As a result of the financial crisis, many countries are trying to regulate their financial markets. Recently the heated debates about bonus taxes, a permanent levy on banks’ balance sheets and a ban on short sales have taken another turn, with France’s President Nicolas Sarkozy promoting a transaction tax. But are policy-makers doing the right thing? Or will they obstruct the self-regulating forces of the markets?
Action is certainly needed, particularly since the classical instruments of Central Banks — money supply and interest rates — have already been stretched to the maximum. Buying Government Bonds is considered to be a last resort. Have Central Banks run out of instruments to fight the crisis? Are their weapons still sharp enough to control the markets?
The use of two fundamental mechanisms from physics, friction and noise, could provide a solution. These mechanisms can be viewed as setting boundaries to the system’s self-organization, rather than impeding it.
What to do?
Top-down regulation of the system does not seem to be working. No institution has god-like insights and power: Frequently there is a lack of information and a delayed response. On top of increasing the level of bureaucracy, some of these policies may actually hamper the self-regulation of the markets. We now have a choice between continuing to operate on a detailed level, trying to control single factors by separate laws and bureaucratic structures, or concentrating on the design of the financial architecture and its interaction rules.
The authors strongly argue for the latter approach. Inspiration can be found in the theory of complex systems. This theory features a variety of methods and models which make it possible to understand and manage not only complex systems in physics and ecology, but social and economic systems as well.
Instability is not a bad thing per se
Financial markets are social institutions which should balance supply and demand worldwide based on decentralized investment decisions. They allow entrepreneurs to collect money from investors and to realize huge projects. Speculation, however, may distort prices. Such speculation bubbles are hard to spot and are usually recognized only when it is too late. We need mechanisms that make bubbles burst before they grow so big that they imply systemic risks. These mechanisms must work without requiring prior knowledge of what is a bubble and what instead is healthy economic growth.
To a certain degree, instability is useful to support the ability of markets to adapt to changing facts. Instability is also crucial to allow innovative forces to operate. Hence, instability itself is not a problem, but the extent of instability is. This suggests that mechanisms should be sought which “automatically” regulate the degree of instability, i.e. which control the system smoothly without requiring continuous reassessment of whether the level of instability is good or bad.
The analogy of a boiling pot of water can stimulate our imagination
This is where physics comes into play. Comparing financial flows with energy or heat flows in a fluid heated from below, we imagine the heat flow transporting energy (capital) from the heat source (the capital providers) towards the surface of the liquid (the firms or countries in need of capital). If the temperature difference between the surface and the heat source is small, normal heat conduction will take place. Above a certain temperature, however, self-organized patterns such as convection rolls occur within the liquid, transporting the heat more quickly than diffusion can. At a certain point, bubbles form and the liquid boils, spilling over the edge of the pot. There are four ways to prevent such bubble-induced spill-overs: Firstly, one can lower the amount of liquid. Secondly, the temperature difference can be reduced. Thirdly, the inner friction of the liquid can be manipulated, for instance by adding another substance. Forthly, noise can be generated by sprinkling sand into the liquid. Noise created in this way will produce many small bubbles and thereby avoid large and dangerously eruptive bubbles.
These four “control parameters” occur in many different technical systems. Interestingly, these parameters are not specific in the sense that they determine which zones get more heat (i.e. who gets more capital). Instead, the system regulates itself. Despite the merit of this strategy, the financial sector nowadays only manipulates the control parameter corresponding to the amount of the liquid and the temperature difference. But reducing the money supply or increasing the interest rate are not real options in today’s markets — the situation is far too delicate.
Adding friction to financial markets
Physical friction is a force that increases with the speed of movement, while simultaneously working against the motion. In fact, the faster a car moves, the greater the friction (for example, due to air resistance) and the greater the related braking effect. In financial markets, the equivalent of movement is capital transaction. In the frictionless world of financial transactions, a friction mechanism should therefore be introduced, which should increase in strength proportionally to transaction volume. It is then up to Central Bankers to decide on the desirable amount of volatility in the system, and, likewise, when such a volume tax would become effective.
Clearly, this is not the same as the much-discussed Tobin tax. The Tobin tax is a fixed tax that applies to every single transaction. In contrast, the “frictional tax” which we propose would not apply under normal market conditions, and in critical situations it would increase with the transaction volume. Two positive effects are expected from this. Firstly, in a state of financial panic, attempts to minimize taxes would slow the pace of transactions. Thereby, market participants and regulators would gain time to take the right decisions. Secondly, the Central Banks would make money just at the time when the system is in a danger of getting out of control.
Thinking out of the box: Adding noise
While the concept of friction fits into ongoing economic debates, the idea of adding noise to the market is novel. Noise lays some “fog” over a signal and thus reduces its information content. This runs contrary to the dogma of maximum transparency, as it is generally believed that having more information would result in more accurate prices. In ideal markets, there is no need for noise — it would even be considered a source of irritation. In reality, however, stock prices may differ considerably from the fundamental value. Two factors are responsible for this: insider information on one hand and herding effects on the other. These factors may lead to price bubbles, as insider information encourages speculation on a large scale. However, when additional noise makes price trends less easy to manipulate, speculation becomes risky and cannot create unfair gains. The same applies to herding effects, which tend to occur whenever a large number of market participants suffer from information overload and therefore orient at some financial gurus and market trends.
Both problems are well known in financial markets, but are still not under control. The authors propose that adding noise may be the solution. Noise in the financial markets could be generated by big players such as the Central Banks, by randomly buying and selling financial assets. Studies show that random investments create returns that are comparable to the market performance and usually greater than those of most funds. It should therefore not be a ruinous strategy for Central Banks to randomly buy and sell. Such random investment would trigger an unbiased “fog of uncertainty,” which would reduce both herding effects and the value of insider information. Bubbles would burst while they were still small and posed no risk.
While friction generates an upper limit to instability, noise on the other hand creates a lower bound. Together, they provide two potent new tools for controlling financial markets, supporting their self-organization ability rather than undermining them.