Over on Rethinking Development Economics, a recent post highlights a provocative speech by Dr DeLisle Worrell, Governor of the Central Bank of Barbados. Worrell focused on the problems with economics today, with much of his talk given over to ‘complexity economics‘.
To quote directly from Worrell:
Our theories can’t deal with reality, so we ignore the real world and spend our time “testing” our theories. If economics is to have any advice to offer which is useful for the management of real economies, we must speak to the reality in all its rich complexity, using all the data we have, all the methodologies we can devise, and all the sources of insight we can borrow. We must dig as deeply as we can, and become sleuths in pursuit of deeper understanding of our economies, even if our search leads us into paths that are dark and uncertain.
Elsewhere, the new book by David Orrell, Economyths, also focuses on why standard economic theory has little to do with reality. The wonderful phrase ‘Neoclassical Logic Piano‘ is Orrell’s label for the outmoded economic paradigm – a cluster of conventions including efficient market hypothesis, equilibrium theory, rational expectations which he happily debunks. Orrell recommends an interdisciplinary approach to a “new economics”, again drawing on complexity theory.
Both of these accounts owe something to the 2006 publication The Origin of Wealth by Eric Beinhocker, which explores complexity economics and related issues to great effect, and which was covered at length last year by Duncan Green on the From Poverty to Power blog.
But what might the ‘dark and uncertain’ alternatives to the Neoclassical Logic Piano look like in practice? And what might be the relevance for development economists – a group which, one suspects, may be even more wedded to the assumptions of the Neoclassical Logic Piano than is the norm for the profession as a whole? In a recent article in the Economist, there was a glimpse of one possible answer.
The article gives a succinct explanation for the growing interest in rethinking conventional economic principles: “Mainstream economics has always had its dissidents. But the discipline’s failure to predict the financial crisis has made the ground especially fertile for a rethink…. [experts have] attacked many of the assumptions, including efficient financial markets and rational expectations, on which these models are predicated. These assumptions were clearly too simplistic. But there is less agreement on what should replace the old ways.”
The article goes on to focus on what is seen as ‘one of the most promising options’ – agent-based modeling (ABM). As Joshua Epstein discusses in the clip below, ABM is a key tool of the complexity scientist and has many diverse applications, including the smallpox example he mentions.
The implications of ABM for economic analysis are set out as follows:
Agent-based modelling (ABM) does not assume that the economy can achieve a settled equilibrium. No order or design is imposed on the economy from the top down. Unlike many models, ABMs are not populated with “representative agents”: identical traders, firms or households whose individual behaviour mirrors the economy as a whole. Rather, an ABM uses a bottom-up approach which assigns particular behavioural rules to each agent. For example, some may believe that prices reflect fundamentals whereas others may rely on empirical observations of past price trends.
Crucially, agents’ behaviour may be determined (and altered) by direct interactions between them, whereas in conventional models interaction happens only indirectly through pricing. This feature of ABMs enables, for example, the copycat behaviour that leads to “herding” among investors. The agents may learn from experience or switch their strategies according to majority opinion. They can aggregate into institutional structures such as banks and firms. These things are very hard, sometimes impossible, to build into conventional models. But in an agent-based model you simply run a computer simulation to see what emerges, free from any top-down assumptions…
ABMs… make no assumptions about the existence of efficient markets or general equilibrium. The markets that they generate are more like a turbulent river or the weather system, subject to constant storms and seizures of all sizes. Big fluctuations and even crashes are an inherent feature. That is because ABMs contain feedback mechanisms that can amplify small effects, such as the herding and panic that generate bubbles and crashes. In mathematical terms the models are “non-linear”, meaning that effects need not be proportional to their causes. These non-linearities were clearly on show in the credit crunch.
The article goes on to list the kinds of things ABMs are good at capturing, from ‘the web of interdependencies created by the use of complex derivatives’ and ‘network-based vulnerabilities’ to the ‘the role of interactions between different sectors of the economy’.
More generally, as an article published by the US National Academy of Sciences in 2002 suggests, there are 4 areas where ABM can be of use:
Flows: evacuation, traffic, and customer flow management.
Markets: stock market, shopbots and software agents, and strategic simulation.
Organisations: operational risk and organisational design.
- Diffusion: diffusion of innovation and adoption dynamics.
(…as an interesting aside, these four areas of flows, markets, organisations and diffusion would seem to form a pretty good framework for understanding the aid system…)
To extend Orrell’s metaphor, the ABM approach seems to take the rule-based neo-classical logic piano and replace it with something more akin to improvisational jazz. In fact, researchers at Cornell have made exactly this link – suggesting that while traditional models have tried to understand social life as a structured system of institutions and norms that shape individual behavior from the top down, agent-based models assume that much of social life emerges from the bottom up, more like improvisational jazz than a symphony.
The clip below in which Mulgrew Millar likens jazz improvisation to language acquisition is a revealing one, especially as ABM has been used extensively in linguistics.
Of course, there are many dangers associated with treating any new approach as a silver bullet for the failings of the old, and ABM is no exception. Patrick Beautement, a good friend of Aid on the Edge of Chaos, and Director of Abaci Partners, has thought long and hard about the ‘real-world relevance’ of agent-based models, and argues that there are some outstanding questions to be addressed.
In particular, it is important to be clear about assumptions, limitations and constraints of ABM (just as with any models), and to use such models as the basis for dialogue and discussion rather than as prediction engines.
The Economist article highlights a suggestion which came from, among others, researchers at the Santa Fe Institute, the leading complexity sciences think-tank, to construct an agent-based model of the global economy, to permit real-time simulation and analysis – “in the manner of global climate simulations, which project various possible futures”.
One would imagine that – just as with climate simulations – such models would be subject to rigorous and extensive scrutiny / debate. One would also hope that the debate would be rather less politicised and dysfunctional than those surrounding climate change, but that’s a different story…
The article closes by drawing a strong link between economic analysis and seismology – (for more on comparisons between earthquakes and economic crises, see a 2009 Aid on the Edge of Chaos post):
Seismologists may not be able to forecast earthquakes precisely but it would be deplorable if they were to resign themselves to modelling just the regular, gradual movements of tectonic plates. Instead they have developed ways of mapping the evolution of stress patterns, identifying areas at risk and refining heuristics for hazard assessment. Why not do the same for the economy?”
Why not indeed? And while we are at it, why not do the same in development economics?