[OPE] Labour productivity

From: Jurriaan Bendien (adsl675281@tiscali.nl)
Date: Wed Jun 04 2008 - 16:14:40 EDT

One of the great mysteries of modern society is that while there is an obsession about productivity, it is perpetually controversial just how one ought to define it. As Ian Wright point out once, which I think is a very important insight, it is not clear how exactly the profit accruing to owners of the factor capital is quantitatively connected to the "productive contribution" of the capital they own, because, in practice, within the capitalist firm, no generally accepted procedure or convention exists for measuring such a "productive contribution" and for distributing the residual income accordingly. 

What the productive contribution is, is therefore mainly just inferred by deducting costs from revenues in some way, but all that says is that by accounting definition (i.e. tautologically) the productive contribution is basically sales less costs, the residual item in the account - yet how this residual item will actually be distributed is not even directly determined by production. This leads to the view that, what profits are, could be interpreted in many different ways (reward for risk, delayed gratification, entrepreneurial energy, rents, results of market imperfections etc.). 

The simple insight I have to offer is just that the "net output values" cited in official statistics or accounts do not really refer to the sums of the unit-prices for products sold which are charged by producers, but rather to the gross value added (which is not even necessarily equal to the value created). Also, it is impossible to tell from data on labour hours paid, just how much of that time is directly implicated in the work done to create the output produced. 

Unit labour costs are often calculated as the share of labour compensation in either gross output or net output (often net output), but if those output measures do not really refer to an aggregated value of product units, then the measure of unit labour costs does not truly measure what it purports to measure - it may measure only the share of labour compensation in gross value added, not the value of the labour share in real product unit-values. That aside, "compensation of employees" in the account is often an inaccurate measure of true labour costs.

So really when macroeconomic measures as used to justify arguments about productivity, the trouble is that by definition mostly the concepts involved are not adequate to the task. That in turn means that a lot of the debates about this are not very fruitful, because in truth there are no standard empirical data which would adjudicate either way. All that happens is that inferences are drawn from flows of costs and revenues about productivity - depending on how the relationship between costs and revenues evolves, productivity is said be be increasing or decreasing. This can cause absurdities, such as when an enterprise which operates at a loss is nevertheless enormously productive. We are in a much better position if we look at the average cost structure of a product unit (the whole value chain) but that is often a closely guarded secret since business people do not want the true mark-ups to be public knowledge.

I doubt for many reasons that it is really possible to calculate a MELT using data on gross output and labour hours worked. Quite possibly, the nearest we get to it empirically is by calculating an average real wage per labour hour worked, i.e. the socially average value of a unit of labour-time.

I think one of the merits of Marx's theory is that he offers an explanation for why it happens to be the case that the very meaning of productivity perpetually remains a controversial topic. 


ope mailing list

This archive was generated by hypermail 2.1.5 : Mon Jun 30 2008 - 00:00:16 EDT