By Greg Fisher
Yesterday INET published a valuable essay written by John Kay entitled “The Map is Not the Territory: An Essay on the State of Economics”. In this blog I would like to discuss how John’s essay is a useful and fair critique of the state of economics and use his framing to demonstrate a viable alternative approach to economics, namely Complexity theory. This is the study of complex systems, which contain a number of potentially unique agents that interact with – and adapt to – each other over time. I have always thought John’s writings to be highly compatible with a complexity approach and his recent essay is no different.
With Complexity theory in mind, the fundamental problem with orthodox economics seems to be that it takes a reductionist approach to making sense of a complex system. Put another way, economics tries to dissect something that is highly interconnected and constantly evolving in to discrete tractable chunks. The only way to resolve this paradox is for economists to re-frame their field as ‘Complex’. Without doing this, they will continue to make a framing error. Moreover, what we mean by “an economy” ought to be viewed as an irreducible part of the biosphere.
Rigour, Consistency, and Universality
The core of John’s essay is to explain that economics orientates itself around rigour, consistency, and universality. This is in the best traditions of classical physics, which was (in effect) smashed to smithereens by the relativity and quantum revolutions of about a century ago, and also (in my view) by the chaos / complexity revolutions of the 1970s and beyond. So, as Eric Beinhocker described in ‘The Origin of Wealth’, orthodox economics is a bit like Cuban cars – lost in time due to the embargo – while the rest of the world has moved on.
As John emphasised, rigour lends itself to a mathematical approach and consistency lends itself to ideology. These result in economics being based on axioms, deduction, and ‘analytical solutions’. Through a complexity lens, we can see that the value of the axioms-and-deduction approach is limited by the concept of emergence. Notably, what Stuart Kauffman refers to as ‘ontological emergence’ means that we cannot necessarily move deductively from the ‘micro’ to the ‘macro’. Some properties will emerge at the macro level from the interaction of the (micro) constituents of the system, which cannot be deduced from those constituents. This can be the case with a large society or even a conversation between two people. Emergence is probably the most powerful concept to have, well, emerged from Complexity theory because it contextualises the value of deduction.
In John’s words, consistency “means that any statement about the world must be made in the light of a comprehensive descriptive theory of the world”. And universality requires “a description of the world that is independent of time, place, context, or the observer.”
The primary problem with consistency and universality is that complex systems are continuously evolving, whereas orthodox economics frames the economy as a static system in which all of the ‘rules’ or ‘laws’ never change i.e. they are independent of time. In technical-speak, economics views its terrain as ergodic, meaning ‘sameness’ over time. By contrast, complex systems are non-ergodic – they are in a constant state of flux (but this is not to say they lack structure, or patterns). The distinction is enormously important for understanding any type of complex system, including social systems. Non-ergodicity means that universality is problematic because any description of the world cannot be independent of time – change is inherent in the system.
Moreover, economics is orientated around systems that are in equilibrium, and in such systems inconsistencies will eventually be ironed out (such as opportunities for arbitrage in financial markets). However, complex systems can exist in a perpetual state of disequilibrium, with apparent inconsistencies. This statement is the Complexity analogue to Keynes’ famous dictum that “in the long run we are all dead”: in the shorter term the economy can remain in a state of disequilibrium. So complex systems can and do contain inconsistencies – this may result in inelegant and intractable views of the world but that is the nature of (real) life, not a flaw of Complexity theory.
Universalism also runs up against another key feature of complex systems: the prevalence of idiosyncrasies. The importance of idiosyncrasies is in fact due to emergence: the combined characteristics in one place, or set of circumstances, may lead to emergent properties in that one place. Another – similar – place, or set of circumstances, may give rise to different emergent properties. These properties may make the two ‘places’ completely different even though they have very similar constituents. For example, one housing estate in the UK may have worklessness rates of 70% whereas another – otherwise identical – estate may have a 10% workless rate. The difference may be due to emergent properties, such as human values, which are impossible to predict ahead of time. This is a problem for universality, which in John’s definition requires independence of place and context.
Inductive Reasoning, Pattern Recognition, and Cognition
The non-ergodic view of the world leads to a very different understanding of complex social systems. Importantly, the contrast between human cognition and how the world works becomes even starker and the intractable nature of reality becomes more obvious. Herb Simon and Friedrich Hayek both emphasised the insufficiency of human cognition relative to the real world: Complexity theory asserts this point further.
Continuous change also contextualises inductive reasoning, which John also referred to. Induction is typically thought of in the same way that Popper referred to it: making general statements out of the singular. It is preferable, I think, to talk about pattern recognition rather than induction because its definition is clearer. To paraphrase Popper, it is obviously silly to infer a general rule from a single instance. But it is possible to observe general (not necessarily universal) patterns from one’s own experiences, including empirical research. Hypothesis formation can be viewed as a form of pattern recognition, which Popper overtly chose to ignore in his “Logic of Scientific Discovery” because he accepted he did not understand it. With a Complexity perspective on the world, we can understand it much better.
John Kay’s critique of orthodox economics was entirely fair. But it is one thing to criticise and something entirely different to offer a replacement. The most obvious counterpart framework to offer in place of orthodox economics is one based on complex systems. That is the paradigm change orthodox economists have to accept: rigour, consistency, and universality were hallmarks of classical physics, which physicists moved past a hundred years ago. It is about time economists moved on too.