Tuesday, 7 June 2011

Value of labour-power & wage differentials

On 6 June, Dave Z put a comment at the end of my article on “What the ‘China price’ really means”.

This note addresses the two main points raised by Dave Z (in blue – if you want to see the full comment, refer to the original article) and answers them. His focus was on the “parts of your theory dealing with the lower wages of the industrializing countries.”

1) “Firstly, I think your argument about productivity is inadequate. There is surely a significant productivity gap between capitalist regions that explain a significant part of the wage gap.”

I agree that there is likely to be a large productivity gap between capitalist countries at very different levels of development. After all, that is an important part of what being a developed economy means – to have high(er) productivity. Where I do not agree is on whether that difference in productivity ‘explains’ most of the wage gap (in the article a gap of 10x, 20x or 30x was noted!), and on how big the relevant productivity gap is (see point 2 below).

A key point is that the level of wages depends on the reproduction costs of labour-power, or what capitalists need to pay the worker to get them to be able to show up for work (not just individually, but also to allow for family costs, etc). This, in turn, depends on subsistence costs as a minimum, plus what Marx called a ‘historical and moral element’. This latter element is based on the social conditions prevailing, including the success or otherwise of working class struggle for higher wages, benefits, etc.

There is not necessarily a direct relationship of wages to productivity. It is true that higher productivity can allow the capitalist to make some concessions on wages and benefits while still making a profit. Equally, low productivity means the capitalist will have to impose harsh conditions in order to survive in competition. However, there is no one-to-one relationship. It depends on the political and social situation. A defeat of the working class can lead to high levels of exploitation and high productivity but low wages. This was true for the west German ‘economic miracle’ in the 1950s, for example, where exploitation of the working class was comparable to that under Hitler.

In periods of crisis-free growth, it is likely that wages will rise, but commonly we find that nominal wages grow less than productivity. The degree to which that happens is not predetermined. Rising productivity is usually an indication of a rise in the rate of exploitation, despite what may be improved living standards (higher real wages) for workers. However, one message in my article on the ‘China price’ was that this mechanism does not work in the same way for workers in the dominating, imperialist countries and for those in a more subordinate position.

In the imperialist countries, the capitalist class may attack living standards, but it has far less freedom to do so than in the dominated countries. In the latter, it is also starting from a lower level of living standards from which to begin exploitation. In this case, the ‘historical and moral elements’ work in capital’s interests. Especially for countries that are newer entrants to the global economy, the more traditional social relationships can substitute for higher wages paid by the capitalist (eg growing some of your own food). Wages are likely to be very much lower than in the major countries, even if productivity in the factory is not that much lower than in the more developed economies.

2) “Secondly, it does not follow that differentials in rates of return on capital invested between regions A and B are the result of higher rates of exploitation in the latter. In fact, we have shown that the differentials are invariant to such distributional differences ….”

I use a lot of statistics in my analysis, but try to treat them with the relevant degree of scepticism. I have not seen your analysis, so I don’t know for sure, but I suspect that there may be several issues invalidating your results, or at least your interpretation of what I am arguing.

a) I agree, differences in rates of exploitation may not be the reason or the only reason for the different measured rates of profit. Tax concessions for foreign capital, or other concessionary deals to attract foreign capital can also be important. One important factor is buying up local productive capacity at knock-down prices (as happened after the 1997-98 Asian crisis in South Korea, for example, with the sale of parts of Daewoo). These issues were not raised in my article, which focused on wages.

b) Measuring productivity is another issue. The national average productivity level may be low, but my argument is that foreign companies invest in, or are supplied by, companies with levels of productivity that are not materially different from those in the major countries. This then highlights the massive gap between wage levels paid in China, India, etc, and the wage levels paid at home. When I say ‘not materially different’, I mean not 3, 5 or 10 times lower than in the major countries. For the same reason, national measures of investment as a share of profit or the growth rate of the total national workforce are not valid factors to explain the rate of profit measured by foreign capital’s activities.

The scarcity of good statistics means that sometimes we have to rely on a good journalist report (as I did for the Bangladeshi textiles example), or a study that may only give a snapshot of developments and which is also limited by its own assumptions (eg the BLS studies of China and India that I cited). In my professional experience, the worse the exploitation, the less you are likely to find consistent, detailed timeseries. No surprise there, really.

The BLS study of Chinese data implies (on my reading) that US corporations invest in the upper level urban companies, and pay the higher-level wages and benefits ($1.47 per hour versus the $0.53 for the TVEs). However, this ‘higher’ wage is a trivial hike in labour costs for a US corporation used to paying more like $30 per hour at home. I find it completely implausible to argue, as you seem to be doing, that the rate of profit on investment in China has little relationship to this fact, and instead is a mix of a range of other factors.

Tony Norfield, 7 June 2011

1 comment:

Dave Z said...

Regarding 2) I think your reply is valid. We have derived the *average* rates of profits for regions; it doesn't follow that individual capitals earn this rate of profit.

My point here was that the average level of profitability are likely to be higher in industrializing countries independently of whether the wage shares are high or low in comparison to the advanced economies.

Regarding 1), there is certainly no deterministic relation between productivity and real wages. As indicated above, I think one must take a statistical view of the operation of a capitalist economy and try to derive its stochastic laws.

The real wage of workers in industrializing countries will firstly depend on their bargaining power vis-a-vis capital, foreign or domestic. Given the balance of forces this power is obviously weak, e.g. with a large rural population flowing towards the cities. This would be the immediate limit to the rise of real wages.

But on top of this you have the limits set by the average productivity level of the economy, regardless of whether the workers are employed in advanced capital-intensive firms such as a Toyota factory in India: If a large fraction of their real wage bundle is domestically produced then the low productivity growth in the labour-intensive backward sectors of the Indian economy will limit the growth of the real wage.

Recall the mechanism proposed in Marx's theory of relative surplus value: productivity improvements work their way through the economy and reduce the labour-value of the real wage bundle.