Sunday, April 6, 2014

John Eatwell on the theoretical lessons from the crisis

It's from 2012, but still very much relevant. Short and to the point about the limitations of the mainstream to understand the crisis. I would put less emphasis on the Sonnenschein–Mantel–Debreu theorem, and more on the Sraffa-Garegnani critique of General Equilibrium, but that's a detail. He is correct in pointing out to Keynes' Principle of Effective Demand (PED), and to the insidious role of finance.


  1. Matias,

    how would you respond to someone who claims that:

    - wealthy people are wealthy because they provide goods and services that people value, therefore they deserve their wealth.

    - poor people don't provide goods and services that people value, therefore they deserve to be poor.


    I come across this sort of reasoning quite a lot.

    1. Hi Philippe. These are misguided comments. First, most poor people actually do work, but their compensation is not determined by productivity or skills (as the neoclassical supply and demand theory suggests), but by their relative bargaining power. They actually do produce the things that people value. So effectively they are exploited. Second, capitalism often creates the conditions to maintain a good chunk of the labor force unemployed (what Marx referred to as the reserve army) in order to reduce their bargaining power. So these are not bums unwilling to work (even if some of those exist, but mind you most of them are wealthy, not poor people), but people that cannot find a job. The problem with these comments is not that they are callous, and that they are, but that they are simply wrong.

    2. By the way, the flip side of the surplus approach response to this is that the wealthy (capitalist) don't provide the goods that people want. They own the means of production, the accumulated wealth that allows them to extract surplus from workers. Note that the single most important thing to determine is someone is wealthy, is whether the parents were too.

    3. Thanks Matias,

      "their compensation is not determined by productivity or skills (as the neoclassical supply and demand theory suggests), but by their relative bargaining power"

      Where can I find more on this in particular? It's difficult to convince people in a neoclassical mindset that wages are not determined by productivity and skills.

  2. Thanks Matias,

    "their compensation is not determined by productivity or skills (as the neoclassical supply and demand theory suggests), but by their relative bargaining power."

    Do you know where I can find more on this in particular? Any papers that make this argument convincingly? It's difficult to convince someone in a neoclassical mindset that wages are not determined by productivity and skills.

    1. Well there are a few post in the blog. But note that the evidence for that is nil. Over the last 30 something years real wages have stagnated while labor productivity has grown quite a bit. See here

    2. Well, I've stumbled on a problem regarding wages and productivity and could not get out of it by myself.. Taking Garegnani's description of the neoclassical theory it seems that one cannot say that productivity determines wages but the other way around. I mean, wages would be determined by its relative scarcity and then go on to determine the marginal productivity of labor and employment. Does the idea that marginal productivity determines wages come from a misconception or I am completely misguided?

  3. Yes, wages are not determined by the methods of production, it is rather the other way around. But wages are note determined by the relative scarcity of labour through supply and demand. As Keynes noted in the General Theory, you need to know the wages in order to know the demand for labour, so you cannot use supply and demand in order to determine the wages without circular reasoning (Keynes is certainly drawing upon Sraffa’s 1925 and 1926 papers, where Sraffa shows the inconsistencies of supply and demand analysis).
    For Garegnani, wages are determined by institutional factors, such as the wage bargain, and the political factors involved. But following a remark that Sraffa makes in page 33 of Production of Commodities, Garegnani came to see the rate of profits as the central variable, determined in turn by the rate of interest, which in turn is determined by monetary policy. So political factors work through their influence on monetary policy which, as Garegnani notes, does not operate in a political vacuum. Once the rate of profits is set, wages receive the remaining part of the surplus, and prices can be determined in terms of the cost of production. Rents are the only thing determined by relative scarcity of land. Garegnani is assuming an economy in which there is no full employment of labour, this is why labour is not scarce.

  4. Yes, that much I got. My problem is regarding the internal logic of neoclassical theory, specially the one that Garegnani used to make his point.

    1. Dear Pedro,

      I don’t know which specific article by Garegnani you have in mind. In any case, Garegnani distinguishes two stages in neoclassical theory, a first one where the central figures are Walras and Marshall, who focus on the normal (long period) position of the economy, and a second one after Hicks culminating in the Arrow-Debreu-McKenzie model, where the economy is in (short period) intertemporal equilibrium.

      Walras knew that savers do not intend a specific good (as in the Arrow-Debreu model) but rather a rate of return on their savings. But once you add a normal rate of profits to a set of inputs, prices are already determined and you don’t need supply and demand curves to find them (the technicalities are solved with great simplicity in Sraffa’s 1960 book, but Sraffa had the basic idea since 1927). After Hicks, this methodology, which focuses on the long period normal position, is abandoned, and economists start focusing on the short period, without using the notion of a normal rate of profit, leading to the Arrow-Debreu-McKenzie model, which presupposes that everyone knows what they want to buy in the future. This assumption avoids the problems brought by a uniform rate of profits, but leads to the well-known stability problems which cannot be avoided once you stop assuming that your model is just a normal or average position, and start assuming intertemporal equilibrium.

      The central message from Garegnani, I think, is the inconsistencies of neoclassical theory, both when it focused on a normal position (long period) as Walras did, and when it focused on (short period) intertemporal equilibrium (which does not avoid the problems that Walras faced, and actually makes things worse).

      As for scarcity influencing wages in neoclassical economics, I think Garegnani say supply and demand analysis as an extension of the Malthusian theory of rent, assuming all factors are scarce, that is, extending a scarcity theory to the study of capital and labour. Walras criticized Ricardo for assuming that labour is available for further production, and assumed full employment. Hence, demand plays a role in setting prices. Garegnani follows Ricardo who assumed that labour is available for further production. In various articles where Garegnani makes several assumptions in order to make an illustration, he always notes that the availability of labour is the crucial assumption that cannot be dropped.

      The idea that marginal productivity determines wages comes from the application of supply and demand curves to the determination of wages, but I don't know which specific article you are thinking about?

      Or maybe I didn't get your point again?


The Political Economy of the COVID-19 Crisis in Latin America

Following my talk on the same topic, on the same venue, now someone that might know a bit more about what's going on, particularly i...