On Economic Thought - 5

Last time, I asserted how financial markets and securities help cut out the tautological subjectivity of value. Well, they go further. The idea of 'rationality' is well defined. 'Efficiency of a market' can be tested sharply because the implication of an 'efficient' market is very objective and can be checked for against market prices. There is a tremendous amount of high quality data available. Finally, even information has some precise definitions in the field of finance.

Among the most celebrated and debated ideas in finance is the efficient market hypothesis (EMH). What does it say? Well simply, that financial markets are rational and hence stocks are fairly priced all the time. In case there is a move away from this ideal equilibrium, it adjusts back to the stable equilibrium state fast enough.

But isn't that absurd? To assume that investors are always rational? Well, that's the catch. The rationality of a market on the whole (or indeed, of the entire economy) has very little to do with the rationality of every individual participant. You see, for the market to be rational, it is not necessary for every, or indeed, even most investors to be rational. All that is needed is that
1) The 'irrationality' is randomly distributed across people and has a mean of zero.
2) There are a reasonable number of 'arbitrageurs' who have the capital and the freedom to make use of those opportunities in the market when stocks are incorrectly priced.

Thus, when Ravikiran asserts here and here that retail investors in India are stupid and hence the EMH may not hold for India, it makes me wonder if he understands the EMH at all. As long as the idiot retail investors in India (or elsewhere) are idiotic in their own randomly distributed ways, the EMH would still hold. The argument is simple - the overestimations of the positive idiots will cancel out the underestimations of the negative idiots. The only problem arises when idiocy is systemic - when idiots make mistakes that are very similar to each other.

Even if the mistakes are systemic, the EMH may not be affected. If there are some few rational people working for a few hedge funds or investment banks who spot these systemic mistakes, they will use their money to capitalize on it to make riskless profits, or arbitrage. These smart guys will make the EMH work by two means
1) The mispricing will correct immediately.
2) If the mispricing does not correct, the arbitrageurs will keep making money and over some time drive out the idiot investors who will be losing their shirts.

Thus, any reasonably mature financial market should be efficient. In this fully developed form, the EMH was among the technically soundest theories to have propped up. It ruled out the efficacy of either technical or fundamental analysis, and implied that your best bet in investing was to buy an index fund, or to passively replicate the benchmark index.

The theory was especially popular in the Chicago School tradition. It fit beautifully with the idea of rational expectations and other such tenets of neoclassical economics. Opposition to this theory has come from various quarters. One of them has been the value investors - the supremely successful stock pickers of the Warren Buffet mould. They rubbish the idea of the market always being fairly priced, buy when things are cheap, sell when they are expensive, and generally make millions doing this. However, it must be noted that they are only aghast at the claim that markets are always efficient. Because, this strategy of buying low and selling high makes money only if prices will someday be correctly priced in. In this manner, the value investor is only performing the role of the rational, efficiency-inducing arbitrageur of the EMH. The value investing school thus quarrels with the more extreme conclusions of the EMH, but has no strong theoretical model to oppose it.

The real challenge to the EMH has come from behavioural finance. Now Avataram probably believes that he has introduced the Indian blogosphere to the existence of Daniel Kahneman, but I stumbled upon these theories while preparing for my summer internship interviews last year. Behavioural finance hit the EMH on its two most important pillars - on the assumption of irrationality being random instead of systemic, and on the assumption of the possibility of profiteering from riskless arbitrage.

Multiple experiments have shown that people across levels of education make choices that are wholly inconsistent with the idea of risk-averse expected utility maximization. The Ellsberg paradox and the Allais paradox are two illustrations of that. More importantly, arbitrage is never truly riskless. There are limits to arbitrage, and the most important one is capital preservation. Arbitrageurs can be driven out of business if the mispricing that they are trying to take advantage of deepens and does not correct as quickly as they'd want to. Their trades will keep losing money, and they run the risk of getting fired for losing money or underperforming their benchmarks. About 50 years before behavioural finance became en vogue, this idea was captured beautifully by Keynes in the quote that I'm sure many have heard - "Markets can stay irrational longer than you can stay solvent". Indeed, if there is one adage that you should know and remember about the financial markets, it is this one. The most striking example of the risk of arbitrage preventing a rational outcome in the markets has been the celebrated case of Royal Dutch and Shell, where stocks of the exact same company split in a 60:40 ratio refused to trade in that ratio for elongated periods of time.

Research into behavioural finance and behavioural economics has over the years produced a rich body of evidence of the systemic cognitive biases that individuals have. The idea of the importance of market microstructre has also gained relevance. I have long believed that if mathematics is at the base of the natural sciences, and philosophy is at the base of the humanities, then psychology is definitely at the base of the social sciences. Since economics is a somwhat scientific social science, I should probably say that cognitive science is at the base of all economics. Some of the ideas presented by the behavioural economists are easily among the most compelling arguments on economic activity that I have come across. But more on that later.