Do Sectoral Rates of Surplus Value Tend to Equalize, and Why Ask?

It is suggested in an earlier post that the presence of complex labor affects value creation in Marx’s theory, including at the aggregate level. The argument starts from Marx’s distinction between concrete and abstract labor. In making this distinction, Marx explicitly identifies productivity as a property of concrete labor and labor complexity as a property of abstract labor. Variations in productivity, since they relate to concrete labor, directly affect the aggregate production of use-values (‘real’ output) but not the aggregate creation of value (socially necessary labor time or nominal value). In contrast, variations in labor complexity, since they relate to abstract labor (the substance of value for Marx), directly affect the aggregate creation of value rather than the aggregate production of use values.

If it is true that labor complexity affects value creation at the aggregate level, then empirical macro analysis from a Marxist perspective – and not just empirical micro analysis – is made more difficult. Nevertheless, the difficulty largely dissolves so long as it is safe to assume that there is a tendency for sectoral rates of surplus value to equalize. Under an assumption of equal rates of surplus value, labor complexity is easily discerned from wage relativities.

Marx did in fact suggest that such a tendency operates:

Other distinctions, for instance in the level of wages, depend to a large measure on the distinction between simple and complex labour that was mentioned already in the first chapter of Volume 1, p.135, and although they make the lot of the workers in different spheres of production very unequal, they in no way affect the degree of exploitation of labour in these various spheres. If the work of a goldsmith is paid at a higher rate than that of a day-labourer, for example, the former’s surplus labour also produces a correspondingly greater surplus-value than does that of the latter. (Capital, Vol. 3, Penguin, 1981, p. 241)

Marx is asserting that the complexity of labor does not affect the proportions in which value added replaces variable capital or represents surplus value. According to Marx, a worker who is paid more, because the labor is more complex, creates “correspondingly greater surplus-value” than a worker performing simple labor. In other words, the rates of surplus value applying to complex and simple labor, in Marx’s view, are the same.

For this to be so, there needs to be some kind of social process that tends to bring wages and labor complexity into line. Whether it is mostly labor complexity that determines wages, or the other way round, is a matter that would seem to depend on institutional particularities.

It is doubtful that the required tendency for wages and labor complexity to correspond can be brought about solely through competition between firms in product markets. According to Marx, a sector’s rate of profit, which in principle is observable, does not correspond to its rate of surplus value except in the socially average case. Instead, sectoral rates of profit diverge systematically from sectoral rates of surplus value according to sectoral compositions of capital. Firms are accountable to their owners for realizing the general rate of profit, making allowances for risk, not for extracting or realizing a particular rate of surplus value. Put another way, firms compete on the basis of prices, not values.

The onus is therefore on some other process to cause an equalization of sectoral rates of surplus value. A likely candidate is competition between workers. The separate roles played, in Marx’s theory, by competition between workers and competition between firms is emphasized in an interesting paper by Jonathan Cogliano.

For each job, the onus will be on competition between workers either to push the wage to the level reflecting labor complexity (via a raising or lowering of wage demands / wage offers) or to push labor complexity to the level corresponding to the wage (via a raising or lowering of entry requirements), or some combination of the two effects. If a job’s wage persistently remains above the level reflecting labor complexity, it is plausible that an excess supply of qualified applicants will be attracted to the job and, depending upon the institutional context, exert downward pressure on the wage and/or upward pressure on entry requirements. In such a scenario it will be worthwhile for a potential future employee, in competition with other potential future employees and incumbents, either to undercut the prevailing wage (if this is possible in the institutional setting) or to undergo education or training beyond what is currently sufficient to fill the role. This incentive is in place because, in Marx’s terms, the ‘real’ wage currently exceeds the value of labor power (the cost to the worker of reproducing labor power of the required complexity). Conversely, when a wage remains persistently below the level reflecting labor complexity, there will be more incentive to hold out for a better wage and/or less incentive to develop labor power of the current complexity.

If operative, this tendency to an equalization of rates of surplus value would take time and so only be a long-run tendency. But, by the same token, a long time has already elapsed in the case of many jobs, so perhaps much of this process has already played out as a product of history. The possibility of causation running both ways (from labor complexity to wages and from wages to labor complexity) suggests that monopolistic impediments to the process might at least partially be countered by variations in the difficulty of obtaining positions (variations in entry requirements), which will imply variations in the value of the labor power involved and hence to the level that the wage, in Marx’s theory, tends. For example, the training required in some categories of skilled labor can be very costly to society, and not just to the worker, to the point where it becomes socially wasteful to train more than the required amount of workers for the roles. University entry into Medicine, for instance, may be intentionally limited to the number of positions a society wishes to create. The setting of doctors’ pay above the level that would strictly reflect the complexity of their labor would not leave the profession open to an extra influx of doctors. Nevertheless, the extra competition to gain places would tend to raise entry requirements at the university level and so require, on average, more effort and/or higher academic (and perhaps extracurricular) attainment from candidates. Likewise, if doctors’ pay fell below the level consistent with labor complexity, entry requirements would tend to become more lax (in particular, university placements would become somewhat less sought after).

Why ask the question?

The primary reason for asking whether sectoral rates of surplus value can safely be regarded as equal is that, if they can, then a ‘real’ labor-time measure of marxian value added can be estimated empirically. If not, and if it is true that labor complexity affects aggregate value creation (as suggested in the earlier post), then the empirical status of marxian value is not much better than that of utility in neoclassical theory. It would still have important theoretical uses, as does utility within the neoclassical paradigm, but would not be amenable to empirical work.

A secondary reason for asking the question is that it would provide a correspondence between the aggregate analyses of Marx and Keynes. The level of effective demand, as outlined by Keynes in chapter 3 of the General Theory, is indicated by the point of intersection between the D and Z curves, both curves relating the level of employment, weighted by skill, to expected proceeds. If it is assumed, for a moment, that all labor is productive, in Marx’s sense, and that sectoral rates of surplus value can reasonably be supposed equal, then there will be a correspondence between the two approaches when it comes to macroeconomics. The level of employment weighted by skill will provide a ‘real’ labor-time measure of total marxian value added, expressed in hours of simple labor, with expected proceeds (if net of depreciation) corresponding to expected total value added expressed in nominal terms.

The productive/unproductive distinction, if applicable, complicates the correspondence between the macro analyses of Marx and Keynes, though not in a damaging way. It becomes necessary to demarcate productive from unproductive labor, which is a nuisance but not an insurmountably difficult task. It makes the labor-time measure of value added smaller than non-Marxist measures of employment. This will just mean that marxian value added, measured in ‘real’ labor-time terms, will be some fraction of total (skill-adjusted) employment.

For what it’s worth, I am skeptical that the productive/unproductive distinction actually applies in modern monetary systems, in part for reasons stated previously. But while dropping the distinction in the case of modern monetary systems would simplify the correspondence between Marx and Keynes at the macro level, retaining it is not fatal.