In praise of complexity economics

The Economist's ranking of the most influential economists of 2014 has been derided, and Tyler Cowen provides a more sensible list. But I wonder: which economists should be more influential than they are?

The glib answer here is: whoever corroborates my prejudices. A less glib answer would be the countless economists who are doing good work which is insufficiently appreciated by the public: I try to publicise some of this in both here and in my day job.

But there's another answer I'd like to offer: those economists who have been working in the field of complexity economics. Perhaps the pioneer here is Brian Arthur, who has written a great short introduction (pdf) to the subject. But I'd also mention Eric Beinhocker*, Cars Hommes, Alan Kirman and, in financial economics, Andrew Lo, whose theory of adaptive markets I find an attractive alternative to the efficient/inefficient markets hypothesis.

One feature of complexity economics is that recessions can be caused not merely by shocks but rather by interactions between companies. Tens of thousands of firms fail every year. Mostly, these failures don't have macroeconomic significance. But sometimes - as with the Fukushima nuclear power plant or Lehman Brothers - they do. Why the difference? A big part of the answer lies in networks. If a firm is a hub in a tight network, its collapse will cause a fall in output elsewhere. If, however, the network is loose, this will not happen; the loss of the firm is not so critical. Daron Acemoglu has formalized this in an important paper, and there are some good surveys of network economics in the latest JEP.  

The question is: why is complexity economics not more influential?

One reason is that it requires different techniques. It can't be studied merely by problem sets (ugh) devoted to standard optimization techniques. Instead, it requires agent-based simulations (here are a couple of examples), laboratory experiments of the sort done by Charles Noussair among others, or close attention to history and the institutional and cultural settings in which markets operate.

And therein lies a second reason why complexity economics is under-rated. For me, one of its big messages is that context matters. Emergent processes sometimes lead to benign outcomes and sometimes instead to inequality and inefficiency, and which turns out to be the case can hinge on quite small differences. The great economists of the 20th century - such as Keynes, Samuelson or Friedman - tried to offer a general theory. Complexity economics doesn't.

There's a third reason why complexity economics is under-rated. It does not give us a means of foreseeing the future. Of course, conventional economics doesn't do so either. But the difference is that complexity theory tells us that such forecasts might well be impossible - which is not what the customer wants to hear. The best it can do is help us understand what has happened. And for me, this is good enough. As someone once said, "Economists have only changed the world; the point, however, is to understand it."

* The Origin of Wealth is like Marx's Capital or Kahneman's Thinking Fast and Slow; it gets a lot better once you get past the first 50 pages or so.

Comments

[Esamjones]

Good post. I'd also like to see more from complexity economics, but my (limited) reading of this literature -- much like evolutionary economics -- is that while it provides a useful metaphor or narrative of economic processes, it is much less clear / distinctive when it comes to the policy implications. Moreover, given the reliance on simulation empirical testing is less straight-forward.

[Barkley Rosser]

As editor of the now more than a decade-old volume, I suggest people check on _Complexity in Economics, Vols. I-III, 2004, Edward Elgar, ed., by me. It is a collection of readings on the subject and covers all fields of econ.

[Thornton Hall]

Humans are bad at predicting the future. Any theory that contradicts this obvious and well demonstrated truth is probably wrong.

[Demostenes]

It is not taught at any of the top-100 US PhD programs that I am aware of. I don't know of any complexity paper published in the top-5 economics journals.
Most mainstream economists think of complexity as a heterodox soapbox. So the onus is on the complexity guys to change this by making substantive contributions--if they can....

[Simon]

If economics cannot give us a glimpse of the future then let's not waste any more time with it. I believe it does. ..Or I would have lost interest in it by now. There are numerous people who have been able to make good predictions based on a good understanding of conventional economics.

Nourial Rubini
Robert Shiller
My favourate Michael Pettis
Even Ben Bernanke can make reasonable predictions.

Martin Wolf. Non economist but better than most
There are lots more. What use is any kind of science that can not provide an informative Model? What's needed is more science and less ideology.

[Deviation From The Mean]

Economics never rises above meta theory, and more often than not never rises above a really superficial and cognitive bias.

Everything useful is abstracted out of the picture and everything mundane is included.

It is peculiar in that any understanding of things seem to have regressed rather than progressed over time. In this way it is quite a unique subject.

TREAT IT WITH SEVERE CAUTION!

[William Janeway]

For a promising exercise in policy analysis based on a rich agent-based model, see G. Dosi et. al., "Fiscal and Monetary Policies in Complex Evolving Economies," at www.science.direct.com/science/article/pii/S016518891400311X

[Metatone]

Go and read Dave Snowden's blog about complexity approaches.

The message is not that the future is inherently a closed book, but that it is not deterministic.

We can do things to improve our lot in the future, even if we can't predict the future in the ways we might all wish for.

Complexity science is not the dumb Hayekian learned helplessness that people like Paul Ormerod are always promoting.