Today, a paper plus a rant on some pet peeves. [It’s my blog after all. Ho-ho-ho]
The human brain was designed to get us safely around the Serengeti, not stock markets. Some of the wisest and most enduring advice for investors then urges us to overcome amygdalas and let prefrontal cortexs do the driving. In recent years much of this advice has crystallized itself into the pseudo science of ‘behaviorism’ [although Chazza Munger and Wazza Buffet were applying many of these rules long before Kahneman, Knetsch and Thaler began trying to get coffee mugs back].
Today most investors are, or should be, familiar with the concepts of endowment effect (the coffee mugs above, http://en.wikipedia.org/wiki/Endowment_effect for more if this is new to you), anchoring, regret, loss aversion and etcetera and do what they can to correct such biases. One Serengeti-amygdala habit though persists unchecked among many otherwise savvy investors; the temptation to see patterns where probably none exist. Our ability to make the intuitive leap from rustling grass to lion persists and in many aspects of daily life, such as driving and avoiding dog poo, this prehistoric tic continues to serve us well.
Where it’s of no use and downright dangerous though is when it comes to investing. Before I get to my main point I want to digress briefly on the issue of our increased tolerance for BS. In recent years we’ve accepted gaming in favor of gambling, adult for pornography, correction for fall, Asian (if you’re British) for Indian and so on. One particularly irritating example of this creeping eupha-misery regularly occurs in my gym. When machines are out of action a sign appears saying ‘Under Maintenance’. Hmm.. Apart from it’s capacity to annoy, this trend I believe, ranges from mildly to very corrosive in terms of how acceptable norms are manipulated. Gambling, for example, is bad plain and simple; but ‘gaming’? Now that sounds like it could be alright, no?
To get now to my main point, and this Sunday’s paper. Serengeti behavior pattern recognition and over-rapid conclusion formation isn’t a problem if you sense a lion but are wrong; when the same reflex gets involved in investment decisions though it’s a sure sure guide to sub-optimal ones. Yet a flourishing and pervasive sub-industry exists to encourage just such behavior.
If we called this process pattern-recognition bias it’d walk on punier legs then the enduring ones its widely accepted euphemism does; I’m referring to ‘technical’ analysis. When I began my career, because data and computers were scarce and expensive, the tribe that followed this pattern recognition false-God did so drawing pictures by hand and were referred to, mostly pejoratively, as chartists. Then, along came the Bloomberg and cheap computers, and with that technical analysis for all. Doesn’t that sound like something you’d want though? Technical, analysis. Serious and much better than the old non-technical analysis from the paper and pencil crowd? Just one slight problem, NONE OF IT WORKS.
The paper I’m highlighting this week is from Professor Ben. R. Marshall who together with colleagues from the Massey University of New Zealand, Mr. Rochester H. Cahan and Mr. Javed M. Cahan,have written ‘Technical Analysis Around the World’. The Professor and his colleagues drilled into 49-MSCI country indexes using over 5, 000 trading ‘rules’ to see if they work. The rules are from four mainstream technical schools; Filter Rules, Moving Average Rules, Support and Resistance Rules and Channel Breakouts. The Professor is generous in conceding in some markets, especially developing ones, a few of these things appear to work. Overall however, especially after the common bias caused by ‘data-snooping’ is removed, they’re a bust.
So, next time you’re recommended something because moving averages have crossed or it’s ‘broken-out!’, be reminded, you’re listening to a fool, a charlatan, or some combination of both; and above all, make sure it’s not yourself you’re listening to.
Happy Sunday and Merry Christmas all!
[Professor Marshall’s full paper, to which I’ve done poor justice with my brief summary above, can be found via this link http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1181367]