David Spencer spreads more academic myths about why wage slavery continues to grow

by Jehu

A new academic monograph on the problem of labor time and its stubborn growth despite the tendency toward technological unemployment that is inherent in capital has been posted on Reddit. The paper is, Fear and hope in an age of mass automation, by David A. Spencer. Spencer argues that,

“[Work] will likely persist, despite and indeed because of the wider use of new technology. The threat to workers from technology is seen to come more from the erosion in the quality of work than from the loss of work. The paper argues that a better future for work and workers ultimately depends on broader changes in ownership.

Work is what Spencer calls wage slavery — apparently chosen for its neutral, even ambiguous meaning. Can’t call it wage slavery, now can we?

Spencer argues the Keynes got it wrong: the constant improvement on the productivity of social labor — Spencer cites studies that estimate labor today is 15 to 18 times more productive than in the mid-18th century — is leading not to the end of wage slavery as Keynes predicted, but to increasingly repulsive employment conditions and a longer working day. Spencer attributes this paradox to two factors: consumerism and declining worker bargaining power.


Although Keynes, in 1930, predicted a fifteen hours work week by 2030, Spencer argues he neglected these two influences on working time:

Keynes’s mistake was to ignore both the effects of consumerism in maintaining a demand for longer work hours and the weak bargaining power of workers in preventing shorter work hours (Friedman, 2017). In the former case, he assumed falsely that consumer wants would have a natural upper limit—he failed to anticipate, in this case, how consumer wants would multiple as capitalism developed ever more sophisticated ways of stimulating consumption (Skidelsky and Skidelsky, 2012). In the latter case, he assumed that capitalist employers would pass on the proceeds of productivity growth to workers in the form of higher wages and shorter work hours—he failed to foresee how capitalist employers would use their superior power to appropriate the proceeds for themselves and deny workers both higher wages and shorter work time.

While rising productivity has created the potential for a shorter working day, it has also created pressure for longer hours of labor. Automation under capitalism has eliminated wage work in industry and agriculture, while expanding it in the low wage services sector. Despite this clearly evident trend, a number of writers continue to promote technology as the path to the end to wage slavery.

Spencer argues this faith in technology is misplaced: technology is privately owned in a capitalist economy. This echoes Marx, who argued, “No capitalist ever voluntarily introduces a new method of production, no matter how much more productive it may be, and how much it may increase the rate of surplus-value, so long as it reduces the rate of profit.” If new technologies force out older technologies at an ever increasing pace, it is because the newer technologies increase profits, not free time away from wage labor. Free time, when it at last arrives, appears for the worker only in the form of unemployment and absolute destitution.

The problem with this line of argument is that free time from wage slavery hasn’t been realized even in the form of unemployment. Not only does wage slavery continue to persist, the population of actively working proletarians since 1939 has grown more than four-fold, as shown in the chart below:

Total Nonfarm Employees, 1939-2012 (Source: Bureau of Labor Statistics)

Total nonfarm payroll in 2012 was 450% of what it was in 1939. Keynes predicted society would increasingly be faced by technological unemployment, which would compel it to reduce hours of labor as productivity increased. Instead, as productivity has increased, so has the aggregate length of the working day. (The aggregate labor day is defined here as the total number of employees times the average hours of labor per employee.)

Spencer wants to explain this staggering development by arguing it can be accounted for by rampant consumerism and the weak bargaining position of the working class, but I think he has several problems with his argument.


The first problem with Spencers argument is that the above chart shows employment growth rising since 1939. This covers a period where worker bargaining power is said to be strong (what the Endnotes collective calls the golden age of the capital-labor compact) and a period where worker bargaining power is said to be weaker (what Spencer calls neoliberalism). The growth of employment doesn’t appear to be affected by any change in the bargaining power of workers nor the presence (or lack thereof) of significant unionization.

Spencer needs to explain how worker bargaining power is a factor that explains employment growth when the data appears to show it has no effect on employment growth.

The second problem with Spencers argument is that employment growth is not continuous. There appear to be periodic troughs in employment growth, as shown in the chart below:

Periods of flat or no growth in Total Nonfarm Employees, 1939-2012 (Source: Bureau of Labor Statistics)


Did rampant consumerism suddenly collapse during these periods? If so, what explains the subsequent explosion of consumerism after each trough? Most of all, what are we to make of periods from 2001 to 2012, when employment growth was (basically) flat for more than a decade?

Third, I want to ask about this rather bizarre argument:

In the period since the 1970s, with the retrenchment of the welfare state and the declining power of unions, workers have faced stagnant or falling wages that has meant they have had to go into debt or secure credit to maintain or increase consumption. Here, the achievement of higher consumption has been at the expense of longer work hours. Generally, the existence and persistence of strong consumption norms, underpinned by mass advertising and marketing, has inhibited the move to shorter work hours (Cowling, 2006).

Forgive me for being obtuse, but Spencer seems to be saying that a weaker bargaining position for the working class led to falling wages. Okay, I can see where this might happen. But then Spencer goes on to state that in the face of falling wages, the working class went into debt to maintain or increase their consumption. I am not a financial adviser or anything, but who lends money to people who have little or no means to repay their debt? Isn’t it a standing joke in the United States that banks lend money to people who don’t need it?

Even if we accept the proposition that banks were lending money to people who clearly could never repay, how do we explain this bizarre practice? Is this a normal characteristic of the capitalist mode of production?


Pardon me for being a complete asshole toward a guy who is just trying to argue for the state to take control of production, (which is Spencer’s real object in this paper), but this paper is trash. In his defense, Spencer didn’t invent these arguments; they are common enough in academia. All he has done is gather them in one place in order to spread them, like a sort of ideological virus, to his unwary students.

David, get a real job; you suck at critiquing political-economy.