October 19, 2026
A recent Reddit post asks: “So, AI takes over, everyone has lost their job and only 10 trillionaires own everything. Now what?” The author goes on: “Meta, Open AI, and Amazon own everything, cool beans! No one can work; therefore, no one has money to buy any of the horse shit temu slop they prime on Amazon now. Won’t everything just implode from there?” The post gives voice to the anxiety and despair that has greeted recent advances in artificial intelligence. Many foresee a future in which America becomes a hellscape of unemployment and penury, in which data centers and ghost towns prevail outside of the fortified coastal retreats where tech oligarchs live.
Does it have to be that way? Sixty years ago, a discussion of computer automation took place that was the opposite in tone and tenor. In 1964, Michael Harrington, Robert Theobald, Robert Heilbroner, and a host of other leading progressive figures of the era issued a memorandum entitled “The Triple Revolution,” in which they predicted that the “cybernetic revolution” would result in “potentially unlimited output … that will require little cooperation from human beings.” Such utopian speculations were common at the time. Marshall McLuhan, in Understanding Media, predicted automation would encourage “imaginative participation in society.”
Needless to say, these dreams of post-industrial abundance didn’t come to pass. But a less well-remembered document of the same period, while affected by the same blithe optimism, laid out a framework for understanding technological change that could help us understand the challenges we face today. In 1969, a small circulation journal, Radical America, published a long essay by the historian Martin J. Sklar entitled, “On the Proletarian Revolution and the End of Political Economy.” Basing his analysis on a neo-Marxist framework that may be relevant today, particularly with the emergence of artificial intelligence, Sklar offered a theory of what he called the “disaccumulation of capital” that provides a window through which to view the liberatory promise, as well as the looming peril, of automation. It is a difficult read—Sklar said it was written for “militant readers”—but it helps explain what is happening to the American economy today.
Sklar took his cue from Karl Marx’s theory of capital accumulation. For Marx, the central engine of growth is the accumulation of capital, which he regards as equivalent to the accumulation of labor time. According to the labor theory of value—which was embraced by Adam Smith and David Ricardo as well as Marx—the exchange value of consumer goods and of capital goods (plant, equipment, raw materials) is measured in the labor time it takes to produce them. There have been numerous byzantine attempts to create a formula that would convert exchange values into prices, but the usefulness of Marx’s theory doesn’t rest on that. The labor theory is a way of describing the labor process and division of labor that takes place in capitalist production.
The accumulation of capital, achieved when the profit from sales is invested, consists of the accumulation of labor time. Thus, during the period of rapid industrialization from the Civil War to the 1920s, there was not just an accumulation of goods and factories, but also of work-hours and of workers. From 1900 to 1920, for instance, the number of manufacturing workers doubled from five to ten million. As Marx stated in Capital, “Accumulation of capital is, therefore, increase of the proletariat.” America looked at the time as if it were going to become a giant factory with a labor force divided between a sprawling working class and small capitalist class. This vision of the future informed the socialism of Eugene Debs as well as the progressivism of Herbert Croly and Thorstein Veblen. This accumulation of capital entailed a Protestant ethic of deferred consumption. The business class cited the deferral of consumption for investment and saving as the justification for its existence and also as the justification for reducing wages or resisting demands to increase wages or to limit the work week. Work, thrift, saving, and efficiency were the moral and economic guideposts to capital accumulation.
But according to Sklar—and he had copious data to prove it—during the 1920s, the output of goods soared but the hours spent producing them, and the number of workers engaged in producing them, stopped increasing. What he called the “disaccumulation of capital” began. He didn’t think disaccumulation would immediately replace accumulation, but that it was becoming a persistent tendency within capitalism. It would replace accumulation only under socialism.
In the 1920s, electricity spread to almost all workplaces, along with new technology and forms of work organization such as the assembly line. The manufacturing workforce slightly shrunk during the decade, and farm employment also continued falling. (According to the Census, the number of manufacturing workers fell from 8,997,921 in 1919 to 8,838,743 in 1929, while the number of farm workers continued falling from 11.3 million to 10.5 million.) The amount of physical capital grew, but capital, as crystallized labor time, disaccumulated. Some economists at the time, including John Maynard Keynes and George Soule, noticed this, and remarked upon its significance. However, this development and what it meant were soon buried under the rubble of the Great Depression, until Sklar resurrected it in his 1969 essay.
Sklar contended that disaccumulation—and the new forces of production behind it—created the potential for an increase in economic and personal freedom. Labor would no longer have to be devoted primarily to the perpetuation, or reproduction, of the working class itself. Disaccumulation, Sklar wrote, “means that less and less labor power is required for the production of the goods necessary for sustaining and reproducing physical and social life. The people are increasingly freed to apply their labor, or life-time, to other pursuits and fields of endeavor.” Disaccumulation also upended the moral logic of capitalism. “The hitherto necessary contradiction, in the absolute sense, of the necessity of deferring immediate consumption as the condition of expanded production capacity, falls away,” Sklar wrote.
There were two potential consequences of this. First, the capitalist class itself—as the enforcer of deferred consumption and the reinvestment of savings—would lose the principal justification for its existence. (Keynes famously predicted the “euthanasia of the rentier, and consequently, the euthanasia of the cumulative oppressive power of the capitalist to exploit the scarcity-value of capital.”) Secondly, the elimination of the need to defer consumption would begin to undercut the basis for the Protestant work ethic and its celebration of self-sacrifice.
Some thinkers of the ’60s foresaw that the onset of automation and the disappearance of drudgery would enlarge the role of play and leisure in people’s lives. In Life against Death, Norman O. Brown called for the embrace of “polymorphous perversity.” But Sklar was more focused on the redefinition of work and on an end to what Marx had called “alienated labor.” Sklar’s views were similar to those he cited of the Young Intellectuals of the 1920s, who “insisted upon the right of all persons to engage in directly self-expressing and self-fulfilling work, and upon the principle that work, properly, constituted not a means to life, but one of the more important expressions of life.”
Sklar foresaw the disaccumulation of capital as a precondition for a full-blown socialism—what the young Marx called communism. In The German Ideology, Marx described a communist society as one in which “nobody has one exclusive sphere of activity but each can become accomplished in any branch he wishes.” In this way, he declared, it becomes “possible for me to do one thing today and another tomorrow, to hunt in the morning, fish in the afternoon, rear cattle in the evening, criticize after dinner … without ever becoming hunter, fisherman, herdsman or critic.” Along more abstract lines, Sklar wrote that “the objective of socialism is a society where people increasingly express their freedom … in dealing with necessity, and where they open greater and greater areas for discretionary, voluntary life-activity.”
Sklar contended that if capitalism were replaced by socialism in the new disaccumulationist economy, the age-old relationship of power between an owner-manager class and a working class or peasantry, which was justified as a means of extracting a surplus, would no longer hold. The new economy, which was being created by the transition from the accumulation to the disaccumulation of capital, would make possible Marx’s vision of a fully democratic society.
“Sklar foresaw the disaccumulation of capital as a precondition for a full-blown socialism.”
“Recreation and simple exercise have become a big business.”
Sklar wrote his essay at a time of political ferment when many people in “the movement” believed an imminent revolution would shake the foundations of capitalism. In his 1969 essay, he wrote of a “swelling movement for revolutionary socialism.”Of course, this revolution was not to be. Sklar recognized, however, that within the social relations of capitalism, the liberatory potential for disaccumulation was being thwarted. That was his key insight, and it remains valid.
The tendency toward disaccumulation helped to bring about the Great Depression of the 1930s. It created a surplus of capital, which was no longer needed for reinvestment in goods production, and which flowed into speculative bubbles that finally burst. Disaccumulation also led to technological unemployment, which increased the ranks of the long-term unemployed. Capitalism finally survived the Great Depression through military spending outside the civilian market during World War II. After World War II, government continued to supplement private production and consumer demand. But more importantly, businesses extended capitalist relations of production “beyond the sphere of goods-production, finance, distribution, and exchange, to all other social spheres.”
During this phase, capitalism—no longer confined to goods production for the perpetuation of the working class and for capitalist luxury—spread to niches that had formerly been outside the market—from fast-food restaurants (which replaced the family dinner at home) to health clubs, theme parks, and retirement villages. Recreation and simple exercise have become a big business. In Marx’s terms, what had been production for use has become production for exchange. As a result, service employment now dominates the economy. In 1943, for instance, manufacturing accounted for 38.9 percent of employment; by the end of 2024, it comprised about 8 percent. About 80 percent of working Americans produce services.
New industries, like aircraft, aerospace, and computer technology, have also arisen, and expanded initially through the accumulation of capital. But Sklar’s theory assumed that they, too, would eventually reach a level of productivity where production would expand, but labor hours would not. Here is what the rise in computer and electronic parts manufacturing looks like over the last forty years:
Here is the rise and then the decline in the number of workers making computer and electric products:
Capitalism changed after the 1920s in ways that reflected the transition toward disaccumulation. From the end of the Civil War through the 1920s, American economic growth was driven by the expansion of the capital goods and transportation sectors. After the 1920s, it depended more on “effective demand” from consumer goods and service production. The failure of wages to rise with productivity over the last four decades has been compensated by government transfer payments and by consumer finance, which has periodically precipitated widespread bankruptcies and financial crises.
The moral logic of capitalism has also changed. During the heyday of industrialization, Americans were still in the strong grip of the Protestant ethic, but as the fulcrum of the economy shifted from producer goods—and the deferral of consumption—to the encouragement of consumer spending, a new consumer ethos emerged, reinforced by the advertising industry. Americans still measured their worth in having a job, but they also measured it in their possessions, lifestyle, and appearance. The great companies of the American industrial revolution were capital goods producers like International Harvester and US Steel. The great company of this century is Amazon.
Sklar argued that the attempt to thwart the promise of disaccumulation would lead to the “malfunctioning, underutilization, and perverted utilization of production capacity” and to “the parasitic engrossment of social wealth.” Disaccumulation would, if allowed to run its course, cause the cost of consumer and capital goods to plummet. Instead, firms hold up prices and underutilize capacity, as has long been the case with big agriculture. Subsidized and highly productive industries, such as pharmaceuticals, gouge the consumers who, as taxpayers, funded them. Financial firms luxuriate in exotic speculation that has no productive purpose; trillion-dollar corporations spend billions buying back their stock; private equity firms buy up small independent businesses, like veterinary clinics, for whose customers they have scant regard. These are the results of a capitalism that has long since left behind its original role of producing the goods and services that allow its workers and owners to live and prosper.
There have been four periods in which prominent thinkers have questioned whether the development of new technology has severed the relationship between economic growth and employment. The first was in the late 1920s and early 1930, typified by Keynes’s “Economic Possibilities for Our Grandchildren.” The second was in the 1960s and driven by the introduction of computer automation in factories. The third was in the first decades of this century and was prompted by the emergence of computer technology and the internet. Unlike the first two periods, thinkers in the most recent period were concerned with the disappearance of lower-level white-collar as well as blue-collar jobs.
In 2013, two Oxford University researchers, Carl Benedikt Frey and Michael A. Osborne, published a highly regarded study in which they concluded that skilled labor and professional work would largely escape the scourge of computerization. “Rather than reducing the demand for middle-income occupations,” they wrote, “computerization will mainly substitute for low-skill and low-wage jobs in the near future.” They went on to say that “high-skill and high-wage occupations are the least susceptible to computer capital.” According to their model, bank tellers and legal secretaries would disappear, but computer systems analysts and multimedia artists and animators would endure.
We have now entered the fourth period, which has been driven by the introduction of Large Language Models. There is an ongoing and currently unresolvable debate about exactly what generative AI’s economic consequences will be, but there are important figures within the tech industry who have warned that while boosting productivity and production, AI would wreak havoc on employment. AI, they suggest, will eliminate many of the jobs that Frey and Osborne thought were insulated from computerization, creating a “permanent underclass.”
Dario Amodei, the CEO of Anthropic, prophesied in 2025 that before the decade’s end, AI would endanger 50 percent of entry-level white-collar jobs. OpenAI co-founder Andrej Karpathy scored exposure to AI for 342 jobs on a scale of zero to ten, with ten being the most endangered. He found that jobs paying more than $100,000 a year were almost twice as vulnerable as jobs paying less than $35,000 a year. Office and administrative workers averaged an eight, software developers a nine. The least vulnerable, he concluded, were janitors, roofers, and home health care aides. In other words, by this account, AI endangers those jobs that had been bypassed by previous waves of computer-driven automation. It would eliminate still more of the service sector jobs by which capitalism had fought off the threat of disaccumulation.
There are three consequences of AI unleashing capitalism’s tendency toward disaccumulation. The first is economic. In 2025, Amodei predicted that AI could bring about an unemployment rate of 10 to 20 percent in one to five years. That prediction already seems much too extreme. Like the growth of the aerospace industry or computer technology, the development of AI has initially created jobs, particularly in constructing data centers. In the September Bureau of Labor Statistics report, the largest job increases were in construction and manufacturing and in fields unaffected by AI, such as ambulatory health services. But the construction boom will eventually abate, and the loss of jobs, measured already in entry-level white-collar jobs, will take hold. That could raise unemployment rates and lower demand, threatening a recession.
Even if, as some critics of the doomsayers argue, new jobs arise that fully replace those lost, these jobs are likely to be lower-wage service sector jobs like those in healthcare that rely on personal interaction. Capital income will also continue to grow at the expense of labor income. In the second quarter of this year, labor’s share of national income had fallen to 52.8 percent, the lowest since 1947. That, too, reduces consumer demand. China, which has an even lower ratio of labor to capital income, gets around the problem of effective demand by its gigantic export surplus, but the United States, which has run a trade deficit for a half-century, doesn’t have that option.
Second, AI is already causing immense anxiety and discontent among young people as they enter the job market. According to a report from the Yale’s Sloan School of Management, “Job market confidence has deteriorated sharply, with the share of US workers who say it is a good time to find a job falling from roughly 70 percent in 2022 to just 28 percent more recently.” The decline in worker confidence about finding a “quality job” coincides with the commercial introduction of AI in 2022. This anxiety about job prospects has political consequences.
Industrial automation, along with shipping jobs overseas, contributed to the growth of right-wing populism and the emergence of the MAGA movement in industrial towns and cities decimated by the loss of factory jobs. The new wave of automation triggered by AI has contributed to the growth of a left-wing populism and socialism among young college-graduates who had expected to find a place for skills in the economy. The base of the Democratic Socialists of America has been described as “downwardly mobile” college-graduates. The economic and political crisis will have to be met by an enlarged role for state and federal governments in the economy. The Great Depression of the 1930s—the first crash that disaccumulation contributed to —let loose powerful political forces on the left and the right. In the United States, the left and the liberal welfare state won out; in much of Europe, the right and fascism did. With Donald Trump’s election in 2016 and re-election in 2024, one can say that the right has won the first round in this conflict. But with Trump now in disfavor, the left could rebound in 2026 and 2028.
Third, the rise of AI and the tendency toward disaccumulation will precipitate, or has already precipitated, personal, psychological, and spiritual crises. Sklar foresaw routine work being replaced by lifetime continuous learning as Americans adapted to the promise of new technology. He envisaged people finding meaning in work that was fulfilling. But the onset of AI raises the possibility that the need for remunerative intellectual labor and much useful work will radically diminish. The ranks of software developers, computer engineers, even scientists and mathematicians will shrink. Much creativity may become the province of algorithms.
The historian James Livingston, who studied with Sklar in the early 1970s, has written extensively on the mechanism and consequence of the disaccumulation of capital. His book Pragmatism and the Political Economy Cultural Revolution, 1850-1940, includes the most comprehensive explanation of how disaccumulation works. Two later books, Against Thrift and No More Work, dwell on the moral and psychological as well as the economic and political consequences of disaccumulation. Livingston believes that the waves of computer automation and artificial intelligence have already reduced the need for employment and will continue to do so. He advocates a government-funded Guaranteed Annual Income that would sever the connection between work and income. That is a position that has also been popular in Silicon Valley among those who believe AI will doom the promise of full employment.
But Livingston also raises the psychological question of where people will find meaning in life if they do not have to work for a living. “And what would society and civilization be like if we didn’t have to earn a living—if leisure was not our choice but our lot?” he asks in No More Work. He doesn’t answer this questions, but it suggests that humanity, at least in the most developed capitalist countries, could be about to encounter a spiritual crisis.
After Sklar’s essay appeared in his 1992 collection, The United States as a Developing Country, a review in the American Historical Review dismissed the essay for its “heavily deterministic Marxist framework” without even summarizing its argument. It has lain fallow ever since. But if there were ever a time for reconsidering—or considering for the first time—its argument, that time is now. We may be at one of those historical crossroads from which America will emerge a very different country. Sklar’s essay, even though written fifty-seven years ago, helps us understand how we got to where we are. It begins to hint at the kind of drastic reformation that would be required to stave off disaster. It also hints at the benefits we could enjoy from artificial intelligence. The freedom that politicians have promised for centuries could have new life. But between the present and fulfillment of that promise lies a great chasm of uncertainty.
John B. Judis
Compact Magazine