Husson’s and Treillet’s call for labor hours reduction: Important but seriously flawed
I just finished reading this article by Michel Husson & Stephanie Treillet on the significance of labor hours reduction, Liberation Through Vacation. I want to offer some thought on why I think it is, on the whole, as important as it is disappointing. I make these points, not because I disagree with what I think was the intended thrust of their article, but because certain folks will go after Husson’s and Treillet’s argument. For instance, A. Kliman has already taken David Graeber and others to task for their weak arguments on labor as just another attempt to rebrand social democracy. (See Kliman’s, Post-Work: Zombie Social Democracy with a Human Face?) My point here is to expose weaknesses in their argument because Husson and Treillet’s main thrust is, after SYRIZA’s election, the most important development to emerge from the crisis in 2015.
The essay begins with an introduction for which I see no use, since it reduces labor time reduction to a measure to promote full employment. The meat is provided by Husson and Treillet, two Left economists. I am not familiar with Treillet but I have seen Husson’s writings before in relation to the law of the tendency of the rate of profit to fall.
1. The critical place of labor hours in capitalist exploitation
The two economists, in contrast to the useless introduction, clearly state that “working-time reduction is central to the history of capitalist exploitation of labor and worker resistance.” Labor hours reduction however has come to a halt in France, the Left has had difficulty making it a mass issue. The economists appear to blame two factors for this: First, the Left appears to believe that labor hours reduction requires a large shift in what they call “the balance of power” — I assume this means between the two classes, although this is never clearly stated. Rising unemployment and layoffs perversely has thrown labor on the defensive. However, the economists also point to another surprising cause:
“we face an ideological assault, also within part of the Left, proclaiming that we must choose between employment and wages and that the current distribution of surplus value is not negotiable.”
To defeat this attitude on the Left, economists suggest we need to get clear theoretically (I hate the term ‘ideologically’) on why labor hours reduction is so critical at this time. The economists then state their first thesis:
“So it is essential to find ways to overcome this situation and resume fighting, including ideologically, for working-time reduction. Contrary to widespread ideas, true full employment is possible. But it requires confrontation with employers.”
Why the economists begin with this argument and not with their initial point is beyond me. Their initial point was that working-time reduction is central to the history of capitalist exploitation of labor. Rather than show why this is true, the first thesis reduces labor hours reduction to just another full employment measure. The thesis has the defect that it places reduction of hours of labor on the same level as a job guarantee or basic income. No matter what the case for a jobs guarantee or basic income, no one has ever argued that either is central to capitalist exploitation. This is like putting a Trident submarine in the same category with an IED as a weapon of mass destruction. Yes, reducing hours of labor can certainly forcibly reduce unemployment temporarily, but it is hardly in the same category as basic income.
Freedom from wage labor is itself human emancipation.
2. Labor hours reduction and the debate over secular stagnation
By far, the weakest part of the argument made by Husson and Treillet is right out of the Marxist playbook: It is inexplicable why Husson and Treillet focus on the impact labor hours reduction has on employment and make no mention of the role labor hours reduction has on the development of the productive forces. Central to Marx’s argument in Capital is the fact that profit is the goad of capitalist production. To extract the largest possible mass of surplus value from the working class in the normal working day, the capitalist constantly improve all of the instruments of production — organization, machinery, science, technology — with an eye to reducing the socially necessary labor time required for production of commodities.
Reduction of hours of labor accelerates this process by setting strict limits on the aggregate hours of labor time available to the capitalist in a single working day. In chapter 16 of Capital, Marx notes that limits set on hours of labor in England in the 1840’s were largely responsible for a greater than 5-fold increase in industrial productivity growth between 1850 and 1856. He writes:
“One fact is sufficient to show how greatly the wealth of the manufacturers increased along with the more intense exploitation of labour-power. From 1838 to 1850, the average proportional increase in English cotton and other factories was 32%, while from 1850 to 1856 it amounted to 86%.”
Marx credits reduction of the working day to ten hours for this explosive increase in the rate of productivity in those years and explains why this is so:
“There cannot be the slightest doubt that the tendency that urges capital, so soon as a prolongation of the hours of labour is once for all forbidden, to compensate itself, by a systematic heightening of the intensity of labour, and to convert every improvement in machinery into a more perfect means of exhausting the workman, must soon lead to a state of things in which a reduction of the hours of labour will again be inevitable.”
Under the capitalist mode of production, the drive to increase profits is so overwhelming that if hours of labor are reduced, capital responds by introducing even more advanced methods to extract surplus value within the new limits of the working day. Two points can be made about this: First, since the new society requires a very high development of the productive forces, it becomes clear that the proletariat can compel an accelerated pace of capitalist development, not by limiting its wage demands, but simply by limiting the aggregate hours it supplies to the capitalists. The general reduction of hours of labor enforced by law, therefore, prefigures communist society itself.
Second, the impact reduction of hours of labor does not just concern future society, but also addresses the present crisis and alters the discussion of so-called secular stagnation in a way that is favorable to the proletariat. What is secular stagnation except the slowing rate of net new capitalist investment? If Marx is correct in his argument, net new investment can be spurred not by higher fascist state deficits, which produce nothing, but simply by reducing hours of labor. A reduction of hours of labor would compel capitals to invest the vast sums of idle capital, amounting to an estimated $5 trillion in 2011, that they are sitting on or simply investing in unproductive state bonds.
Labor hours reduction is likely the only measure that can compel these private firms to employ their excess capital productively by introducing more advanced means for wringing surplus value out of the working class.
3. Does increased productivity produce unemployment?
According to Husson and Treillet:
“Nothing could be further from the truth than claiming that greatly increased productivity is the cause of unemployment.”
In fact, this is one of the most important arguments of labor theory of value: the rising organic composition of capital produces excess capital and a surplus population of workers. If rising productivity does not lead to unemployment, Marx’s entire argument on the impact of the falling rate of profit is disproven.This is why, as the economists assert, reducing hours of labor requires a confrontation with the employers. The employers introduce improved machinery as means to reduce the portion of their outlays of variable capital. Which is to say, they seek to introduce improvements in productivity to render some portion of the existing labor force redundant.
As evidence for their odd argument, the economists compare employment immediately post-war to employment today. The period immediately post-war saw a very high rate of improvement in productivity and very low unemployment; while, at present, we have relatively low rate of improved productivity with very high unemployment.
In the first place, I am astonished they would make this comparison, as if the war itself did not leave Europe in ruins. Considering the damage done to the productive forces during the war, it is no surprise we find both high demand for labor power and a sharp rise in productivity of this labor power. Whatever other defect of neoliberalism has had on advanced societies since the 1980s, it has in no way wrought the damage produced by fascism, which produced two global wars of redivision resulting in 100 million dead. Based on this bizarre comparison they conclude: “when productivity gains slow down, unemployment explodes.”
In fact the conclusion drawn by neoliberals like Larry Summers is that, if you really want to have low unemployment and high rate of increase in productivity, leveling Europe helped end the Great Depression:
Alvin Hansen proclaimed the risk of secular stagnation at the end of the 1930s only to see the economy boom during and after World War II. It is certainly possible that either some major exogenous event will occur that raises spending or lowers saving in a way that raises the FERIR in the industrial world and renders the concerns I have expressed irrelevant. Short of war, it is not obvious what such events might be.
4. Productivity and output
The economists know that their argument makes no sense, so they try to explain it away.
“This paradox [high unemployment and low productivity] disappears if we remember that employment depends not just on the overall level of production and the productivity of labor, but also on working time.”
According to the economists, labor increases more or less at the same rate as output. More or less? Why not exactly? This is because, without admitting it, the writers are using two incompatible definitions of productivity. There is the technical improvement in productivity, which is measured in some unit of physical use values. This could be stated, for instance, in so many cars per hour; an increase in the number of cars equals an increase in technical productivity. But there is another sort of productivity at work here: the productiveness of labor as producer of surplus value. In the capitalist mode of production — and in most government statistics with which I am familiar — labor productivity refers to values. It does not refer to the production of use value themselves, but to the exchange values of these use values.
To give an example that might show the difficulty here: An increase in the raw number of cars produced per hours does not necessarily means an increase in the mass of surplus value. Moreover, at a certain point these two — technical and value productivity — are not only NOT correlated but inverse to each other. An increase in the number of cars produced in a given period of time must, lead to a fall in profits, not an increase. By one measure, productivity has increased; while, by the other, productivity has fallen. The result is that there is no necessary relation between the increase in labor productivity and output. As Husson knows or should know, there is no relationship whatsoever between any use value and its exchange value. You cannot simply double the quantity of use values and expect the sum of their exchange values to double with this. If it did, there would be no reason for labor theory.
5. The employment effects of labor hours reduction
In Marx’s Capital, reduction of hours of labor results in the increase in the technical productivity of labor; however, it simultaneously results in the fall of the rate of profit. To raise the rate of profit again, the capitalists have to increase the employment of labor power, which adds to employment. But the increase in employment is a short-term effect; while the long-term effect is introduction of improved machinery, etc. Which means any reduction of hours of labor has to reduce employment in the long-run.
Thus a reduction of hours of labor accelerates the introduction of improved labor methods which then requires another reduction of hours. The point of hours of labor reduction is not to create full employment, but to abolish wage labor. No reduction of hours of labor — like France’s 35 hours work week — prevents rising unemployment in the long-run, because, on the one hand, the productivity of labor continues to improve, while, on the other hand, the reduction of hours of labor itself accelerates the improvement in the productivity of labor.
According to the two economists, labor productivity increases at about the same rate as output, net job creation depends on reduction of hours of labor. To support this claim, they look at what is allegedly the total improvement in productivity during the 20th Century. According to the writers, productivity improved nearly 14-fold; GDP rose 10-fold, hours fell by 44%. At the same time, total employment rose by 26%, while total hours of labor fell by 30%. Frankly, I have the slightest idea what these statistics mean; perhaps I am missing something here. In any case, the two economists assert that if hours of labor had not been reduced, unemployment would have reached staggering levels.
The argument is unsupported by the empirical data or by theory, and the economists never show why there is anything more than a correlation. Employment has increased steadily throughout the whole of the 19th and 20th centuries and likely shows no necessary relation to labor hours. It is true that, at any given level of productivity of labor, a given quantity of physical output will require a given mass of employed workers working for a given duration. And it is also true that a reduction in the duration of labor under these conditions will require an increase in the number of workers employed.
However, it is by no means true that an increase in the number of workers employed requires a reduction in the duration of hours of labor. An increase in the number of workers employed could have just as easily come from an increase exports or state spending. There are four variables here that each must be accounted for: productivity, output, labor force size, duration of labor. We can, theoretically, hold two variables constant and ask ourselves about the relation between the other two, but actually pulling this data from fascist state statistics is difficult. This is further complicated by the fact that, in the capitalist mode of production, employment and output are not determined by the raw technical productivity of labor but by its capacity for producing surplus value.
Moreover, in the capitalist mode of production, the capacity of labor to produce surplus value is itself determined by the duration of labor. The longer the working day, the more surplus value that can be produced and, consequently, the more labor power capital will employ. But this is true only up to a point, then the thing suddenly and without warning inverts and millions of workers are thrown into the streets. In the crisis, productivity, output, employment and hours all fall together for some mysterious reason. It is as if the development of the forces of production are suddenly thrown back a decade or even a generation.
Husson and Treillet appears to be on firmer ground when they discuss France’s experiment with the 35 hours work week. Between 1975 and 1997, private sector employment shows a moderate rise before sharply increasing between 1997 and 2001. It then levels off with the 2001 recession and resumes its upward rise into the present crisis. Unfortunately, even here the argument is weak, since in the United States, where there was no reduction in hours of labor, unemployment also decreased to very low levels. In fact, the US labor force participation rate grew until just about where it began to level off in France also and then began to plummet.
It would be helpful if Husson and Treillet could isolate how much of the private sector employment increase could be explained specifically by reduction of hours in France.
6. Surprise! Unemployment also reduces of hours of labor
The question, say the two economists, is not whether hours of labor will be reduced but how this reduction will be accomplished. A reduction of aggregate hours can either be accomplished all at once for the entire working class; or it can be accomplished by unemployment or partial employment of some definite portion of the class. In the latter case, some have jobs, some are only partially employed, while a third group is left entirely without jobs. A general reduction of hours of labor has far different effects on the distribution of income than the latter case.
Unfortunately, the writers do not follow up on this very important point. In first place, we know from labor theory that the unemployed serve to depress the wages of the employed. Since, in any case, labor hours must fall as the writers insist, a reduction that takes the form of unemployment depresses even the wages of those who remain employed. Moreover, it is not simply that the unemployed depress the wages of the employed, according to labor theory the employed depress their own wages as well.
Under the threat of redundancy, and by addition of improved methods of production, the labor of the employed is intensified, and makes possible a further reduction of employment. The threat to the wages of the employed is not in first place competition with the unemployed, but the labor of the employed themselves, who render their jobs redundant by their own labor. It cannot be forgotten that those first unemployed were displaced from employment by their own labor and by introduction of machines, etc. The process whereby the unemployed depress the wages of the employed begins with the employed themselves, not the unemployed. Even if we began with zero unemployment, a definite portion of the working class would be rendered redundant by their own labor.
This problem cannot be fixed by a basic income or a jobs guarantee, since it arises from wage labor itself. Unemployment is not necessary to depress wages, since the reduction of the value of labor power is the flip-side of production for profit. The continuous reduction of the socially necessary labor time required for production of commodities is simply the continuous reduction of the value of labor power itself. Wages do not fall because of competition between employed and unemployed, but because it take less labor to produce the value of labor power. In labor theory, competition does not lead to a fall in wages, as the bourgeois simpletons think, but itself arises because the value of wages has already fallen.
Having reduced the socially necessary labor time required for production of labor power, the capitalists reduce the labor force accordingly. Employment is not an end in itself in the capitalist mode of production, but is solely for production of surplus value. No capitalist will ever lay out his capital on more labor power than is necessary for production of surplus value. If the intensification of labor and introduction of machinery allows for less labor power in production, lower outlays of variable capital, the capitalists will reduce their work force. In other words, as the economists explain, with labor hours reduction or without, the aggregate socially necessary labor time required to produce the total mass of commodities in society will fall.
7. Not reducing hours of labor finances state deficit spending, precarity, financialization and crises
Moreover, at present the fall in the socially necessary labor time required for production of commodities outpaces any possible new uses for labor. (See Keynes, 1930.) This has given rise not just to unemployment, part-time employment, and increasing precarity, it has also led to excess profits, the monstrous public sector deficit spending, financialization, speculative bubbles and crises. While the economists note the relation between hours of labor, precarious employment and financialization, they completely ignore the deficit spending of the public sector and the growth of the public sector generally. In fact, we know that not reducing working hours not only produces precarity, unemployment, financialization and crisis, but also requires ever rising state sector deficits.
How do we know this?
In labor theory, the total volume of surplus value is a function of the duration of the actual working day. The volume of surplus value increases, on the one hand, with the increase in the total duration of labor, understood as the total mass of employed labor power times the average hours of labor of these workers. As the population of employed workers increase, assuming labor hours remain constant, the total mass of surplus value should increase. On the other hand, as the productivity of labor increases, the total volume of surplus labor should increase as well. As the exploitation of labor rises both extensively and intensively the mass of surplus value seeking a place in production increase as well. This increase, according to Minsky (1992), must lead to ever increasing deficit spending by the fascist state.
Thus, it becomes clearer why, according to the two economists, the state has been resistant to the 35 hours work week in France:
“the Aubry Laws have created permanent jobs, but almost nobody, at least in the Socialist Party, defends this record. In addition, the rules adopted at the time have worsened the living conditions of broad segments of the working class.”
The socialist party has been quietly engaged in reducing the living standards of the working class to boost the production of surplus value. It is not just the capitalist class that benefits by the reduction of working class subsistence, but the state sector itself. The reduction of the subsistence of the working class not only places an ever larger mass of surplus value at the disposal of the capitalists, the state indirectly benefits when the excess profits, which cannot be employed productively must be lent to it.