Conclusions

Conclusions
The fundamental building blocks of Post Keynesian theory are: (i) the nonneutrality
of money; (ii) the existence of non-ergodic uncertainty in some
important decision-making aspects of economic life; and (iii) the denial of
the ubiquitousness of the gross substitution axiom. One way that humans
have coped with having to make decisions where pay-offs will occur in the
unforeseen and possibly unforeseeable future is via the development of the
law of contracts by civilized societies, and the use of money as a chartalist
means of discharging contracts. The abolition of slavery makes the enforcement
of real contracts for human labour illegal. Accordingly, civilized society
has decided not to permit ‘real contracting’ no matter how efficient it can be
proved to be in neoclassical economics.
Keynes’s revolutionary analysis, where money is never neutral and liquidity
matters, is a general theory of an economy where the complete unpredictability
of the future may have important economic consequences. By contrast, neoclassical
optimization requires restrictive fundamental postulates regarding
uncertainty and hence expectations regarding future consequences that Keynes’s
analysis does not. The analyst must therefore choose which system is more
relevant for analysing the economic problem under study.
For many routine decisions, assuming the uniformity and consistency of
nature over time (that is, assuming ergodicity) may be, by the definition of
routine, a useful simplification for handling the problem at hand. For problems
involving investment and liquidity decisions where large unforeseeable
changes over long periods of calendar time cannot be ruled out, the Keynesian
uncertainty model is more applicable. To presume a universe of discoverable
regularities which can be expected to continue into the future and where the
neutrality of money is therefore central (Lucas, 1981b, p. 561) will provide a
misleading analogy for developing macro policies for monetary, production
economies whenever money really matters and affects production decisions
in the real economy.
Economists should be careful not to claim more for their discipline than
they can deliver. The belief that in ‘some circumstances’ the world is
probabilistic (Lucas and Sargent, 1981, pp. xi–xii), or that future prospects
can be completely ordered, will tend to lead to the argument that individuals
in free markets will not make persistent errors and will tend to know better
than the government how to judge the future. Basing general rules on these
particular assumptions can result in disastrous policy advice for governmental
officials facing situations where many economic decision makers feel
unable to draw conclusions about the future from the past.
However, if economists can recognize and identify when these (non-ergodic)
economic conditions of true uncertainty are likely to be prevalent, government
can play a role in improving the economic performance of markets.
Economists should strive to design institutional devices which can produce
legal constraints on the infinite universe of events which could otherwise
occur as the economic process moves through historical time. For example,
governments can set up financial safety nets to prevent, or at least offset,
disastrous consequences that might occur, and also provide monetary incentives
to encourage individuals to take civilized actions which are determined
by democratic processes to be in the social interest (Davidson and Davidson,
1988). Where private institutions do not exist, or need buttressing against
winds of true uncertainty, government should develop economic institutions
which attempt to reduce uncertainties by limiting the possible consequences
of private actions to those that are compatible with full employment and
reasonable price stability.