Technical analysis and market psychology
This article examines the relevance of technical analysis to today’s financial markets. Its genesis springs from two branches, Dow theory and the Elliott wave principal.
Originally, Dow theory sprang from an attempt to discern the state of the economy through the construction of representative stock indices for railroads and industrials. Elliott came into existence in the 1930s, but only really became widely followed in the late 1980s.
Both Dow and Elliott were intended to help investors make profits by reading equity stock indices, though their use has spread to other financial assets. The validity of this wider use is questioned.
And finally, Elliott wave theory has some disturbing implications for current equity markets. But how valid is Elliott or is it little more than a financial ouija board? And what has technical analysis to say about cryptocurrencies?
There are two basic branches of technical analysis, that is to say the art of financial prediction based on chart patterns and various derived indicators. Today, chartists have their own favourite indicators, the enormous variety of which only really became possible with computers. But the origin of technical analysis can be traced back to the Dow theory by Charles Dow —the founder of the Dow-Jones financial news service, inventor of the Dow stock market averages and before 1902 editor of the Wall Street Journal.
Dow was primarily interested not in using his findings for stock market speculation, but more as a barometer to gauge general business trends. Before they were quantified by Dow’s stock averages, businessmen, speculators, and even the British economist Arthur Pigou were aware of the interesting phenomenon that stocks tended to rise and fall together, and that the few that bucked this underlying trend were unlikely to do so for very long.
So far as we can tell, they missed the reason for this cyclical behaviour, which had its foundation in the cycle of bank credit expansion and contraction. At the time we are discussing, money was meant to be sound, being anchored to gold standards, and the fact that banks created credit out of thin air was poorly understood. Figure 1 shows the notable crises of the nineteenth century, and their regularity of ten or eleven years on average is striking.
Later research would have answered Dow’s curiosity. We now know that the root cause of business trends was —and still is —money; occasionally through new gold discoveries, but regularly through the expansion and contraction of bank credit which plagues us to this day.
Not only does banking behaviour create economic cycles, but it also drives stock markets, and it was Charles Dow who gave us the tools to judge the effects.
Basic Dow theory
Instead of pursuing business cycles, Dow gave birth to a new art form, which can be best described as based on the recognition of investor psychology. However, it remains wholly subjective within its framework. If it was to be otherwise, all investors would make money consistently by trading substitutes for the averages and no one would have to engage in production of the goods we would wish to consume with our new-found wealth. Logic tells us that this can never be the case and experience also confirm that despite Dow theory and the whole field of what is now called technical analysis of stock trends, the majority of traders still lose money and investors still tend to buy high and sell low.
In 1897 Dow commenced his investigation by taking the average of 20 railroad stocks, being the dominant corporate enterprises at the time, to compute what we know today as the Dow Jones Transportation Average. He also calculated an industrial average comprised of twelve stocks, which was later expanded to 20 in 1916 and then 30 in October 1928. This latter average is now the bell-weather Dow Jones Industrial Average.
Dow’s germ of an idea was further developed by his successor at the Wall Street Journal, William Hamilton, into Dow theory as we now know it. As listed in Edwards & Magee’s Technical Analysis of Stock Trends, which is taken by most Dow theorists as the definitive work on the subject, there are some basic tenets that apply:
1. The averages discount everything, except “acts of God”, because they reflect the combined activities of thousands (now millions) of investors, many of which are skilled and well-informed.
2. The market follows three discernible trends: major trends that last for a year or more resulting in appreciation or depreciation of 20% or more, interrupted by secondary trends in contrary direction, and minor trends in either direction, or day to day fluctuations which are unimportant.
3. In a bull market, primary uptrends are usually comprised of three phases. The first is an accumulation phase following a bear market, where far-sighted investors are willing to pick up shares offered by discouraged and distressed sellers. At this early stage, financial reports are still bad, and the general public is getting out of stocks. The second phase usually accompanies the early stages of economic and business recovery and offers the skilled trader the best environment for profitable investment. The third phase sees the general public return to the market, encouraged by a combination of bullish news and a desire to not miss out on sure-fire opportunities.
4. Edwards & Magee described a bear market also consisting of three phases. Initially, in a distribution phase far-sighted investors sense things have gone too far and sell their stocks to the public, who still tend to be eager buyers, not realising that valuations have over-discounted evolving prospects. This is followed by a panic phase, as buyers begin to thin out and sellers become more aggressive. And lastly, there is a third phase, when the business news deteriorates rapidly, and inexperienced investors who held out through the panic phase lose all hope and finally accept their losses.
5. The two averages of rails (transportation) and industrials must confirm. In other words, if the rails fail to make new highs while the Industrials are achieving them, the industrials are not yet in a bull market. Equally, if the industrials are still in a bull market while transportation fails to confirm, a warning signal has been given. Edwards & Magee fail to explain this phenomenon adequately, other than to opine that those investors who ignore the confirmation rule usually live to regret doing so. But we can suggest that transportation stocks are the first to detect an end to a business slump with volumes and freight rates stabilising before turning higher. Equally, they are early indicators of an economy losing the momentum of expanding production as freight rates begin to slip and trans-shipment volumes begin to decline.
6. Trading volume goes with the trend. In other words, as a trend is more progressively established, volumes increase, but on countertrends they diminish.While these tenets have stood the test of time, there is still room for error. Dow theory is often seen by modern analysts as too blunt an instrument. But having established that there is a rhythm of bull and bear markets, further analysis of price patterns have been developed on the back of Dow theory. First came patterns such as head and shoulders, signifying reversals. These were followed by continuation patterns, such as flags and pennants. And as computing power became available, moving averages followed by various other complex indicators were developed. Furthermore, technical analysis expanded from indices to stocks to commodities, currencies and bonds, where similar psychological factors driving prices were perceived by practitioners.
But perhaps the ultimate development of Dow theory was the consequence of independent observations by Ralph Elliott in the mid-1930s. This is our next topic in the subject of technical analysis and market psychology.
Elliott waves and Dow theory
Though it appears to have been developed independently, Elliott wave theory is essentially a further elaboration of Dow theory, aimed at quantifying the bull and bear trends identified by Dow.
While Dow theory describes a bull market as consisting of three phases, Elliott counts the intervening corrections as well to give five waves of lesser degree. His description of a bear market differs from Dow, by claiming it should be regarded as breaking down into three waves, two downward interrupted by a corrective phase. Elliott goes even further, by separating all waves of different degrees into impulses and corrections, with impulses comprised of five waves of lesser degree and corrections generally comprised of three, but where they overlap, they can be more complex.
An important innovation was the introduction of Fibonacci relationships and ratios. This number series starts at 1, with the following number comprised of the sum of the two that precede it. The series progresses as follows:
1, 1, 2, 3, 5, 8, 13, 21, 34, 55, 89, 144, etc.
From this series, relationships between the numbers are derived in ratios which quickly settle down to 1.618 between a lower number and the next higher in the series, and 0.618 being the ratio between a lower number preceding the next higher in the series. The multiple between a number and the next but one higher is 2.618, and its reciprocal is 0.382. These ratios are used to predict subsequent waves based on the measurement of the extent and duration of previous waves.
Elliott is not confined to the primary, secondary and minor trends of Dow theory, but its structure is repeated to larger degrees of pattern and also to smaller. Timescales between a few minutes and even several centuries are deemed possible, labelled from minuette to grand super-cycle. Under this notation, it could be argued that a super-cycle commenced in 1932, following the Wall Street crash, and having run for 89 years (a Fibonacci number) is due to end in 2021 — this very year. Further confirmation is found from the 1987 crash, which marked the fourth (corrective) wave of cycle degree, and which happened 34 years ago —another Fibonacci number. And the peak of the dotcom bubble was 21 years ago, another Fibonacci number. Furthermore, the pre-Lehman crisis top was just over 13 years ago. You’ve guessed it — yet another Fibonacci number.
While this evidence from Elliott wave theory is that stock markets appear to be on the edge of a new devastating bear market, devastating because all the excesses since 1932 are due to be washed out of the financial system, it should be noted that playing around with Fibonacci numbers and ratios can yield radically alternative outcomes. Furthermore, the theory relies heavily on the analyst’s belief in it and there is insufficient theoretical explanation as to the overall validity of the method. But while the reasoning behind ratios requires an occult-like faith, Elliott’s development on trading patterns is a meaningful advance on Dow theory.
One of the benefits of such a radical approach is to encourage an investor’s imagination to consider market outcomes that are truly outlandish. For example, who, in 1974 when the Dow Jones Index was at 572, thought that it would subsequently increase not just fivefold to 3,000, but sixty times to the current 34,600?
Further descriptions of Elliot wave patterns and the use of Fibonacci ratios is beyond the scope of this article. But for those readers interested, probably the best book on the subject is Frost and Prechter’s Elliott Wave Principle first published in 1978.
Before moving on to more general comments about technical analysis, a brief mention of Gann theory is warranted. Or rather the ragbag of trading rules invented by William Gann, who was a successful stock trader between 1905—1950. Many of his “never failing rules”[i] were common sense, and applied to individual stocks, unlike Dow and Elliott wave theories intended for divining wider market trends. For the purpose of this article, which is not about individual stocks, Gann will be ignored.
The distortions of money
We would like to think that a rise in the indices by which stocks are measured is due entirely to human progress in manufacturing and the provision of services. But for this to be true requires money to be stable in its purchasing power, which has never been the case in the history of investment. Before 1971, when the Bretton Woods agreement was abandoned, the dollar was theoretically tied to gold at $35. So, it can be argued, the progress from the Dow’s 8 July 1932 low of 41.22 to the 900 level in 1971 — an increase of nearly 21 times was real. But during the same period, GDP, which is primarily a reflection of money creation, increased from $58.7bn to $1127.1bn —an increase of 18 times. So yes, the stock market outperformed the creation of broad money, but not by much spread over such a long time. Furthermore, the trend for redeployment of funds from activities on Main Street to Wall Street probably accounts for the rest.
Before the 1970s there were as always two sources of monetary expansion. The larger of the two is normally taken to be the expansion of bank credit. But from the dollar devaluation in January 1934 and the financing of the Second World War, to the provisions of dollar loans to Europeans and the subsequent war finance for Korea and Vietnam, these were identifiable instances of central bank and government funding by the creation of dollars on a massive scale. It was facilitated by the loosening of the gold standard under Bretton Woods and led to the failure of the London gold pool in the late 1960s, before all the dollar’s ties to gold were abandoned in 1971 under market pressure. The point is that monetary inflation fuelled the stock market from the depths of the depression to the final abandonment of any link to gold.
Since the 1971 abandonment of the Bretton Woods agreement, the monetary environment has been demonstrably more inflationary. M3 broad money supply has increased 32 times, while the Dow has risen 37 times – again a modest outperformance for the Dow spread thinly over fifty years and easily explained by the redeployment of funds to purely financial activities rather than being a reflection of genuine overall economic progress.
Thus adjusted, Charles Dow’s original quest to use an index of stocks to assess the state of industrial conditions makes a bit more sense. But the ups and downs of what his subsequent followers described as primary trends may have failed to identify business trends in the latter part of the twentieth century, where it would have succeeded in the nineteenth, illustrated in Figure 1 above. Following the 1930s, the bank failures every ten years or so that evidenced earlier credit cycles have been less reliable punctuators of economic activity, though there have been localised exceptions that escaped central bank rescues —most notably the Lehman crisis.
Applications extended to other financial assets
We have noted that both Dow and Elliott, the twin pillars of technical analysis, are flawed by ignoring changes in stock prices that come from the money side. For investors, this defect has been irrelevant because they are interested in maximising their financial wealth measured nominally in their currencies of account. Obviously, both technical methods add value, otherwise they would have been discarded long ago.
More questionable is the extension of these analyses beyond stock prices. There is a case, perhaps, for taking the evidence from Dow theory and discerning a cycle of interest rates and bond yields, because that fits in with the cycle of credit. But that has become less relevant as central banks have tightened control over interest rates. And here, Elliott is simply inappropriate because if it was, bonds would be in a permanent five-wave bull or bear market, an impossibility. With respect to commodities, Elliott competes with Kondratieff, whose followers claim they are collectively in a 50 to 55-year cycle of their own. But with the increasing dominance of financial derivatives over the physical, commodity prices have become more financialised and therefore influenced by followers of technical analysis.
Other than a similar confirmation bias of technical analysis being applied by large funds, currencies have a life of their own, where the actions of governments relative to one another are the principal influence. A build-up of speculative positions will have effects within that context, but that is not cyclical in nature. Any anyone who attempts an Elliott count on one in a currency pair ignores the fact that it cannot apply to the other, because the bullish currency cannot move in fives while the other bearish currency moves in threes.
It is by falling into this trap that many speculators in gold are ensnared. They persist in treating gold as if it were an equity, when it is priced on the markets either as a commodity or as a form of money, depending on the influences at the time.
The weight of money following technical analysis can have short-term technical effects on non-equity markets of all types. For this reason, a follower of Elliott wave theory can claim some success predicting gold prices, not because it is a valid transposition of method, but because a whole community of investors is influencing prices using charts. In recent years, this trend has increased through the spread of commodity ETFs and similar investment vehicles.
But perhaps one category that should fit Dow theory and Elliott wave principles is the new cryptocurrency boom. It is too early in their history to make this statement with confidence, but the same human factors and emotions that drive both Dow and Elliott are manifestly evident. The peaks of investor emotion seen at the end of a bull run have recently been observed in bitcoin, which at the time of writing has more than halved from a high of $65,000 to as little as $30,000 in only six weeks, despite continuing claims from many investment gurus that the rise in bitcoin prices has only just begun.
The timing of the cryptocurrency boom has coincided with the accelerated debasement of fiat currencies, prompting aficionados to claim bitcoin is the new money. What is ignored is a likelihood that bitcoin is the new Mississippi venture, the new South Sea bubble, marked, as Charles Mackay memorably described the stock market bubble phenomenon, by as little more than the madness of crowds.
Both Dow and Elliott wave theories presume free markets. In other words, market prices are not distorted by government intervention. But since the First World War this has increasingly not be the case. Monetary debasement commenced in earnest in the roaring twenties, not just in Europe but by the new Federal Reserve Board under the direction of Benjamin Strong. Government intervention escalated in the depression years. And following the Second World War, the suppression of interest rates began in earnest. Furthermore, the relationship between markets and business conditions changed, with individuals’ savings as the sole source of business investment being destroyed as a deliberate neo-Keynesian policy, particularly in the US and UK.
In the US, business and banking lobbying progressively displaced free markets, while in the UK there was widespread nationalisation following the Second World War, only partially reversed during the Thatcher years. And finally, the ending of the US Glass-Steagall Act in the 1980s gave US megabanks an additional licence to print money by combining commercial and investment banking, while London swapped self-regulation for that of the state.
As identifiers of investment trends, both Dow and Elliott have continued to be valuable to investors even though progressive state intervention has turned markets into tools of government. Despite its bewildering subjectivity, Elliott has perhaps accommodated state intervention better than Dow theory in its original form, insofar as Elliott admits to larger cycles of human behaviour than one confined to cycles of bank credit. According to an Elliott analysis, the current super-cycle started with the Dow industrial average bottoming on the 8 July 1932 at 41.22. That phase ended with the Second World War, following which a new expansionary phase commenced, which ended at about the time Bretton Woods ended in the early 1970s. The fourth wave of investor behaviour was driven by the consequences of the end of any gold standard and its replacement by pure dollar fiat currency as prices, and then interest rates, were driven into the low twenties during the 1970s. And then the final leg started, fuelled by the financialisation of the American and British economies.
Pedantic counters of Elliott waves will find much to dispute in this timing of five waves, three impulses separated by two consolidations. But that is to miss an important point. In the broadest, even philosophical sense, it has identified an 89-year cycle of human behaviour, a far larger cycle than that of Dow identified on the back of bank credit. And we must also consider the implications, that the correction can be expected to be significantly greater than that of a normal banking and financial crisis.
Frost & Prechter also postulated that there was an even larger cycle, a grand super-cycle which they say started in the late eighteenth century, though it must be admitted that the evidence reflected in stock prices is somewhere between non-existent and unreliable. But judging it by the evidence of economic history clarifies matters somewhat. There is no doubt that the coincidence of the Mississippi and South Sea bubbles in about 1715-20 was a major economic event, arguably being the top of a medieval grand super-cycle. The collapse led into the earliest stages of the European industrial revolution, ending in the second currency collapse suffered by the French in the 1780s, a revolution and the Napoleonic wars. The third broadly based wave was the second industrial revolution, taking the world up to the First World War. The fourth and corrective phase was the consequence of millions being killed by war and the flu pandemic, inflationary collapses in major European economies, and the Russian revolution — ending in the Wall Street crash. From there our current 89-year final cycle commenced.
There is a nice distinctiveness in it all. The last grand super-cycle ended with the collapse of two stock bubbles and the destruction of the French fiat currency instrumental to inflating the Mississippi bubble. Today, led by the Fed, major central banks are repeating the use of unlimited monetary inflation to puff up a similar bubble —but on a global scale. The lesson of all bubbles is they eventually pop, and if they are inflated by a fiat currency, the currency collapses as well. The only comfort is these events are 300 years apart. And that is not a Fibonacci number!
p.s. Three centuries is a Fibonacci number, if a century is taken as a unit of time.
[i] See WD Gann: 45 Years in Wall Street
Find more: Contributing Authors