ABSTRACT
Introduction We have earlier claimed that, on the microscopic level, the neoclassical theory may be a reasonable approximation to the behaviour of the agents. Given a set of epistemic cycles, we have said, there is no reason, but in abnormal cases, to expect the agent to be irrational in the sense of the three axioms defining rationality in the neoclassical theory. We may look at the rationality concept of the neoclassical theory as a very basic form of animal rationality which is executed when all the environmental, social and temporal considerations are done. Animals which in general are seen as honourable have this kind of rationality; they react upon the precise current needs, which make them both ‘efficient’ and ‘naïve’. Thus we still regard the assumption of individual optimization behaviour as a valid tool for microeconomic analysis. To assume agents consciously sacrificing achievable welfare improvements is indeed a remarkable assumption of deeper philosophical significance than assuming temporal and local rationality. We suspect that an assumption of non-rationality is due to implicit norms by the analysts. Optimization is, however, a sort of temporal and local equilibrium, but since we reject the additive aggregation there exists a multitude of partial and temporal states of equilibrium where the aggregate situation is more or less stable. These states are attached to specific sets of epistemic cycles and may be contradictory or even contrary to other epistemic cycles, and may also internally contain actual or potential contradictions although they appear stable, since normally any epistemic cycle underlying a complex decision is to be seen as an aggregate of epistemic cycles. Thus, eventual local and temporal states of equilibrium in a mathematical sense may very well exist but are normally dissipative states of more or less inertia. We have in the preceding chapters rejected two central results deducted from the axiomatic structure of the neoclassical theory: Say’s law in all its forms and additive aggregation, both closely linked to the reflexivity axiom. Furthermore we have discussed genuine uncertainty in earlier chapters and come to the conclusion that it is difficult and even not possible to attach a probability distribution to future sets of epistemic cycles. From a company’s point of view this implies that investments in inert capital and organizations are risky and must be protected with appropriate measures.
