387.
The programme of economics
Here I start the series on economics that I announced before.
As in other sciences, in economics there are diverse
schools of thought: neo-classical, which is the mainstream, evolutionary, institutional
and post-Keynesian. Here, I give an introductory survey. In subsequent items, I
will discuss things in more detail. First, I will focus on the mainstream.
How does one represent
a school of thought? Imre Lakatos[i] proposed
the notion of a research programme. That
has a ‘core’ of fundamental principles, assumptions and directions for
research, which must be protected from falsification at all costs, by means of
a ‘protective belt’ of subsidiary assumptions that supplement or implement the
core principles. When something comes up that falsifies the whole, it is
attributed to the subsidiary assumptions, and a replacement is sought there to
make the core work better.
Isn’t such tenacity to a
core unscientific? There is an argument for it. If something has performed well
you will not give it up at the first sign of imperfection. That is an economic
argument, but also an epistemological one. All theory is abstraction, imperfect
and incomplete. It is by sticking to a programme, and milking it for all it is
worth, that one discovers where its real limits lie, and finds indications for
improvement. It does constitute a form of conservatism, but if someone is unsatisfied,
he can start a new, competing programme. And that is again an economic
argument: the argument for competition. As I proposed in this blog:
imperfection on the move.
This conservatism can
derail into dogmatism, and that happened to mainstream economics. One reason
for it is that unlike natural sciences, the falsifiability of economics is
dubious. I will discuss that in another item.
There are two dominant
characterizations of economics: optimal allocation of scarce resources, and exchange
(through markets). In mainstream economics, the core assumptions are: rational
choice by autonomous agents, in the calculation of optimal choice, and the
operation of markets that yields equilibrium between supply and demand. The
protective belt gives subsidiary assumptions of legal conditions (e.g of
ownership), technology of production, infrastructure, the role and availability
of information, etc.
Research seeks ‘forbidden
events’ (falsifications), to repair and improve the subsidiary assumptions. The
process entails what Thomas Kuhn called ‘normal science’, solving puzzles
within the programme, such as, in economics, finding yet another, more
sophisticated ‘production function’ to model production technology.
Game theory brought a
major transformation, as a tool to model strategic interaction between agents,
with the central notion of a Nash Equilibrium: an outcome of interaction that
is stable, in that every player wants to maintain its present strategy, as the
best in view of the strategies adopted by the others. It is a useful device,
but the basic principle of optimal choice remained the guiding principle. It is
assumed that the set of strategies players can chose from, as well as the
‘pay-offs’ of combinations of the strategies of players, are given.
A later, more
fundamental change, towards ‘behavioural economics’ allowed a relaxation of the
principle of optimal rational choice, in allowing for decision heuristics from
social psychology that are not substantively rational, not yielding optimal
outcomes, though they may be rational in the face of conditions. A
methodological advantage was that experiments could be made in laboratory
settings, often with students of the researcher. However, it sits somewhat
uneasily in the research programme of mainstream economics, as a more or less
separate appendage. In a collaboration in this between economists and applied
psychologists, a complaint of the latter is that the former cannot desist from
forcing the heuristics back into optimal choice.
In economics, a
distinction is made between risk and uncertainty. With risk one knows what can
happen, and one can then append probabilities to calculate optimal choice, such
as the one with highest expected outcome (outcomes multiplied with their probabilities).
Under uncertainty, by contrast, one does not know all that can happen: that is
not given prior to choice but emerges after choice, in action.
There, economists stand
empty-handed, cannot ply their trade of calculation, so they ignore or neglect uncertainty.
However, in innovation and relations uncertainty is routine. Radical innovation
is uncertain, and the most fruitful relations are the most uncertain: they
yield the surprise of novelty that goes beyond present insights. One engages
most fruitfully with others who bring in things one could not before have
imagined. One does not know in advance even of oneself how one will respond to
unforeseeable events. There, also game theory falls short. In technical terms: there
is no longer a matrix of strategies of agents and the values of outcomes of
their combinations, because those mostly emerge during strategic interaction.
Here, economics short-changes
itself. If the most valuable relations are the most uncertain, incalculable in
advance, then shying away from them for that reason foregoes opportunities for
value, which is an uneconomic thing to do.
An exception is Keynes, who
recognized uncertainty and used it to explain herd effects in the economy that
produce booms and busts: if one cannot calculate optimal choice, then it is
reasonable to follow choices others are making, especially others who claim
that they know what they are doing, even if they don’t. Keynesian economics has
not replaced the mainstream, but it remains as a minority stream, in
post-Keynesian economics.
Mainstream economics is
focused on optimal outcomes, not on processes that may or may not yield those
outcomes. Evolutionary economics recognizes uncertainty and models economics as
an evolutionary process of more or less random initiatives, selected by markets
and institutions, and transmission of what survives.
Institutional economics
recognizes that markets do not work automatically and require institutions to
function, and, more fundamentally, deviates from the assumption of autonomous
individuals, recognizing that they develop their knowledge and their preferences
in interaction with their environment.
papers volumes 1 and 2, J. Worrall and G.
Curry (eds), Cambridge University Press.
No comments:
Post a Comment