High-Yield versus Government Bonds
Another confidence relationship can be obtained by comparing the high-yield IBOXX Corporate High Yield Bond Fund (symbol HYG) to the Barclays 20-Year Trust, a government-bond ETF (symbol TLT). A rise in the ratio means that the prices of low-quality bonds are outperforming the relatively safe government sector. Such action indicates growing confidence by bond investors. Charts 28.7 and 28.8 compare the ratio to the ups and downs in the S&P Composite.
CHART 28.7 S&P ETF, 2007–2010, and a High Yield/Government Bond Ratio
CHART 28.8 S&P ETF, 2010–2012, and a High Yield/Government Bond Ratio
Most of the time, both series are moving in the same direction. It is when they are not that a subtle warning of a change in trend is given. Usually, these prove to be negative indications, but the positive divergence in 2009 shows that it is not a one-way street. Once again, it is always useful to obtain some confirmation from the price as flagged by the various trendline combinations. Since the HYG has only been around since 2007, this relationship does not have a long-term track record. However, a similar relationship that has been around since the early twentieth century is the ratio between Moody’s BAA corporate bonds and the constant 20-year maturity of U.S. government bond.
Chart 28.9 shows the actual ratio, with the arrows flagging extreme point reversals.
CHART 28.9 S&P Composite, 1923–2012, and a Government/Corporate BAA Ratio
Solid arrows indicate bullish signals, dashed ones bearish indications, and dotted ones failed signals. The chart clearly indicates that there is a definite relationship between bond market sentiment and equity prices. The trick is knowing where the extreme points of the ratio are. But how do we know they won’t become more extreme? A useful answer is to take the 12-month rate of change (ROC) of the ratio and set up some overbought and oversold zones at +15 and –15. Reversals that take place from a position beyond these levels then serve as our mechanism of primary-trend equity reversals. This is shown in Chart 28.10, where the extreme movements of the 1920s and the post-2007 period have been excluded so that the normal range can be appreciated. Once again, the two dotted lines indicate failed signals.
CHART 28.10 S&P Composite, 1940–2007, and Government/Corporate BAA Momentum
Chart 28.11 shows the post-2007 period that involved the 2008 financial crisis. These swings were enormous and certainly reflected the dramatic swings in investor confidence during these troubled times.
CHART 28.11 S&P Composite, 1997–2007, and Government/Corporate BAA Momentum
Finally, Chart 28.12 shows the ratio on a weekly basis, where divergences show up every few months or so. Also, changes in direction of the intermediate KST have provided useful signals of smaller trend reversals. I should add that prior to 2007, this relationship did not work as well as it has since, so it is possible that the 6 years of price behavior shown in this chart could turn out to be an aberration. Historical data for these and other bond series can be downloaded from the Federal Reserve website under “H.15 Selected Interest Rates Download.”
CHART 28.12 S&P Composite, 2006–2012, and Two Indicators
Using Brokers as Market Leaders
The market as a whole discounts the economy, but brokerage stocks, which obtain their profits from market conditions, have a tendency to lead the market in both directions. For example, during bull markets, volume and therefore brokerage commissions tend to increase because traders and investors find it easier to take profits than losses. When these folks are making money, they also tend to make faster, more careless decisions. That also inflates trading. By the same token, more companies go public during the course of a bull market because they obtain a higher price for their stock. Underwriting fees can represent a large share of brokerage income. Finally, interest rates have a tendency to lead the stock market. That means that brokerage carrying costs decline at the start of an equity bull market and begin to rise prior to its ending. The bottom line is that a rising market means greater profits for brokerage companies and investment banks, and a falling market less so.
Each sector discounts or anticipates developments in its sector of the economy, so the leading role often played by the price of brokerages is no exception to this rule. There are two principal ways in which it is possible to follow these stocks. The first is the Amex Brokerage Index (symbol XBD) and the second is the Dow Jones Broker Dealer, or more practically through the ETF that uses it as a tracking index, the IAI.
There are several ways in which the broker/market relationship can be analyzed. The first arises from positive and negative divergences. Chart 28.13 compares the S&P to the XBD.
CHART 28.13 S&P Composite, 1981–2012, and the Amex Brokers Index
The arrows show that brokers typically top out ahead of the overall market at primary-trend peaks. The leads are not constant, of course, but change from cycle to cycle. The width of the arrows roughly flag the lead times, and there does not appear to be a connection between the size of the divergence and that of the ensuing decline. For example, the divergence that took place in the 1989–1990 period was relatively large but the decline was fairly truncated. This compares to the devastating 2007–2009 decline that followed the very small 2007 negative divergence. Positive divergences, where the brokers fail to confirm new equity market lows, exist but are far more infrequent than negative ones. Chart 28.14 compares brokerage action to the market again, but this time, it reflects their relative action.
CHART 28.14 S&P Composite, 1978–2012, and the Amex Brokers Index Relative Action
The first thing to notice is the failure of brokers to identify the 1981 peak since the top in the RS line developed as the bear market was in its final stages. Other than that, the RS action, judging by the width of the arrows, provides a longer-term warning of tops than the absolute price data in the previous chart.
Chart 28.15 compares the Dow Jones Broker Dealer ETF, the IAI, to the S&P. Here you can see the negative 2007 divergence and the 2008–2009 positive one. Also featured in the chart is the July/August 2010 divergence, as flagged by the backward-pointing arrow on the IAI. This was an example of where the ETF lagged the market, thereby creating a disagreement. In such cases, price trumps everything, which means that when such disagreements develop, they are cancelled in the event that both series reverse their prevailing trend. In this case, it was a downtrend which was canceled by two upside trendline breaks.
CHART 28.15 S&P Composite, 2007–2011, and the Dow Jones Broker ETF
The brokerage/market relationship is far from perfect, but it often gives a hint as to whether investors and traders are optimistic or pessimistic about the stock market’s future performance.
Using Inflation-Protected versus Regular Bonds as a Commodity Barometer
Another useful relationship is to compare the prices of inflation-protected bonds (Barclays TIPS ETF) to the Barclays 20-Year Trust, or the TIP to the TLT. If the ratio is rising, it means that investors are favoring bonds protected by inflation over those that are not. If that is the case, it should mean that bond investors are anticipating inflation. Were they expecting deflation, the ratio would decline in view of the fact that the noninflation-protected instruments were outperforming their inflation-protected counterparts. In this respect, Chart 28.16 compares the ratio to the performance of the CRB Spot Raw Industrial commodity index. Prior to 2006, there was not much in the way of a relationship as commodities rallied and the ratio fell. However, at the start of that year both series violated trendlines and signaled higher commodity prices. Later in 2006 they started to move in similar directions, though the magnitude of each one was different as the more volatile commodity measure grabbed the majority of the price moves in both directions.
CHART 28.16 CRB Spot Raw Industrials, 2005–2012, and the TIP/TLT Ratio
The joint trendline breaks indicate signals for both trends reversing, and the two arrows indicate the confirmed negative divergence that developed in 2008. The more recent convergence of
the price action can be better appreciated from Chart 28.17, which compares the intermediate KST for the two series.
CHART 28.17 Commodity Momentum versus the TIP/TLT Momentum, 2005–2012
It would be convenient to be able to conclude that one series is a consistent leader, but that is not the case, as the solid arrows show ratio leadership and the dashed ones when the commodity KST turned first. Turning points with no arrows indicate a coincidental relationship. Starting in 2001 there are many periods where it is not possible to distinguish between the two series because their trajectories are so similar. When you consider that the ratio contains such completely different data from the commodity index, the recent similarities between the two series serve as a sharp reminder that commodity and bond market participants are clearly on the same page.
Since the history of this relationship has only been available for a few years, it would be incorrect to place a great deal of weight on these conclusions. However, the fact that the connection between the ratio and commodity prices appears to be growing suggests that this form of analysis should be closely monitored as it appears to be quite promising.
Summary
1. Confidence ratios, such as the relative action of defensive stocks or quality bond market spreads, typically turn ahead of market averages such as the S&P Composite.
2. They can be used to trigger buy and sell signals through trendline analysis or momentum relationships.
3. Brokers have a tendency to lead the stock market at tops, less so at bottoms. Useful buy and sell signals can be derived from joint trendline violations.
4. In recent years there has been a strong relationship between the trend of commodity prices and the ratio between inflation-protected and regular bonds.
29 THE IMPORTANCE OF SENTIMENT
I find more and more that it is well to be on the right side of the minority since it is always the more intelligent.
—Goethe
During primary bull and bear markets, the psychology of all investors moves from pessimism and fear to hope, overconfidence, and greed. For the majority, the feeling of confidence is built up over a period of rising prices so that optimism reaches its peak around the same point that the market is also reaching its high. Conversely, the majority is most pessimistic at market bottoms, which is precisely the point when it should be buying. These observations are as valid for intermediate-term peaks and troughs as they are for primary ones. The difference is normally of degree. At an intermediate-term low, for example, significant problems are perceived, but at a primary market low, they often seem insurmountable. In some respects, the worse the problems, the more significant the bottom.
The better-informed market participants, such as insiders and stock exchange members, tend to act in a manner contrary to that of the majority by selling at market tops and buying at market bottoms. Both groups go through a complete cycle of emotions, but in completely opposite phases. This is not to suggest that members of the public are always wrong at major market turns and that professionals are always correct; rather, the implication is that, in aggregate, the opinions of these groups are usually in direct conflict.
Historical data are available on many market participants, making it possible to derive parameters that indicate when a particular group has moved to an extreme historically associated with a major market turning point.
Unfortunately, there are several indexes that worked well prior to the 1980s but have been partially distorted because of the advent of listed options trading in 1973 and the introduction of stock index futures in 1982. The reason is that the purchase and sale of options and index futures substitute for short selling and other speculative activities that had been used as a basis for the construction of sentiment indicators.
Generally speaking, long-term data relating to market participants that have not been unduly affected by options trading before the early 1970s are limited. As with any data series of limited duration, a greater degree of caution should be exercised in its interpretation. Since a description of technical analysis would not be complete without some reference to investor sentiment, some of the more reliable indicators are considered here. Use of three or four indexes that measure sentiment is useful from the point of view of assessing the majority view, from which a contrary opinion can be taken.
Momentum as a Substitute for Sentiment
Individual stocks and many markets do not have published sentiment data from which indicators can be derived. In such instances, it is possible to substitute oscillators since there is a close correlation between overbought conditions and those of excessive bullishness and vice versa.
In this respect, the bottom panel of Chart 29.1 shows two momentum series.
CHART 29.1 S&P Composite, 2009–2012, Comparing Momentum for Price and Sentiment
Both are calculated by dividing the weekly close by a 13-week moving average (MA). The solid line is derived from the S&P Composite and the dashed one from the weekly percentage of bearish letter market advisors as published by Investors Intelligence.com. This latter series has been plotted inversely to correspond with the direction of equity market price movements. Thus, one is a momentum indicator derived directly from fluctuations in the S&P Composite and the other from sentiment data, which are, statistically speaking, completely independent. It is fairly evident that there is an extremely close correlation between them. The two trajectories are not identical, of course, but are close enough to prove the point that rising prices attract fewer bears and vice versa. We can also see a similar relationship in the bond market. Chart 29.2, for instance, compares a 10-week MA of a 14-week relative strength indicator (RSI) for bullish bond market traders as published by Market Vane against a 10-week MA for a 14-week RSI for the inversely plotted, government constant 20-year maturity bond yield. Again, the similarities between them are very close indeed.
CHART 29.2 Government 20-Year Yield (Inverted), 2003–2012, Comparing Momentum for Price and Sentiment
The fact that sentiment and momentum indicators are closely related should come as no surprise because rising prices attract more bulls and falling ones more bears. I am not suggesting that every sentiment indicator and oscillator have this close relationship. However, it does point up the fact that if sentiment indicators are not available, momentum series can be useful substitutes.
Insider Trading
Stockholders who hold in excess of 5 percent of the total voting stock of a company and corporate officers or other employees who have access to important corporate information are required to file with the Securities and Exchange Commission (SEC) any purchases or sales within 10 days. As a group, these “insiders” are generally correct in their decisions, having a tendency to sell proportionately more stock as the market rises, and vice versa. An 8-week MA of the weekly insider sell/buy ratio is shown at the bottom of Chart 29.3. The chart shows that as prices work their way higher, insiders accelerate their sales as a percentage of purchases. Market peaks are signaled when the ratio rises for a period of a few months or more and then reverses trend. In this respect, a rise above the 70 percent level and a subsequent reversal in the direction of the index are sufficient under normal circumstances to induce a decline.
CHART 29.3 DJIA 1978–2001 Specialists and NYSE Total Short Position.
At market lows, the 60 percent level appears to offer the best warning of an impending advance. If the index either falls below the 60 percent level and then rises above it, or even just declines to briefly touch it, as in early 1978 or March 1980, a rally usually results.
Advisory Service Sentiment
Since 1963, Investors Intelligence (www.investorsintelligence.com) has been compiling data on the opinions of publishers of market letters. It might be expected that this group would be well informed and would offer advice of a contrary nature by recommending acquisition of equities at market bottoms and offering selling advice at market tops. The evidence suggests that the advisory services in aggregate act in a manner completely opposite
to that of the majority and, therefore, represent a good proxy for an “anti-majority” opinion.
The most popular way to display this data is to compare the number of bulls or bears each week to the S&P Composite or some other average. That approach, though, does not take into consideration those looking for a correction. Chart 29.4 attempts to get around this problem by calculating the number of bulls minus the number of advisors who are bearish, i.e., the plurality of those with a firm opinion on the direction of the main trend.
Technical Analysis Explained Page 42
