Tony Lawson & Hashem Pesaran Keynes’ Economics: Methodological Issues Croom Helm 1985.
Edited papers from a conference organized by the Cambridge Journal of Economics to mark the centenary of Keynes’ birth, 1983.
2. Persuasion, Expectations and Limits to Keynes
This notes the important role of institutions in the formation of expectations.
3. Animal Spirits and Rationality
Alexander and Sheila Dow
If individual entrepreneurs are viewed as acting according to reason … then ‘in principle’ it may be possible to capture that reasoning in a complete theory of long-run expectations, endogenising animal spirits. [If] all decision-making is subject to the exogenous influence of expectations shifts, economists must retreat into nihilism. [The] narrow use of the rationality concept employed within the orthodoxy rules out the possibility of the type of indeterminacy implied by animal spirits.
Animal spirits in their spiritual home
Perceptions of how expectations are formed are themselves not objective, but the product of a particular way of looking at the world. … The mode of perception and analysis on which Keynesian theory is built is quite different from the all-encompassing closed system derived deductively from a set of axioms which is the basis for orthodox theory. [An] alternative theoretical structure can be built using a variety of logical chains (expressed with varying degrees of formality depending on the content) with different starting points, and taking different parts of the system as endogenous. … Thus it was in a sense from the traditional Western mode of scientific thought that he had his ‘long struggle to escape’.
Kants’ economic work is underpinned by his Treatise on Probability, which recognizes that, even ‘in principle’, there is no possible ‘reason’ that could be ‘captured’ in a complete endogenous theory. Yet his theory is not at all nihilistic, offering a way for governments to avoid or escape from depressions. As the Dows observe, the problem – which Keynes addressed – lies with the narrow, orthodox, concept of rationality.
It is also worth noting that ‘the spiritual home’ that the Dows identify, above, is one of complementarity. Thus Keynes has much more in common with modern science (post Whitehead) than the traditional practices that he criticised, and which are still common today.
5. The Slippery Transition
Johannes J. Klant
The falsifiability of Economic Theories
Theories which merely yield tautological non-predictions … may be useful as consistent systems of definitions, boxes of tools and elastic supports to world pictures. They comply perfectly with the idea of economics as a kind of formal system, a branch of logic which enables us to talk about the world in a specific and consistent way. …
But most economists … assume that the structural parameters are ‘stable’, that is, more or less constant over time, at least during a certain period within a specific area.
‘Economics is a science of thinking in terms of models joined to the art of choosing models which are relevant to the contemporary world.’
Economics as Art
The economists do not land in worlds unimaginable through their research. They always stay at home.
[Relations] which are characterised by universal numerical constants … are missing in basic economic theories. It makes them non-falsifiable. That is presumably not a symptom of immaturity but reflects the nature of the domain, where ‘motives, expectations and psychological uncertainties’ reign. Historical change in the economists’ domain is much more rapid than in those of the natural sciences, biology included. Human beings learn and create, they change their world all the time in an unpredictable way.
Considering econometrics and forecasting:
We should handle them for what they are, with scepticism, modesty and care.
The editors regard this chapter as being anti-Keynes. But if we distinguish general theories from specific models, then it is no error if a general theory does not provide what is wanted from a specific model. What it does do is to put models into context. This is not a simple matter. According to many, one should establish the best model that one can, and then to act as if one believed in it. The value of Keynes’ theory, as with Klant’s remark above, is that it points up the folly of such a ‘pragmatic’ approach. Keynes actually goes much further, in showing how orthodox rationality can break down, as in financial crashes.
7. Keynes, Prediction and Econometrics
This provides a good, relatively accessible, overview of Keynes’ approach to modelling and their interpretation. The focus is on induction.
“[The] question as to whether a particular hypothesis happens to be propounded before or after their examination is quite irrelevant.”
“[We] can only define [a ‘remarkable occurrence’] as an event which before its occurrence is very improbable on the available evidence.”
Inductivism v Falsification
[For] Keynes, the ‘testing’ of a hypothesis has more to do with putting into effect an economic policy that is based upon it.
Keynes’ Comments on Tinbergen’s Method
“If we were dealing with the action of numerically measureable … we might be able to use the method of multiple correlation with some confidence … .
In fact we know that every one of these conditions is far from being satisfied … .”
“In [the] natural sciences the object of experiment is to fill in the actual values of the various quantities and factors appearing in the equation or formula; and the4 work when done is done once and for all. In economics this is not the case, and to convert a model into a quantitative formula is to destroy its usefulness as an instrument of thought.”
Final Comments and Conclusion
If Keynes’ account, or indeed any analytic/a priori account, is adopted then there must, it seems, be some intuition provided as to why, in the given context, the ‘material can be treated as constant and homogenous’ through time. … Among other things it is difficult to see how by this account a research programme can avoid being committed to investigating the real possibility of frequent ‘structural breaks’ given any preconceived theoretical framework.
The problem is that if one draws upon too rich a family of descriptions then one of them may fit the data quite by chance. The more conventional approach is to split the data, find a good fit to one part, then check the fit against the other part. Keynes is concerned with complex systems, such as economies, for which this conventional approach yields little, if any, confidence that the model will remain valid. In contrast, Keynes looks for some general theory, with some a priori grounds for supposing it to be probable. Then a fit to the evidence gives some grounds for confidence in the theory, and that it may survive a change in policy or other shift.
8. Keynes on Econometrics
Hashem Pesaran and Ron Smith
Keynes’ judgement – ‘I have a feeling that Professor Tinbergen may agree with much of my comment, but that his reaction will be to engage another ten computers and drown his sorrows in arithmetic’ – was to prove true for econometricians in general … .
Keynes’ Critique: the Methodological Issues
While he [Keynes] welcomed ‘numerical and diagrammatic methods’ for the purpose of summarising the important features of large bodies of data, he strongly warned against mistaking ‘the statistical description for an inductive generalisation’.
Action and Induction
Econometric practice expanded … because it happened to fill a vacuum that was created by the need to formulate and implement the type of interventionist policies that Keynes himself had advocate. As a tool of economic analysis, Tinbergen’s method serves an important purpose and it is unlikely that it will be discarded. The challenge lies in recognising it limitations and in using it with care.
How true. And how much these insights were needed circa 2005.
9. Keynes on Cause, Chance and Possibility
Carabelli draws on the collected works and early drafts of Keynes’ Treatise on Probability to shed some light on Keynes’ epistemic thinking.
[Keynes] wrote that the English empiricists,
‘have entirely lost sight … of the relational nature of the conception [of probability]. [They] have sought in probability a quality belonging to the entities of phenomenological experience and have imagined that events have probabilities just as men belong to nations. This realist view has been one of the most dangerous disillusions in the past and is not even now eradicated from English philosophy.’
Chance: in contemporary science
‘The statistical result is so attractive in its definiteness that it leads us to forget the more vague, though more important considerations which may be, in a given particular case, within our knowledge.’
Probability as organic
Writing of probability:
The ‘objective quality measured may not … posses numerical quantitativeness.’
[Keynes] seemed to see his work on probability as a sort of anti- Discours de la methode, based on probability, ordinary discourse and common sense rather than on certainty and analytical reason. [His] work appeared as a general approach to cognitive procedures involving uncertainty both in everyday experience and in the moral as well as the natural sciences.
‘As living and moving beings, we are forced to act … Yet we must be guided by some hypothesis. We tend, therefore, to substitute for the knowledge which is not attainable certain conventions.’
In A Treatise on Money (1930), discussing the cause of the depression, he maintained that it was neither inevitable nor a result of a natural force or an accident!
‘If then, these are the causes, was the slump avoidable? And is it remediable? The causes to which we have assigned it were the outcome of policy, and in a sense, therefore, it was avoidable. Yet it is evident that the policy could not have been radically different, unless the mentality and ideas of our rulers had also been greatly changed. That is to say, what has occurred is not exactly an accident; it has been deeply rooted in our general way of doing things.’
Carrabelli, like many others, seems to regard Keynes’ approach as not being mathematical, logical or analytic. Yet the Treatise was approved by Whitehead for a mathematics Fellowship. It would be more reasonable to say that Keynes’ approach was not conventionally mathematical, logical or analytic, but reflected the more modern approach of Whitehead and Russell. For example, Keynes did not think that probability is always ‘just a number’, but that doesn’t make him un-mathematical.
The first and last quote from Keynes, above, seem appropriate to the Great Depression 2008-.