The Global Regulatory Consequences of an Irrational Crisis: Examining ‘Animal Spirits’ and ‘Excessive Exuberances’

Abstract

What does it mean to describe the financial system as ‘irrational’? And what would be the global regulatory implications and consequences for the financial system if it were thoroughly ‘irrational’? These are the issues pursued in this article. The article sets out to explore the nature of both the financial system and the economic models deployed to price the main products that are traded in the system, like options, derivatives and collateralized debt obligations (CDOs). The underlying assumptions associated with these economic models are examined and the failure of the markets to track risks assessed. The article moves on to review several alternative and radical theoretical approaches that draw attention to the nature of the potential irrationality of markets and decision making in the financial sphere, like ‘excessive exuberances’ and ‘animal spirits’. Finally, the article assesses the consequences of this kind of analysis for regulation. A key claim is that the financial system might be more profitably considered as one that works in a similar way as do natural disasters like earthquakes, tsunamis, or volcanoes. Natural disaster planning is thus an intellectual resource that needs to be brought into play to manage and regulate the financial system. This is linked directly to the issue of ‘irrationality’ considered in its existential forms as highlighted by the radical philosophical literature reviewed earlier. Thus a completely new and different framework for considering financial regulation is suggested in contrast to the current emphasis on rational top down initiatives emanating from a global calculating centre like the BIS, the IMF's Financial Stability Forum or the G-20.

¿Qué significa la descripción del sistema financiero como ‘irracional’? ¿Y cuáles serían las implicaciones regulatorias y las consecuencias para el sistema financiero si fuera completamente ‘irracional’? Estas son los problemas que estudia este artículo. El artículo parte de la exploración de la naturaleza tanto del sistema financiero como de los modelos económicos implementados para poner el precio a los productos principales que se negocian en el sistema, como las opciones, derivados y las obligaciones de deuda colaterales (CDOs, por sus siglas en inglés). Se examinan las suposiciones subyacentes asociadas con esos modelos económicos y se evalúa el fracaso de los mercados para rastrear los riesgos. El artículo continúa revisando varias alternativas y métodos teóricos radicales que generan interés a la naturaleza de la irracionalidad potencial de los mercados y la toma de decisiones en la esfera financiera, como la ‘exuberancia excesiva’ y el ‘espíritu animal’. Finalmente, el artículo evalúa las consecuencias de este tipo de análisis para la regulación. Una afirmación clave es que el sistema financiero puede ser más lucrativo, considerándolo como uno que trabaja de una manera similar a los desastres naturales, tales como los terremotos, tsunamis, o volcanes. La planeación de los desastres naturales es por lo tanto, un recurso intelectual que necesita ponerse en juego para manejar y regular el sistema financiero. Esto está directamente relacionado al problema de ‘irracionalidad’ considerado en su forma existencial destacada por la literatura filosófica radical analizada anteriormente. Por lo tanto, se sugiere un esquema completamente nuevo y diferente para considerar la regulación financiera, en contraste con el énfasis actual sobre las iniciativas racionales de arriba abajo que emanan de un centro calculador global como el BIS, el Fórum de Estabilidad Financiera del IMF, o el G-20.

Keywords:

• financial crisis
• options pricing
• derivatives
• CDOs
• risk assessment
• natural disaster planning
• global regulation

Acknowledgements

I would like to thank the editor of the special issue, Barry Gills, and Mike Pryke and an anonymous referee for the extensive comment on earlier drafts which have significantly contributed to the clarification of the argument. Any remaining errors are, of course, attributable to myself alone.

Notes

The complacency within orthodox opinion is demonstrated by Woodford (2009) who argues that there is now a general agreement amongst macroeconomists that dynamic stochastic general equilibrium models have become the accepted norm. See Blanchard (2009) for a more reflective account.

The object of critique here is not, then, neoclassic economics as a whole but only those models of financial product pricing that were at the forefront of the recent surge in the financialization of economic activity.

One that will soon be matched by the four volume, 2,600 page Wiley Encyclopedia of Quantitative Finance (Cont, 2010).

In 2000 there were 104,000 classes of options traded on the Chicago Board Options Exchange (MacKenzie, 2006, p. 201).

The term ‘existential’ is put in quote marks to indicate its shorthand descriptive character. Later quite what this might mean in an analytical sense is developed. But as a shorthand description its is meant to indicate that the system lacks overall purpose, meaning or authentication, leading to a sense of anxiety, disorientation, and confusion in the face of the seeming randomness, absurdity, and volatility of events.

I concentrate upon Black–Scholes here whereas the classic model is often referred to as the Black–Scholes–Merton model to indicate the importance of Merton's (1973) contribution. In this contribution Merton ‘generalizes’ the Black–Scholes model to some extent by reconciling American-type and European-type option contracts, dealing with dividend distribution before the contract matures, and introducing interest rate variation into the model. But as far as I can judge he does not deal directly with the main issue taken up in a moment, which is asset price volatility in various forms. Indeed he endorses the idea of its stability: when discussing the variables on which an option price depends he remarks ‘It does depend in the rate of interest (an “observable”) and the total variance of the return on the common stock which is often a stable number and hence, accurate estimates are possible from time series data’ (Merton, 1973, p. 161, emphasis added). Note the tentative nature of this assertion.

So an important distinction to keep in mind is the obvious one between the markets themselves and the modelling of the products traded in those markets. Quite how these are linked (if at all) is the key, and something discussed later in the main text.

A Gaussian normal distribution is a bell shaped distribution of events which is assumed to be highly stable so no single element of what is being measured can significantly affect the average. This is combined with the idea of a random walk; each event is discrete and not impacted by what has gone before. The combination of these two features does not allow for scalable variability or increasing error rates, something that ‘bubbles’ and herding behaviour clearly imply.

This is complicated because at the point of ‘maturity’ of an options contract its price should be zero! What is hinted at here, however, is that closing an option ‘before’ maturity is the issue.

Such a resort to mimetic rationality conceptions is also a central feature of some quasi-Marxist approaches to explaining the financial crisis, see Marazzi (2008).

CDOs are financial products constructed out of previous bonds like a bundle of mortgage backed securities or credit card liabilities, the income stream from which passes to the CDO. These are ordered into tranches from bottom to top, where any default is felt by the bottom tranche first. CDOs are also traded as derivatives of CDOs (CDO2s, CDO3s).

This echoes Donald Rumsfeld's comment when US Defence Secretary in February 2002: ‘There are known knowns. There are things we know that we know. There are known unknowns. That is to say, there are things that we now know we don't know. But there are also unknown unknowns. There are things we do not know we don't know.’ Although this was pilloried in the popular press, it is not as stupid as suggested—see Wikipedia entry ‘Unknown unknown’. Expected unknowns as expressed here are rather like Rumsfeld's known unknowns.

Brunnermeier suggests an interesting measure to cope with contagion in these settings designed to indicate to the ‘value at risk’ of any organization's balance sheet which is co-varied with other organizations in the network (‘CoVaR’). This is designed as part of a better assessment of network stress (Adrian & Brunnermeier, 2009).