Investor 'prediction addiction' now most relevant

SERIOUS MONEY:HUMAN BEINGS are hard-wired to detect patterns and identify causal relationships amidst the constant stream of new information. This behaviour can be traced to our ancestral past on the African savannah many millennia ago, where the ability to shape expectations from small samples of data enabled our hunter-gatherer ancestors to successfully forage for edible fruits and seeds, stalk prey, avoid predators, find shelter and seek mates.

Scott Huettel, a neuro-economist at Duke University, explains that, “The brain forms expectations about patterns because events in nature often do follow regular patterns: when lightning flashes, thunder follows. By rapidly identifying these regularities, the brain . . . can expect a reward even before it is delivered.”

The ability to anticipate outcomes from regular patterns undoubtedly helped our ancestors to flourish, but Huettel warns that “in our modern world, many events don’t follow the natural physical laws that our brains evolved to interpret”. The human brain is designed to conserve scarce neural resources, and so much so that it requires only a single confirmation to anticipate a recurring pattern. Huettel notes that, as a result, “The patterns our modern brains identify are often illusory . . .”

Pattern recognition and subsequent tactical buy or sell decisions are part and parcel of active investment management. Investors, however, often place too much emphasis on the recent past when forming expectations about the future. Top-down analysts make tactical calls based on recent economic data, while technical analysts divine the future on historical patterns in stock prices.

Investors’ “prediction addiction”, as the behaviour has been called by financial columnist Jason Zweig, is particularly relevant today. Economists are busy shaving their economic growth forecasts for both this calendar year and next, following a string of disappointing data that fell well short of expectations.

Meanwhile, technical analysts are arguing for a reduction in equity allocations, given price action in the major stock market averages that confirms a change in the underlying trend.

Due to recession fears, investors are in need of guidance that will preserve capital and/or yield profits, amid the uncertainty. Indeed, anticipating turning points in the business cycle and adjusting asset allocation accordingly is central to successful top-down investing.

Unfortunately, economists have a patchy forecasting record at best, having failed to anticipate every one of the last five recessions. Monthly publication Blue Chip Economic Indicatorsnoted in July 1990 that "the year-ago consensus forecast of a soft landing in 1990 remains intact" – but economic expansion peaked that very month.

More than a decade later, in March 2001, fewer than 5 per cent of economists anticipated there would be a recession that year, even though a downturn was set to begin just days later. More recently, during the spring of 2008, the calls for a soft landing were almost deafening, despite the fact the deepest recession since the 1930s was already under way.

Perhaps the study of historical price patterns performs better. After all, stock price data is not reported with a lag and, unlike economic data, is not subject to revisions that continue several quarters after the fact. As William Hamilton, the fourth editor of the Wall Street Journalwrote in his 1922 classic, The Stock Market Barometer– "The market represents everything everybody knows, hopes, believes, anticipates, with all that knowledge sifted down to . . . the bloodless verdict of the market place."

The study of historical price patterns suggests a further decline in the major market averages may lie in wait. The 18 per cent fall in stock prices from their recent late April peak resulted in a bearish “Death Cross” signal on August 12th, as the stock market’s 50-day moving average of closing prices dropped below its 200-day moving average.

The “Death Cross” is considered by technical analysts to be a portent of future weakness, but is the signal’s presumed ability to anticipate turning points and enhance investment performance supported by the historical record? To find out, the “Death Cross” and its converse, the “Golden Cross”, are employed as tactical sell and buy signals, respectively, for a simple long/short strategy and the investment results – excluding dividends – are compared to those generated from a straightforward buy-and-hold strategy.

The historical record shows that before the most recent “Death Cross” there had been 63 tactical signals since the summer of 1949 – 32 buy and 31 sell signals – which gives weight to the late Paul Samuelson’s criticism in 1966 that “The stock market has predicted nine out of the last five recessions.”

The buy and sell signals resulted in 33 winning trades and 30 losing trades, which is not much better than a coin toss. More importantly, the price return generated by the long/short strategy saw an initial $10,000 investment compound to $537,000 over the period, as against $775,000 for the buy-and-hold strategy.

The historical evidence suggests the “Death Cross” adds no value to the investment process. However, a more complete examination of its credentials reveals it subtracts from investment performance during secular bull markets, which are characterised by powerful up-trends with only the briefest of interruptions, while it adds to performance during secular bear markets, which are characterised by a protracted sideways pattern punctuated by violent downward price swings.

The bearish indicator provided ample warning to investors of impending danger, close to a market top in both the autumn of 2000 and the winter of 2007. Has the “Death Cross” sounded an early warning bell once again? Time will tell.

www.charliefell.com