Saturday, March 9, 2013

Kaldor on Economics without Equilibrium

If one were state the difference between Post Keynesianism and mainstream neoclassical theory, it might be summed up with the idea that Post Keynesian theory is “economics without (Walrasian) equilibrium.”

Nicholas Kaldor wrote a short monograph called Economics Without Equilibrium (Armonk, N.Y., 1985), and it still makes rewarding reading. Now it is true that some of Kaldor’s observations apply more to the full employment era of 1946 to the 1970s (the golden age of capitalism), but there is much that is still highly relevant.

I summarise Kaldor’s arguments below.

I. Prices in the Real World versus Walrasian Equilibrium Theory
For Kaldor, we need empirical investigation of the real world market economies. If our received theories do not fit the facts, the theories need to be abandoned or properly modified.

In many markets in the real world, the “sellers are price-makers and quantity-takers, and not, as Walrasian equilibrium theory supposes, price-takers and quantity-makers” (Kaldor 1985: 31). Many prices are cost-determined, and demand has considerably less influence on prices than neoclassical theory supposes (Kaldor 1985: 31).

Kaldor identifies the major flaw in Walrasian neoclassical theory in relation to prices:
“… since Walras first wrote down his system of equations over 100 years ago, progress has definitely been backwards not forwards in the sense that the present set of axioms are far more restrictive than those of the original Walrasian model. The ship is no nearer to the shore, but considerably farther off, though in a logical, mathematical sense, the present system of derived tautologies is enormously superior to Walras’s original effort.

Perhaps for that reason general equilibrium theory retains its fascination for teachers and students of economics alike. Indeed, judging by the number of Ph.D. students working on the implications of the rational expectation hypothesis, it is gaining ground, at any rate, in America. One reason is the intuitive belief that the price mechanism is the key to everything, the key instrument in guiding the operation of an undirected, unplanned, free market economy. The Walrasian model and its most up-to-date successor may both be highly artificial abstractions from the real world but the truth that the theory conveys — that prices provide the guide to all economic action — must be fundamentally true, and its main implication that free markets secure the best results must also be true. (This second proposition was indeed demonstrated but under assumptions so restrictive that Professor Hahn turned the argument around and suggested, in his inaugural lecture, that the importance of general equilibrium theory lies precisely in showing how stringent the conditions must be for ‘free markets’ to secure the results in terms of welfare that are naively attributed to them. This may well be true, but if so, it is truth bought at a very high cost.)

But the basic assumption in all this — that prices are very important in the working of a market economy — is rarely, if ever, questioned. Yet it is precisely this over-emphasis on the role of the price system that I regard as the major shortcoming of modern neoclassical economics, particularly the Walrasian version of it.”
(Kaldor 1985: 13—15).
The major flaw in neoclassical theory – and indeed in Austrian theory as well – is the failure to really understand the role of prices in market economies.

In many markets, changes in supply are caused by “quantity signals,” not price signals. A “quantity signal” can be one of the following:
(1) a change in the amount of the stock/inventory* a business carries,
(2) a change in demand: changes in the sales volume or orders, or
(3) a combination of (1) and (2). (Kaldor 1985: 23).

* Note that, for many producers, stocks consist of factor inputs and not just output stock.
What, generally speaking, induces changes in production is a “quantity signal”: changes in demand or in stock. In the latter case, when a producer’s stock falls, he has the incentive to raise production to restore his stock to a normal level (Kaldor 1985: 25), and, if demand changes are expected to be large and permanent, more labour and capital goods will be employed.

Kaldor’s conclusion is that in normal times (outside, say, a severe recession) “in actual adjustment of supply and demand, prices play only a very subordinate role, if any [sc. role]” (Kaldor 1985: 25; my emphasis). Moreover, for industries with increasing returns to scale, a large increase in demand can indirectly result in falling prices (Kaldor 1985: 25).

All this means that the fundamental neoclassical ideas must be rejected: the notion of a market economy where prices are entirely or mainly set by the dynamics of supply and demand curves is wrong. Prices are often not adjusted to clear supply by gravitating to some equilibrium, market-clearing level. The main process by which an economy is supposed to have a tendency to a general equilibrium state is thus shown to be a fiction – and this is before we get to theoretical problems with the idea that market-clearing prices exist for all markets, just waiting to be discovered by Walrasian tâtonnement.

II. Supply and Demand
The firm producing a commodity in a fixprice market relies on the carrying of stocks (whether of the finished good or stocks of its factor inputs) and its sales orders (that is, the demand for the firm’s output). Price signals are less important than neoclassical theory assumes.

The normal state of a capitalist economy is unused excess capacity and a certain level of idle resources: therefore production is mostly demand-constrained, not resource constrained (Kaldor 1985: 35). Even at the state Keynesians call “full employment” – which is not zero unemployment but somewhere in the order of less than 1% up to 4% unemployment – capitalist economies are still, generally speaking, demand constrained, not resource constrained. In the case of unemployment, even advanced capitalist economies have “disguised unemployment” in the sense that demand for more labour in manufacturing (or higher paid jobs) tends to pull people from low paid jobs in services, but the loss of employment here does not really result in any significant loss of output and certainly not if workers’ productivity can rise to compensate (Kaldor 1985: 36).

Kaldor does not deny that excessive demand can cause resource constraints in an advanced capitalist economy: it shows up in bottlenecks, delays in delivery and unavailability of certain goods (Kaldor 1985: 37).

But the economies faced with serious resource constraints are more likely to be developing nations, and historically in the 20th century were actually Communist command economies that began as developing nations and that tended to engage in a level of investment beyond their resources.

III. Inflation
The source of inflation is mainly higher factor input costs: a higher wage bill or other factor input bill tends to induce higher prices (Kaldor 1985: 53). Changes in demand and in selling volume, on the other hand, are overrated as causes of price fluctuations.

The general stability of mark-up policies of firms in fixprice markets is dubbed “mark-up rigidity” by Kaldor (Kaldor 1985: 41). The profit mark-up in fixprice markets is constrained by the fear that a producer has that his competitors will not raise prices and achieve greater sales at his expense (Kaldor 1985: 40). The mark-up can also be determined by the desire to improve on market share by choosing a low mark-up in relation to other firms.

Many firms need to take care of customer relationships with regular clients, so their pricing behaviour is also constrained by the need to build up and maintain a regular clientele which can rob them of the neoclassical motive of pure profit maximisation (Kaldor 1985: 26, 48).

An important cure for inflation is the use of buffer stocks of important factor inputs that can be used to stabilise the price of these commodities when short-term supply side factors influence prices. By means of this policy measure, there will be created the real expectation of long-run price stability (Kaldor 1985: 79). The other solution to inflation is to stop excessive increases in money wages in relation to productivity growth (Kaldor 1985: 79).

Obviously, the need to stop excessive demand during booms is also a factor, but probably the least important one in the arsenal of inflation-controlling policies.

IV. International Trade
Increasing returns to scale are an important characteristic of modern production (Kaldor 1985: 68). The advantages of economies of scale are largely confined to manufacturing (Kaldor 1985: 70).

Walrasian theory is flawed by its assumption of constant returns to scale, and is thus a flawed theory for analysing problems of international development and trade (Kaldor 1985: 75).

Kaldor contends that free trade may damage the growth of industry in some nations, and that free trade is not necessarily beneficial to all parties (Kaldor 1985: 71).


Kaldor, Nicholas. 1985. Economics Without Equilibrium. M.E. Sharpe, Armonk, N.Y.


  1. Nice post.

    Kaldor was touched by Genius!

    About free trade he said: " “success breeds further success and failure begets more failure.”:

    Traditional theory, both classical and neoclassical, asserts that free trade in goods between different regions is always to the advantage of each trading country, and is therefore the best arrangement from the point of view of the welfare of the trading world as a whole, as well as of each part of the world taken separately. [footnote: The latter part of this proposition abstracts from the possibility that a particular country possesses some degree of monopoly power and thereby can turn the terms of trade in its favour by means of a tariff even after retaliation by other countries is taken into account.] However, these propositions are only true under specific abstract assumptions which do not correspond to reality. Under more realistic assumptions unrestricted trade is likely to lead to a loss of welfare to particular regions or countries and even to the world as a whole – that is to say that the world will be worse off under free trade than it could be under some system of regulated trade …

    … Owing to increasing returns in processing activities (in manufactures) success breeds further success and failure begets more failure. Another Swedish economist, Gunnar Myrdal, called this ‘the principle of circular and cumulative causation’.

    - “The Role Of Increasing Returns, Technical Progress And Cumulative Causation In The Theory Of International Trade And Economic Growth”, Further Essays On Economic Theory And Policy, 1981.

  2. Oh, man... where do I start. Kaldor's short book is the source of basically everything I write and think about economics.

    LK, I really recommend this paper by two obscure characters: an Irish philosopher of mathematics and an Irish Post-Keynesian.

    If you can get your head around it I think it tells you EVERYTHING you need to know about neoclassical theory in general and General Equilibrium analysis in particular. I riffed on the paper and dragged out the main points here:

    Mine is definitely more readable and makes for a good introduction, but check the paper out in the original afterwards. Its quite profound. I don't know why I don't hear more about these two of my fellow countrymen...

    1. Philip-

      "Divine mathematics?" A fairly witty dis, but I would think the obscurity of the reference would leave most readers scratching their head. Am I to understand that the bizarre pseudoscience of Elijah Muhammad and Clarence 13X has travelled as far as Ireland?

  3. "In all markets dominated by speculation of course prices must be flexible."

    Very interesting.

    From Kalecki's 'Theory of Economic Dynamics':

    "The situation of raw material is different [to manufactured goods, which are fix-priced]. The increase in the supply of agricultural products requires a relatively considerable time. This is true, although not to the same extent, with respect to mining. With supply inelastic short periods, an increase in demand causes a diminution of stocks and a consequent increase in prices. This initial price movement may be enhanced by the addition of a speculative element. A primary rise in demand which causes an increase in prices is frequently accompanied by secondary speculative demand." (P. 11)

  4. I find the current ramblings about the 'productivity paradox' in the UK most amusing. Once again one you realise that firms buffer internally, the paradox resolves itself.

    The apparent reduction in productivity measures in the UK is a function of the calculation method. Just like the quantity of theory assumes all money is in circulation when it decomposes Turnover to find Velocity, the productivity measures assumes all labour is deployed when it decomposes Turnover to find labour output.

    Neither of those calculations take into account the buffer stock - whether it is the stationary money stock or the stationary labour stock.

    Whatever expectations are keeping the labour stock up thank heavens for it. It's keeping the money circulation up!

    1. I don't get you. I thought the falling productivity in the UK was due to a Kaldor-Verdoorn's Law dynamic.

    2. I'm talking about the recent 'unusual' high level of UK employment that confuses commentators that believe firms operate marginally.

      That then affects the productivity stats due to the way they are calculated.

  5. This was a great analysis of a great book. It was a series of lectures given in honor of Okun. I plan to digitize it but I'm glad you and Pilkington were already able to get a copy.


  6. Yikes!

    1. Kaldor's book is valuable, but nowhere near that valuable!

    2. HA! You think that's bad

  7. Great post LK, lots of zombie ideas called out. When we stop hearing conservatives blame recessions on 'structural supply side' issues we'll have achieved success.

    Do you know where Kaldor's book is available at an affordable price? I found one seller offering it for $1,500 on Amazon

    1. Your best bet may well be to borrow it from a good library, I think.

  8. Forgive me the grave-digging. Although I fully endorse Kaldors views I am confused given the forelast argument:

    Walrasian theory is flawed by its assumption of constant returns to scale, and is thus a flawed theory for analysing problems of international development and trade (Kaldor 1985: 75).

    What about Krugman's 1979 account of increasing returns and monopolistic competition? Does the treatment of the labor market disqualify the model from being considered a general equilibrium solution in a Walrasian economy?

    As I have no access to Kaldor (1985) I'd be thankful for any response.

  9. re. Anonymous

    New trade theory is not Walrasian. It's just that Krugman confuses everyone by trying to keep the faith with DSGE.