Monday, April 14, 2014

Substitutes or Complements: Marx and Brad and Me

by Suresh Naidu

Since Brad Delong has attributed some thoughts on Marx to me, and I have gotten some emails inquiring whether or not I did say them, I thought it would be useful to publically air what I understand to be the context. Brad has been a mentor and advisor for 10 years now, and is one of the very few economists broad and open enough to know anything about what Marx wrote. I have met very few other people in mainstream economics departments who cared enough to read Marx, let alone enough to assign the Communist Manifesto and Wage Labor and Capital in a mandatory first-year economics Ph.D. course (even I don’t do that).

The context of the long-running conversation has been trying to establish a dialogue between Marx and modern growth theory. Inside the modern production function there is a pretty undifferentiated view of “K” (which leads it into some troubles as bad as any in the labor theory of value). Marx on the other hand distinguishes (at least) machines, technology, and money-qua-productive-input as different from each other conceptually. The fact that these are rolled into an aggregate production function by mainstream growth theory is not Marx’s fault. And so when somebody is trying to translate Marx into modern economics, the slippage between what is “K” and “what Marx meant” can get confusing. I am still thinking about these issues, and won’t pretend to have figured it all out, but let me lay out some preliminary thoughts.

Brad in one sense is correct. I believed and still believe that Marx’s view of capitalist technological change is labor-saving. There is no way to read the chapters on machines and modern manufacturing without thinking that Marx was pretty convinced that technologically advanced machinery would displace labor and lower wages. 
But we have already seen that, with every advance in the use of machinery, the constant component of capital, that part which consists of machinery, raw material, etc., increases, while the variable component, the part laid out in labour-power, decreases.
But on the flip side, the increased use of machines can increase the demand for labor. More factories means more labor demand, even if those factories are 90% robot. So the relationship of capital and labor is both of these things: the net substitutability of labor by capital, holding output constant, and the gross complementarity of capital and labor while output changes.
We also know that in no, other system of production is improvement so continuous, and the composition of the capital employed so constantly changing as in the factory system. These changes are, however, continually interrupted by periods of rest, during which there is a mere quantitative extension of the factories on the existing technical basis. During such periods the operatives increase in number.
In terms of wages, Marx I think can be read as saying that that medium-run wages are pinned down by institutions and norms and the reserve army of labor. Long-run wages could be determined by political conflict and changing economic institutions, and short-run wages could be determined by the business cycle and labor-saving technological change. This is a pretty rich theory of wage-setting, and I think focusing on the technological change dimension of it is missing a lot of the full Marx story.

When "capital" is understood as money and financial services enlisted in circuits of production (the M-C-M' increments) or the quantity of investment, I think you can find passages where Marx thinks of them as complements to labor (i.e. they increase labor demand, even if marginal products don’t exist and wages are institutionally set). For example, here is a passage where (a young) Marx talks about capital and labor as complements, and perhaps even thinks of wages at marginal products.
To say that "the worker has an interest in the rapid growth of capital", means only this: that the more speedily the worker augments the wealth of the capitalist, the larger will be the crumbs which fall to him, the greater will be the number of workers than can be called into existence, the more can the mass of slaves dependent upon capital be increased.
In fact, rhetoric aside, this passage is completely consistent with, say, vintage optimal tax theory, which says that the best way to increase workers' crumbs wages in the long run is to not tax capital ever. And I think it is illuminating about the manifold criticisms Marx makes of capitalism. Perhaps Marx thought that even if workers’ standard of living increased under capitalism, it would not undo the exploitative/unfree/ undemocratic institutions inherent in the labor market. Maybe some of those criticisms are empirically wrong or don't resonate with contemporary ethical intuitions. In any case, it is not just any particular analysis of inequality and business cycles, but instead the unrelentingly political and conflict-ridden view of the economy that is what is precious to me in Marx. 

But all this said, I think Brad wrote a good column, even if the question of “Was Marx Right?” is fundamentally silly. The larger point is that Marx’s fertile mind generated many ideas, distributed over a lifetime of thinking and writing exactly as capitalism was transforming itself and the world. The theological endeavor to discern and police "what Marx really meant" is the worst kind of intellectual program. Marx, like Smith and Ricardo and Keynes, is a place to draw on for inspiration and hypotheses, not the last word on anything. So, at the end of the day, I don’t really care if “Marx was right”, and I certainly don’t begrudge others the right to find support for a variety of arguments on the invaluable www.marxists.org. The howls of outrage from the online guardians of the one true Marx in response to the NYT panel were more irksome than anything written by the panelists. Marxist economics is much larger than Marx, and textual exegesis of the original manuscripts, as opposed to reinventing and redeploying the concepts in new historical and intellectual environments, does it a great disservice. Brad is right to look to Capital  for inspiration to understand our possible robot future, even if Marxists don’t like it. To paraphrase Joan Robinson, the critics have Marx on their screens, not in their bones.

17 comments:

  1. If I were to misinterpret this post and then dismiss it as garbage, Suresh has a right to insist on a more accurate interpretation without me then dismissing him further as a zombie theologically obsessed with textual exegesis.

    Right?

    I really don't think that Joan Robinson quote means what everyone post-Mike Beggs thinks it means. Having Marx in one's bones did not mean "translate Marx into modern economics." Robinson is having a conversation that already assumed a rejection of neoclassical economics.

    I think it is intellectually dishonest to either dismiss Marxian economics or dismissively appropriate it (just optimal tax theory with a bunch of rhetoric) and then call Marxist economists who question your "creative" interpretation names.

    If I refer to contemporary neoclassical macro as Marxian economics with some mathematical rhetoric, or misinterpret/dismiss the theory, and a macro person responds by pointing out what neoclassical macro really is, do I get to call them zombies with a "theological endeavor to discern and police" what modern macro really is? Do I get extra points because I assigned a Brad Delong article to a class once? I really did! I'm super open-minded.

    ReplyDelete
    Replies
    1. 20 years before Mike Beggs, I attended an URPE conference or two and it quickly convinced me there wasn't such a thing as Marxist economics. There were a bunch of heterodox schools that disagreed about lots of things (except the bankruptcy of neo. econ) including some very fundamental things. So at least Beggs gives a convincing argument for why this is so, i.e., because Capital was a useful critique of earlier political economy but not a well-targeted critique of post-Marshallian econ (inevitably so) nor a full blown systematized economics (though lots of self-described Marxists seem to have faith that it is one or the other or both). And the tendency of some Marxists to believe they inhabit/possess a philosophically/methodologically superior framework is not helpful either, imo. All that said, I do understand that academic Marxist/ian economists are tarred, at times unfairly, with the same brush that political (sectarian) Marxists provoke.

      Delete
    2. Joe,

      Of course I agree. Any collective intellectual project requires a shared language, a consistent set of premises and methods of analysis. Marxism is just like any other specialized intellectual work in this respect, this nothing "theological" about it. Now, it is true that we on the heterodox side pay more attention to our classic texts. But that's not out of some kind of veneration, but just because we have more need of the classics as a coordination mechanism, because we don't have the institutional resources of the mainstream.

      I really don't think that Joan Robinson quote means what everyone post-Mike Beggs thinks it means. Having Marx in one's bones did not mean "translate Marx into modern economics." Robinson is having a conversation that already assumed a rejection of neoclassical economics.

      Right on. Somewhat relatedly, I was amused recently to find Joan Robinson telling Roy Harrod that "you are trying to turn Harrod's theory into a pre-Keynesian theory. I will not have this... You can criticise me for misunderstanding Harrod. But I maintain that I understand him a lot better than Sir Roy does." So I think the famous Marx line was also about Joan Robinson's confidence that she understood everyone else's economics better than they did. Which in her case, was probably true! but is a dangerous assumption for the rest of us to make.

      Delete
    3. "Having Marx in one's bones did not mean 'translate Marx into modern economics.'"

      Actually, I'm afraid it sort of did -- a project that Maurice Dobbs was doubtful about. But it is ironic, isn't it, that Beggs had to resort to "taking down the volume and looking it up" for the Robinson quote to make his point? (ha ha).

      Dobb to Robinson: "I think my trouble about this translation business you spoke of the other night is twofold. First, I feel that in the case of writers who have something important to say, it is much more important (at any rate, of prior importance in time) to learn their language than to translate _{_I’m assuming of course that by “translate” something more radical is meant than simply substituting a few words here and there, like utility for “use-value” etc}; and the more important the more one’s own conventional ways of thoughts and modes of expression diverge from theirs. Otherwise, unless one has really got inside t’other chaps head and his skin (ie felt problems within the context that he viewed them), one is almost sure to miss, or even distort, the meaning in translation, as is so commonly done by historians of thought. Secondly, I feel that the “poetic” element—shades of meaning inherent as it were in style, construction, emphasis—is important in all economic languages certainly, and perhaps in all languages outside the rarest and most refined discourses of Logical Positivists. This is just what gets lost in translation from one poetic convention into another. And most of it, I suggest, is not just irrelevant “moral” stuff, but is concerned with the slant that theory has on reality—with the completeness or incompleteness of the picture of the real world it affords, with the perspective and “projection” and dimensions it is employing, with what it throws into relief as causally important and what it relegates to the shade. Whether these meanings could or could not ultimately be reduced to a propositional system I don’t feel competent to say. But I feel quite sure they usually can’t be rendered in half-a-dozen or a dozen simple propositions. All of which means that I think I’d plump for Marx, inter alia, being approached and studied so far as possible in the original language, despite its inconveniences, rather than in modernised translation (without, of course, ruling out the virtue of translating special points for purposes of elucidation)."

      Delete
    4. Prue Kerr, "Joan Robinson and Maurice Dobb on Marx," Contributions to Political Economy (2007) 26, 71-92.

      Delete
    5. By the way, can't resist "taking down the volume and looking it up" one more time to find Lord J. M. Keynes himself vanishing into the swamp of the dreaded LTV: "I sympathise, therefore, with the pre-classical doctrine that everything is produced by labour..." http://econospeak.blogspot.com/2014/04/j-m-keynes-disappears-into-swamp.html

      Delete
    6. That Dobb quote is fantastic. Wow. I will be using that.

      Delete
    7. it is ironic, isn't it, that Beggs had to resort to "taking down the volume and looking it up" for the Robinson quote to make his point? (ha ha).

      This is a good point also. Obviously, one uses quotes like that for two reasons: First to avoid duplicating the work someone else has already done to express the point you want to make, and second, to signal membership in a particular intellectual community. People quote Marx for the same reasons Mike quotes Robinson in that piece. (Which I admit I did not like very much.)

      Delete
    8. I missed these interesting replies. I'll have to research the issue of translation between Dobb/Robinson more, but (for now) I still think it is true Robinson's willingness to "translate" into neoclassical language was much more limited than these invocations suggest. "Modernised language" is not modern economics. There is a huge gap between the modernized Marxism of Duncan Foley and the modernized Marxism (sic) of DeLong.

      Delete
  2. DeLong at Project Syndicate:

    "The economist Suresh Naidu once remarked to me that there were three big problems with Karl Marx’s economics. (...) And, third, Marx was fixated on the labor-theory of value".

    DeLong at NYTimes:

    "I have long thought that Marx's fixation on the labor theory of value made his technical economic analyses of little worth."

    So, DeLong has long thought what Suresh Naidu once remarked to him.


    Naidu: "The larger point is that Marx’s fertile mind generated many ideas, distributed over a lifetime of thinking and writing exactly as capitalism was transforming itself and the world."

    DeLong: "Thus he vanished into the swamp, the dark waters closed over his head, and was never seen again."

    No need to explain, Suresh, I only wanted to understand.

    ReplyDelete
  3. @Suresh Naidu

    "Marx’s view of capitalist technological change is labor-saving."

    This is certainly true in relative terms - Marx believes that in proportion fixed and constant capital will grow more than variable capital (i.e. labor).
    However it isn't obvious that this is true in an absolute sense - that is, that less labor will be used overall. It might just mean that more stuff is produced (what you called "more factories").

    DeLong in the article you link says that Marx didn't get this difference, and believed that labor would have not only a falling share of income, but a falling total of income in absolute ("real") terms. But I don't think this is true, Marx is quite ambivalent (plus he was also a polemicist, so I think that it is natural that he used the most aggressive possible language).

    A falling wage share of income is already a problem, expecially if one believes that workers have an higer propensity to consume than capitalists.
    Marx not only believed this, in fact he assumes that capitalism can't work in a "simple reproduction" regime (when capitalist consume all that they get). So I think that we can have at least three interpretations of Marx:

    1) an institutionalist/Piketty/Rodbertusian (gee! I just learnt about this guy: http://en.wikipedia.org/wiki/Rodbertus on wikipedia!) view where "capitalists" just get to eat most of the income, because they control the institutions; I don't think this is the correct view (both in terms of being what "Marx really meant" and in being useful in the world of today) but it seems the most fashonable today.

    2) A "Kaleckian" view in wich the falling wage share causes a crisis of underconsumption unless it is balanced by an increase in credit. I think this would be the most promising interpretation. But, what would be the added value of Marx? I think that, since Marx thinks in terms of LTV, he implies that capitalism, to be stable, needs the rate of profit to be higer than the rate of growth (since in LTV terms the rate of growth is always 0). But this is ultimately impossible. This means that capitalism will have period of growth punctuated by crises that "destroy" capital (but mostly financial capital). The role of policy then would be that of preventing crises to destroy real capital too.

    3) A "real business cycle" interpretation of Marx, where cycles are due to an increase of the amount of "real" capital, and financial and rentier capital just follow. When "real" capital reaches the point where additional capital goods are no more productive, we have the crises, because the additional demand caused by the production of additional capital disappears. I think that this interpretation would be closer to "what Marx really meant", but it is problematic because, while the concept of "real" capital makes sense under the LTV, it makes no sense out of the LTV.

    ReplyDelete
    Replies
    1. Marx believes that in proportion fixed and constant capital will grow more than variable capital (i.e. labor).
      However it isn't obvious that this is true in an absolute sense - that is, that less labor will be used overall. It might just mean that more stuff is produced


      Right. I think this is consistent with what Suresh wrote.

      in fact he assumes that capitalism can't work in a "simple reproduction" regime

      I don't think this is correct. Can you elaborate?

      falling wage share causes a crisis of underconsumption unless it is balanced by an increase in credit. I think this would be the most promising interpretation. But, what would be the added value of Marx?

      This is a very popular view right now. There may be some truth to it, but I think it rests on shaky foundations. And, as you suggest, there is a serious tension between this idea and Marx. Marx -- I think -- is clearly thinking of a world where demand constraints may be a problem at moments of crisis, but are not a factor in the long-term trajectory of the system.

      I think -- as Tom Walker (aka Sandwichman) suggests below, that we cannot understand what Marx really meant in purely "economic" terms. The qualitative transformation of productive activity is critical not just for the motivation but for the positive content of Marx's analysis.

      Delete
    2. RE: simple reproduction

      First, in the worst tradition of Marx debates:
      The entire character of capitalist production is determined by the self-expansion of the advanced capital-value, that is to say, in the first instance by the production of as much surplus-value as possible; in the second place however (see Buch I, Kap. XXII) [English edition: Ch. XXIV. — Ed.] by the production of capital, hence by the transformation of surplus-value into capital. Accumulation, or production on an extended scale, which appears as a means for constantly more expanded production of surplus-value — hence for the enrichment of the capitalist, as his personal aim — and is comprised in the general tendency of capitalist production, becomes later, however, as was shown in Book I, by virtue of its development, a necessity for every individual capitalist. The constant augmentation of his capital becomes a condition of its preservation. But we need not revert more fully to what was previously expounded.

      Capital Volume II - Chapter 2: The Circuit of Productive Capital - II. Accumulation and Reproduction on an Extended Scale
      http://www.marxists.org/archive/marx/works/1885-c2/ch02.htm#2

      I read the sentence "The constant augmentation of his capital becomes a condition of its preservation" as: either capital grows or it collapses (as in a business cycle crisis). I suppose that there are other possible readings, but I read it this way.

      Why did Marx think that (If he really tought that)?
      My understanding is that Marx, who thinks in LTV terms, implicitly thinks that actors in his system also think in LTV terms, so if capitalist act in order to increase profits, but thinks in LTV terms, he tries to increase profits above the rate of growth (that is already discounted in the LTV).
      I think that this actually makes sense also in pratical terms: suppose that Apple produces a new, better iPhone (thus with a higer "real" value). But Apple doesn't use iPhone, it is only interested to sell them at a good price, so the increase in the real value of production is totally non important to Apple, other than as a tool to increase its market share. Most of the increase in "real" terms is eaten by the consumers. If, in the meanwhile, another capitalist invested in medical equipment, the "real" value of his profits would be increased by the new iPhone as much as those of Apple (if we exclude the increased market share), as the dollar that consumers of medical equipment use to pay capitalist B's medical equipment have now an higer "real" value.
      So it is quite logic for the "capitalists" to think in relative, LTV-like terms and discount growth from their profits, IMHO.

      Delete
  4. "Substitutes or complements" could refer to the relation between capital and labour or to the relationship between the thought of Marx, DeLong and Naidu, respectively. Magpie highlights an instance of the latter: at Project Syndicate DeLong attributes to what Suresh once remarked to him the views that at the New York Times he had himself "long thought." Suresh pursues this intellectual program of substitution and complementarity in his contrast between "what Marx really meant" and the conversation/translation in which reinvention and redeployment occur.

    With regard to the substitution/complementarity between capital and labour, Suresh posits the labour-saving (substitution) argument in one paragraph but then devotes the next four to the "flip side." How does this complement/substitute for DeLong's account? In his 2009 essay, "Understanding Karl Marx," DeLong wrote:

    "Marx believed that capital is not a complement to but a substitute for labor. Thus technological progress and capital accumulation that raise average labor productivity also lower the working-class wage. Hence the market system simply could not deliver a good or half-good society but only a combination of obscene luxury and mass poverty. This is an empirical question. Marx's belief seems to me to be simply wrong."

    Not a lot of "flip side" there to play around with. According to DeLong, Marx was "simply wrong" in his one-dimensional view of capital as a substitute for -- rather than a complement to -- labour.

    Imagine doing a word-for-word dictionary translation from, say, French into English and then declaring the author to be a dunce at grammar. That is the kind of "substitution" that DeLong performs on Marx's analysis. The result is neither complementary nor complimentary. It is incoherent. But that incoherence derives from the mechanical ineptitude of the translation, not from any grammatical failure of the source. Marx's "belief that capital is not a complement to but a substitute for labor" was not "simply wrong." It was not Marx's belief, period.

    "Substitute" and "complement" are rather elementary concepts compared to the analysis that Marx was aiming at in Capital. Do the terms "formal subsumption" and "real subsumption" translate grammatically into "substitute" and "complement"? Hardly. Yet it is in the realm of the distinction between formal and real subsumption that Marx's views about the "complementarity" of labour and capital can be discerned. It is not a simple complementarity in that the labour process is transformed in the movement from formal to real subsumption. For Marx, it was not a matter of either/or (substitution or complementarity) nor was it merely a matter of "both." It was a question of transforming and revolutionizing the labour process into a means specifically (rather than incidentally) for the production of exchange values. Here's a small sample:

    "The capitalist mode of production develops the productivity of labour, the amount of production, the size of the population, and the size of the surplus population. With the capital and labour thus released, new branches of business are constantly called into existence, and in these capital can again work on a small scale and again pass through the different developments outlined until these new branches of business are also conducted on a social scale.... Large-scale industry throws as large a mass of human beings into the branches not yet subjected to it, or creates in these branches as large a relative surplus population, as is required for the conversion of handicrafts or of the small, formally capitalist business into a large-scale industry."

    ReplyDelete
    Replies
    1. Somehow I'd missed the "Understanding Karl Marx" essay before now. It's very bad. Marx certainly did not believe that "technological progress and capital accumulation that raise average labor productivity also lower the working-class wage."

      Now, you could say that Marx believed that the real wage, as opposed to the wage share, was relatively constant, and that this belief was inconsistent with the empirical evidence for the rich countries between, say, 1870 and 1970. But that's not what DeLong writes.

      Do the terms "formal subsumption" and "real subsumption" translate grammatically into "substitute" and "complement"? Hardly.

      Right. The real issue for Marx is the reduction of our capacity to transform the world and ourselves, into an instrument for the production of surplus value.

      Delete
  5. Maybe this helps explain some of the discrepancies here:?

    "Larry Summers explains Washington to Elizabeth Warren in one sentence:"

    "In the spring of 2009, after the panel issued its third report, critical of the bailout, Larry Summers took Warren out to dinner in Washington and, she recalls, told her that she had a choice to make. She could be an insider or an outsider, but if she was going to be an insider she needed to understand one unbreakable rule about insiders: 'They don’t criticize other insiders.'"

    http://www.newyorker.com/arts/critics/books/2014/04/21/140421crbo_books_lepore?currentPage=all

    ReplyDelete
  6. Fav line:
    "Perhaps Marx thought that even if workers’ standard of living increased under capitalism, it would not undo the exploitative/unfree/ undemocratic institutions inherent in the labor market."
    Is there a chance the idea doesn't "resonate with contemporary ethical intuitions." Hell, it is very very relevant; it especially deals with the "our poor people have cable tv" argument...

    ReplyDelete