trendline on the RS line, there was no such possibility for the price itself, which literally reversed on a dime.
The Russell Indexes
The Frank Russell Organization, among other things, publishes three important indexes: The Russell 3000, 2000 and 1000. The Russell 1000 is a composite capitalization-based series containing the 1,000 largest stocks in the country. The Russell 2000 represents the next 2,000 issues based on capitalization. Finally, the Russell 3000 is a composite index of the other two. It represents in excess of 95 percent of the investable U.S. equity market. These indexes are plotted in Chart 21.17. Normally, they move in gear with each other. It is when they disagree that the discrepancies can sometimes be very revealing. In October 1999, all three succeeded in violating important down trendlines, and the joint break indicated a rally lay ahead. On the other hand, the Russell 2000, which is often used as a proxy for the low-cap sector, experienced a sharp rally going into February 2000. All three indexes then retreated, but the Russell 2000 was unable to rally to a new high, unlike the other two. Thus, we have what had previously been the leader no longer leading. Such leadership failures are often a sign that the prevailing trend is running out of steam and throws up a definite red flag.
CHART 21.17 Three Russell Indexes 1999–2000
Major Technical Principle When several closely related securities are being led by one of the group and that leader fails to signal a new high (or low in the case of a declining trend), this is usually a sign of exhaustion and is followed by a trend reversal.
In this case, the April rally proved to be the top of the bull market. Finally, we see that the Russell 1000 rallied back to its spring high in September 2000 but the Russell 2000 was unable to confirm. When all three violate their (dashed) up trendlines a little later, the divergences were confirmed and a major decline followed.
The relationship between the Russell 2000 (low cap) and Russell 1000 (blue chip/high cap) can also be helpful because it can provide a clue as to which category investors should favor. Chart 21.18 shows that the relationship can be quite cyclical in nature. This can be seen from the long-term Know Sure Thing (KST). Sometimes it is possible to augment KST MA crossovers with trendline breaks in the ratio itself.
CHART 21.18 Russell 2000/1000 Ratio, 1987–2012 and Long-Term Momentum
This was the case in 1991 and 1995, but the drop was too steep in the late 1990s to construct a line. The next breakout, flagged by the dashed arrow, developed with the sharp rise in 2000, which turned out to be a whipsaw. The reason was the dramatic first-quarter run up in the technology sector, which temporarily dominated the Russell 2000. Later on, if one was prepared to ignore this false move, it was possible to observe a subsequent breakout from the extended base at the end of 2000.
The KST peaked in 2002, fully three years before the 2005 top, which was confirmed with the violation of a dashed up trendline. Even that followed a second lower KST peak. A small decline followed, and the next upleg in the secular bull market of this relationship was signaled by the price violating the solid down trendline in 2008. The KST followed with a lag, which is unusual, and the ratio continued to rally into early 2011.
Global Equity Indexes
MSCI, Dow Jones, and FTSE are the leading index providers for international equity indexes, but since the vast majority of widely traded international ETFs fall under the Morgan Stanley Capital International (MSCI) banner, we will focus on two of their offerings. The MSCI World Stock Index has been available since the mid-1960s and includes over 6,000 stocks from developed countries. A related index, the MSCI All Country World Index, incorporates both developed and emerging countries. It is the tracking index for an ETF (symbol ACWI), and is used here as a proxy for global equities.
There are also many regional and individual country indexes that are too numerous to mention. However, the MSCI Europe Australasia Far East Index does deserve mention, since for all intents and purposes, it represents 22 developed countries, excluding its biggest component, the United States. Canada is also excluded. In eff
ect, it reflects the rest of the non-U.S. world. This ETF carries the symbol EFA and is useful when calculated as a ratio between the SPY and itself. A rising ratio trend then indicates that the United States is outperforming the rest of the world and vice versa.
Global Bond Indexes
The most comprehensive U.S. bond ETF is the Barclay’s Aggregate Bond (symbol AGG). AGG holds bonds across the spectrum: Treasury notes, Treasury bonds, corporate bonds, utilities, U.S. agencies, and more. About 40 percent of its holdings at the end of 2012 were invested in bonds with a maturity greater than 5 years. It serves as a proxy for the overall U.S. credit market. A good proxy for long-term U.S. government bonds is the Barclay’s 20+ year Treasury bond ETF (symbol TLT).
Internationally, the Barclay’s Capital Global Treasury Ex-U.S. Capped Index is a useful benchmark for the world, with the exception of the United States. It includes government bonds issued by investment-grade countries outside the United States, in local currencies, that have a remaining maturity of 1 year or more. Since the tracking index for this fund includes a large group of countries, it could be adversely affected by questionable sovereign debt. The symbol for this ETF is BWX.
Commodity Indexes
The commodity index that you will find used a lot in this book is the CRB Spot Raw Industrials published at CRBtrader.com. This series is constructed from 18 raw industrial commodities, none of which except cotton are traded on the major exchanges. This series is very useful for intermarket and interasset analysis since it is not driven by weather, but by economic developments. Therefore, it better reflects true inflationary conditions as they arise in the commodity markets thereby affecting bond yields and equity prices.
There are two principal commodity funds based on tracking indexes. The first is the DB Commodity Fund, which tracks the DB Commodity Index. At the end of 2012, this index comprised just over 50 percent energy with 22 percent in grains and the balance in metals. The weightings are determined by the liquidity of the various contracts rather than by their economic importance. The second is the Dow Jones UBS Commodity ETN, the DJP, which tracks the Dow Jones UBS Commodity Index. The rough commodity sector ratings in December 2012 were energy 30 percent, agriculture 32 percent, industrial and precious metals 32 percent, and a small 6 percent in livestock. Of the two funds, the DBC is the more liquid.
Summary
1. There is no perfect index or average that consistently and truly represents “the market.”
2. There are two principal methods of calculating market averages: those that use capitalization and those that incorporate an unweighted formula.
3. The technical indicators described elsewhere in this book can be applied to market averages.
4. Most of the time, market indexes move in gear with each other. It is when discrepancies develop and are confirmed that reversals in trend are signaled.
1Robert W. Colby and Thomas A. Meyers, The Encyclopedia of Technical Market Indicators, Homewood, IL: Dow Jones-Irwin, 1988.
2Investors Press, Palisades, NJ, 1968. The actual rule used for buy signals was as follows: “If the 200-day (40-week) average line flattens out following a previous decline, or is advancing and the price of the stock penetrates that average line on the upside, this comprises a major buying signal.”
22 PRICE: SECTOR ROTATION
Chapter 2 discussed the relationship between the three key asset classes—debt, equity, and commodities—and the business cycle. It was established that there are certain periods when they move in concert, but more often, their trends diverge. The combination depends on the maturity of the business cycle. The most important point to remember is that deflationary forces predominate during the early stages of the cycle, whereas inflationary pressures come to the fore as the recovery matures. No business cycle ever repeats itself exactly, and the leads and lags between the peaks and troughs of the various financial markets differ from cycle to cycle. In spite of this drawback, the concept of the chronological development of the debt, equity, and commodity cycles works quite well in practice.
This chapter takes a description of this process one step further by pointing out that specific industry groups are sensitive to different types of economic conditions, in effect categorizing them according to their sensitivity to deflationary or inflationary forces, i.e., leading or lagging characteristics. Since the cycle itself is continually moving from deflationary to inflationary conditions and back again, it follows that the various industry groups also undergo a rotation. Unfortunately, this categorization is far from an exact process. First, many industries do not conveniently fall into an inflationary or a deflationary category. Second, equities rise and fall in reaction to the outlook for profits and, also, what is more important, in response to investor attitudes to those profits. Because interest rates are a significant, but not necessarily dominant, influence on the profits of interest-sensitive stocks, it follows that the price performance of certain interest-sensitive issues may, from time to time, become unlatched from or independent of the price movements in the credit markets. For example, savings and loan stocks declined in 1989 because of a financial crisis in the industry. Normally, they would have been expected to rise because interest rates fell during most of that year.
In spite of such drawbacks, the theory of sector rotation serves two useful functions. First, it can provide a framework within which to assess the maturity of a primary trend. For example, if there is technical evidence that the stock market is deeply oversold when the primary trend signals a reversal from bearish to bullish, it would be very useful to know that some of the groups that normally lead market turns have failed to confirm new lows made by the market averages or have established an uptrend in relative strength. On the other hand, in a situation in which the technical picture is indicating the possibility of a market top, it would be helpful to know that leading industry groups had made their highs some weeks or months earlier, and that stronger relative performance was concentrated in industry groups that typically lag the stock market cycle.
Second, the sector rotation theory is helpful in determining which groups, and, therefore, which stocks, should be purchased or pared back. This aspect is discussed in greater detail in Chapter 32.
The comments in this chapter refer to the U.S. stock market, but the concept of sector rotation can be extended in principle to other stock markets. Every country experiences business cycles, and there is no reason why Italian or Japanese utilities should not respond to changes in interest rates just as U.S. equities do. Indeed, it is possible to take this concept one step further by saying that markets heavily weighted to the resource area, such as Canada, Australia, and South Africa, ought to perform best at the tail end of the global economic cycle, and in most cases, they do.
Major Technical Principle A bull market is an extended period, usually lasting between 9 months and 2 years, in which most stocks move up most of the time. A bear market is an extended period between 9 months and 2 years in which most stocks decline most of the time.
The Concept of Sector Rotation
In Chapter 2 we established the fact that the stock market in the form of the S&P Composite discounts the economy and peaks and troughs with the economic growth path. That’s the theory, but the reality is closer to that set out in Chart 22.1. The lower panel features my Master Economic Indicator, which is constructed from the momentum of several forward-looking economic indicators. Equity market lows are consistently identified with the growth path of the economy bottoming out. Most tops are as well, but those at 2000 and 2007 were preceded by a series of declining peaks in the economy’s growth path. Nevertheless, the chart’s message is that there is a definite relationship between equity prices and business activity in most cycles.
CHART 22.1 S&P versus the Master Economic Indicator, 1966–2010
Since the economy can be divided into sectors that experience a chronological sequence as the cycle unfolds, it follows that equity sectors should, like the S&P, discount their sectors of th
e economy in a rotational way. Chart 22.2, for example, compares the S&P Homebuilders Index with national housing starts data.
CHART 22.2 Homebuilders versus Housing Starts, 1966–2012
There is not a lot that can be gained from such a raw comparison. However, Chart 22.3 shows the Know Sure Thing (KST) for both series. The dashed line reflects the homebuilders, and the solid one housing starts. There is no question that the dashed homebuilder line leads the solid housing start momentum line. The relationship is not exact and, of course, the magnitude and lead times vary from cycle to cycle. However, there can be no disputing the fact that homebuilding stocks lead their industry. The same principle can be applied to other stock groups and industries, and the result is sector rotation. In this scheme of things, the S&P Composite or some other aggregate market measure represents a coincident indicator for the stock market as a whole, just as gross domestic product (GDP) is for the economy. If the S&P is a coincident indicator, it follows that there are sectors that typically lead it and those that bring up the rear.
CHART 22.3 Homebuilder Momentum versus Housing Start Momentum, 1989–2012
For example, Chart 22.4 shows a key intermarket relationship—that between brokers and the stock market. It is based on the idea that the profits of these companies expand as the market rises. That happens because rising prices mean more customer profits and when customers are making money, they trade more, thereby generating greater commissions. Higher prices also attract companies planning on going public, so the number of underwritings rise along with the fees they generate and so forth. Since brokerage stocks anticipate brokerage profits and typically rise and fall with market prices, it follows that brokerage stocks have a tendency to lead the overall market.
Technical Analysis Explained Page 31
