The Transformation Problem and The Elasticity of Demand

In Marx's initial formulation of the transformation of exchange values into prices of production, he sets out five different spheres of production with different organic compositions of capital. Implicit in the model are a number of assumptions. The composition of the capital in each sphere is given in percentage rather than absolute terms. Consequently, it is implicit in this that the quantity of capital employed in each sphere is the same size. There is also an assumption here that the capital in each sphere turns over at the same rate. Marx sets out elsewhere that where capital in a particular sphere turns over at a slower rate than the average, its price of production will be above the exchange value, and vice versa.

The basis of the initial formulation is to add together the capital in each sphere, which gives a figure for the total advanced capital, and to total the surplus value from each sphere. On that basis an average rate of profit can be determined as s/c+v. Marx and Engels describe the process by which this average rate of profit then acts to establish prices of production for commodities, around which market prices fluctuate, rather than the situation in pre-capitalist commodity production, where market prices fluctuated around the exchange value of commodities.

As Engels describes, in his Supplement to Capital III, it is the historical development of this process which explains why, as soon as capitalist commodity production begins, even on a fairly limited basis, in the 15th century, this brings to an end the exchange of commodities according to their exchange values, even where those commodities are not produced capitalistically. A capitalist producer will always seek to invest their capital where it can obtain the highest rate of profit. If we stick with the assumption that capital in each sphere turns over at the same rate, that will be where the organic composition of capital is lowest.

The process by which exchange values are then turned into prices of production, and by which the average rate of profit becomes established is fairly straightforward. As capital is accumulated in those spheres where the rate of profit is highest, the supply of commodities in that sphere will rise. This rise in supply, with no rise in demand, causes the prices of those commodities to fall below their exchange value. The supply of these commodities will continue to rise, and the price of them fall, so long as capital continues to enjoy a higher rate of profit in that sphere than in some other.

The reason that this process, as Engels says, means that all commodities cease to exchange at their exchange values, is also then quite straightforward. Suppose that cotton is produced by peasant producers. Its exchange value is, say £100 per ton. Cotton is an input, for peasant producers of yarn, and yarn is also an input for peasant producers of cotton cloth.

Suppose that cotton production offers capital the most profitable area for its investment. It provides capital with say 30% rate of profit, whereas the next most profitable sphere for investment would only return 25%. Capital then invests in cotton production. As other capitalists see the potential to obtain this 30% rate of profit, they too invest in this production. Consequently, the supply of cotton rises, and as it rises, the price of cotton progressively falls from £100 per ton, to £80, per ton, to £60 per ton. 

At £80 per ton, the rate of profit for capital invested in this sphere might have fallen to say 27%, and when it falls to £60 per ton, the rate of profit will have fallen to 25%. If more capital is invested in cotton production, that would push the price below £60 per ton, and would reduce the rate of profit below 25%. But, capital can obtain 25% rate of profit, in the next most profitable sphere, say pottery manufacture, and so, rather than further capital accumulating in cotton production, it would begin to be invested in pottery manufacture, and the same sequence would unfold, in that sphere.

However, as the price of cotton falls, as a a result of increasing amounts of capital accumulating in that sphere, and the supply of cotton thereby continuing to rise, the peasant producers of yarn, now benefit from this lower price for their main material input. The peasant yarn producer, sees the price of yarn fall from £100 per ton to £80 per ton.

Suppose, the peasant producer of yarn originally laid out £20 for cotton, and added £20 of additional value by their labour. The exchange value of their output would then be £40. Now, capitalist production of cotton has caused its price to fall by 20%. The peasant yarn producer now only lays out £16 for their cotton, and continues to add £20 of value by their own labour. They sell their output of yarn, therefore, for £36, rather than £40. This is not a price of production for yarn, because yarn is not currently produced capitalistically. But, nor is it simply the exchange value of yarn, which continues to be £40, which is the value of the cotton consumed in its production, plus the value of the labour they add to it.

The £36 new price for yarn is a modified exchange value that arises, because the output of the capitalist cotton producer, is simultaneously the input of the peasant yarn producer, and as the price of cotton moves increasingly towards its price of production of £60 per ton, so this output price of cotton simultaneously modifies the input prices, and thereby also the output price of the peasant yarn producer. As Marx puts it,

“The foregoing statements have at any rate modified the original assumption concerning the determination of the cost-price of commodities. We had originally assumed that the cost-price of a commodity equalled the value of the commodities consumed in its production. But for the buyer the price of production of a specific commodity is its cost-price, and may thus pass as cost-price into the prices of other commodities. Since the price of production may differ from the value of a commodity, it follows that the cost-price of a commodity containing this price of production of another commodity may also stand above or below that portion of its total value derived from the value of the means of production consumed by it. It is necessary to remember this modified significance of the cost-price, and to bear in mind that there is always the possibility of an error if the cost-price of a commodity in any particular sphere is identified with the value of the means of production consumed by it.”

(Capital III, Chapter 9)

When capital has become fully invested in cotton production, so that any further investment would cause the rate of profit to fall below the average rate of profit of 25%, the price of cotton will have fallen to £60 per ton, which is its price of production, i.e. the cost of production, c + v, plus the average profit on the advanced capital. The market price of cotton may fluctuate above and below this level due to short term variations in supply and demand.

At this level, the peasant yarn producer's output will have fallen in value to £12 for cotton, and £20 value added by labour = £32. But, the peasant yarn producer, is, in turn, a provider of inputs to the weaver, who producer cotton cloth. Just as the price of production of cotton, simultaneously modifies the market value of yarn, so this modified market value of yarn simultaneously modifies the market value of cotton cloth.

If initially, the yarn producer consumed £20 of cotton, and added £20 of value by their labour, and sold this to the weaver, for £40, who, in turn, added £20 of value by their labour, to produce cloth with a value of £60, so the transformation of the value of cotton into a price of production, not only simultaneously modifies the market value of yarn, but also thereby the market value of cloth. The cloth producer sees the price they pay for yarn fall first to £36, which modifies the price of their cloth to £56, and then to £32, which reduces the price of their cloth to £52.

But, this continues to apply, when yarn production and cloth production themselves succumb to capitalist production. The only difference then is that instead of the price of their output being made up of the cost of their constant capital, plus the value added by labour, it too becomes a price of production comprising the cost of their constant capital, and their variable capital, plus the average rate of profit on their advanced capital.

In the same way as described above, therefore, if the rate of profit in yarn production was higher than in cloth production, capital would accumulate in the former, the supply of yarn would rise, pushing yarn prices down towards their price of production, at which point, only the average rate of profit would be produced.

The process, then by which this average rate of profit becomes established – never, as Marx points out, as an established fact, because it is a continual process, of moving towards it, as changes in social productivity continually change values, and the rate of profit – is then, a continual movement of capital away from spheres where the rate of profit is low, and into those where it is high. That movement reduces supply in the former, and raises supply in the latter. That raises the prices of commodities in the former, and reduces them in the latter.

However, on the basis of this understanding, its quite clear that in moving beyond the initial formulation of the question of the transformation of values into prices of production, the original assumption about equal amounts of capital being invested in each sphere, can no longer hold. If there are just two spheres of production A and I, then if the starting point is that there is £100 of capital invested in each, whilst the rate of profit in A is 25%, and the rate of profit in I is 20%, then, because the equalisation of the rate of profit involves a movement of capital to A from I, it cannot be the case that, as this process unfolds, both spheres continue to be the same size.

In Marx's initial model, the capital in each sphere is indicated in percentage terms, so that the total capital in each sphere totals to 100, and that continues to be the case once the values are transformed into prices of production. But, this representation of the capital in percentage terms, obscures the fact that in absolute terms, this process necessitates that the actual mass and capital in each sphere is thereby also changed. In the above, example, it might be the case that in order to bring about the average rate of profit of 22.5% ((25% + 20%)/2), the capital invested in A would have to rise from £100 to £120, thereby increasing the supply of A commodities, and reducing their price to the price of production, whilst the capital employed in I, would have to fall to £80, reducing the supply of I commodities, and so raising their price towards the price of production.

Yet, there is no way of determining on the basis purely of values, how much capital would need to be accumulated in A, in order to bring about the necessary reduction in price to reach the price of production. That would depend upon the price elasticity of demand in A. Similarly, the capital consequently withdrawn from I, that migrates to A, will cause a reduction in the supply of I commodities, but there is no way of knowing purely on the basis of values, what effect any given reduction in capital, and so supply will have on the prices of I commodities, because again that will be dependent upon the elasticity of demand for I commodities.

Marx himself makes this point in Theories of Surplus Value, Chapter XII. He writes, in respect of price changes resulting from a movement of capital in agriculture, between different classes of land,

“For the price of IV to rise to to £1 12s., the individual value of III, the demand would not have to rise by 75 tons. This applies especially to the dominant agricultural product, where an insufficiency in supply will bring about a much greater rise in price than corresponds to the arithmetical deficiency in supply.” 

In other words, Marx is basically describing a situation where, for a commodity that is a necessity, the staple food commodity, the price elasticity of demand is relatively inelastic. As supply falls short of demand, the price rises, but the rise in price does not choke of demand by a corresponding proportion. In order to bring demand and supply into balance, therefore, the price will need to rise by a proportionally greater amount than where the price elasticity of demand is relative elastic, so that any given rise in price would cause a proportionally greater reduction in demand.

Marx well understood the concept of elasticity of demand, long before it was theorised by the marginalists. For example, he also writes in Theories of Surplus Value, 

“The same value can be embodied in very different quantities [of commodities]. But the use-value—consumption—depends not on value, but on the quantity. It is quite unintelligible why I should buy six knives because I can get them for the same price that I previously paid for one.” (TOSV3 p 118-9)

Unfortunately, at the time Marx was writing, he did not have the analytical tools that the marginalists later developed to be able to measure such elasticities.

Suppose we take an economy with the two spheres A and I above. In I, its capital is comprised:

c 50 + v 50 + s 25 = 125.

A's capital is comprised:

c 60 + v 40 + s 20 = 120.

In that case, the rate of profit in I is 25%, and in A 20%. The average rate of profit is 22.5%. In order for both capitals to obtain this average rate of profit, capital must leave A, and migrate to I. The supply of I commodities will then rise, reducing their price to the price of production, and similarly the supply of A commodities will decline raising their price to the price of production.

In order to examine this, its necessary to also consider the actual mass of capital employed, rather than just its percentage composition, and to consider the actual volume of output. For ease of calculation, we can assume that these initial proportional relations are also absolute amounts of capital. In other words, in I, £50 of constant and variable capital is employed, and in A £60 of constant capital, and £40 of variable capital. (It doesn't matter whether this is £'s, $'s, or €'s, or whether this represents thousands, millions or billions)

We can also assume that the actual output of I is 125 units, and of A is 120 units. In that case, the price per unit of I commodities is £1, and also the price per unit of A commodities is £1. However, we know that the rate of profit in I is above the average, and is below the average in A. Put another way, if both spheres are to produce the average rate of profit, the unit price of I commodities must fall, and the unit price of A commodities must rise.

If we examine the value composition of a unit of output in each sphere, it is, in I:

c 50/125 + v 50/125 + s 25/125 = c £0.40 + v £0.40 + s £0.20 = £1.

In A, it is:

c 60/120 + v 40/120 + s 20/120 = c £0.50 + v £0.33 + s £0.17 = £1.

In order for I to obtain the average rate of profit of 22.5%, the unit price of its commodities must fall. The price of production for each unit will be:

c £0.40 + v £0.40 = k £0.80 + p £0.18 = £0.98.

Similarly, for A to obtain the average rate of profit, the unit price of its output will have to rise:

c £0.50 + v £0.33 = k £0.83 + p £0.19 = £1.02

But, as Marx says, above, how much supply of I will have to rise to reduce the unit price from £1.00 to £0.98, will depend upon what type of commodity I is, and what the price elasticity for it is. If I is a “dominant agricultural product, where an insufficiency in supply will bring about a much greater rise in price than corresponds to the arithmetical deficiency in supply”, then similarly, it may not require a large rise in supply, in excess of demand, to cause the price to fall by the required amount.

Suppose, however, that, in order to reduce the price of I from £1.00 per unit to £0.98 per unit, the supply must rise by 10% to 137.5 units. In that case, where previously in I, the capital employed consisted of £50 of constant capital and £50 of variable capital, this would also now have to be increased in absolute terms by 10%, to effect this 10% increase in the supply of I commodities. The actual situation in sphere I would then be:

c 55 + v 55 = k £110.00 + p £24.75 = £134.75.

The price of production of I's output then is £134.75, and this now represents 137.5 units, with a unit price of £0.98. 

The capital employed in I has then risen by £10, but this implies that the capital employed in A falls by £10. The capital employed in A would then be:

c 54 + v 36 = k £90.00 + p £20.25 = £110.25

The total capital employed then remains £200 as under prices determined by exchange values; and similarly the total profit is £45, as before, and the total prices of commodities is £245, the same as the total value of commodities.

However, there is a problem here. The increase in the capital in I was determined on the basis of the need for the unit price of I commodities to fall to £0.98 per unit, and, in turn, on the basis that to effect this reduction, supply would have to rise by 10%, requiring £10 of capital to be transferred from A to I. But, there is no reason why £10 of capital having been removed from A, and causing a 10% reduction in the supply of A commodities will result in the price of A commodities rising from £1.00 to £1.02, which is what is required for capital employed in A to obtain the average rate of profit.

Exactly how, the price of A will respond to a 10% reduction in supply, will again depend on the elasticity of demand for A commodities, just as the elasticity of demand for I commodities, determined that the supply of those commodities would have to rise by 10% to effect the fall in price from £1 per unit to £0.98 per unit. 

It could be argued that the elasticity of demand for A and I, as the only two commodities, mutually influence each other. But, that would be to make a variant of the error of Say's Law. In fact, in a money economy, A and I here are, of course, not the only commodities, because in addition to them is the general commodity money. If 10% of capital is withdrawn from the production of A, the supply of A commodities will fall by 10% from 120 to 108. But, unless we know what the elasticity of demand for A commodities is, its impossible to know what effect this reduction in supply will have on the price of A commodities.

On the one hand, if demand for A is relatively inelastic, the initial demand for 120 units may require the price to rise substantially, in order to choke it back to the 108 units of supply. If, in order to reduce demand to the available 108 units of supply, the price per unit rises not to £1.02, but to £1.05, then that would mean that there would be insufficient monetary demand required to buy the 137.5 units of I, at the price of production of £0.98. That would mean that the unit price of I commodities would have to fall below £0.98.

In that case, in terms of realised profits, A would be making a higher rate of profit than I, which would lead to a flow back of capital from I to A, increasing the supply of A, and reducing the supply of I, whilst raising the price of I and lowering the price of A.

On the other hand, if the demand for A is relatively elastic, the demand may be cut from 120 units to 108 units, even if the price rises to only £1.01. At that price, capital employed in A will still make less than average profits. In order to raise the price per unit to £1.02 it may be necessary to reduce supply by a larger amount. Suppose that at a price of £1.02 per unit for the 108 units supplied, demand falls to just 100 units, there would then be an overproduction of A, and the immediate consequence would be a fall in the market price of A.

Capital would be withdrawn from A, so as to reduce supply to 100 units, so that demand was met. It could be that this capital would migrate to the production of I, but as set out above, that in turn depends upon there being sufficient demand for the resultant increased demand for I commodities, which again depends upon the elasticity of demand for I.

Alternatively, it could be that the consequence is that capital is simply withdrawn. Consumers faced with the higher price of A commodities might reduce their demand from 120 to 100, but they may not increase their demand for I commodities correspondingly. Instead they might decide to simply hold on to money.

As Marx puts it in Theories of Surplus Value,

“At a given moment, the supply of all commodities can be greater than the demand for all commodities, since the demand for the general commodity, money, exchange-value, is greater than the demand for all particular commodities, in other words the motive to turn the commodity into money, to realise its exchange-value, prevails over the motive to transform the commodity again into use-value.” (TOSV2 p 505)

In other words, the actual process whereby the price of production of commodities is determined, and that capital is reallocated from one sphere to another to bring about such prices of production, and consequently an average rate of profit, across all spheres of production, is far more complicated once the analysis moves from the abstract level of pure value relations, and the consequences for supply, to include the effects of competition and demand, which thereby effects the determination of actual market prices.

The Transformation Problem Once More

Over the years, the issue of the transformation of values into prices has caused considerable controversy, including in Marx's time itself. Bohm-Bawerk believed that the divergence of values from prices proved that Marx’s theory was wrong. Some of Bohm's own students, such as Von Borkiewitz demonstrated that it was mathematically possible to use Marx's method, set out in Capital III, Chapter 9, to arrive at a set of prices, including input prices, that were transformed from values. In more recent years, there has been debate between the neo-Ricardians, who believed that the work of Pierro Sraffa made the LTV irrelevant, because it was possible to arrive at a set of prices without it, and Marxists who argued that the LTV remained fundamental. There has been debate amongst Marxists, too, over the interpretation of Marx's theory, and his method of transforming values into prices. A lot of this debate is misplaced because it is based on misconceptions of what Marx actually said.

For example, a large part of the debate has been about the fact that Marx does not, in Chapter 9, transform input prices alongside output prices. Some argue that this failure was due to Marx not recognising the need to do so; some that he did, but didn't get round to it; and others that Marx was right not to transform input prices, simultaneously with output prices. Alongside this is the supposed need to maintain a set of constants. For example, its argued that the amount of surplus value remains constant under a regime of prices of production compared to one where prices are determined directly by exchange values. This last claim is false. Marx directly refutes it. I want here to look at what Marx actually does say, in respect of this, and how it relates to what Marx says in general, which shows that he did recognise the need to transform input prices simultaneously with output prices.

What Marx does say, and what must follow from his theory, is that, at an aggregate level, the sum of values must equal the sum of prices of production. In other words, the sum of c + v + s is the same, whether it is measured by values or prices of production. In fact, c + v + s, is merely a quantity of abstract labour-time, so whether it is measured by value – which is nothing other than labour-time – or prices of production, derived from those values, it is tautologically true that the result must be the same. The only difference is how this labour-time is distributed under one regime as opposed to another. Nor can this be affected, as some claim by the introduction of credit money. The measurement of values and prices of production, using some money token, cannot change the underlying relation. A sum of value of £10, is equal to a sum of prices of production of £10, and if the value of money is halved by the introduction of credit money, this does not change this underlying equality. It simply means that the relation is expressed as a sum of value of £20, and a sum of prices of £20!

But, its precisely because of this equality, and because Marx recognises that transformed output prices also cause transformed input prices that the equation of surplus value under one regime as opposed to the other cannot follow, and Marx sets out precisely why that is. That also means that the rate of surplus value cannot be the same under the two regimes, which in turn means that the rate of profit cannot be the same under the two regimes. But, as Marx demonstrates, none of this undermines his theory one bit, or the conclusions he draws from it.

There are several places where Marx makes clear that he understood perfectly well that input prices must be transformed simultaneously with output prices. But, even were that not the case, its necessity is not only practically obvious, but both Marx and Engels, in their historical analysis of how prices of production come into existence, describe the actual situation in which both exchange values and prices of production exist side by side, because non-capitalist producers exist side by side with capitalist producers, although increasingly the former must become modified by the latter, precisely because non-capitalist producers are forced to accept market prices determined by capitalist producers on the basis of prices of production, and because non-capitalist producers themselves have to buy inputs from capitalist producers at market prices determined by prices of production.

I've set out the process by which Marx arrives at prices of production here - Prices Of Production – so I do not intend to rehash that explanation. But Marx, also specifically sets out that input prices must be simultaneously transformed alongside output prices. So, he writes,

Non-capitalist producers bought inputs from capitalist producers.
The inputs are priced according to Prices of Production, so even
when the peasant producer sells according to exchange value, a
part of their own output price has been determined by an input
price determined by prices of production.

“The foregoing statements have at any rate modified the original assumption concerning the determination of the cost-price of commodities. We had originally assumed that the cost-price of a commodity equalled the value of the commodities consumed in its production. But for the buyer the price of production of a specific commodity is its cost-price, and may thus pass as cost-price into the prices of other commodities. Since the price of production may differ from the value of a commodity, it follows that the cost-price of a commodity containing this price of production of another commodity may also stand above or below that portion of its total value derived from the value of the means of production consumed by it. It is necessary to remember this modified significance of the cost-price, and to bear in mind that there is always the possibility of an error if the cost-price of a commodity in any particular sphere is identified with the value of the means of production consumed by it. Our present analysis does not necessitate a closer examination of this point.”

Capital III, Chapter 9 

Marx insists that a characterising feature of capitalist production
is that its continuous.  Capital exists, he says simultaneously
in each of its different forms, and stages, so that inputs
incorporated into the final product, already simultaneously exist
in their reproduced form at the other end of the process, as well as
in the production process itself.  Taking capital as whole, the total
social capital, outputs exist simultaneously as inputs.

In other words, Marx is saying openly here that a capitalist who buys commodities as inputs, buys them not at their values, but at their price of production, i.e. at the output price of some other capitalist producer, and, because capitalist production is a continuous process, output prices are then simultaneously input prices. Because buyers purchase inputs at prices of production, even if this buyer is a non-capitalist producer, who would base their prices upon exchange values, their own prices are necessarily modified by the fact that they have bought inputs at prices of production not exchange values. This is why, Engels says in his Appendix to Capital III, that from the 15th Century, when capitalist production commences, the determination of prices by exchange values ceases. Exchange values can continue to form a large part of the prices set by non-capitalist producers, but the very fact that they buy inputs from capitalist producers means that their own output prices cannot be pure exchange values.

Marx elaborates on this in Capital III, Chapter 12, and here, not only does Marx set out why input prices must be transformed simultaneously with output prices, but he describes why this means that the surplus value differs in one regime compared to the other. He writes,

“We have seen how a deviation in prices of production from values arises from: 1) adding the average profit instead of the surplus-value contained in a commodity to its cost-price; 2) the price of production, which so deviates from the value of a commodity, entering into the cost-price of other commodities as one of its elements, so that the cost-price of a commodity may already contain a deviation from value in those means of production consumed by it, quite aside from a deviation of its own which may arise through a difference between the average profit and the surplus-value. 

It is therefore possible that even the cost-price of commodities produced by capitals of average composition may differ from the sum of the values of the elements which make up this component of their price of production. Suppose, the average composition is 80 c + 20 v. Now, it is possible that in the actual capitals of this composition 80 c may be greater or smaller than the value of c, i.e., the constant capital, because this c may be made up of commodities whose price of production differs from their value.”

Capital III, Chapter 12 

There can be no clearer statement by Marx than this that he recognised that transformed output prices become simultaneously transformed input prices, and that these transformed input prices, thereby form a part of the cost-price of the commodity in whose production they participate. In fact, as Marx says here, this means that even the commodity/industry that represents the average when measured in terms of prices of production may not be so when measured in terms of value, and vice versa.

But, what Marx then goes on to say, draws out the implications of this in relation to surplus value. Marx writes,

The prices of wage goods might be higher or lower when
 determined by prices of production, compared to if they
 are determined by values, because depending upon the
 composition of capital used in their production they may
 obtain a greater or smaller share of the total surplus value
 produced.  But, if the result is higher prices, workers
must perform more necessary and less surplus labour,
 and vice versa.  That affects the amount of surplus
value itself produced, and thereby affects the rate of profit.

“In the same way, 20 v might diverge from its value if the consumption of the wage includes commodities whose price of production diverges from their value; in which case the labourer would work a longer, or shorter, time to buy them back (to replace them) and would thus perform more, or less, necessary labour than would be required if the price of production of such necessities of life coincided with their value.”

So, if prices become determined by prices of production rather than exchange values, the consequence may be that the price of wage goods rises. If the price of wage goods rises, then as Marx says,

“the labourer would work a longer, or shorter, time to buy them back (to replace them) and would thus perform more, or less, necessary labour than would be required if the price of production of such necessities of life coincided with their value.”

The worker thereby produces less surplus value than would be the case if prices were based upon values. So, the idea that there must be an equality between the sum of surplus value where prices are determined by values, and where they are determined by prices of production is false. Marx here directly refutes it. But, this has further ramifications.

The total of c + v + s remains the same under both regimes. But, v + s, must remain constant, because v +s is nothing other than the new value created by labour. It is equal to the labour-time performed by that labour. In fact, its because of this that, as Marx says above, if v is higher when determined by prices of production as opposed to values, then s must be smaller – the rate of surplus value must also then fall. But, if c + v + s remains constant, and v + s remains constant, then c must also remain constant.

This might seem to contradict what Marx says above in relation to the average capital, but it does not. The, average capital, under a regime of prices of production, is not the same capital as that which is the average capital under a regime of values, precisely because input prices are transformed simultaneously with output prices. The total of c, across all capitals, remains constant, but is differently distributed, across different capitals, as a result of the transformation of input prices.

But, if c remains constant, then any change in v, as a result of the transformation into prices of production, thereby increases or decreases both c/v, the organic composition of capital, and c + v, which means that it changes the rate of profit. If v increases, this will be combined with a fall in s to bring about a fall in the rate of profit, and vice versa.

But, none of this undermines Marx’s theory, or the conclusions drawn from it. As Marx points out these laws do not flow from the measurements of different components of capital by value rather than prices of production. They flow rather from the proportions of these components of capital one to another. It does not matter whether v is measured in value terms or in terms of prices of production. If v rises, s falls and vice versa, given a fixed length and intensity of working day. Whether we use values or prices of production, it remains the case that the rate of profit is determined by the ratio of s to c + v. Moreover, the laws that Marx described in Chapter 11, showing the effects of wage rises and falls on capitals of different compositions remain entirely in force, so that a wage rise for capitals of average composition cause no change in the price of production, whereas they cause the prices of commodities produced under conditions of a lower organic composition to rise, and those produced under conditions of a higher organic composition to fall.

“For instance, in a capital of the given composition 80 c + 20 v, the most important thing in determining surplus-value is not whether these figures are expressions of actual values, but how they are related to one another, i.e., whether v = l/5 of the total capital, and c = 4/5. Whenever this is the case, the surplus-value produced by v is, as was assumed, equal to the average profit.”

Value Theory, The Transformation Problem, & Domestic Labour - Part 1

Value Theory And The Transformation Problem

I was reading a review of “The Value Controversy” by Ian Steedman et al (NLB), and “Value: the Representation of Labour in Capitalism” Diane Elson Ed. CSE Books, recently. The review titled “The Remystification of Value” was by Barbara Bradby in Capital & Class 17 (Summer 1982). The two books essentially deal with the same subject, the debate over Marx's Labour Theory Of Value, and its relevance. Steedman was the champion of those Sraffian economists who argued that the LTV was no longer necessary, because Sraffa had shown how the same results could be achieved using Input-Output tables of physical quantities provided some basic assumptions were made. I want here to deal with one aspect around, which much of that debate centred – the question of the Transformation of Values into prices – but also to take up some of the points that Bradby herself makes in relation to Value Theory from a feminist perspective.

Some years ago I began to look again at the so called “Transformation Problem” myself as part of an ongoing analysis of Theories of Value. The task was made easier – particularly given that I am not an advanced mathematician – by the development of powerful tools as standard within spreadsheet programmes such as Excel. The basic problem is this. Marx throughout Capital argued on the basis that commodities exchanged at their Exchange Value. However, a contradiction arises, that he and other Classical Economists were aware of. That contradiction is that in reality different branches of industry operate with markedly different ratios of Constant (Machinery, Building, Materials) Capital to Variable (Labour-Power) Capital. On the assumption that all Labour-Power is exploited at the same rate, then, if all commodities sold at their exchange values, this would result in large variations in the rate of profit. For example assume two companies producing each 10,000 units:

1) C2000 + V4000 + S4000 = K10,000 = R66.6%; Price per unit = £1
2) C4000 + V2000 + S2000 = K8,000 = R33.3%; Price per unit = £0.80

In short, the higher the organic composition of Capital, the higher C in relation to V, the lower the rate of profit. This contradicts the idea of an average rate of profit, and the observable fact that Capital will move to where the rate of profit is higher so that Supply and Demand will tend to bring about this average rate, as market prices adjust accordingly, and thereby diverge from Exchange Values.

Marx was criticised by Bohm-Bawerk on this basis, and other critics believed that it showed his theory was flawed. In fact, even before he published Volume One of Capital, he had already solved this problem. The solution was quite simple:

Marx argued that the process of the circulation of Capital has to be viewed as a whole. On that basis each individual Capital obtains a share of the total amount of Surplus Value extracted from workers, and does so in proportion to the amount of Capital it throws into that overall procees. This is important theoretically also, because it provides a basis for the solidarity of Capital as a whole as against Labour. If we do that in the above example we get:

C6000 + V6000 + S6000 = K18,000 = R50%. If each Capital then makes this average 50% profit we get:

1) C2000 + V4000 + S3000 = K9,000 = R50%; Price per unit = £0.90
2) C4000 + V2000 + S3000 = K9,000 = R50%; Price per unit = £0.90

Now, Marx realised that this solution was only a partial solution, because these changed prices have only been changed as far as Output Prices, but these outputs, are also inputs. A full resolution requires that the values of C and V also have to be changed to reflect these prices rather than values, and this will in turn change the organic composition of Capital, and so on. It was, in fact, one of Bohm-Bawerk's students – Von Bortkiewicz – who first put forward a mathematical solution to this problem, and later Francis Seton provided a further mathematical solution. But controversy continued over the matter. In the review referred to above an article by Anwar Shaikh in the latter book essentially puts forward the solution that I came to in resolving the equations whilst maintaining the basic requirements of the theory – that total Exchange Value Equals Total Prices, that Total Surplus Value equals Total Profits, that the Rate of Exploitation (Ratio of Surplus Value to Variable Capital) is held constant, and finally that real wages are held constant (achieved by the requirement that Money Wages continue to buy the same quantity of wage goods).

I set up a simple two Department model on this basis, and then used the “Solve” function on Excel to arrive at a set of prices that achieved all these conditions. Methodologically, I prefer this iterative solution to soluitons based on solving simultaneous equivalent, because it reflects what I believe is the way in which the problem, the transformation of Values into prices actually takes place in the economy. I will have to set this up again as the original spreadsheets have disappeared. There was no problem in achieving this objective, which demonstrates that the “problem” is solvable, because I had not set up any specific relations or quantities to achieve the result. The only thing that did not conform to Marx's theory was that the Rate of Profit, calculated on prices, differed from the Rate of Profit calculated on Exchange Values. But, as Shaikh argues, this is no objection, precisely for the reason I have set out previously – the Average Rate of Profit only exists as an abstraction not a reality. At any one time there is no single Rate of Profit that applies even across sectors, let alone across all firms. It is a purely abstract calculated figure to indicate the idea that Capital making profits below this level will tend to move away from it, and into areas where above average rates are being made, and indeed it is this very process that is required to establish any materiality to the abstraction, as changes in the level of Supply that result from such movements bring about changes in market prices, and profits.

Forward To Part 2

Prices, Profits and Capital

Over the last couple of years I have been considering two main questions of Economics - Theories of Value, and what is known as the Transformation Problem, the means by which Exchange Values are transformed into market prices. I'm not ready to write anything definitive about either yet, but continuing my reading of Mandel's, "Marxist Economic Theory", See: Mandel's Mistakes in Marxist Economic Theory , I find myself needing to again correct what I consider to be another error in this respect on Mandel's part.

The Transformation Problem

Exchange Value

Its first necessary to explain what the Transformation Problem is. Marx like the other Classical Economists set out how Exchange Value is determined by the amount of abstract, average, simple labour-time that is socially necessary to produce a commodity. This Labour-time can be divided up as follows. First of all a certain amount of past labour-time has gone into the production of the materials used in producing the commodity. In addition, the machinery used for producing the commodity gives up some of its value - what modern Accountancy calls depreciation - because the act of production also causes it to wear out. The cost of its replacement has to be recovered. But, these expenditures of Labour-time have occurred, they have established the Exchange Value of this material, and of the machines used. It cannot produce any new Exchange Value, only pass on the value it possesses into the new commodity. For this reason Marx calls it Constant Capital (C). It would also include things such as factory buildings as well as ancillary materials such as oil to lubricate machines etc.

The other thing that goes into the production of a new commodity is the labour-power that works up this material using the machines etc. This Labour Power like any other commodity has its Exchange Value determined by the labour-time required for its production. In other words how much time is taken to produce the workers needs for food, clothing and shelter and all of those other cultural and physical needs taken as standard for the particular country and time. However, just as was the case for say a slave in the past or a medieval peasant the worker can work for longer than is required to produce this necessary amount. They can produce a surplus. If a worker produces Exchange Value equivalent to this necessary amount in say 4 hours, and receives that value back in the form of wages, but works for 8 hours, they have produced a surplus Exchange Value equivalent to 4 hours. The new Value they have created is equivalent to 8 hours. Four hours of this is merely transferring the value of their own Labour Power into the commodity in the same manner as the Constant Capital, but the 4 hours of Surplus Value they produce is new value, additional value, Exchange Value that previously did not exist. This quality of labour power to create new Value rather than simply transfer existing value into the commodity is what leads Marx to describe it as Variable Capital (V). Technically, this is incorrect. The worker produces a new USE-Value, but as stated above the Exchange Value of this new Use Value that is required to cover the workers reproduction - called necessary labour-time - is not new Exchange Value. It already exists in the form of the workers Labour-power. The new product produced during this time simply reproduces it. It is technically speaking the new value over and above this, the Surplus Value which can properly be described as Variable. But, Marx wants to show that it is the Labour-power which is the genesis of this new value so his use of the term is understandable. The Surplus Value created by this Labour-power is labelled as (S).

We now have the formulae for the Exchange Value of the product. C + V + S. It is important to understand that this breakdown of that value is not like the way in which orthodox economics determines value from the costs of the factor inputs. The Value of the Commodity is not simply a summation of costs, so much for Constant Capital, so much for wages, so much for profits. These are objective quantities of expenditure of Labour-time. The exchange Value of the Constant Capital IS the amount of labour-time required for the production of the various components. Added to this is the average, socially necessary labour-time required for producing the new product - let us say 8 hours. That gives us the total value of the product. But, that sets the limit within which the distribution of that 8 hours is divided between the Capitalist and the worker. S is not an amount of profit simply added on to the costs of production.

Generalised Exchange

Now, when commodities first began to be produced and generalised exchange begins to take place things are not the same as they are under Capitalism proper. As I have shown in my previous blog in relation to the mistakes of Mandel under peasant or artisanal production, in fact, what we have is not the production of surplus value, but the production of surplus use values, that as a consequence of generalised exchange take on the form of commodities that have Exchange Value. The peasant or artisan is not seeking to maximise profit, because they have no compulsion to accumulate Capital. They merely seek in the case of the peasant to sell those goods excess of their own needs in order to buy those things they cannot produce or themselves, and to pay their taxes etc. The artisan does the same except the whole of their production is to be sold to achieve that result, because they are specialists. This is important.

Capitalism Proper

Under Capitalism proper the Capitalist is forced to maximise profit, because that is the way to accumulate Capital, and the Capitalist has to accumulate Capital, because that is the only way to stay ahead of your competitors. If you fail to do that ultimately you go out of business, and become a worker.

So, some important relationships arise under Capitalism, between C,V, and S. The Capitalist wants to maximise S, but because C is fixed, can only do so, by squeezing V i.e. the worker. This can be done by increasing the length of the working time of the worker. In the 19th century workers hours were more than doubled compared to the time the average peasant or even day labourer worked in previous centuries. Today, the retirement age is raised to have the same effect over a worker's lifetime. This means of raising S is called ABSOLUTE Surplus Value. The same effect can be obtained by squeezing more out of workers during the same given period. Achieved by speeding up production lines, telling postmen they have to walk faster etc. The other means of raising V is by cutting V. That can be done by reducing the cost of all those things that the worker has to buy. that was why the capitalists wanted to abolish the Corn Laws against the resistance of the Landlords. By cheapening the price of bread the Capitalists could justify cutting wages. This is called RELATIVE Surplus Value. Capitalism's continual improvements in technology and technique, bring about this cheapening of the worker's requirements. In fact, by this means they can not only reduce V in real terms, but it can be accompanied by actual improvements in workers living standards, as cheaper prices for certain goods mean that workers have money left over to buy a wider range of goods. All that is required is that this cheapening is greater than the actual reduction in workers wages. This is what Marx called the "Civilising Mission" of Capitalism, and stands in complete contrast to the Staliinist and Lassallean notion of "absolute immiseration", replicated by others on the Left like the AWL, who proclaim that "Capitalism creates poverty", in complete contradiction to the observable reality.

So, Capitalism is concerned with the ratio of S/V or the rate of exploitation. But, increasingly, because of technological development, and because Capital seeks to increase Relative Surplus Value, C increases at a faster rate than S. For the Capitalist, increasingly he is concerned with the not only how much he can screw his workers, but with how much profit he makes in proportion to the total amount of Capital he has to lay out to achieve it i.e. with S/C+V, the Rate of Profit. This is important, not because like with say a saver who wants to earn the highest rate of interest on their savings they want to increase their spending power, but because it is the index of how much they can increase their business, and thereby out compete their rivals.

But, this is where a difference with those pre-capitalist forms comes about. Not only does the peasant or artisan not produce Surplus Value per se, but even were we to designate the Exchange Value of their Surplus Product as Surplus Value, they would not be concerned with maximising it, because they were not in competition with other peasants or artisans to the extent of needing to expand their operations or die. Moreover, the amount of C used by such producers was quite small. There could be little variation of S/C+V, even could we correctly apply those terms. That is not the case with Capitalism. Capitalists move their Capital to where they can maximise not just the amount of profit, but also the Rate of Profit.

But, when C begins to expand significantly this has consequences. If we take two firms with the same amount of Capital employed, and assuming that the workers in each are exploited to the same degree we will find that the Rate of profit varies depending upon the ratio of C to V - called the Organic Composition of Capital.

C100 V 400 S 400 Rate of profit 400/500 = 80%

C400 V 100 S 100 Rate of profit 100/500 = 20%.

So it can be seen that firms that are more developed and have more C to V, a higher organic composition of Capital, would on the basis of Exchange Values make less Surplus Value, and a lower rate of profit. Now where this happens in a given line of production what happens is that Exchange Values are determined as the average across all firms in that line of production. So, the firm with the higher organic composition of Capital would not sell its commodities at Exchange Values determined by its own costs, but those of the average. It would have a competitive advantage, and would in fact make a higher profit, and rate of profit than the average firms. The firms with a lower organic composition of capital, who would be less competitive would make less profit, and a lower rate of profit. Over time the less efficient firms go out of business, and so more or less the same organic composition of Capital will apply across an industry.

Values Become Prices

But, that is not the case between industries. The degree to which such an equalisation can occur will depend upon the nature of the industry, the degree to which technological developments can be introduced etc. For example, agriculture for a long time remained very poorly capitalised compared to industry. But, if these variations across industries cause differences in the Rate of profit, and if Capitalists will always seek to move their Capital to where it brings the best return, causing profit rates to be averaged over time, how is this average profit to be brought about. Marx resolved this problem, by showing that in establishing an average rate of profit Capital creates market prices that differ from exchange values, such that these prices cause this average rate of profit to be achieved across the whole of Capital. This is the transformation of Values into prices.

If we take the example above:

C100 V 400 S 400

C400 V 100 S 100

Then the total C is 500, V 500, and S 500. The average rate of profit is 500/1000 = 50%. In order to achieve this average the prices received for its products for the first firm have to be reduced, and those of the second increased.

Previously, total Exchange Value of firm 1 was 900, and this has to fall to 750, a reduction of 150, to bring about a rate of profit of 50%.

This surplus value is now transferred to firm 2. The Value of its product was previously 600, and becomes 750 also producing a rate of profit of 50%.

Mandel's Account

All, of the above more or less mirrors Mandel's explanation of this process other than the fact that I do not accept that Surplus Value is produced other than by wage labour, or at least absent wage labour forming a major part of both production and consumption. I also broadly agree with the first part of how Mandel explains the way in which this averaging of the rate of profit arises.

On page 158, he relates how when after the end of the Napoleonic wars the price of coffee in Europe rose steeply, whilst whereas the price of sugar fell as a result of the introduction of beet sugar, sugar producers in Java, Cuba, Haiti, and San Domingo switched over their plantations to coffee. The fall in the supply of sugar caused its price to rise, whilst the increase in the supply of coffee caused its price to fall, and profit rates were once more equalised. This to me is the correct interpretation as to the way in which Capital is allocated to bring about an average rate of profit, and also the way in which Exchange Values give way under pressure of shifts in demand and supply to market prices.

But, Mandel goes on to give a further explanation, which to me is wholly incorrect. First of all he introduces a concept which I think is unnecessary not to say contrived. He argues that because Capital moves to those areas where the Organic Composition of Capital is low it will bring about an increase in that composition, because the increased competition will encourage a greater use of machinery, and more efficient methods. So it will tend to equalise the organic composition of Capital. I think this is largely nonsense. The example he gives of coffee and sugar shows why. Its unlikely that there was any increase in the organic composition of Capital in Coffee production as a result of more Capital being employed there. The fall in price has nothing to do with any such increase in the organic composition of Capital, but is purely and simply a consequence of changes in demand and supply. In reality what we have here is the following. If we begin from a situation where prices are the same as exchange values, and demand and supply are in dynamic equilibrium then supply will respond both to changes in demand, and changes in profitability. If, there arises a change in the organic composition of Capital in one branch of industry this will cause a divergence in rates of profit. Capital will move from where the rate is lowest, and towards where it is highest. But, this means that Supply falls in one and rises in the other. If Supply and demand were previously in balance then this must mean that it is thrown out of equilibrium. There will be an oversupply in one and an undersupply in another. But, oversupply and undersupply are relative terms. Equilibrium means only that the amount suppliers are prepared to supply at a given market price is equal to the amount consumers will demand at that price. Crucial here is supply. The amount producers are prepared to supply is a function of the profit that can be made at the given market price. As at the original Exchange Values profits either above or below average profits were made this meant in effect that at market prices rather than Exchange Values some commodities were being over supplied and others under supplied. Capitalists are not interested in Exchange Values, but market prices, because it is market prices that they are paid. It is these market prices and the resultant money profits they produce which determine how much they are prepared to supply. It is not under Exchange Values that market equilibrium is attained, but under market prices.

Elasticity of Demand

The amount of Capital that has to exit one sphere and enter another is dependent upon another factor known to Marx, what today Economists call the Price Elasticity of Demand, or in other words the extent to which demand changes in response to changes in prices. Speaking in this regaard Marx talks about the steel industry in Sheffield, and the production of cutlery. He remarks that a cut in prices will result in more knives and forks being demanded, but he continues there are only so many knives and forks people need, however cheap they become. If we take an industry where the Organic Composition is high, and where Exchanges Values lead to below average profits then the degree to which prices and profits rise will depend upon this elasticity of demand. Where demand is inelastic this means that demand does not change much as a result of changes in price. So if prices rise by 10%, firms revenues will rise by 10% if demand is perectly inelastic. In contrast where demand is elastic a rise in price will cause demand to fall a lot, so that the revenues firms take in will not increase much, and might even fall, so profits will not rise much. Consequently, where demand is inelastic less Capital will need to leave that sphere than where demand is elastic.



The above diagrams illustrate this. In Fig 1 to the left we see that from an original position of equilibrium represented by the intersection of SS and DD the price is P, and the quantity demanded and supplied is Q. If we assume that the price has to rise to P1 in order that average profits are achieved then this results in a new equilibrium point at S1S1 DD. At this equilibrium level Supply has to shrink all the way back to Q1. This is because demand is elastic, and as price rises demand falls back sharply as shown by the shallowness of the demand curve DD. By comparison if we draw in a demand curve showing relatively inelastic demand the difference is marked. Fig. 2 to the right shows that now a price rise from P to P1 results in a new equilibrium level at the intersection of S2S2 and D1D1. But now the equilibrium level is achieved with demand and supply only falling back to Q2.

It is apparent then that in industries or for products where price elasticity of demand is higher more Capital will have to be withdrawn (because the level of Supply is a direct function of the Capital employed) in order for an equilibrium level to be achieved that produces average profit.

Socially Necessary?

Keen to show that an average rate of profit exists across the whole of Capital short of the need for this equalisation of the organic composition of Capital, Mandel comes up with what I think is a thoroughly contorted argument. He says that because Exchange Value is based upon socially necessary labour any industry where the Organic composition of Capital is lower than the average is in effect wasting labour-power. Whilst, this argument holds WITHIN a given industry, it is ludicrous to apply it across industries. In what way is it sensible to say, for example, that agriculture is wasting labour, because it has a lower organic composition of Capital than Steel-making? It isn't. Socially necessary means socially necessary given the available techniques, and technology at that particular time. It makes no sense to say that labour is being wasted in agriculture or some other industry because it is not using machinery that does not exist!!!

Its undoubtedly the case that over time increases in demand leading to high prices that attract new Capital will encourage new ways of producing more efficiently. This has nothing directly to do with the fact that there are variations in the organic composition of Capital, and the lower prices are caused by the increased supply not some kind of punishing by consumers of suppliers that don't use enough Constant Capital!! Generally, this works backwards from the consumer to the producers of Constant Capital. But, it is not necessary to argue that the movement in prices is a result of producing below some imaginary socially necessary labour-time that would be the case if non-existent machines did already exist.

There are a number of examples which demonstrate the point. When the price of wool rose, largely due to increased trade bringing about an increased demand for woollen garments, Capital moved into the farming of sheep. This didn't to any significant degree entail an increase nin the organic composition of Capital in sheep farming. It did involve a greater volume of Capital leading to an increased supply of sheep, and consequently lower prices. As the price of wool fell as more wool became available, and as weaving became more profitable Capital moved into weaving, new machines were developed, but now not enough wool could be spun. Prices for woollen thread rose causing Capital to move into spinning. New spinning machines were developed.

The Average Rate of Profit and Mysticism

Mandel is led to these arguments because he sees an average rate of profit as being a feature of Capitalism. It is the average rate of profit itself that MUST be applied to all Capital that then determines prices. This is clearly not true WITHIN industries. Some firms that are more efficient earn higher rates of profit, others lower rates of profit than the average. But, it is not true ACROSS industries either. At any one time some industries will have higher rates of profit than others. It is possible as with almost anything else to take the total profits, and divide them by the total Capital to arrive at an Average Rate of Profit, but in and of itself that is meaningless. To argue as Mandel does that this average can be applied to determine prices is bordering on the mystical. Moreover, the argument Mandel uses here contradicts the correct argument he commenced his exposition with. Indeed to a certain degree it removes the transformation problem altogether. If prices are magically formed in the way that Mandel suggests by the application of the average rate of profit, so that in some way unexplained consumers reduce teh prices they are prepared to pay for goods produed by industries with a low organic composition of capital, and vice versa then the resultant average raate of profit resulting from these prices will give no reason for Capital to move frm one sphere to another! He removes the very mechanism by which the average rate of profit is formed. The real average that Marx is concerned with is the average that arises via competition, and the movement of Capital from those areas where the Rate is low to those where it is high, which of itself through the working of supply and demand brings about changes in market prices, and averages out the rate of profit. But, Capital does not flow easily from one area to another, there are necessary frictions, and so the averaging of the rate of profit can never in reality be achieved. Even besides the continual changes in technology, of the composition of Capital and so on, the average rate of profit is always a moving target which can never be reached. Modern Capitalism achieves it as best it can not in he sphere of production, but in the sphere of fictional Capital, through the instantaneous revaluation of Capital values on the world's Stock Exchanges.

The Concentration and Centralisation of Capital

One final point. Mandel links this argumeent with the concentration and centralisation of Capital, which is an undeniable feature of Capitalism. But, Mandel has in my mind too mechanical a view of this. Its true that Marx talked about this centralisation and concentration of Capital, and the necessary drive towards Monopoly. But Marx not only said that Competition breeds Monopoly, he also said that Monopoly breeds competition. I think that Marx's centralisation and concentration of Capital really refers to that centralisation and concentration in the hands of a few Capitalists.

For example, the Dow Jones Index is an index of the US's 30 biggest companies. Not one single company that was in that index at the beginning of the twentieth century is in it today. Meanwhile companies that did not even exist 10, 20 or 30 years ago today make up some of the biggest companies. So it is clear that the argument about the centralisation and concentration of Capital caannot be used mechanically to argue that some given set of very large companies must dominate the economy. In any one industry it is inevitably true that the largest companies tend to dominate - though again not absolutely as the demise of GM and Ford proves (not though to be replaed by new small companies but by other very large companies like Toyota, VW etc.) - but over time, new industries develop that repalce the old. These new industries are almost inevitably dominated by small not large companies, but these small companies grow rapidly to become large companies. Contrary, to Mandel's assertion, it is usually these new, small dynamic companies that have the highest rates of profit, not the old large companies. It is precisely, this higher rate of profit of these small companies that enables them to grow rapidly.

And again contrary, to Mandel's argument in relation to the organic composition of capital these companies usually have a very low organic composition of Capital. When Bill Gates, Steve Ballmer, and Paul allen started Microsoft they did so with a very small amount of Constant Capital. It amounted to little more than use of Bill Gates parents garage. They did have a large amount of Variable Capitalin the form of their respective skills - even if initially those skills were not remunerated in high salaries. But, it was not the existing giant of he Computer industry - IBM - that was to be the dominant player in the following years, but this tiny start up, which made huge profits, and a very high rate of profit. Nor is that experience untypical. It can be applied to almost any new industry you care to mention.

It is the basic idea behind the Harvard Business School Model of the product Life Cycle. New products are generally developed in developed economies where consumers are more likely to be early uptakers. But, also new products rely on innovation and development. They tend to have a high level of skilled labour in their research and development. Small production runs mean that this labour component tends to be a high proportion compared to any capital used in production. It is only as the product becomes established and demand rises that larger production runs enable a profitable implementation of new production techniques, and eventually as the product becomes mature to the introduction of new machinery, mass production etc. and unskilled labour. As the model shows this is a large part of the explanation of the way multintionals are able to utilise cheap labour economies for the production of commodities in that phase of the product cycle.

What is clear is that although large companies come and go as new industries develop and old ones decline, the ownership of Capital itself becomes more centralised and concentrated in the hands of fewer and fewer people. To get into the wealthiest 1% in he US you need a Net Worth of just $3 million. Most of the peeople in this top 1% are not really Capitalists. The large majority are people who are over 50 years old, nearly all of their wealth is in the form of their home, and their pension fund. In short it is not in the form of productive wealth over which they have direct control. They are basically middle class people, who have had a decent income and been able to buy a nice house, and save for their retirement. By contrast, Bill Gates has a personal net worth of around $40 billion, or more than 10,000 times that of most of these people. The real ownership of productive wealth in the US is probably concentrated in the hands of not the top 1%, but the top tenth of 1%, or even less than that. In other words in the hands of around 250,000 people. Very few of these people with that extent of wealth will find themselves being cast down into teh ranks of the middle class let alone the proletariat. 1% interest a year on $40 billion would give you an annual income of $400 million!!! But, the entrance every so often of the odd Bill Gates into this exclusive club can hardly change its nature as a stable ruling class. Even 10 Bill Gates every decade hardly changes the nature of a group of 250,000.