Christian Arnsperger and Yanis Varoufakis (A&V) have argued at considerable length that neoclassicism defends itself by refusing to be defined. A&V propose ...
three crucial axioms which, in conjunction with one another, as we shall claim, underpin all (and only) neoclassical theory 1And they maintain that, if economics’ mainstream ever admits to definition by these axioms, it would have no option other than a straight-up fight to the finish with its heterodox antagonists.
Our evaluation of these axioms agrees with A&V’s rather strong assertion of inclusivity:
It is hard to imagine how any standardly trained economist could deny that her theoretical practices digress from the three methodological motives mentioned above: METHODOLOGICAL INDIVIDUALISM, METHODOLOGICAL INSTRUMENTALISM and METHODOLOGICAL EQUILIBRATION. For simplicity we shall henceforth refer to them as the neoclassical meta-axioms. Whether it is general equilibrium theory, evolutionary game theory, non-Walrasian equilibrium theory, social choice theory, industrial economics, economic geography, new political economy, analytical Marxism, public choice economics etc., all mainstream approaches in these fields remain loyal to the three meta-axioms above.And we cannot honorably deny that these axioms also particularize SFEcon. We cooperate with economics’ neoclassical mainstream insofar as our explanation for the ordering of material affairs presumes a general tendency for marginal revenues to approach marginal costs. We also reference the most visible conventions of neoclassical analysis, e.g.: production and utility functions, Say’s Law, homo economicus, etc. SFEcon demonstrates these particulars to be just sufficient for control of our engineering dynamic models of economic adjustment, while we fail in our efforts to imagine alternative premises upon which comparable demonstrations might be made.
A&V’s first axiom of METHODOLOGICAL INDIVIDUALISM asserts that ...
any socio-economic phenomenon under scrutiny is to be explained via a synthesis of partial knowledge derived at that individual levelwhich invokes homo economicus: the hedonistic, perfectly informed individual, whose only activity is the securing of gain, and who has indeed ...
been implicit to neoclassical theorising since the first stirrings of marginalism.A&V allow SFEcon one avenue of escape from the application of this axiom:
Methodological individualism is vacuous without a theory of what motivates individuals.SFEcon has no theory of the individual, but it would not be gallant to quibble on this point. We have our homo economicus, he is motivated according to the necessities of neoclassical theory, and he does control our models. The quibble would be that our economic man composes only that fraction of humanity who are professional securities analysts. The agents of our macroeconomic input/output model are economic sectors described in terms that engage the global competition for capital controlled by SFEcon’s homo economicus.
SFEcon presents the second axiom of neoclassical economics, METHODOLOGICAL INSTRUMENTALISM, in the most direct and conventional form. Where A&V have it that ...
neoclassical economists [have] increasingly sought to distance themselves from the assumption that preferences are fixed and exogenousour preferences are expressed in familiar production and utility tradeoffs that are fixed and exogenous from the standpoint of the basic theory’s operation. (Utility parameters only become variables when embedded in more advanced models.)
Neoclassical economics’ third axiom of METHODOLOGICAL EQUILIBRATION is presented by SFEcon whenever one of our emulators is run under constant parameters for a sufficient period. This experiment corresponds to a hypothetical situation in which all technical and demographic change has ceased for so long as to remove any expectation of change in the future. It eventuates in a steady state expressing Pareto's optimum. We cannot imagine any other behavior that might be acceptable as a dynamic theory’s response to this peculiar specification of its parameters.
Matters of agreement having been settled, we must take issue with the labeling of neoclassicism's characteristics as axioms. It would be much more informative to say that INSTRUMENTALISM forms the boundary conditions of our model; INDIVIDUALISM provides its causality; and EQUILIBRATION is among its behaviors. This seemingly small matter of vocabulary is critical in that ...
while the first two meta-axioms allow for rich depictions of socioeconomic phenomena, they lead to an unquenchable indeterminacy.Equilibrium is one of the plethora of dynamic responses possible for an SFEcon model. These responses are something a model does, not something that a model is. Formal emulators of the SFEcon algorithm are presented so that one might observe our theory’s behavior as it navigates alone on the dimension of time, guided by nothing beyond INSTRUMENTALISM and INDIVIDUALISM. We submit these behaviors as efficient, stable, and realistic.
As with any determinant mapping of a dynamic thought structure onto exterior reality, SFEcon tolerates no specification whatever of its behaviors — especially the equilibrium implicit in its parameters. Our models are true finite state machines: they are always operating on the verge of an unknown future; they never know anything more than their current state and their rules for arriving at their next state; they never compute the general optimum they discover; and they do not know when they have discovered it.
If the behaviors of a dynamic system are not allowed to unfold out of free articulations among its elements, then the structure simply has no time-variant behaviors. It will either (mathematically speaking) explode or it will remain inert. Given this certainty of dynamic systems analysis, it is odd to find that economists ever expected to acquire mathematical determinacy by imposing equilibrium:
The third feature of neoclassical economics is, on our account, the axiomatic imposition of equilibrium. The point here is that, even after methodological individualism turned into methodological instrumentalism, prediction at the macro (or social) level was seldom forthcoming. Determinacy required something more: it required that agents’ instrumental behaviour is coordinated in a manner that aggregate behaviour becomes sufficiently regular to give rise to solid predictions. Thus, neoclassical theoretical exercises begin by postulating the agents’ utility functions, specifying their constraints, and stating their ‘information’ or ‘belief’. Then, and here is the crux, they pose the standard question: “What behaviour should we expect in equilibrium?” The question of whether an equilibrium is likely, let alone probable, or how it might materialise, is treated as an optional extra; one that is never central to the neoclassical project.But do we really know that it is the interaction of INSTRUMENTALISM with INDIVIDUALISM that causes indeterminacy? Here A&V offer only a reference to the admittedly powerful authority Hugo Sonnenschein has had over this discussion:
neoclassicism cannot demonstrate that equilibrium would emerge as a natural consequence of agents’ instrumentally rational choices.while passing by the fact that the findings of Sonnenschein and others on this point only represent their failures to make the objective demonstrations they rightly demand. Our experience is that such demonstrations have been available for as long as the computer technology required for their creation (circa 1970); and economic science veered away from the possibility of this demonstration’s realization just as it appeared on the technical horizon.
Reference to SFEcon would suggest that, in certain implementations, the INSTRUMENTALISM and INDIVIDUALISM that A&V describe as rich depictions of socioeconomic phenomena are just sufficient to specify working analogs to economic adjustment. In our view, the first requirement of neoclassicism’s macroeconomic project is to express INSTRUMENTALISM with INDIVIDUALISM so as to make equilibrium redundant as an element of theory. We submit our Model 0 as having accomplished this task, thereby providing a basis for the discussion A&V say they wish to have.
So: if heterodoxy sincerely wishes for neoclassicism to stand still and fight, they have here
a neoclassical installation that is open to attack — one exhibiting actual behaviors
that might be compared with heterodox models comprehending the same variables, but controlled by
tendencies other than a resolution of marginal revenues with marginal costs. If the heterodox impulse
has given rise to such models as might provide a basis for the comparison, we have not yet been able
to draw them out.