Saturday, January 11, 2014

Does the Market Tend to Drive Profits to Zero?

The short answer is “no.”

The reason is not just that competition in real world markets falls far short of the imaginary “perfect competition” of neoclassical theory, but because most prices are mark-up prices and businesses are almost always setting them to make a profit. Even new entrants into a particular market where mark-up pricing exists will generally set their price to make a profit as a mark-up over costs of production. The profit will not be competed away.

As Lee notes,
“… business enterprises do not consider profits as part of costs. Consequently, the typical statement made by Sraffians and Marxists that prices equal their costs of production (which includes a uniform rate of profit) in long-period positions has no conceptual correspondence to the concepts of costs and prices used by business enterprises. Without this link, the Sraffian and Marxian theory of prices cease to be grounded in the real world inhabited by actual business enterprises and their costing and pricing procedures.” (Lee 1998: 204, n. 10).
But it is not just Sraffians and Marxists who make serious errors on this subject: neoclassicals and Austrians also think prices tend to be driven towards marginal cost in competitive markets, and profits will be competed away in a convergence towards an equilibrium state (which itself will not be reached), whether it is the Austrian “final state of rest” or Walrasian general equilibrium.

Lee, Frederic S. 1998. Post Keynesian Price Theory. Cambridge University Press, Cambridge and New York.


  1. The problem is : what your definition of profits? If profit is defined as the difference between the value of sales and money costs of labour (i.e wages or variable capita in marxist languagel), circulating capital and consumption of fixed capital (i.e "constant" capital), there's nothing in marxian economics that drives total profits to zero. Homogeneous rate of profit between sectors is not equivalent as zero profit! For Marx, when firms are making profits higher than the average, he calls this "extra surplus value" (or extra-profits)...but it doesn't mean that an average of profit is a zero profit rate! Marx called cost of production the sum of constant capital and variable capital.
    So, the real question would be : is there an inherent tendency of convergence towards an uniform rate (positive) rate of profit? In that case, the response implies a long term empirical study like this one :

  2. Yeah, Lee isn't quite getting that point on Marxians; the thing he is referring to is closer to Marx's "prices of production," which equal the costs of production (or "cost-price") plus the average rate of profit in the industry. But market prices rarely if ever equal prices of production (which are a social abstraction), much as they rarely ever equal their values (in the sense he employed the term) -- that is, a commodity's individual value generally diverges from its social value. In Marx's theory, the difference between price and value is where all the action is, so to speak.

    For a good overview of price theory from a Marxian perspective, you might check out Carchedi's Frontiers of Political Economy.

  3. Companies have to put a non-zero mark up, since no one would invest in an enterprise if it does not return a profit. The only exception I could imagine, are consumer cooperatives, which are non-profits "businesses". But even consumer cooperatives will use (small) mark ups to cover for future uncertainties.

  4. Alternatively, you could look at the sector balances (e.g., Kalecki profit equation). There's nothing in the macroeconomy to constrain the variables in the equation to zero.

  5. The neoclassic model of perfect competition doesn't state that "normal" profits drops to 0, it actually states that "extraordinary" profits drops to 0. Normal profit is an homogeneous amount that all the firms in the competitive market earn (a good measure of this could be the return to assets ratio, since all the firms have the same cost structure and total revenues given the assumptions of the model).

    For a deeper insight, take a look to Varian's "Microeconomics". Btw, congratulations for the blog, I like it a lot.

    1. Does Wikipedia have this wrong?

      "In order not to misinterpret this zero-long-run-profits thesis, it must be remembered that the term 'profit' is also used in other ways. Neoclassical theory defines profit as what is left of revenue after all costs have been subtracted, including normal interest on capital plus the normal excess over it required to cover risk, and normal salary for managerial activity."

      It seems to agree with LK on the definitions used by neo-classicals here.

    2. No, it is correct, but by "profits" they (the Neoclassics) mean "extraordinary profit", in other words, profit derived from market power. That's is only possible when the demand function has a negative slope, but in the perfect competition model firms take the price as a cosntant and they can't set their own price (so the demand function is a constant too and equals the marginal cost for all firms).

  6. I think much of this comes out of the marginalist school conclusion, which simplifies to: competition and profit maximization means output will be driven to the point where all of the marginal product produced is claimed by labor. The capitalist has no incentive to order and organize and finance the production of another unit of produced output. A volume of profits still exists on all the produced goods up to that last increment unit, even if that last produced unit is entirely claimed by labor.