The uncomfortable thing about models of the sort of system that I am
referring to, is that there is no necessary proximate major cause for a
sudden shift in the aggregate state. Similarly, in social or economic situations,
when a major and sudden change occurs in the state of the aggregate
economy, there are no culprits to blame and no easy remedies to
prevent similar occurrences in the future. If we stick to the physical analogy
there is no warning that the system will go through a phase transition.
There is, of course, an alternative view which might be more appealing
for economists. It is that the system may reorganize itself in such a way
that it becomes vulnerable to small changes in the environment. Neither of these alternatives is encompassed by modern macroeconomic models. The
relations between the variables are essentially fixed and the system functions
at equilibrium in a very mechanical way. The only thing that may
perturb the evolution of the economy is an external shock from which
the economy adjusts, by assumption, to a new equilibrium. How it does so
is typically not analysed. Out of equilibrium, dynamics are not a central
issue in economics and Gerard Debreu said explicitly that their analysis
was too difficult and that was why he had never ventured in that direction.4
So, the most interesting aspects of economics, if the economy is
viewed as a complex interactive and adaptive system, are absent in macroeconomic
models based on the General Equilibrium view (for an interesting
discussion of the role of the equilibrium notion in economics, see
Farmer and Geanokoplos 2009)