The U.S. stock market began the New Year where it left off in 2010; prices continued their upward march to fresh post-crisis highs and not surprisingly, the perennial bulls on Wall Street believe further gains are in store. Fundamental analysts are almost universally of a bullish persuasion and have pencilled in economic growth of up to four per cent for the current year, which is expected to generate a double-digit increase in corporate earnings and a commensurate advance in stock prices.
The bulls may well prove right, but their stock market targets are proffered as near-certainty and fail in their entirety to recognise the fact that the variability of potential outcomes remains extraordinarily wide. Meanwhile, their esteemed colleagues that practice the art of technical analysis note that the stock market generated a bearish signal known as the Hindenburg Omen on December 14th, a warning of impending trouble that was confirmed the following day.
The Hindenburg Omen is a technical indicator that is believed by many analysts to portend a stock market crash. It was developed a number of years ago by Jim Miekka, a blind mathematician, and was subsequently dubbed the Hindenburg Omen by Kennedy Gammage, a technical analyst and newsletter editor, in 2005, after the ill-fated passenger airship that crashed and burned in 1937, killing 36 people.
The Omen was preceded by, and appears to be based on, the High Low Logic Index (HLLI) developed by Norman G. Fosback, editor of Fosback’s Fund Forecaster, in the 1970s. Fosback explained in his classic textbook Stock Market Logic that:
The rationale behind the (High Low) Logic Index is simple. Under normal conditions, either a substantial number of stocks establish new annual highs or a large number set new lows – but not both ... When the Index attains a high level, it indicates that the market is undergoing a period of extreme divergence ... Such divergence is not usually conducive to future rising stock prices.
Miekka’s technical indicator requires a number of conditions to be satisfied for a stock market crash to be deemed imminent. First, the daily number of new 52-week highs and new 52-week lows on the New York Stock Exchange (NYSE) must both be sufficiently high such that the lesser of the two is greater than 2.2 per cent of the total number of issues traded on the NYSE that day. This requirement is clearly based on Fosback’s HLLI, though his indicator requires a far higher threshold of not less than five per cent of NYSE issues traded, and even though this is the only condition that must be satisfied for the HLLI to emit a bearish warning, the high cut-off points means it is extremely rare.
Second, the Hindenburg Omen can only be triggered in an advancing market and this condition is considered satisfied if the NYSE Index is higher than its level at any time during the previous ten weeks on the day the first condition is observed. Third, the McClellan Oscillator, an indicator of market breadth developed by Sherman and Marian McClellan in 1969, must be negative on the same day – the oscillator is measured as the (19-day exponential moving average of advances less declines) minus the (39-day exponential moving average of advances less declines), and a negative value indicates that fewer issues are participating in the market’s upward momentum.
The fourth condition requires that new 52-week highs on the NYSE are not more than twice the number of new 52-week lows, though the reverse is allowed given that it suggests faltering momentum. Finally, for the Hindenburg Omen to be considered ‘official’, the first four conditions must be satisfied on the same trading day again and within a 36-day period following the first warning.
The requirements for an official Hindenburg Omen would appear to be quite onerous, and given that the crash indicator was triggered in the middle of December, it would appear to suggest that caution is appropriate in today’s runaway market. However, it should be noted that the signal has been refined over and over again to fit the market’s historical gyrations, and thus, may be nothing more than a prime example of data-mining with little relevance for future market behaviour.
Investors will be quick to ask how the indicator has performed in practice and the results are instructive. There have been 39 previous Hindenburg Omens since 1985 and a crash warning has been generated before every substantial market decline registered during that period including 1987, 2000, 2001, 2002, 2007 and 2008. However, almost six in ten observations have been followed by a subsequent market drop of less than ten per cent and the median decline for all previous Omens is less than nine per cent. Only one in seven observations has occurred immediately before price declines of more than 20 per cent, the traditional definition of a bear market, which suggests the indicator is hardly deserving of its name.
The historical evidence indicates that the Hindenburg Omen may well portend a soft January, but the indicator is certainly not sufficient of itself to prepare for an imminent crash and an end to the 21-month long cyclical bull market. An examination of sentiment indicators paints a similar picture; investors are excessively bullish and a pullback is warranted to restore the market’s overall health. A normal market correction would appear to be in the offing and though this columnist believes the secular bear market is not over and will resume at some point, preparation for that event should not be based on some dubious indicator, but a close examination of the underlying fundamentals.
Disclosure: I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.
