Saturday, September 11, 2010

Nature and us

(Translated from "La nature et nous")

We can see Nature in at least two different perspectives:

- We think it is the physical world, whose order obeys immutable laws and that is external to human knowledge and action;
- Or we think Nature is the “state of things", that is to say everything that confronts our intentions and desires as an obstacle or a tool, and that is transformed by the expansion of our knowledge and by our action.

The first view considers only the physical nature whose complexity is a challenge: our knowledge, cutting a finite sphere in an unlimited space, can never be complete and absolute. Physical nature is as unknowable as God in Judaism.

The second point of view adds to the physical nature the sphere of our present knowledge, skills and artifacts, a sphere whose gradual extension transforms the obstacles and tools that nature presents us. The art of navigating transforms the ocean, a road transforms a wilderness, building a house transforms the ground, our waste pollutes soil, water and atmosphere.

This later point of view transforms the concept of Nature in including, in addition to physical nature, human nature and social nature. It is in this sense that Durkheim said “we must consider social facts as things".

Assume there is a major oil field under the ground of a country. This is a physical fact, hence independent of knowledge. But the day this field is discovered, this country is facing new opportunities and new dangers: in this sense, we can say that for this country nature has changed.

The physical properties of silicon are what they are and have been for ever. But when we learned to build integrated circuits we discovered a world of new possibilities: it has transformed nature insofar it is offered to our action.

*     *

Economic theory is build on three pillars: the utility function expresses the needs, the initial allocation distributes ownership of resources, the production function allows to transform the resources in useful products. Then this theory expresses how we can “do best with what we have", manage production and exchange in such a way that utility is maximized (“Pareto optimum"). Economic theory is a theory of efficiency.

It branches out depending on how one specifies the three pillars, and on additional assumptions that must be chosen in order to explore the diversity of situations : thus unfolds from a simple starting point a tree whose complexity I do not intend to explore here.

I just want to emphasize one fact: economy theory starts from a given state of natural resources, obstacles and tools or, if one accepts my vocabulary, from nature considered under the second point of view above : it comes to do as best as possible with what is at hand, once “what is at hand" given. But it is bad equipped to consider changes in the hand.

Let us consider the case of an oil field. Economic theory can say how to organize a desert country, deprived of natural resources and populated by Bedouins who live in tents. It knows, too, how this country can take advantage of oil once oil is discovered. But it is poorly equipped to explain the transformation this country suffers immediately after the discovery, and it will take time to passe from the first model to the second model – a delay that can be very long if the development of the second model is hindered by sociological or intellectual obstacles.

This theory knows what to do with a static nature, but it is uncomfortable to consider a changing nature – and if it has some dynamic growth models like Ramsey's model, or a model of endogenous growth like Romer's model, all these models are based on a static conception of nature.

*     *

Nature changed in the seventeenth century England: the emergence of experimental science opened new territories to action, and the production system was progressively mechanized. Economic theory, entirely present as a seed in the work of Adam Smith, is nothing else that the intellectual expression of a mechanized society in search of efficiency. This society and this theory absorbed thereafter the improvements which have been brought by chemistry and electricity.

Hence economic theory was from the beginning connected with a technical system (see Bertrand Gille) based on the synergy between mechanics, chemistry, electricity and oil: it has explored its implications and followed its development.

But now comes a new technical system, based on the synergy between microelectronics, software and the network – or, put another way, here comes the alloy of the brain of the human being inserted in an organization and the ubiquitous programmable automaton provided by computerization.

The emergence of this alloy changes the games, modifies the conditions of practical action and brings new dangers: in short, it transforms the nature. But it is disconcerting for the economic theory, which has been constructed in order to express a different nature. Then economists do not know how to think this emergence, nor how to draw conclusions.

*     *

Theory experienced in the 1930s a similar confusion: while it had previously overlooked the uncertainty inherent to the future, it was suddenly forced to assimilate it: this was the contribution of Keynes, whose work is essentially a critical analysis of anticipations.

The current disarray is event deeper and more radical, because the fundamental of the theory themselves are changed. The initial allocation, the utility function, the production function are no more expressed, in a computerized society, as they were in a society whose technical system was based on mechanics, chemistry, electricity and oil.

For sure, the distribution of expertise and ability to organize was already part of the initial allocation, but now it is preponderant. Admittedly, the quality of products was already one of the arguments of the utility function, but it roses to the forefront. Certainly, the production function contained almost always, for small quantities, an area of increasing returns (sometimes large enough to determine a natural monopoly), but automation has expanded this area so that at equilibrium monopolistic competition, formerly a theoretical curiosity, supplants presently in almost all the sectors the old canonical pair formed by perfect competition and monopoly.

The consequences of computerization are blindingly obvious: deformation of the structure of employment and of international trade, lower transportation costs, globalization, financialization, outsourcing, hybrid products (now packages of goods and services) and hybrid production (done by several firms organized in partnerships), crucial role of the information system, last extreme violence of competition associated with the emergence of a large scale predation.

Of course the economists see these consequences, and they describe them in detailed monographic works. But they do not relate them to the cause that explains them all, they do not go back to the change of technical system that is the spring of the consequences. They just describe rather than explain and, when they try to explain, they obey a professional sociology that compels them to search exclusively in the vocabulary of economics the explanation of everything in economy (as well as sociologists claim to find in sociology itself the explanation of everything that is sociological).

Therefore they will try to explain the phenomena they observe by economic conditions, the economic cycle, the behavior of economic agents, and not by a transformation of nature that modifies the conditions of practical action, the opportunities and the risks.

They will thus explain the financial crisis by greed, fear, excessive self-confidence, moral collapse, pursuit of immediate benefit etc. (Hersh Shefrin, Beyond Greed and Fear, Oxford University Press, 2007).

But the do not explain why such behaviors have emerged: this would have required to identify the material cause which, having first made them possible, made them eventually mandatory, those who do not adopt these behaviors being ejected from their profession.

This material cause is the computerization of finance: it has unified the world market, facilitated automation and devalued all human control, unbalanced risk-return arbitrage by reducing the perceived risk, and it resulted a foolish performance race. What followed is history.

*     *

Economic theory can probably overcome its own limits: it would suffice for it to assimilate the historical approach of Bertrand Gille (Histoire des techniques, Gallimard, La Pléiade, 1978), therefore to assimilate the transformations of nature, natural resources, which result from the transition from a technical system to another.

But the corporation prefers so far to stay in its familiar rut, as did the economists of the 30s when they refused to listen to Keynes. Talk indeed to French economists of computerization, and see how they are contemptuous. "This consists in petty technicalities", they say (in their language, whatever is "technical" is "petty"), because they think to see things from above, and more broadly, than do "technicians" with ancillary knowledge.

In the 1980s, at a conference at the Ecole des Ponts et Chaussées in Paris, I said that the network was becoming a marketplace. "There's nothing new here, declared one of the most distinguished economists: businessmen already use the phone for a long time." In the 1990s, the Commissariat général du Plan refused to launch a study on computerization: "computers belong to technics and not to economics", said his experts.

Yet another anecdote. Speaking one day of the production function, I was interrupted by an economist bearing a famous name. "The production function," he said with a scornful pout, is an engineering concept, it's not an economical one. The only thing that matters in economics is supply and demand." Wanting to know only the partial equilibrium, he probably ignored that supply and demand are obtained by differential calculus from the three basic elements of the theory mentioned above – and therefore they depend, in particular, of the production function.

The condescending contempt of these economists towards everything "technical" is for them a handicap to both sociological and intellectual points of view: it prevents them from understanding and thus explaining the phenomena they observe.

*     *

For sure, the term "new economy" makes knowledgeable people shrug: the speculative bubble that formed in the 1990s around the Internet has made many dupes and this has left its mark. But that would be a shame if the accident which gave such a bad reputation to "new technologies" prevented us from seeing the emergence of a really new economy and of taking the bearings needed to act effectively.

This involves making the effort to understand computerization and not be content with the superficial knowledge that is gained by using a computer: we need to delve into the contents of the process of computerization, information systems, network architecture, the art of programming etc.. It's a real work and it is in a sense understandable that economists are reluctant: having had to work hard in order to understand their own discipline, they think they know enough to speak with authority.

Well, this is not the case. How they use the word "numérique" (French for digital) to refer to the computerization shows that they do have it all wrong. Although it is true that all documents and all programs are translated into 0 and 1 to go through the processor, the "numérisation" which takes place well in the lowest layers of the machine does not summarize the operation of any upper layer that computerization also includes: operating system, compilation, programming language, applications, business processes, semantics – and one must also add layers (psychology, sociology, values) and geopolitical layers. The word "numérique" masks the reality of computerization and bend too conveniently at intellectual laziness.

The economists must get to work! They must attentively read Histoire des Techniques by Bertrand Gille (out of print and not translated into English, which is a pity), Hackers by Steven Levy, The Soul of a New Machine by Tracy Kidder; they have to read De l'informatique (this book was written for them), to immerse in The Art of Computer Programming by Donald Knuth, to learn to reason in a layer model etc. : Thus they will gain gradually the knowledge they need to be able to develop the economical models necessary for a computerized society.

And then, among them, will perhaps finally arise the Adam Smith or the John Maynard Keynes who will adequately express the new economy.

No comments:

Post a Comment