You are here: Home » Opinion » Columns
Business Standard

Parthasarathi Shome: Financial market behaviour

Models are still restricted in their capability to predict market behaviour

Parthasarathi Shome 

A familiar issue is how to gain an edge over international financial markets to obviate or Should capital controls be imposed by countries or under multilateral understandings or agreements? Should there be an international financial regulatory agency to initiate action under stipulated conditions? Should financial transactions be taxed? Some countries are making forward-looking suggestions; some are resisting; others are being Humpty Dumpties.

To address price volatility, price determination needs to be understood first. Traditional financial economics literature says the outcome of the market is a “random walk” in prices, or, all available information is already reflected in the current price, so that any future price movements would only result from random exogenous shocks. However, information has always been scarce. Thus, even quite early, such premises were already challenged. Ratti and Shome (1977) demonstrated that, in the presence of uncertainty, the usual solution for an auctioneer, to search and find a set of relative prices through a tatonnement process, at which there is no excess demand in any market, is no longer achievable. Thus, exchanges are not perfect Walrasian auctions. More recent evidence by Lo and Mackinlay (1999) showed that stock market prices do not follow a random walk; there is volatility that cannot be explained as an outcome of exogenous or uncontrollable factors alone.

Related questions arise on whether traders are rational; whether it matters that the marketplace does not determine a price through an equilibrating tatonnement process; and whether volatility is a market failure that needs to be addressed. (Click here for Diagram)

Therefore, to develop a coherent policy in an international framework, we need to first understand the behaviour of players in financial markets.

Consider diagrams I and II. The dotted straight lines indicate the line of “fundamental values” or the true underlying long-term trend values of a stock or share. The lines in blue indicate price movements in the share market with speculators, while the lines in red indicate the price trajectory with no speculators.

The question is: will speculation bring prices towards, or take them away from, the fundamental values? The old algorithm was that “stabilisation is stabilising”. Thus, in diagram I, when prices are high, speculators consider it a good opportunity to sell, and do that. Consequently, the prices decrease, which results in speculation bringing prices towards fundamentals. So, speculation becomes stabilising.

Diagram II shows the opposite. When prices rise, operators buy in the hope that prices will rise even further. This type of “noise” trader is said to trade often, usually in a herd, who ignores market fundamentals. This is compounded by “technical or automated trading” that is software-driven and takes place on the basis of recent price and trade volume information rather than on any analysis of underlying economic data. The outcome is destabilising speculation — prices being pushed further away from fundamentals. The relative strengths of the two behaviours – stabilising and destabilising – determine the outcome of whether market activity leads prices towards or away from fundamental values.

Technical trading mimics destabilising speculation. This is because technical traders buy until a certain ceiling is reached when prices are rising, and sell when prices fall below a certain minimum level. Beinhocker (2007) has given an interesting example from Farmer et al (2004), on the impact of a bid-ask spread. They studied one trade in AstraZeneca, a pharmaceutical company. AstraZeneca’s “limit sell order” – offers to buy and sell that are conditional on a price – was set at £31.84. The next limit order was set at £32.30. When a small “buy order” of £16,000 came in, the asking price jumped from £31.84 to £32.30. This was an increase of 46 pence, which now represented the bid-ask spread; the share price moved up by 23 pence. This added £374 million to AstraZeneca’s market value, though there was no policy or performance change of AstraZeneca on that day. Thus, a £16,000 buy order had generated a £374 million valuation jump for the company, reflecting solely the way market price recording and clearing take place in technical trading.

To curb volatility, those who recommend controlling financial transactions through capital controls or by taxing capital flows claim that such instruments would curtail technical intra-day trading or short-term transactions; and, in turn, the market would move more towards fundamentals or the underlying long-term trend. The opposite argument is that the reduction in transactions could mean that, when people do trade, they trade in larger amounts in a “thinner market” with fewer participants. This leads to bigger gaps between the limit orders in the market makers’ order books. These gaps increase price volatility whenever a market order – an order to buy and sell with immediate fulfilment regardless of price – is placed.

Unfortunately, models of noise trading (De Long et al, 1990) are generally unable to adequately explain behaviours. They fall short of establishing one-to-one relationships between a trader and a behaviour or, for that matter, an “attitude to speculate”. This is probably because of the multiple personalities of traders and investors. As a result, Grundfest et al (1991) have even claimed that a tax on financial transactions would affect both (short-term) traders and (long-term) investors.

Thus, instead of modelling speculators’ behaviour, would it not be more revealing to model financial markets themselves? One possibility is to model the chaos and complexity endemic in financial markets. The theory of chaos and complexity is best explained through an example of throwing a pebble in a pond. The ripples have no pattern and are, therefore, chaotic, as are financial markets. At the edge of chaos, where the ripples are dying out, however, complex patterns may be discerned. This is the Mandelbrot set, named after the scientist who programmed it.

The complexity theory is being used to explain biological cell growth and galactic formation, so why not financial markets? Global financial equilibrium has successfully escaped traditional analyses and prescriptions. We need new methodologies to take corrective action. If complex behaviour patterns are discerned in otherwise chaotic financial markets, we can observe those patterns and introduce policy to manipulate them. A new generation of economists could surely pioneer such methodologies and bring them to the policy table?

The writer is director and chief executive,
All opinions are exclusively those of the author