Jacques Sapir’s recent books include Les trous noirs de la science économique. Essai sur l’impossibilité de penser le temps et l’argent (2000) and Pays de l’Est vers la crise généralisée? (2000)
Comment on Bernard Guerrien’s Essay
For Guerrien… and beyond
Two Perspectives to Guerrien’s Question
Towards a New Economics
Yes, There is Something Worth Keeping in Microeconomics
Can we please move on? A note on the Guerrien debate
Once Again on Microeconomics
Bernard Guerrien’s provocative paper in PAE Review no. 12 raises some important issues. Along with other heterodoxical French economists I am greatly indebted to Guerrien’s frequently illuminating work about neoclassical economics. However, I feel that this particularly recent contribution was ill-conceived and perhaps pointless.
I agree with Guerrien that standard microeconomic assumptions are not relevant. I too criticized them in my Les Trous Noirs de la Science Économique. In addition, I agree that the Arrow-Debreu model has nothing to do with a decentralized economy. However, part of Guerrien’s argument reveals an unhelpful bias against abstraction itself. One of Spinoza’s famous arguments is helpful here. Spinoza taught us that the concept of dog doesn’t bark nor bite but the concept of dog is nevertheless necessary to understand a world where real dogs can do both.
Guerrien’s misplaced bias against abstraction is very clear when he states that Akerlof, Spence and Stiglitz have no new findings and have simply restated old ideas. Actually what these Nobel Prize winners did was to restate certain ideas at a higher level of abstraction. To return to Spinoza’s argument, real dogs may have predated the concept of dog, but such a concept was still necessary and useful in the development of human culture. Unfortunately, the way Guerrien orients his quest for reality throws us back to the nomilaism which pertained pre-Ockham.
In addition, rejecting Stiglitz’s contribution to the future demise of the current orthodoxy in economics, be it his theoretical one or his political economy one, will not help to achieve the needed breakthrough. The PAE movement would probably not have been possible without Stiglitz’s theoretical and political work in the last decade.
To acknowledge the positive effect of Stiglitz’s work does not mean agreeing to accept it as offering the limits to change. What is needed is a shift from the information paradigm - plainly positivist - to a new knowledge paradigm. The best way to move forward on this path is probably to destroy, step by step, old microeconomics, and not just to forget them. We now have the possibility of opposing new arguments to every argument voiced by the neoclassical traditionalists. This is particularly true when we compare the traditional way of thinking about individual preferences to what we have learnt during the last two decades (Slovic’s preference reversal or Tversky’s framing effect). It is also true of our thinking about rationality and, last but not least, about the necessary of making a distinction between signals and information. Actually the information Stiglitz is so fond of does not exist as such. Information is always a processed signal, which means that knowledge predates information and that knowledge is at the center of the paradigm, not information. And don’t think that this is just a matter of terminology. If we acknowledge the centrality of knowledge, then trying to collect as many signals as possible can be as detrimental to decision-making as being starved of information. Increasing competition, which is Stiglitz’s cure for most of our problems, would then mean a considerable increase in signal gathering that could overwhelm our signal processing ability and lead to bad decisions and imperfect resources allocation.
It is probably important for me to state explicitly that I do not belong to the methodological individualism school. I think we have to reclaim holism but in a non-deterministic framework. To understand that the individual agent is part of a whole does not tell us how and why this agent is making specific decisions. And, if we need and want to understand how agents decide to understand interactions and transactions, then we need to devise some radically new kind of microeconomics. Guerrien’s nihilistic stand here does not help us to move forward.
Neoclassical microeconomics has to be taught (in an abbreviated form of course) to allow students to move forward, but also to equip them to understand the language the autistic economist uses. After all, you could never understand history of the XIIth to XIVth Century without a good knowledge of the Catholic dogma, which does not mean you have to believe in it, or even in God.