This is a blog about some interesting books that I read on my extended Easter holidays. First up is Eric Beinhocker’s wonderfully accurate summary – well before the credit crunch – of how economies really work. (My thanks to Andy Hopkins for suggesting it.) The subtitle shows that we need to think of economic systems – just like biological systems – as complex adaptive systems or networks of interacting elements that evolve over time according to non-trivial but broadly knowable principles. I would like to hope (but doubt) that all economics students are exposed to this way of thinking. There is thus much to be gained by bringing together our knowledge of natural and other evolution (including evolutionary algorithms – introduction ) with that of the operation of social and economic systems, and I have some plans in this regard.

I also enjoyed two spendidly iconoclastic and somewhat related books by Nassim Nicholas Taleb. Much of these – as with Beinhocker’s book above – are a critique of equilibrium theory and so-called rational behaviour in economics (see also a previous blog), whereby it is assumed that markets move randomly, and swiftly adjust prices from any such excursions to reflect their ‘true’ values. Despite all evidence to the contrary, this is apparently still widely believed and taught. A particularly striking statistic in the latter book is that “in the last fifty years, the ten most extreme days in the financial markets represent half the returns”. Large crashes and bank bail-outs are, sadly, not a new occurrence. “Those who cannot remember the past are condemned to repeat it.” – see the original by George Santayana.

Network biology – as a subset of systems or network analysis generally – shares many attributes with social networks, a subject summarised by Linton Freeman that has evolved rather independently of it (and somewhat stochastically). A chief point is that much of empirical social research has been dominated by the sample survey, a strategy that is likely to ignore completely network-based interactions; this is equivalent in systems biology to the study of ’omics alone (as concentrations of RNAs, proteins or metabolites) without including analysis of the biochemical networks in which they are embedded. Systems biology is much more than omics. A second point that emerges from these studies of social networks is that the structure (i.e. topology) of a network alone, with little or no knowledge of its parameters, provides a huge constraint on the kinds of behaviour that can be exhibited by such a network. This is certainly true in metabolic networks, where there are also stoichiometric and thermodynamic constraints, but is also seen in some developmental and signalling systems (where such robustness to parameter changes might reasonably be seen as an evolutionary goal). One might suppose robustness to be a desirable goal of economic systems too, implying that systems biology (and the analysis of biological network structures) might be a useful source of economic and social ideas.

I also dipped into Science without Laws, which focuses on the use of models in biology (a personal interest), re-read The Two Cultures – whose 50th Anniversary is coming up on May 7th – and finished up with The Third Culture, a very interesting set of mutual commentaries that despite being published 14 years ago presaged much of the thought about complex adaptive and evolutionary systems in biology and elsewhere, including economics. Truly, the economy is the network.

Related posts (based on tags and chronology):