Stuck in the neoclassical world

Nick Rogers bases his economic analysis on a reality which had ceased to exist even in Engels’s time, writes Arthur Bough

In his review of Andrew Kliman’s The failure of capitalist production Nick Rogers writes:


On the question of historic versus current replacement costs, it seems to me that the application of a temporal perspective to the returns on the prices that have actually been paid for inputs offers at least the prospect of measuring the real rates of profit of real capitalists.1


The problem is it then becomes a subjectivist study of capitalists, not an objective study of capital. Different adherents of the ‘temporal single-system interpretation’ (TSSI) of Marxist political economy have their own version. The basic argument is that the value of inputs may have changed since they were bought, and this is reflected in the value of outputs. But in calculating the amount and rate of profit, it is the money price paid for inputs that has to be used to define the cost of production.

Most Marxist economists reject the TSSI. Fred Moseley surveyed all Marx’s writings referring to the valuation of constant capital and concluded:


... Marx was consistent throughout his writings on this topic. Marx assumed throughout that constant capital is valued at current reproduction costs ... The value of constant capital is not determined at the point in time when the means of production enter the production process. The value of constant capital does not remain unchanged after the means of production enter production if there is a change in the value of the means of production before the output is sold. The stock of the constant capital, as well as the flow of constant capital, is revalued as a result of a change in the value of the means of production.2


I will quote one of the numerous examples (this one from Capital volume 3):


If the price of raw material, for instance of cotton, rises, then the price of cotton goods - both semi-finished goods like yarn and finished goods like cotton fabrics - manufactured while cotton was cheaper, rises also. So does the value of the unprocessed cotton held in stock, and of the cotton in the process of manufacture. The latter because it comes to represent more labour-time in retrospect and thus adds more than its original value to the product which it enters, and more than the capitalist paid for it.3


Nick is correct that the replacement cost method does not internalise capital gains/losses, whereas the TSSI does. So if we have C 1,000 + V 1,000 + S 1,000 = K 3,000 and assume the value of C changes to 2,000, then replacement cost recognises the labour-time required for producing the commodity as 4,000 (made up by

C 2,000 +V 1,000 + S 1,000, using Marx’s method above, where C is retroactively revalued to 2,000).

There is only one logically consistent way that I can see to internalise capital gains/losses. Only if inputs are valued differently to outputs can you capture gains or losses arising from changes in their value: ie, if you value inputs according to the historic price paid, and outputs according to current necessary labour-time. So: C 1,000 + V 1,000 +S 2,000 = K 4000.

Total output has risen to 4,000 because that is the labour-time now required for its production. C remains at the 1,000 historic cost. V has not changed, nor has the rate of exploitation. As the totals on each side of the equals sign must be the same, the only way this is possible is if S rises to 2,000 (now comprising 1,000 surplus created by labour, and 1,000 capital gain).

Profit is the money form of surplus value, so now labour is no longer here the only source of surplus value. Part is attributable to capital! This is not quite a marginal productivity theory of factor contributions, but it is clearly a step towards it. Logically, here labour is not required at all. Capitalist A who buys asset X at t1 for £1,000, and sells it to B at t2 for £2,000 has made a £1,000 ‘profit’, purely as a result of time, speculation, good fortune or however you like to describe it. Replacement cost recognises that individual capitalists make capital gains/losses, but argues that non-inflationary gains/losses are merely redistribution of the surplus value made in production, and must cancel each other out from the perspective of capital in general. A’s capital gain is cancelled by B’s capital loss on their money capital relative to X.

Marx uses replacement cost of capital for valuation for a good reason, related to this point of difference between trading profits/losses, and capital gains/losses. His analysis is based on the idea of capitalist production as a continual process, of ongoing production. The TSSI only makes sense if you view capitalism from the perspective of the capitalist as an individual, and if you view the dynamic of capitalism as the desire by each capitalist to maximise their own money returns on a period-by-period basis, rather than being capital as self-expanding value.

It is a non-dialectical view, which stems from the syllogistic method and conception of time used by the TSSI.

Carchedi and de Haan say:


In fact, given (t1-t2), the producer of A can sell it at t2 at its transformed price and buy it at t1 at its untransformed price only if t1 and t2 coincide: ie, only if time is abolished.4


In this syllogistic view t1 and t2 are two distinct points in time - singularities - and t1 cannot be t2. However, t1 does not exist as a point in time, but itself has duration. No matter how infinitely small t1 is made, it has a beginning and end. It is itself divisible into a t1 and t2. Consequently for any t1, it is logical from a dialectical perspective to have an input produced at the beginning of that period, also appearing as an output later during that period. t1 is merely an artificial construct to overcome our inability to comprehend the infinite. So t1 can logically, from a dialectical perspective, as easily represent a period of six months as a millisecond. If t1 begins on January 1, and t2 does not begin until July 1, who would deny, looking at the reality of capitalist production, that an output produced on January 1 is clearly capable of appearing as an input on January 2?

Carchedi and de Haan note that “The value of A as the input of B cannot be transformed at (t2) since at (t2) A as an input is not a commodity any more.” But Marx makes clear by contrast that capital is comprised of commodities. He makes clear that all commodities existing at the same time have the same social value: “Each individual commodity, in this connexion, is to be considered as an average sample of its class.”5

Marx uses current replacement cost because capital does have to reproduce itself, and its ability to do so is determined by what has to be paid now. Suppose Marx’s capitalist above saw the apparent increase in profit as a windfall. If they believed their consumption could increase, or their production expand faster, they would be disappointed. Replacing the cotton at its current price would eliminate the additional profit!


Nick says: “In brief, Kliman’s thesis is that the root cause of the crisis that began in 2007-08 was insufficient profitability.” Nick links this to the law of the tendency of the rate of profit to fall, and supports the idea that this has caused capitalism to stagnate.

However, Michael Roberts provides a chart of US profitability.6 The chart shows the rate of profit (RoP) in the US for non-financial companies has risen! It reaches a low point of around 3% in the early 1980s and rises to approximately 7% last year. The argument that a falling RoP is causing stagnation seems contradicted by the data - the RoP has risen, and global output rose sharply.

Most calculations, not just the TSSI, understate the RoP. To be fair, there is difficulty obtaining the relevant data from official figures. Marx and Engels describe the effect of the rate of turnover of capital on RoP. Rate of turnover measures how quickly commodities get sold, replacing the constant and variable capital consumed in their production. They demonstrate the real RoP is S x n/C+V, where n is average number of turnovers of capital per year. Any analysis of RoP that does not take account of changes in the rate of turnover is meaningless. Even a modest rise in this rate will have a significant effect on the RoP.

Marx and Engels point out: “The chief means of reducing the time of production is higher labour productivity ... The chief means of reducing the time of circulation is improved communications ...”7 The transformations they witnessed are nothing compared to recent developments. Robotics, computers, the internet, new types of materials, new production systems, etc have brought epochal transformations in productivity, and massively reduced the value of constant capital, revolutionising the pace of circulation, and the efficiency of the capital.

Assume the early 1980s rate of turnover was 10, and today is 12. These are just hypothetical figures, but it is likely any increase has been considerably more than 20%! Calculating the real RoP puts the former at 30%, and the latter at 84%. RoP only just doubled, on the original calculation, but on a real basis it almost trebled - with just a 20% increase in the rate of turnover.

The RoP is shown as 8% in 1952, only marginally higher than 2011. However, any rise in the rate of turnover during the last 60 years is sufficient to bring about a higher real RoP today.

Moreover, output has not stagnated. Bill Jeffries has pointed out:


Kliman ignores the significance of the fact that, according to Maddison, during the noughties world GDP per capita growth rates exceeded those of the 1950s and 1960s. Indeed the Groningen Growth and Development Conference (GGDC) recently noted that world per capita GDP growth had doubled during the noughties.8


According to the ILO, in that period the world’s labour force also grew by around 30%, or 500 million workers! Global fixed capital formation per year doubled from $7 trillion to $14 trillion. This is not an indication of a stagnant economy.

Nick states:


This narrative [the rate of accumulation can fall below the RoP] rests on what seems to me the frankly implausible hypothesis that, despite reaping higher profits than ever in the productive sectors of the economy, capitalist investors have been alerted by 20-20 foresight to the risk that any additional increase in production in these sectors would run smack-bang into a brick wall of insufficient consumer spending power. In the neoliberal economy, apparently, the rate of profit no longer determines the direction of the flows of capital.


High rates and volumes of profits have not been fully invested for a variety of reasons, including the possibility of speculative gains for capital in assets such as property or shares rather than productive investment. In the last couple of years intense fear over the euro zone debt crisis has caused financial repression. Vast sums sit on corporations’ balance sheets; cash floods into near zero-risk assets, clearly contradicting the crude, determinist claim that capital at all times is forced to maximise profits! The watchword is return of capital, not return on capital.

That is why capitalists deposit their billions with BNY Mellon, and pay the bank to hold it for them or buy German bunds issued with a zero coupon; why Microsoft keeps more than $40 billion uninvested, and takes advantage of being able to borrow at 3% to add to its cash position; and why US corporations have $15 trillion waiting to be invested. Yet, despite all this hoarding, there has also been continued productive investment, and continued growth, which itself is an indication of the mass and rate of profit extracted!

Disappeared world

When the credit crunch stopped economic activity in 2009, companies like Honda in Swindon slowed down production, and got their workers to take holidays. They did not say, ‘We have to maximise profits so let’s ramp up production to secure a bigger market share’ - which is how a crisis of overproduction classically developed, according to Marx. The capitalist state helped smooth out the dislocation through ‘cash for clunkers’ schemes. The Chinese state provided vouchers redeemable against consumer goods. In other words, Nick’s view of capitalism seems stuck in an early 19th century neoclassical world, which had even disappeared in Engels’ last descriptions.

Engels knew this more developed capitalism could base its decisions on such foresight, and criticised the Erfurt programme for not recognising it. He wrote:


Capitalist production by joint-stock companies is no longer private production, but production on behalf of many associated people. And when we pass on from joint-stock companies to trusts, which dominate and monopolise whole branches of industry, this puts an end not only to private production, but also to planlessness.9


Throughout the 20th century capital tried to remove risk as much as possible through the introduction of planning. Central banks were introduced. Fordism recognised mass production required mass consumption, which meant higher wages, related to rises in productivity. The capitalist state utilises Keynesianism and monetarism. The welfare state ensures and regulates the reproduction of labour-power. At enterprise level there is a high degree of business and financial planning.

As Simon Clarke said 20 years ago,


Indeed it would be fair to say that the sphere of planning in capitalism is much more extensive than it is in the command economies of the Soviet bloc … The extent of coordination through cartels, trade associations, national governments and international organisations makes Gosplan look like an amateur in the planning game. The scale of the information flows which underpin the stock control and ordering of a single western retail chain are probably greater than those which support the entire Soviet planning system.10


Point-of-sale systems control, in real time, stock control for retailers, and, via ‘just in time’ and other Toyotist systems, control, in real time, production decisions. None of this is ‘socialist planning’. It is no more a planned production of use values than is market capitalism. It remains planning to ensure profit maximisation, but profit maximisation in a longer-term, more planned sense than was the case with early 19th century free-market profit maximisation.

The basic contradiction of capitalism is that it expands production faster than it can expand the capacity to consume that production, at prices that ensure the capital consumed can be reproduced: ie, at a profit. Marx understood the principle of demand elasticity. Fordism/welfarism, and the introduction of planning, provided a partial solution for that during the period of the post-war boom. It is continuing to do so in China, and other economies, where vast new labour forces are being recruited, with vast new consumption needs.

But it is the long wave that explains this. To overcome the limits of consumption, it is not higher wages that are required, as the underconsumptionists argue (which means reducing surplus value to pay for it), but the development of whole new industries, new arenas for capitalist investment, producing new use values that can be sold profitably. Capitalism has not become crisis-free, but to understand the basis of current crises it is necessary to analyse it as it is today, not according to the dynamic of more than 150 years ago.


I would like to thank Bill Jefferies and Graham Ballmer for comments on earlier drafts.


1. ‘Value, profit and crisisWeekly Worker July 5.

2. www.mtholyoke.edu/~fmoseley/CONCP.htm.

3. www.marxists.org/archive/marx/works/1894-c3/ch06.htm.

4. G Carchedi, W De Haan, ‘From production prices to reproduction prices’ Capital and Class autumn 1995.

5. www.marxists.org/archive/marx/works/1867-c1/ch01.htm#S1.

6. http://thenextrecession.wordpress.com/2012/06/30/us-profitability-which-way.

7. www.marxists.org/archive/marx/works/1894-c3/ch04.htm.

8. www.permanentrevolution.net/entry/3388.

9. www.marxists.org/archive/marx/works/1891/06/29.htm (original emphasis).

10. ‘Crisis of socialism or crisis of the state?’ Capital and Class winter 1990.