Economic models ‘not up to task’
by George Hoguet
Policy-makers and investors seeking to deepen their understanding of the dynamics of global capital markets and recurring financial crises would benefit from reading Richard Bookstaber’s The End of Theory: Financial Crises, the Failure of Economics, and the Sweep of Human Interaction.
Bookstaber argues that neoclassical economics is not up to the task of modeling and explaining financial crises. In his view, crises should be the ‘litmus test’ for validating theory. He believes we need a new framework to describe the world and to model crises.
The book reflects the author’s many years on Wall Street as a risk manager. He has testified before Congress on financial stability and was an adviser to policy-makers at the US Treasury and at the Securities and Exchange Commission, where he helped design the Volcker rule and Dodd-Frank Act.
The End of Theory begins with a brief history of neoclassical economics and a critique of the axioms on which it is built. The 19th-century British economist William Stanley Jevons led the way. He argued that economics was ‘purely mathematical in character’. ‘We cannot have a true theory of economics without [mathematics’] aid.’
The desire for mathematical tractability in modelling economic phenomena contributed to – in the words of Gary Becker, the Nobel prize-winning economist – ‘the combined assumptions of maximising behaviour, market equilibrium and stable preferences’ being used ‘relentlessly and unflinchingly’.
But in a crisis, Bookstaber suggests none of these assumptions hold. We need to abandon such restrictive conditions and develop more robust theory.
Bookstaber, who holds a PhD in economics from the Massachusetts Institute of Technology, discusses four phenomena that undermine current economic models. He begins with emergent phenomena, which explain how ‘the actions of the system differ from the actions of the agents that comprise the system’. An example would be where one person running triggers a stampede. What is locally stable can be globally unstable.
The second phenomenon is non-ergodicity. Current economic theory is ergodic. It assumes that processes stay fundamentally the same over time. But in the real world, they don’t. A major financial crisis is more than a fat-tail event; it is a completely new event. The world is non-ergodic. This brings us to the third phenomenon – radical uncertainty. There are some things we cannot know. We cannot model ‘unknown unknowns’.
The fourth phenomenon, computational irreducibility, rejects the notion that the economy can be reduced to rigorous mathematical models based on solid axioms.
Having eviscerated the foundations of neoclassical economics, Bookstaber develops the theory of agent based modelling (ABM). Neoclassical economics assumes a ‘representative agent’ with stable preferences operating under a specified probability distribution. In the ABM framework, there is no representative agent and no general model; there is a complex adaptive system whose states cannot all be delineated. Multiple agents operate with individual heuristics. Their actions change the environment and each agent subsequently adjusts their behaviour. The system is reflexive. As George Soros said, this is one in which ‘house prices are going up because banks are lending, and banks are lending because house prices are going up’.
Bookstaber then applies ABM to financial crises, which he defines as situations in which the demand for liquidity exceeds supply. The final chapters provide a discussion of liquidity freeze-ups, cascades and contagion.
Bookstaber believes that we cannot model the financial system on the basis of traditional assumptions, because it is so complex and interconnected. Hence his adaptive framework. ‘When there is a high degree of complexity, you have to figure it out as you go along.’
This book is an important contribution to our understanding of how markets function. The discussion of crisis liquidity dynamics is particularly valuable. Bookstaber surely is right in focusing on the incentive structure of individual market participants, complex network structures, liquidity freeze-ups and their interaction. But the theory of ABM is not fully developed – perhaps because it can’t be.
The End of Theory is a sobering read, reminding us how difficult it is to understand and model global financial markets.
George Hoguet is a trustee of the CFA Institute Research Foundation.