Saturday, April 11, 2015

Sraffian Long-Run Equilibrium Prices of Production and Post Keynesianism

I find this whole issue to be very interesting because it seems clear to me that Sraffian long-run equilibrium prices (on which, see Brinkman 1999: 44) are fundamentally the same concept as Marx’s “average prices” or long-run “prices of production” (for this concept in Marx, see Fred Moseley’s excellent paper “Marx’s Concept of Prices of Production: Long-Run Center-of-Gravity Prices”), which in turn are the same concept as Ricardo’s long-run “prices of production” or “natural prices”. In all cases these are long-run equilibrium prices based on cost of production and a uniform rate of profit and they are anchors or centres of gravity for the price system around which prices fluctuate.

Now a starting point on this subject is the relation of Sraffianism to Post Keynesian economics.

I quote John King below who argues that Sraffian economics is today no longer even considered part of “broad tent” Post Keynesianism:
“Almost no one today regards ‘Post Keynesian-Sraffian’ economics as a single coherent school of thought (an exception is Luigi Pasinetti, 2007). By the end of the last century the Sraffians had been expelled (or, perhaps, had expelled themselves) from the Post Keynesian tradition, and in 2012 it is not at all clear whether the classical surplus approach to political economy (as its few remaining practitioners prefer it to be known) will long survive the retirement of the first post-Sraffa generation of theorists like Heinz Kurz and Neri Salvadori.” (King 2012: 314).
We can also note that Pratten (1996: 439), Arestis, Dunn and Sawyer (1999), Dunn (2000: 350), Minsky (1985; 1990) and Mongiovi (2003: 218) seem to agree with this assessment.

One of the most serious points of disagreements, so it would appear, is nothing less than the empirical relevance of Sraffian long-run equilibrium prices:
“The main characteristic of Sraffian economics relevant here is the use of long-period analysis where there would be an equalization of the rates of profit and full capacity utilization in the long period. The assumption that there are persistent forces that drive the economy toward a normal or long-period position when the world is characterized by uncertainties, nominal contracts, and path dependency sits rather uncomfortably with the general thrust of Post Keynesian economics” (Arestis, Dunn and Sawyer 1999: 544).
Lavoie (2010: 11) is even more explicit: he states that Sraffian long-period equilibrium and long-run equilibrium prices are “the cause of all the troubles” (Lavoie 2010: 12) between Sraffians and Post Keynesians.

Although some dissident Sraffians, Lavoie notes, have given up this long period equilibrium obsession (Lavoie 2010: 23, citing Roncaglia 1995: 120), it is the Marxian Sraffians following Garegnani who appear to be most extreme in their defence of the Classical view that competition really brings about a tendency towards a uniform rate of profit in capitalism, and that long-period prices of production are centres of gravity for market prices (Lavoie 2010: 14).

Many Post Keynesians reject Sraffian long-period equilibrium prices and the tendency to such equilibrium (see Lavoie 2014: 176), because of
(1) Joan Robinson’s arguments against long-run equilibrium positions (Robinson 1978a and 1978b; 1979: 179–180);

(2) the role of fundamental uncertainty in economic life,

(3) the severe barriers to entry that exist in real world capitalist economies (including aggressive use of capacity utilisation as a barrier to entry), and

(4) the significant, persistent differences in profit mark-ups in different sectors and industries.
On (4) one can refer to the work of the late Frederic S. Lee (Lee 1994: 325–327; Lee 1998: 226, n. 17; Lee and Jo 2011: 868–869).

He argued that the rate of profit mark-up in each industry and business will be determined by many factors such as custom, convention, different desires and needs for various profit rates, different levels of competition, and what mark-up the market will bear, etc., and these factors will themselves vary in different times and places (Lee 1994: 325–326). Consequently the Marxist and Sraffian notion of a real world tendency in capitalism to long-run prices of production with a uniform rate of profit is untenable (Lee 1994: 326–327).

So therefore what rational reason is there to believe in such a long-run tendency as an empirical reality if it is never observed and countervailing forces always thwart it in the short run? And advocates of the alleged long-run tendency to a uniform profit rate cannot evoke the analogy of gravity here, because – while the evidence for a force called gravity is overwhelming – they have not even proven that their long-run tendency actually exists. They cannot move to analogies from the natural sciences until they can demonstrate that long-period equilibrium prices really are a force as well supported as gravity. But they have not done so, and there is much empirical evidence against it. So the alleged long-run tendency seems to reduce merely to an unrealistic assumption in an overly analytic, abstract model set in logical time (Lee and Jo 2011: 868–869), where ultimately it can only be assumed by definition to be true.

So, all in all, if most Post Keynesians reject Sraffian long-run equilibrium prices, then logically they should reject Marx’s long-run equilibrium prices of production too.

Further Reading
“Matias Vernengo on Marx’s Labour Theory of Value,” April 3, 2015.

“Lavoie on ‘Should Sraffian Economics be dropped out of the Post-Keynesian School?,’” June 19, 2014.

“Sraffians versus Kaleckians versus Fundamentalist Post Keynesians,” June 17, 2014.

Arestis, Philip, Dunn, Stephen P. and Malcolm Sawyer. 1999. “Post Keynesian Economics and its Critics,” Journal of Post Keynesian Economics 21.4: 527–549.

Brinkman, Henk-Jan. 1999. Explaining Prices in the Global Economy: A Post-Keynesian Model. E. Elgar, Cheltenham, UK and Northampton, MA.

Dunn, S. P. 2000. “Wither Post Keynesianism?,” Journal of Post Keynesian Economics 22.3: 343–364.

King, J. E. 2012. “Post Keynesians and Others,” Review of Political Economy 24.2: 305–319.

Lavoie, Marc. 2010. “Should Sraffian economics be dropped out of the Post-Keynesian School?,” Paper prepared for the Conference at the University of Roma Tre, 2–4 December.

Lavoie, Marc. 2014. Post-Keynesian Economics: New Foundations. Edward Elgar, Cheltenham.

Lee, Frederic S. 1994. “From Post-Keynesian to Historical Price Theory, Part I: Facts, Theory and Empirically Grounded Pricing Model,” Review of Political Economy 6.3: 303–336.

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

Lee, Frederic S. and Tae-Hee Jo. 2011. “Social Surplus Approach and Heterodox Economics,” Journal of Economic Issues 45.4: 857–875.

Minsky, Hyman. 1985. “Sraffa and Keynes: Effective Demand in the Long Run,” Hyman P. Minsky Archive, Paper 321

Minsky, H. P. 1990. “Sraffa and Keynes: Effective Demand in the Long Run,” in Krishna Bharadwaj and Bertram Schefold (eds.), Essays on Piero Sraffa: Critical Perspectives on the Revival of Classical Theory. Unwin Hyman, London. 362–371.

Mongiovi, G. 2003. “Sraffian Economics,” in J. E. King (ed.), The Elgar Companion to Post Keynesian Economics. Edward Elgar, Cheltenham. 318–322.

Moseley, Fred. “Marx’s Concept of Prices of Production: Long-Run Center-of-Gravity Prices”

Pratten, S. 1996. “The Closure Assumption as a First Step: Neo-Ricardian Economics and Post-Keynesianism,” Review of Social Economy 54.4: 423–443.

Robinson, Joan. V. 1978a. “A Lecture delivered at Oxford by a Cambridge Economist,” in J. V. Robinson, Contributions to Modern Economics. Blackwell, Oxford.

Robinson, Joan. V. 1978b. “History versus Equilibrium,” in J. V. Robinson, Contributions to Modern Economics. Blackwell, Oxford. 126–136.

Robinson, J. 1979. “Garegnani on Effective Demand,” Cambridge Journal of Economics 3: 179–180.

Roncaglia, A. 1995. “On the Compatibility between Keynes’s and Sraffa’s Viewpoints on Output Levels,” in G. C. Harcourt, A. Roncaglia and R. Rowley (eds.), Income and Employment in Theory and Practice. St. Martin’s Press, New York. 111–125.


  1. Hi LK,

    Thanks for consistently bringing up interesting PK topics in your blogs. I read when I can, but I didn’t really get a chance to catch up on all of your posts and comments on classical/Marxian theories of value, so forgive me if I’m just rehearsing past debates.

    I think almost all Sraffians would agree with the claim that ‘the rate of profit mark-up…will be determined by many factors such as custom, convention, different desires and needs for various profit rates, different levels of competition, and what mark-up the market will bear'. However, Sraffians would also emphasise that the ability of firms to set target rates of return is subject to the constraints imposed by the input-output relations of a complex, industrialised economy and the cost-price interdependencies of such a system, as well as the inverse relationship between the real wage and the (competitive, “riskless") rate of return to capital. Thus, the Kaleckian notion that firms can independently choose the mark-up or the profit rate according to particular targets and consequently determine the aggregate mark-up and income distribution is problematic, as pointed out by Steedman (1992).

    Furthermore, Sraffians do not believe that there is a tendency towards the establishment of a single uniform rate of profit across all industries. Even under competitive conditions profit differentials exist due to differential risk and illiquidity. Under the more general and realistic case of non-competitive conditions, profit differentials would also be attributed to the ’severe barriers to entry that exist in real world capitalist economies’ that you referred to, specific to each industry. Profit rates in each non-competitive industry can therefore be broken down into two components: (1) the competitive minimum rate of return (interpreted by most Sraffians to be the riskless rate of return on government securities) and (2) a margin calculated with reference to the competitive rate (the “allowances for monopoly players” that Phil referred to in the comments on the last post). The competitive rate serves an ‘anchor’ for the non-competitive target rates of return. This can all very easily be incorporated into Sraffa price systems (see e.g. Steedman 1984).

    As for the methodological critique of Sraffa price systems as being ‘ahistorical’ or ‘timeless’ for assuming that there is a tendency towards the establishment of a uniform, competitive minimum rate of return (the so-called ‘gravitation’ of actual prices towards normal prices), this assumption is no less ahistorical or timeless than the assumptions of uniform commodity prices and uniform wage rates that are embedded in mark-up pricing theories. Both are equilibrium constructions in the sense that the uniformity of the variables is the outcome of competitive processes that equalise returns from the sale of homogeneous commodities and labour.

    You should check out the first three essays to the third volume of Palumbo et al. (2013) by Tony Aspromourgos, Marc Lavoie and Heinrich Bortis for the “lumpers'” perspectives (to use Lavoie’s terminology) on the relationship between Sraffians and other post-Keynesians. Particularly Aspromourgos’ contribution, which I believe puts to rest the only objection to the Sraffian project that was not refuted in Lavoie (2010).

    Thanks again for all of your insightful posts!


    Steedman, I. 1977. ‘Natural prices, differential profit rates and the classical competitive process’, The Manchester School, 52: 123-40.

    Steedman, I. 1992. ‘Questions for Kaleckians’, Review of Political Economy, 4:125-51.

    Palumbo, A., Stirati A., and Levrero, E.S. 2013. Sraffa and the Reconstruction of Economic Theory: Volume Three, Palgrave Macmillan, Basingstoke.

  2. Long run equilibrium prices and the tendency for profit rates to equalise across industries are two different concepts.

    1. But the argument against long run equilibrium prices surely works against a real world tendency to uniform profit rate.

      Also, if we are talking about net profit (= residual from: total revenue minus all explicit costs such as costs of production, interest on money loans, and depreciation) then surely that is strongly related to mark-prices.

    2. The tendency for profit rates to equalise across industries under competitive conditions is a necessary condition for the establishment of long-run equilibrium prices in the Sraffian sense. Therefore arguments against the uniformity of the competitive rate of return necessarily invalidate Sraffian prices of production, not the other way around.

      As for whether 'the significant, persistent differences in profit mark-ups in different sectors and industries' present an empirical challenge to the classical notion of a tendency towards the establishment of a uniform rate of return, it clearly does not. All classical and Marxian economists recognised that different rates of return persist across different industries, owing to differential risk and illiquidity. Modern classical and (some) Marxian authors also recognise that the margin separating the industry-specific rate of profit from the competitive minimum rate of return may be attributed to monopoly power. The tendency towards uniformity of the rate of return refers to the competitive, minimum rate of return which is an anchor for all other rates of return.

      Arguing that the existence of different rates of return in different industries invalidates the classical tendency towards the establishment of a uniform (competitive, minimum) rate of return is simply attacking a straw person of a classical economist who believes that this tendency will manifest itself in reality as uniform rates of return across all industries, irrespective of risk differentials or market structure. Classical economists as early as William Petty recognised that rates of return will be different in different industries, and this can easily be incorporated into Sraffa price systems.

    3. "Arguing that the existence of different rates of return in different industries invalidates the classical tendency towards the establishment of a uniform (competitive, minimum) rate of return is simply attacking a straw person of a classical economist who believes that this tendency will manifest itself in reality as uniform rates of return across all industries,"

      If it doesn't manifest itself in the real world, the long run "tendency" is itself likely to be a figment of the imagination -- a property of abstract models.

    4. LK: Many thanks for your reply. I note that you omitted the last part of the sentence that you quoted, referring to differential risk and market structure. My point was that no classical economist (from Turgot to the modern Sraffians) has ever believed that the tendency towards the establishment of a uniform, competitive, minimum rate of return means that a single rate of profit will prevail in every industry, even in hypothetical fully-adjusted long-period equilibrium positions.

      What they did believe was that a hierarchy of profit rates exist in capitalist economies, and that all of the differential profit rates are reducible to a single, competitive, minimum rate of return (which most Sraffians now interpret as the riskless rate of return on government securities held to maturity). The difference between the competitive rate of return and the other rates of return in the hierarchy are due to differential risk and (in the modern classical approach) oligopoly/monopoly.

      When, in both the Marxian and Sraffian equations for prices of production, a single, uniform rate of profit (“r”) is presented as prevailing in every industry, the author is abstracting from differential risk and non-competitive conditions in order to lay bare the fundamental conclusions of the classical political economists, e.g. the inverse relationship between the real wage rate and the competitive minimum rate of return as well as (in Sraffa’s equations) the simultaneous determination of relative prices and the non-exogenous distributive variable. Including an extra-profit term that captures the additional profits accruing to oligopolistic firms into the price equations does not change these results, provided that the extra-profit term is of a definite magnitude and cannot be conceived as independently fixed by firms.

      The uniformity of the competitive, minimum rate of return (which is, in Lavoie’s preferred terminology, the “benchmark” rate in a hierarchy of rates of return) in classical prices of production is no less observable than the assumption of uniform wage rates and uniform commodity prices in mark-up pricing theories. Both of these are equilibrium constructions that are the outcomes of equilibrium processes; processes which are not always rigorously specified or empirically verifiable.

      Indeed, every theoretical model requires abstraction. What is important in political economy is that the assumptions that we make are realistic. And, as noted by Phil in the comments on your previous post, the classical hypothesis of a tendency towards equalised profit rates (abstracting from risk and oligopoly) across industries isn’t a terrible assumption because it approximates the logic by which profit maximising wealth holders allocate their financial wealth between different capital assets, thereby systematically structuring the rates of return across those assets.

    5. It is this sort of thinking -- use of excessive abstract models -- that would also allow neoclassicals or Austrians to justify the natural rate of interest, or long run price flexibility or all sorts of other neoclassical ideas remote from reality. Where do you draw the line?

      What this issue boils down to is this: I want an empirical justification that there is a real long run tendency for profits to equalise, not wishful thinking and abstract models.

    6. I want empirical examples of wildly different profit rates. And by the way, the line you draw where you need to draw it depending on the problem at hand.

    7. Heyo, been away for a few days. Just want to chime in mostly to say I agree with Phil up top re: equilibrium vs tendencies. Perhaps the easiest illustration of this is the explicitly non-equilibrium framing of Marx my interpretive camp embraces.

      A few quick notes on profit rate equalization: First off, it's not "always" countervailed; it's just not something that we can claim with any certainty will be dominant from one period to the next, given the huge range of factors affecting it. But its general-case existence is nevertheless obvious; for instance, just consider the alternative -- if capital flooding from low-return industries to those with returns far above average did not exert downward pressure on the latter, then we're supposing that virtually any share of society's total investment can see above-average profits -- for all intents and purposes, it'd be an "infinite money machine."

      And lastly, one can just hunt around for strictly and specifically empirical justifications, too. This Levy Institute link, for example, wasn't far from the top of a google search: "The central finding of the paper is that 14 out of the 20 industries studied have no significant competitive profit rate differentials in the long-run; in other words, their profit rates are equalized in the long run with the general profit rate, after allowing for risk and other premia. Over the period under study, these industries accounted, on the average, for about 75 percent of net fixed capital stock and 72 percent of profits in the manufacturing sector. These results suggest that profit rate equalization may be considered as a dominant, long-run tendency in the U.S. manufacturing sector."

  3. Question unrelated to your current post series on Marxism.

    Do you have an analysis on the Finnish economy or know an heterodox Keynesian/institutionalist who has done a different analysis of it than the mainstream ones which they suggest that the cure to the country's current recession it's more "belt-tightening"?

    1. I regret I don't know of any Post Keynesian analysis of Finland. I would imagine its problems are the familiar ones of austerity and being tied to the Eurozone.

    2. There may be many Finland specific problems but surely the uncertainty of nominal incomes caused by euro does not help. Moreover, you live close to Sweden and Norway and they have their currency and will over the medium term outperform your economy (especially during recessions such as now in 2015). Last but not least, you live near countries whose existence is being threatened by Putin's regime (your existence isn't threatened in the immediately only because you've an army). This is my modest "Keynesian" diagnosis.

  4. LK, i think the key issue is by how much the profit rates vary among industries and producers. If variation is small, then the theory is a good approximation, and it may very well have some practical value. On the other hand, logical (not empirical) arguments for profit equalization do NOT work because of strategic interactions and discrete nature of industry.

    I would also like to add that a theory with more unknowns (like mark-up pricing with many different mark-ups) is more robust again being proved wrong, but much less useful. The result isn't necessarily better.