When a laborer goes to work for an eight hour day in our manufacturing sector, it takes the sale of only a portion of what he produces to cover his total compensation for the day. The rest of what he produces belongs to the owner of the business and is called surplus value. From this share the owner must pay his business expenses (less labor costs) and gain his profit.
The market value of what a worker produces in a given amount of time that belongs to the owner divided by what he is paid for that time, is the rate of surplus value. For example, if our laborer is paid a dollar a day, and in that time he makes two widgets worth a dollar each, then the rate of surplus value is (2-1)/1 = 100%. If our owner hires ten workers at a dollar each for a day who together make sixty of the widgets, then the rate is (60-10)/10 = 500%. Since neither the amount that labor is paid nor the market value of what is produced is set in stone, the rate of surplus value varies over time.
Deriving an accurate measure of the rate of surplus value in our economy is no mean feat. It is not something discussed in polite company, and when it is addressed, it usually is couched in terms of the 'increasing productivity of labor'. However, information about trends in the rate for the US manufacturing sector can be gleaned from data collected and aggregated by the US Census Bureau and published in the Annual Survey of Manufacturers. (The data sets for what follows can be found here, here and here.)
The following graph is based on the relationship of production worker wages paid to value added in the production process and shows a rough estimate of how the rate of surplus value has changed in US manufacturing since 1949:
By this estimate, the rate of surplus value has increased several hundred percent over that time. Its corollary, that production workers' share of production has decreased, is shown in the following graph covering the same time period:
Here the observant reader will object that there is more to production than wage labor and that the rate as represented by the graphs must be inflated. In particular, what of the value added by salaried employees?
The following two graphs are based on the relation of annual payroll to value added:
Just as the earlier graphs overestimate the rate of surplus value, these two underestimate it. They include problematic amounts such as, for example, salaries paid to corporate officers, no matter how inflated or divorced from actual production they might be. Reality must lie somewhere between the two estimates.
There are additional weaknesses with this analysis, perhaps most notable its failure to account for legacy expenses and taxes. As for taxes, and keeping in mind the recent use of tax dollars to bail out and loan to businesses, we must ask ourselves not only who pays them, but also, to whom and in what percentages do their benefits actually flow — owners or workers? And given owners' ability and penchant to renege on legacy commitments when times get tough, as evidenced for example, by Kodak's history, it is not clear how seriously these expenses should be treated. Further complicating the picture is the question of how the money that is supposed to cover them is raised.
In any case, the trend is clear. Overall, owners have taken a steadily increasing share of the value of production; overall, workers have been unable even to hold their own, let alone increase their share over the years.
So does it make sense to blame unions for the sorry state of our now near bankrupt manufacturing industries?