The Real Movement

In this post I show the surprising results of the empirical literature on hours of labor reduction: Contrary to the commonsense assumption, if hours of labor are reduced, the reduction will probably have no impact on wages at all.

Borsch-Supan states there are arguments for the idea reducing hours of labor will reduce the rate of unemployment, but these arguments rest on assumptions that go against the empirical evidence (which he calls “counterfactuals”). The question he asks us to ponder is whether, based on empirical evidence, there is a case for the idea employment will increase if hours of labor are reduced.

By contrast, there is no basis for such an argument within labor theory, because the expansion of employment in labor theory is determined by the rate of surplus value and, all else held equal, the quantity of surplus value produced is a function of the duration of labor. If the duration of labor is reduced, the mass of surplus value will also be reduced; if the mass of surplus value is reduced, the rate of expansion of the existing capital (new employment of living labor) must also fall.

At the same time, a reduction in the mass of surplus value produced means the mass of profits have fallen, since profits are nothing more than the mass of surplus value divided by the total capital invested. Since profit is the motive of capitalist production, the capitalist will again take steps to increase profits by further reducing wages. How this is done is not important at this point, but in labor theory there is no reason to assume fewer hours of labor increases employment — just the opposite: fewer hours of labor accelerates the reduction of living labor in production.

This presents the advocates of shorter hours of labor with a problem: the mode of production steadily reduces the expenditure of living labor per unit of output, leading to unemployment. Reducing hours of labor does not compensate for this process, but actually accelerates it. Thus, if the aim of a reduction of hours of labor is not to end wage labor itself, (i.e., accelerating capital’s demise), it makes little sense.

And this is the result Borsch-Supan argues is supported by the empirical evidence: fewer hours of labor cannot increase the employment of wage labor but only reduce it. However, he admits to a rather interesting additional result of reduced hours of labor: Based on Borsch-Supan’s review of the literature, reduction of hours of labor seems to free up more disposable time for the worker but has no impact on real wages.

“The reduction of working time has provided German workers with more leisure during, and a longer retirement after, working life. There are also no signs that German workers have suffered from income losses due to reduced work hours, given output. This is an important social achievement and has made life much more pleasant for the workers. However, there is little evidence that a reduction in working time has reduced unemployment, while there is some evidence that is has reduced output and thus macroeconomic growth. We have no reasons to believe that the underlying assumptions will change in a way that will make these conclusions less relevant in the future.”

The number one argument on the Left against reducing hours of labor is the fear a reduction might reduce real wages. Yet, it appears the empirical data from Germany does not support this argument at all. Germany has both fewer hours of labor and a more generous social safety net than the US, as well as competitive wages. Thus, at least in the case of Germany, the workers had more free time away from labor without a corresponding fall in their subsistence.

Borsch-Supan attributes the relative stability of wages to several causes:

1. a reduction in labor hours may leads to an increase in overtime: This overtime, however, is calculated on the basis of fewer hours: in the case of a reduction from 40 hours to 32 hours overtime now begins at 32 hours;
2. The capitalist have to deal with the cost of training and supervising new workers;
3. To contain rising labor costs, the capitalist have greater demand for improved machinery, which expands production in capital goods; and
4. These effects can be magnified by stipulating that any reduction in hours of labor be fully compensated in gross wages.

His reasons for the relative stability of wages aside, Borsch-Supan cites the works of another researcher, Hunt (1996), who found a proportional increase in wages for a reduction in hours of labor — the reduction in hours of labor is more or less completely offset by an increase in wages. This finding is fully consistent with labor theory, which assumes any reduction in hours of labor must first lead to a fall in profits, not wages.

Obviously if profits did not fall, there would be no incentive for the capitalist to invest in constant capital to recover his profits; however there is another, more persuasive argument, for this within labor theory. Since the price of every capitalistically produced commodity is equal to the value of the constant capital (c) plus the living labor that entered into its production (v+s), a fall in the surplus value (s) produced must result in a fall in the price of the commodity, (c+v+s).

According to Borsch-Supan, however, the behavior of prices in response to a reduction of hours of labor is the area of economic adjustment that is least understood by simpleton economists.

“Price changes provide an important general equilibrium mechanism that will reduce output in response to an hours reduction, in addition to any straight-forward reductions in output due to lower labor input measured in worker-hours. We do not have good econometric evidence on the output reaction to the hours reduction.”

Borsch-Supan therefore substitutes certain assumptions for actual empirical evidence to see how they hold up. The first assumption is that any increase in labor costs might be passed on to the consumer, but this assumption runs into the problem that when prices of commodities rise, the demand for them falls. Empirical evidence suggests that demand for a good is negatively correlated with its price.

The problem here can be clarified by pointing out that Borsch-Supan is putting the cart before the horse in his discussion of price and its response to a reduction in hours of labor. Once hours of labor are reduced there has already been a fall in demand for commodities. The reduction of hours of labor presupposes not only a fall in the duration of labor, but also in demand for all the commodities upon which this labor is exercised. In a given social labor day, so much constant capital is consumed in the labor process; with a reduction in the duration of the labor day there is a proportional reduction in the constant capital used up in the labor process.

Since the demand for the commodities produced has already fallen proportional to the reduction in the duration of labor, this fall must be expressed in a proportional fall in the prices of the commodities employed in production. Howard Nicholas argues this fall in prices is how the new condition of capitalistic reproduction (reduced labor time) is imposed on the labor process going forward.

Thus, employing labor theory, we do not have to speculate about the prices response to reduced hours of labor. Borsch-Supan has difficulty nailing down the likely price effect of a reduction in hours of labor, because he sees price movement as the beginning of the adjustment, rather than as a response to a material change in production that has already taken place: the reduction in the duration of labor.

In the capitalist mode of production, the extension of hours of labor itself is the source of the poverty of the workers. Reducing hours of labor does not add to this poverty, but is the precondition for abolishing poverty.