Return-Path: Delivered-To: andy@mira.net From: JulioHuato Date: Mon, 8 Dec 1997 14:51:47 EST To: andy@mira.net Subject: Re: economic theory Organization: AOL (http://www.aol.com) Dear Andy: Regarding, "what is needed," do we really have the foresight to determine the extent and depth of the economic science necessary to overthrow capitalism? Scientists complain about pressures to do more "applied" and less "pure" science. Funders invest because they anticipate direct benefits (eg, Citicorp). But, if they only expect immediate benefits, they may be defeating their own narrow aims (Citicorp seems pretty relaxed in the story, but what's a million more or less). Revolutionary movements, under much harder constraints, may face similar dilemmas. I must admit that I'm not immersed in the revolutionary movement, nor have I clear criteria, to venture a judgment on this. At a purely speculative level that you'll excuse, I lean towards giving leeway to science vis-a-vis the "current needs of the movement." My knowlege of the history of socialism is superficial. You'll have more references on this. In a sub-chapter included in Mehring's biography of Marx, R. Luxembourg wrote that Marx's work had a depth that exceeded by far the practical needs of socialism at the time. In Deutscher's biography of Trotsky, the author says that the political triumphs of the II International (particularly, the German branch) before WW I gave its leaders the impression that they didn't need to go beyond vol. 1 of Capital. Deutscher implies that such view exacted a dear cost from the International and the workers movement. To be honest, it is unclear to me how ignoring vols. 2 and 3 connects to the failure of the II International. Theorizing is a means to an end, but why should we route enquiry to make it avoid mathematical modelling? In terms of content, mathematical models are not better, per se, than "discoursive" economics. Mathematical symbols and operations are "mere abbreviations for words" (Phillip Jourdain). But, precisely, abbreviations are shorter than words. The reason we may want to use mathematical models is to make our theories more compact and economical, easier to handle. As long as we are aware of assumptions and of the fact that mathematical "grammar" embeds formal logic, we are safe. Risks are involved, but mathematical models may end up being necessary for the very reasons you allude. As to access and resources, high quality data (the same used by the WB, governments, etc. to run their models, just a bit outdated) is available gratis in the Internet. For example, the Penn World Tables and a wealth of data in the site of the NBER. WORLD AND NON-U.S. DATA . When it is not free, the data is relatively unexpensive. Universities have access to the best data banks (including Citibase, Citicorp's data bank), and it's easy for almost anyone to download data at will. As to the computers necessary, you have a point here. Models like those ran by the econometricians in the Federal Reserve Bank or Citicorp are certainly huge. However, what PC's can do now shouldn't be underestimated. They have increasing power and speed to deal with huge databases, and the trend is encouraging. I would believe that the bottleneck is not on the side of material means but on the side of people. What kind of economics is required for workers managing a transition economy? "[T]he technological revolution taking place is transforming the presuppositions in relation to economic planning, workers control and the self-organisation of the working class and the relation between them." I think this is a very interesting theme. Let me pose these questions. What's the character of such transformation? Is it making it harder or easier for workers to build socialism? Where easier, where harder? There's also a broader question, What is the role of economic science in the movement against capitalism? What does it need economics for? For one thing, it should "inform political strategy" ... and tactics. Class struggle takes place in real time and contexts. You better know where you stand at. Economic analysis at a very concrete level would be required. It'd also help to wage the war of ideas against the ideologues of the status quo. More concretely, in the fight to recruit the best minds for the cause, a clear intellectual edge over the theories of the adversary tends to weigh heavily. And you mentioned (in your article on planning & workers control) that the movement needs theorizing about the transition economy even before the workers seize power, which makes a lot of sense to me. As to "equilibrium." Let me try to state my ideas a bit more clearly. The essential purpose of Capital is to "throw light on the historical tendencies." Or, in Marx's words, it is to uncover the "laws of motion" of capitalism. In his exposition, he starts with the "commodity," which leads to commodity production. Capitalist production is "generalized commodity production." Now, capitalism, as an economic system, is more than production. The production of surplus-value is its essence, but Marx goes on to show how things appear on the surface (circulation, and the synthesis of production and circulation, and more concrete phenomena assigned to further works). Once the secret of capitalist production is uncovered (unpaid labor as the content of surplus-value, surplus-value production as the dynamic law of the system), Marx goes on with accumulation, which only becomes evident when you change your "point of view" -- instead of looking at a simple process of production you look at its recurrence: "re-production." Commodity is use value and value, commodity production is use-value production and value production, capitalist production is production in general and surplus-value production. Likewise, capitalist reproduction is both reproduction in general and specific capitalist reproduction. The former is the reproduction of the material basis of society, any society. The latter is the reproduction of specific capitalist production relations. Two entirely different things. The study of the material reproduction of capitalism is relevant to economic science insofar as it interplays with the reproduction of capitalist relations. Marx does address this interplay in vol. 2 of Capital. Obviously, the critique of capitalism presupposes the historical viability of capitalism. If capitalism is unviable (if it cannot reproduce the basis of society), it becomes impossible and its critique is superfluous. So, capitalism does reproduce the basis of society (at least, it has done it so far), and for such reproduction to take place, it has to meet some conditions. By themselves, these conditions fall in the field of theory of history, technology, and engineering. They are only relevant to economic science because of the interplay I refer to. Again, these are the "material" conditions I refer to. Basically, they are conditions of "proportionality." Even under capitalism, society has to allocate its resources (ultimately, labor time) to create, replace, expand, etc. so that ultimately the needs are met continuously. Such allocation must be somewhat "proportional" or else the process will not go through. Enough machines should be manufactured to supply all industries, etc. I use the unfortunate word "equilibrium" imported from mechanics to designate this, because that's the term used in economics. Marx used "proportionality." He exposed the conditions for an stationary economy and for a growing economy (he had a bit of a hard time with the algebra in the latter case). In any event, he saw this as clearly as anyone could. However, modern economics has more than one thing to say about this. Marx's work and the work of Soviet economists heavily influenced by Marx evolved into a whole area of economics called input-output analysis, still in use. Moreover, there have been even more recent developments in mathematics that have taken things far beyond classical optimization (with calculus), linear programing, and research operations, eg, dynamic optimization, recursive methods, and optimal control theory, widely used in modern economic analysis. All this connects directly with your concerns on planning and workers control. The material conditions of reproduction must be fulfilled by any society, including a transition society, if it is to function. That's the substance of economic planning. My point about general equilibrium is that its results are useful on this regard. I'd add that complexity models could become helpful to illuminate the other aspects -- unstability and disproportionality (that feeds itself gradually and suddenly blasts in crises), etc. Now, we would still need a unifying framework. Regarding planning, let me reproduce extensively a passage in Grundrisse seldom quoted, relevant not only to the "pure" communist case, but also to transition economies: "On the basis of communal production, the determination of time remains, of course, essential. The less time the society requires to produce wheat, cattle etc., the more time it wins for other production, material or mental. Just as in the case of an individual, the multiplicity of its development, its enjoyment and its activity depends on the economization of time. Economy of time, to this all economy ultimately reduces itself. Society likewise has to distribute its time in a purposeful way, in order to achieve a production adequate to its overall needs; just as the individual has to distribute his time correctly in order to achieve knowledge in proper proportions or in order to satisfy the various demands on his activity. Thus, economy of time, along with the planned distribution of labour time among the various branches of productions, remains the first economic law on the basis of communal production. It becomes law, there, to an even higher degree. However, this is essentially different from a measurement of exchange values (labour or products) by labour time." (1993, Penguin, pp.172-173.) Since I'm in this theme, let me say that there are many related problems that only the experience of the Soviet Union, etc. could have raised. Let me mention "quality." It seems to me that on Marx's time, the "universalization of needs" (and their evolution, driven by competition) to which he refers was a slow process, particularly if we compare it to 20th century and postwar dynamism. The concept of "use value" contains the question, but there are edges that have later become very relevant. I'm not an expert on the late Soviet system, but everyone mentions that their planning system (maybe for the reasons that you mention in the article) had huge loopholes with regards to quality. Current capitalist obsesion with "customer service," "customer- perceived quality," etc. cannot overcome the fact that it's a means to the profit end. In a communist society, "customer-perceived quality" (and moreover, the alignment of customers perceptions with the objective fitness of products to do the jobs they are created to do) will be paramount. With regards to supply-and-demand analysis, I think you're right that such analysis is not excluded by Marx's theory. Actually Marx did sketch a supply- and-demand analysis (vol. 3, ch. 10). Had he published vol. 3 himself, I imagine, he would have done more than such sketch, which even in its raw form stands way above Mill's treatment. Question: Does a revolutionary movement need this extension? Are capitalists sharply interested, but workers have no vested interest in this dynamics? Again, we're back to the problem above ... "[T]here is room for some thinking about the relation between Hegel's logical system and the passage from mathematical model to emergent property." Could you be more explicit here? I think of Marx's (and Hegel's?) theory of alienation when I think of "emergent properties." For instance, in the case of "commodity fetishism," independent private producers who own and control their private production, when interacting through markets, find themselves immersed in a hardened reality beyond their control and turned against them. Money, as a social relation, is an "emergent property" of a commodity producing society. Now, this "emergent property" becomes objective, hard- wired to the extent that its basis (private ownership and independent production) stays put. Individuals may try to deny power to a piece of bank paper, but at their own risk. This is entirely different from the modern conventional concept of money as a convention. I may be wrong, but, in my mind, the computer model is an analogic device. It replicates (in a simplified fashion) the individuals, programs a setting and/or assigns "local" rules of behavior. Then the emergent patterns are identified by looking at the statistical results on the "aggregate." What analytical models (analytical in the sense of "mathematical analysis," as opposed to simulation models) do is to spell out (or try to) the mechanism by which such "local" rules of behavior translate into aggregate outcomes. The rules are the "axioms" and the aggregate outcomes are the "theorems." I suspect that the problem is that mathematical analysis is often a hostage of the rules of formal logic AND phenomena come in "levels," as you say. The "transformation of quantity into quality," the passage from level to level, etc. may get lost on the way. These things are not clear to me, and I'll read avidly all you may write about it. "* therefore, more often than not, so far as the "real" issues are concerned, such mathematical modelling is simply tautological proving what was assumed in the premises, even though there may be real value in terms of emergent phenomena along the way. "* it is a hopeless task to think that a mathematical model based on one level of phenomena can adequately model "input" from phenomena at another, or that one can deterministically compute phenomena at one level from those at another. * Every mathematical model has a material theory implicit in it and that should be explicit, not just implicit." All these are very good points. "* I suspect that the project of adequately modelling capitalist crises sort of assumes what has to be proved - capitalism would have to be abolished before we would have the material conditions to solve the problem." Which shouldn't keep us from trying it. This remark applies to all social phenomena. How concrete the mental appropriation (theoretical solution) of capitalist phenomena should go in workers minds before they are able to (and during) overthrow it? We're back to the problem above. Thanks, Andy. And forgive my verbosity. Julio ***************** Return-Path: Delivered-To: andy@mira.net From: JulioHuato Date: Mon, 8 Dec 1997 08:41:31 EST To: andy@mira.net Subject: Re: Complexity Organization: AOL (http://www.aol.com) Andy, I'm glad this reading had such effect on you. :) I also find Michael's article and the book very thought provoking, but I'm afraid I don't feel apt yet to comment on the issues systematically. I'll limit my comments to the areas where I feel more confident. "I worry that mathematics (like software) is a kind of drug. I am very prone to that temptation to abstract, simplify things so that you can do maths with it, and lose the complexity in fact." Consider this paraphrasis: "There's nothing more concrete than a good abstraction." I'll try to comment on this and other issues in my detailed response to your other note. J.