One of the most delightful sights in nature is, on a late summer evening, to watch a flock of swallows (or other birds) rise up and then dance across the skies, wheeling back, forth, up and down in an exhibition of sheer exuberance. We marvel that there is no apparent director or choreographer of this performance but instead the fantastic ever-changing shapes are created when each individual bird changes its direction and orientation in reaction to movements by its near neighbours. A slight change in direction by one bird causes a ripple effect across the flock, flexing the shape and flight path of the entire flock.
After several minutes of frantic flight, the flock then comes to rest for another night. This , on the face of it, is an incredibly wasteful activity, feeding all day just to expend all that energy for fun (?). The serious purpose, of course, is to train new chicks in how to fly as a flock in preparation for the long and dangerous migration (north or south) to their winter-feeding grounds. Flocks of birds are incredible 'machines' for undertaking very long journeys, until they get sucked into jet engines that is.
But what has bird watching to do with Risk Management?
The Efficient Market Hypothesis (EMH) has been the dominant paradigm in global financial markets for the last two decades. EMH basically states that 'strongly efficient' markets are good at turning new information into changes in asset prices. If a piece of new information enters an efficient market, it ripples through the market so quickly that no one (or only a few) can take advantage of it. EMH, however, does not measure the efficiency of information dissemination directly but looks at its impact; postulating that the more 'random' price changes are, the better a market is at absorbing and propagating information.
Of course, an alternative explanation to such an inference might be that investors behave as a 'herd', as suggested by theories of Behavioral Finance, described by Nobel Laureate Daniel Kahneman. In this theory, investors are like a flock of birds or herd of cattle, waltzing across the markets driven purely by the behaviour of the group as a whole, rather than any underlying investment logic.
The one 'prediction' of EMH, that no investor can make better returns that anyone else in a strongly efficient market, is like saying that a contrarian bird, who leaves ahead of, or behind, the main migrating flock may not make it to the breeding grounds, but on the other hand may do so. There is a flaw in the EMH logic in that if someone found a foolproof way of beating the market, they would not tell anyone about it (remember it was not the efficient market that caught Bernie Madoff!)
One of the problems with EMH is its use of the word 'efficient', which is a 'good' word and hence a concept that all sensible investors would want to embrace. The theory would not have gained such support if, for example, it had been called the 'Gossipy Market Hypothesis', which is equally descriptive. A bigger problem is the unjustified inferences that have been drawn from the hypothesis. Because EMH has demonstrated that certain markets have exhibited a degree of 'efficiency' over a period of time (in the past), some people have, often for political purposes, inferred that ALL markets by definition are ALWAYS efficient.
This of course defies logic. All new markets experience teething problems (as for example the Interest Rate Swaps markets of the 1980s), and some markets are hatched and then die before maturity (remember Enron and the Broadband market). Markets are, by nature, inefficient until they are not; it is the classic 'learning curve' in action. Markets can also become suddenly and dramatically inefficient (witness the 1987 stock market crash). In short, inefficiency, not efficiency, is the norm?
With Behavioral Finance, a researcher can follow the traditional scientific method, i.e.: propose a hypothesis; construct and conduct an experiment to test the hypothesis; and then draw general inferences about the hypothesis from experimental results (notwithstanding the tricky problem of scaling up a small experiment to real life situations).
On the other hand, it would be unethical to use such scientific methods to test the Efficient Market Hypothesis, because we are dealing with real money belonging to real people.
But that is exactly what investment banks did during the sub-prime CDO market!
Only created in 2004, the sub prime CDO market had, by 2007, become a fully-fledged global market with all the attendant paraphernalia, of ratings and market indices and, of course, huge bonuses for the market participants.
With hindsight, we have found out that this market was far from efficient, as vital information about the actual performance of the underlying sub prime mortgages did not flow through to investors. Vital data got clogged up in key market information propagators such as rating agencies and securities firms. All the time, investors were behaving like a herd blindly buying and selling securities they did not understand, merely because others were doing it. There was perceived 'safety in numbers' and no one wanted to be last onto the gravy train.
For risk managers this is real dilemma as one of the key judgments they must make, is 'how good is the information underlying the prices of the assets that the firm is holding?' The more uncertainty about the underlying information, the more risk the firm is running and the higher return the firm must obtain. Of course, a market trader is going to say something like, 'look at the market, everyone is trading at that price, it must be right'. Wrong! Investors may be behaving like a 'herd' and just going with the flow. Traders love investors who herd; bonused by volume and volatile they make money either way the market moves, just as long as it moves.
Risk managers must become more skeptical and assume that markets are inefficient unless proven otherwise. This is turning the Efficient Market Hypothesis on its head with inefficiency being the default presumption.
A new 'Inefficient Market Hypothesis' might state that:
Markets are inefficient (at turning information into price changes) unless and until the following conditions are met: (1) mechanisms exist to propagate new information quickly throughout the market; (2) mechanisms exist to correct and remove 'bad' information in the market; and (3) investors exhibit 'herd behavior' (forget 'rational investors'). Markets are dynamic and can subsequently revert to inefficiency if any of the conditions above fail to continue to be met.
To push an admittedly silly analogy to its silliest. The global economy is the airliner; the financial markets the jet engine; and the CDO sub-prime market the bird strike that brought the airliner down. We spend billions on ensuring that individual passengers getting onto an airplane cannot cause it to crash, but much less on ensuring that birds don't do what comes naturally and fly into the engine. Like those introduced at airports, maybe risk managers should be the 'hawks' that will stop bad markets building up before they become a danger to us all?
