First off, I want to apologise for the diary being late this week; a series of unforeseen and unimaginable adventures with EDF energy (the former electrical board) led to my increasing hysteria and inability to get the diary out yesterday. Sorry!
Last week’s diary was a discussion of the structure of classical economic theory, the theory of profits and the role of the labour theory of value in Ricardo’s analysis.
Today we are continuing our discussion of classical economic theory by examining the labour theory of value of David Ricardo and the problems introduced into the theory due to his modifications.
Given the data of classical economic theory [the level and composition of output (P), the technical conditions of production and the physical real wage (w)], we were enabled to determine the quantity of labour employed to produce the output and the amount of total output that was required to reproduce the economy at the same level, aka necessary consumption or N). The surplus product was determined as a residual of the total output minus necessary consumption. In Ricardo, the rate of profits (r) is then determined by this relation:
r = P-N/N or r = s/v (in Marx’s terminology).
From this we are able to determine natural prices or exchange values:
Pn = wL + (1+r)K
Or
Pn = (wL + K) + rK
This looks fine, what is the problem? The problem is that the total output of commodities and necessary consumption are composed of a list of heterogeneous commodities, not all of which are the same, so we cannot subtract or divide them (it is literally dividing apples and oranges by apples). This mean that we need to treat these goods as value magnitudes, that is by their exchange value.
But the problem is, that in order to determine the prices of these goods, we need to know the rate of profits and in order to determine the rate of profits, we will need to know the rate of profits ... a nice bit of circular reasoning. This leads us to Ricardo’s labour theory of value. Since we know the technical conditions of production (the quantity of labour, land and capital required for the production of output) we do know the quantity of labour required for the production of every commodity and since we know this prior to the determination of profits, we can use the quantity of labour required for production of the total output and necessary consumption to determine the rate of profits and then the whole theory works. Right? Well, unfortunately, not really as we shall see below.
Let's look at Ricardo's theory of exchange value. The initial version of this theory, prior to Ricardo's modifications is a simple labour theory of value, by this I specifically mean the labour determinant. Note that I am separating off the labour embodied measure and the labour source (which we discussed last week and since it is non-quantifiable, I argued that in Ricardo, it is non-analytical) and that I am not including them in my definition of a labour theory of value.
Ricardo on the determinant of value
Ricardo opens his Principles of Political Economy and Taxation with his discussion on the exchange values of commodities. Ricardo did not agree with Smith that the exchange values of commodities were determined by the summation of the natural rates of wages, profits, and rents (this is known as the adding-up theory of price). In fact, Ricardo argued that rents were price determined, that means that rents would exist, only if the exchange values of commodities were high enough. Also, as we have seen Ricardo determined the rate of profits on the no-rent land used in agricultural production. Ricardo's discussion of exchange value is concerned with the primary determinants of price which are wages and the rate of profits.
Ricardo begins his theory of value by stating that Smith was incorrect to argue that once capital had been accumulated and land appropriated that the quantity of labour used in production no longer determines the price of commodities. Ricardo argued that profits (or accumulation) could not independently affect the exchange values of commodities without any regard to the different durability of capital. Smith's argument implied to Ricardo that in the transition from the rude and early state to the capitalist stage, that the accumulation of capital meant that the exchange values of commodities had to rise in order that profits could exist, and hence that the quantity of labour no longer determined the exchange values of commodities. Ricardo argues that this is incorrect and that the quantity of labour embodied in commodities determines its exchange value.
The Labour Theory of Value
Ricardo opens his discussion of exchange value by distinguishing between the use value and exchange value of commodities. According to Ricardo, all commodities which are produced have use value or utility -- that means that a commodity must in some way contribute to our enjoyment or a commodity would not be produced and hence would not have exchange value.
Possessing utility or use value, commodities derive their exchange value from two things: their scarcity and their exchange value. Some commodities have their exchange value determined solely by their scarcity alone. In this case, no labour can increase the quantity of such goods and their value can not be decreased by a greater supply of the commodity. These types of commodities, such as rare wines, paintings by great masters, are of a very limited number according to Ricardo and their value is totally independent of the quantity of labour required for their production.
The largest amount of commodities which are objects of desire are those that can be increased by the application of human labour; they can be produced as plentifully as we want if we are willing to provide the labour to produce them. When Ricardo is discussing the exchange values of commodities he is solely concerned with those commodities which can be increased in quantity by the application of human labour and on whose production competition operates without restraint.
Ricardo begins his analysis by stating that in the early stages of society (equivalent to Smith's rude and early state), the exchange value of produced commodities is almost exclusively determined by the relative quantity of labour which is required for each of their production. If the quantity of labour realised in commodities regulates their exchange value, then every increase in the quantity of labour required for their production must increase the exchange value of that commodity and every decrease must lower it.
The first section of chapter one, "On Value" contains a criticism of the measure of value chosen by Adam Smith and Malthus (Malthus also uses a labour commanded measure of value or the quantity of commodities commanded by labour, i.e., the wage). Ricardo begins this criticism with a discussion of the source or foundation of value, where he appears to link the source or foundation to the regulator or determinant of value. I will quote the relevant parts from the 3 paragraphs:
It is natural that what is usually the produce of two days', or two hours' labour, should be worth double of what is usually the produce of one days', or one hours' labour (Ricardo, 1821, p. 13).
In the next section, he is discussing the labour determinant and the labour source as he speaks of what regulates exchange value -- exchange value should relate to the difficulty of obtaining objects.
That this is really the foundation of the exchangeable value of all things, excepting those which cannot be increased by human industry, is a doctrine of the utmost importance in political economy; for from no source do so many errors, and so much difference of opinion in that science proceed, as from the vague ideas which are attached to the word value (Ricardo, 1821, p. 13).
now he clearly discusses the labour source and labour as providing the basis for the exchange value of commodities:
Adam Smith, who so accurately defined the original source of exchangeable value, and who was bound in consistency to maintain, that all things became more or less valuable in proportion as more or less labour was bestowed on their production, has himself erected another standard measure of value, and speaks of things being more or less valuable, in proportion as they will exchange for more or less of this standard measure. Sometimes he speaks of corn, at other times of labour, as a standard measure; not the quantity of labour bestowed on the production of any object, but the quantity which it can command in the market; as if these were two equivalent expressions, as if because a man's labour had become doubly efficient, and he could therefore produce twice the quantity of a commodity, he would necessarily receive twice the former quantity in exchange for it (Ricardo, 1821, pp 13-14).
Ricardo moves from the source to the determinant to the measure of value, where he argues that Smith has put forward an inconsistent argument, as it is only when wages are proportional to what the worker has produced that the quantity of labour commanded and the quantity of labour embodied are equivalent. In other words, these measures are equivalent only when the worker receives the whole product of his labour, which only occurs in the early and rude state.
Ricardo then argues that the quantity of labour commanded and also the price of corn is variable, they are not constant over time. In order to maintain the proportionality between labour commanded and labour embodied (in order to examine the impact of technical change on prices) Smith had to assume a constant value of corn.
However, given Ricardo's (and Malthus's for that matter) theory of rent, the price of corn is changing over time due to revolutions in techniques of production in corn (intensive differential rent) and the use of land of lesser and lesser fertilities -- thereby leading to a rising price of corn over time. In this case, Smith's proportionality breaks down and hence cannot serve as a standard invariable measure of value.
Ricardo then demonstrates that changes in the quantity of labour used in production would, if the money wage remains constant, enable the workers to buy a more varied consumption bundle. In this situation, if wages remain constant, there is no change in the quantity of labour commanded, while the value of the commodities composing the wage bundle would have changed in terms of the quantity of labour embodied.
Then Ricardo argues the case for a measure of value which would enable us to determine when there has been a change in the relative values of two commodities, in which commodity's value there actually had been an alteration. If we have a commodity whose value is constant over time, that is if the quantity of labour required in its production is reasonably constant over time and then that is taken as the measure of value, then we would always know, if we were to compare the exchange values of other commodities to that commodity, that the other commodity has changed in value, that is, that it required less labour in its production relative to the invariable measure of value. Is this clear? Is it clear why Ricardo wants an invariable measure of value, he wants to be able to compare relative values, such that changes in the relative exchange value of commodities can always be traced back to the commodity whose price has changed.
Since Ricardo is cognizant of the fact that there are differences in the qualities of labour available, he argues that the different qualities of labour stand in proportion to each other and that the relative differences would not affect the exchange value of commodities, as the differences in quality stand in relation to each other on a scale and do not alter over time substantially.
Ricardo next considers the capital goods used in production, which as he points out exist in the rude and early state as well as in capitalism proper. He argues that not only the direct labour used in production regulates exchange value, but the labour which has already been embodied in the capital goods used in production (the indirect labour) would affect the exchange values of commodities. Ricardo formally argues that the direct and indirect labour used in the production of commodities determines the exchange values of commodities.
Thus, the labour embodied in the different capital good must be taken into consideration -- now capital goods are not all the same, some require more labour for their production, some capital goods last for a longer time than others (they are more durable). Ricardo argues that capital goods transfer their value to the commodity during production. Thus capital goods, which are seen to be composed of indirect labour, transfer that labour which is embodied in them to the commodity during production. Those capital goods that last more than one production period only transfer a portion of the labour embodied in them during the production process. Thus, a capital good which is more durable will transfer a smaller portion of its value during production than a capital good which is less durable.
Ricardo argues that commodities exchange in proportion to the relative quantities of labour used in their production, and that labour includes the total quantity of labour (direct and indirect) required to produce them and to bring them to the market. The aggregate sum of the various kinds of labour determines the quantity of other things for which a commodity (say socks) will exchange and the various kinds of labour used to produce another commodity (say pants) will determine how many socks we can obtain for one pair of pants. Is that clear?
If we develop an improvement which will decrease the quantity of labour required in any of the intermediate processes through which our socks must pass, from raw cotton up until the socks are brought to market what will be the result? If less people are required to cultivate cotton, if fewer hands were responsible for transporting the cotton to the factories, if less labour was used to create the factories or the machines used in them or if these machines were made more efficient (like the power loom was in the 19th century), then the exchange values of socks would fall in value and hence would obtain less of other commodities in exchange. Is that clear, do you understand? They would fall relative to other commodities because a lesser quantity of labour is needed for their production relative to other commodities whose labour required for production has not changed.
If we assume that we are producing two commodities and that the capitals required in production use the same quantity of labour to produce them, and have the same durability, in this circumstance, the relative values of these two commodities would be determined solely by the quantity of labour required to produce them, whatever the quantity of each commodity produced and whatever the rates of wages and profits. Thus, if we assume equal proportions of capital and labour used in the production of two commodities, those commodities will exchange according to the relative quantities of labour used for their production. Is that clear?
Given the same situation as before, if with the same quantity of labour we could produce more of one commodity than the other, the exchange value of the one we produced less of would be greater than the one which we produced more of. Is that clear?
Now, one again back to the invariable measure of value. If there was a commodity which was invariable in its value, that is it always required the same quantity of labour in its production, then we could compare other commodities to that and see that if there was a change in the relative value of socks and pants, we would be able to see in comparison with our invariable commodity, which of our two commodities had altered in price, do you understand?
This invariable commodity would serve the role of a measure or standard of value, it would be the numeraire to which all other commodities would be compared. What we have been considering is a simple labour theory of value, which always works, when the commodities that are produced require the same capital\labour (K/L) ratio in their production, or as Ricardo describes it, they require equal proportions of capital and labour in their production. Another way to put the same thing, is that they require capital goods of equal durability which require the same quantity of labour to produce them. What happens when that is not the case? Ricardo considers this next and is forced to modify his theory of value.
b) The Modifications
Ricardo considers the case where commodities require the use of machinery, and fixed and durable capital in their production in section IV of his discussion of value. This forces him to modify considerably his theory of value, many would argue that these modifications overturn his labour theory of value.
Ricardo recognizes that the tools, implements, machinery and buildings used in the production of different commodities may be of different durabilities and may require different proportions of capital and labour to produce them. The differences in durabilities and in the proportions of capital and labour used in the production of different commodities introduce an additional cause, besides the quantities of labour used in production, which will affect the relative exchange values of the different commodities and the variations in their relative values.
Ricardo distinguishes between circulating and fixed capital; circulating capital is used up in the production process (as such it transfers is value directly to the commodity), fixed capital will last for more than one production period, it transfers its value to the commodity over many time periods.
But although commodities produced under similar circumstances would not vary with respect to each other, from any cause but an addition or diminution of the quantity of labour necessary to produce one or other of them, yet compared with other not produced with the same proportionate quantity of fixed capital. They would vary from the other cause [...] a rise in the value or labour, although neither more not less labour were employed in the production of either of them (Ricardo, 1821, p 33).
The differences between commodities produced with differing proportions of capital and labour or capital goods with more or less durability affects the exchange values of the commodities independently of the quantity of labour required for their production. How does this happen?
There can be no rise in the value of labour without a fall of profits [...]. Suppose then, that owing to a rise of wages, profits falls from 10 to 9 percent [...], the manufactured goods in which more fixed capital was employed, would fall relatively to corn or to any other in which a less portion of fixed capital entered. The degree of alterative in the relative value of goods, on account of a rise or fall of labour, would depend on the proportion which the fixed capital bore to the whole capital employed. All commodities which are produced by very valuable machinery, or in very valuable buildings, or which require a great length of time before they can be brought to market, would fall in relative value, while all those which were chiefly produced by labour, or which would be speedily brought to market would rise in relative value (Ricardo, 1821, p 35).
Commodities which use a greater proportion of capital relative to labour or more durable capital goods would react differently than commodities which use proportionally more labour relative to capital or less durable capital goods when distribution alters. Those commodities which use proportionally more capital would fall in exchange value when the wage/rate of profit (w/r) ratio rises relative to commodities that require more labour. Those commodities which use more labour proportionally to capital will rise in price when the w/r ratio rises relative to commodities which require more capital proportionally. Can you see why this would be the case? Commodities which require more capital relative to labour are more sensitive to changes in the rate of profits, while commodities that require more labour relative to capital are more sensitive to changes in wages. Thus, their exchange values will react differently to changes in the w/r ratio.
In terms of durability, the less durable a fixed capital good is, then the closer its nature becomes to the situation where our commodity uses relatively more labour than capital. Thus, a rise in the w/r will cause that commodity produced with less durable capital to rise in price relative to a commodity which is produced with more durable capital. Is that clear?
Think of our price equation abstracting from depreciation for a moment:
pa = wLa + rKa
pb = wLb + rKb
We are examining relative prices of commodities, let's say that commodity a requires proportionally more capital relative to labour (compared to commodity b) and that commodity b requires more labour relative to capital as compared to commodity a. If there is an increase in w/r, then the commodity which uses more labour would rise in prices relative to the commodity which requires more capital.
We can see this simply:. let's say that the initial w/r was .3/.7; commodity A uses K/L as 7/3 and commodity B uses K/L as 3/7. So, initially pa/pb = 5.8/4.2. Let's change w/r to .4/.6, now, pa/pb = 5.4/4.6.
So with no change in the quantity of labour required for production, changes in distribution (w/r) have caused changes in relative prices (pa/pb).
Different proportions of capital and labour used in a commodity's production and different durabilities of the capital goods used in production will modify the exchange value of commodities independently of changes in the quantity of labour required for its production. Changes in distribution will affect exchange values, because of the different proportions of capital and labour used in the production of the different commodities.
Moreover, and this is something Ricardo does not consider but which is discussed in depth by Sraffa, the differences in the proportions of capital and labour used in the production of the capital goods, will make the effects on exchange value due to changes in distribution indeterminate.
What does this mean for Ricardo's labour theory of value, what is the difficulty if there are different capital-labour ratios used in the production process? Changes in distribution will change the prices of commodities, thus commodities which use capital of shorter duration or more labour in their production are more sensitive to changes in the wage, while commodities which use capital goods which last longer or more capital relative to labour are more sensitive to changes in profits. Thus a change in the w/r relationship will affect different commodities differently. In this situation, the exchange values of commodities are not determined by the quantity of labour solely, in fact the labour theory of value no longer works.
The labour theory of value only works if there are equal proportions of capital and labour entering the production of every commodity and the capital goods used have the same durability. When there are equal proportions of capital and labour used in production, then changes in distribution will affect all commodities equally, and exchange values will not deviate from the quantity of labour required for their production. Do you understand?
For fans of the Cambridge capital controversies: you should notice that this is the same problem which vitiates Samuelson's surrogate production function once he deviates from equal capital-labour ratios used in the production of the consumer and intermediate goods. The same problem arises in classical theory, that is, the specification of the capital goods independently of distribution -- however, the problem is manifested differently -- in the case of Ricardo, different proportions of capital and labour used in the production of different commodities mean that the labour theory of value -- the labour determinant -- no longer holds.
Commodities do not exchange according to the quantity of labour embodied in their production, exchange values are affected by changes in distribution. Again, how does this affect Ricardo's theory of value and distribution? Remember, Sraffa argued that Ricardo used the labour theory of value in the attempt to determine the general rate of profits and hence demonstrate the inverse relation between wages and the rate of profits.
Ricardo had thought that the quantity of labour used in production, which is known prior to the determination of distribution, would serve to relate the heterogeneous commodities comprising the rate of profits, such that the rate of profits could be determined. What happens if distribution affects exchange values independently of the quantity of labour used in production, can we determine the exchange values of commodities comprising the rate of profits independently of distribution -- in other words can we determine the rate of profits? This brings us to part III of Ricardo’s theory of value, the discussion of an invariable standard/measure of value and the notion of absolute value.
Suggested Readings
Garegnani, P (1984) "Value and Distribution in the Classical Economists and Marx," Oxford Economic Papers, 36, pp. 291-325.
Ricardo, D. (1815) An Essay on the Influence of a Low Price of Corn on the Profits of Stock, in Pamphlets and Papers, 1815-1823, Volume IV of the Collected Works and Correspondence of David Ricardo, ed. By P. Sraffa, Cambridge University Press, 1951.
Ricardo, D (1821) Principles of Political Economy and Taxation, Volume I of the Collected Works and Correspondence of David Ricardo, ed. By P. Sraffa, Cambridge University Press, 1951.
Ricardo, D. (1823) Absolute Value and Exchangeable Value, in Pamphlets and Papers, 1815-1823, Volume IV of the Collected Works and Correspondence of David Ricardo, ed. By P. Sraffa, Cambridge University Press, 1951.
Sraffa, P. (1951) Introduction to Principles of Political Economy and Taxation, Volume I of the Collected Works and Correspondence of David Ricardo, ed. By P. Sraffa, Cambridge University Press, 1951.
Sraffa, P (1960) Production of Commodities by Means of Commodities, Cambridge University Press.
For alternative perspectives and a recent discussion:
Hollander, S (1979) Studies in Classical Political Economy/II The Economics of David Ricardo (Toronto: UTP and London: Heinemann).
Blaug, M. (1999)" Misunderstanding Classical Economics: The Sraffian Interpretation of the Surplus Approach, History of Political Economy 1999 31(2):213-236.
Peach, T. (1993), "Interpreting Ricardo", Cambridge University Press.
Garegnani, (2002) "Misunderstanding Classical Economics? A Reply to Blaug," History of Political Economy 2002 34(1):241-254
Previous diaries on the history of economic thought in this series are:
Smith and the Physiocrats on Economic Growth: http://www.dailykos.com/...
Smith on Wages:
Part I: http://www.dailykos.com/...
Part II:http://www.dailykos.com/...
Part III: http://www.dailykos.com/...
Ricardo on Profits and Value: http://www.dailykos.com/...