Groucho Marxism

The labour theory of value, henceforth LTV, posits that the exchange value of a commodity is proportional to the socially necessary labour time required to produce it. In a 1993 article, the Australian economist Steve Keen launched an attack on the LTV. In this blog post I will summarize his argument. Keen begins by rightly pointing out that, although Marx was the greatest champion of the LTV, many (although not all) Marxists have since abandoned it. He also points out that Marx did not immediately adopt the LTV, but developed an acceptance of it over a period of time, in parallel to developing an understanding of the concept of use value. The concept of use value did not play a role in Marx’s early writings.

It was only later that Marx came to understand that use value and exchange value are inseparable dialectical aspects of the commodity, which itself is central to the analysis of capitalism. He then applied this dialectic to provide an explanation of the origin of surplus value. Marx argued that the exchange-value of commodities themselves cannot be the source of surplus value, as exchange involves transfer of equivalents. It follows that the dialectical opposite of value, use value, is the only possible source of surplus value, so the source of surplus value must lie in the quantitative difference between the use value and exchange value of labor-power. However, Keen points out that this is just one source of surplus value, not necessarily the only source.

Keen argues that Marx reached the conclusion that labour power is the only source of surplus value by contradicting basic premise; namely, that the use value and exchange value of a commodity are unrelated. Specifically, Keen argues that Marx attempted to forge an equality between the use value and exchange value of the means of production, by equating the depreciation of a machine to its productive capacity. This is equivalent to asserting that in the case of machinery and raw materials, what is consumed by the purchaser is not their use value, as with all other commodities, but their exchange value. This suggests that Marx reached the result that the means of production cannot generate surplus value by confusing depreciation with value creation.

In practice, there is no reason why the value lost by a machine – i.e. depreciation – should be equivalent to the value added by it. Depreciation can be equated to exchange value, while a machine’s contribution to production is its use value. The use value of a machine will differ from its exchange value; and, as with labor, its use value may be significantly greater than its exchange value. Keen goes on to argue that In his algebraic explorations of value creation, Marx compounded his previous logical errors by using the same magnitude for the exchange value and the use value of the means of production, whilst using different magnitudes for the exchange value and use value of labor power. Let us go through this algebra in a bit more detail.

The gross output of the production process is given by C+V+S, where C is ‘constant capital’ – the exchange-value of the means of production – V is ‘variable capital’ – the exchange-value of labor power –  and S is surplus value. Keen argues that Marx identifies the use value of labour power as V+S, and the use value of the means of production as C. The latter identification clearly contradicts Marx’s fundamental and oft-repeated proposition that use value and exchange value are unrelated. Keen concludes that Marx’s claim that labour power is the only possible surplus is based on this false identification of use value and exchange value. Moreover, Keen claims to have reached this conclusion by applying Marx’s own logic.

To make this a bit more rigorous, consider an economy which produces m commodities using n types of labour. For such an economy we can define a commodity vector as a vector with positive elements of length m, and a labour vector as a vector with positive elements of length n. The economy is defined by an activity set, with the interpretation that an element (x,u,x’) of this set represents a possible configuration of commodity inputs x, labour inputs u, and commodity outputs x’. We may also define a  price vector as a vector with positive elements of length m; a depreciation matrix as an mxm diagonal matrix with values between 0 and 1 on the diagonal, and zeroes elsewhere; and a wage vector as a vector with positive elements of length n.

The surplus value associated with activity (x,u,x’), price vector p, depreciation rate matrix D, and wage vector w, is given by: S = p(x’-x)-pDx-wu. The quantity p(x’-x) corresponds to gross output, and the quantities pDx and wu correspond to constant capital C and variable capital V (see above). The associated profit rate is given by: r = S/(C+V); rearranging gives: p(x’-x) = (pDx+wu)(1+r). The production price vector associated with the activity (x,u,x’), wage vector w, depreciation matrix D, and profit rate r, is the price vector p satisfying this equation. If n = 1, so there is only one type of labour, the associated labour values may be defined as the vector v satisfying: v(x’-x) = (vDx+u)(1+r). We then have wv(x’-x) = (wvDx+wu)(1+r), so p = wv and production prices are proportional to values.

Production prices can be thought of exchange values expressed in monetary terms. Thus, in the case were there is only one type of labour, we have effectively defined labour values in such a way that exchange values are proportional to them, with the constant of proportionality equal to the average wage. Under this definition, therefore, the labour theory of value holds. However, a critic might argue that this definition of labour values is contrived and chosen specifically to produce the result we want. I will return to this point in a future blog post.