Skip to content

The Making of a New Working Class


Gabriel Winant (@gabrielwinant) is Assistant Professor of US History at the University of Chicago and author of The Next Shift: The Fall of Industry and the Rise of Health Care in Rust Belt America.

This post introduces a symposium on The Next Shift: The Fall of Industry and the Rise of Health Care in Rust Belt America. Read the entire series here.


When the Supreme Court constitutionalized collective bargaining in the 1937 Jones & Laughlin ruling, it described steel production around Pittsburgh as “the heart of a self-contained, highly integrated body” stretching from the West Virginia coalfields to the Minnesota Iron Range—a kind of corporeal image of the industrial working class itself. Hospital workers, however, were apparently not part of that body: when a group of workers at Western Pennsylvania Hospital began organizing a union with the CIO three years later, 26 area hospitals collectively sought an injunction. “Hospitals,” they argued, “are not employers, nor are persons connected with them employees.” Eventually, the Pennsylvania Supreme Court agreed, finding that even private hospitals were a form of public service, “not an industry”—a finding ratified by the 1947 Taft-Hartley Act.

Fast forward 73 years, to when another union attempted to organize Pittsburgh hospitals. Even though the industry was now vastly larger and more consolidated, the University of Pittsburgh Medical Center made a familiar-sounding argument, claiming that it “has no employees.” Rather than contest the jurisdiction of the law by industry, UMPC’s denial of responsibility relied upon its supposed fissured employment structure: UPMC was, it argued, merely a parent company, and the union should take its business separately to each subsidiary hospital.

Despite these differences, the two arguments represent a continuity: the institutional subordination of the work of social reproduction. The mechanisms of this subordination vary historically, but without fail in U.S. history they involve hierarchies of race and gender. Identifying how those mechanisms have worked in turn becomes necessary for any wider exploration of the institutional and social forms of class inequality and class conflict in a given moment. Class never exists as an abstract polarity, but rather is always found in particular agglomerations of historical experience: encoded in law and habit; inscribed on the body and in the built environment; and encased and contested in social institutions and cultural and symbolic practices. The organization of working-class social reproduction—narrowly, the making and maintenance of labor supply—is a central question in all these arenas.

In capitalist societies, the institutions of social reproduction typically hold a distinctive, contradictory position as a semi-separate sphere—constructed as institutionally exterior to the economy of which they in fact form an integral part. The family, of course, is the classic instance, but hospitals, schools, and—in another way—prisons and police all exemplify it. While the New Deal state did not create such a structure ex nihilo, it did renovate and reinscribe it—particularly as the institutions of the welfare state drifted rightward in the 1940s, under the pressure of the Cold War. In particular, access to social rights such as health care and retirement became increasingly privatized, setting health care—now a private social benefit—on a distinct growth path.

In The Next Shift, I carry out a social history of economic change in Pittsburgh from 1950 to 2000, showing how the institutions of this compromised, public-private system gradually turned the health care system into a key site for the negotiation of class conflict and the absorption of social disturbance. As it came to play its anomalous role, the health care system grew—and grew, and grew. But it grew, as it were, on the exterior of a social formation organized around industrial production: health care provision was constructed as an adjunct to the industrial economy, which was the site of economic regulation and disbursal of social citizenship—hence the NLRA’s coverage of steel mills but not hospitals.

These distinctions came to matter tremendously as the giant industrial body described in Jones & Laughlin gradually aged, got sick, and died. Deindustrialization made the population older, poorer, and sicker. The region’s hundreds of thousands of steel industry employees, dependents, and retirees then formed a booming health care market. While an aging population seasoned by years in toxic mills and smoky valleys carried distinct health burdens, their high-quality collectively-bargained insurance allowed them to use the health care system for a broad portfolio of services, combining medicine with what we might now see as social work and long-term care functions. By 1979, on the eve of the Volcker Shock, the Pittsburgh region generated 1.6 hospital inpatient-days per capita—compared to 1.2 nationally (and .56 nationally today). By my rough calculation, between one-quarter and one-third of hospital beds at any given time in the late 1970s would have held a patient covered by the steelworkers’ Blue Cross plan. Borrowing on municipal bond markets on the strength of this revenue stream, hospitals expanded rapidly.

This broad profile of hospital services proved especially useful as women increasingly sought wage work in the 1970s. Thanks to the contraction of manufacturing, their husbands increasingly lacked access to the well-paying steelwork that their fathers had performed—a transition that shifted black women especially rapidly into low-wage service work. As the breadwinner wage declined and women entered the workforce in greater numbers, the supply of free care in the home diminished. Families came to increasingly rely upon institutional health care, which many could afford thanks collectively-bargained insurance. Care labor was thus transferred from home to hospital. In 1979, those 1.6 inpatient-days were serviced by 17.3 hospital workers per 1,000 population—compared to 13.8 nationally.

Hospital workers’ own efforts to renegotiate the terms of care within the expanding health care system were largely met with defeat, however, as the institutional and legal status of hospitals as external adjuncts to industry threw up obstacles to workers’ organization. Fully legal anti-union coercion largely prevented unionization in Pittsburgh. Even in cities where unions were more successful, labor soon encountered the problem that the health care system, although administered privately, was fundamentally a creature of public policy—with responsibility fragmented and distributed beyond the employer. “It can be very frustrating,” explained a labor leader in 1975, “to negotiate with management and realize what they are offering you is contingent on funding from another source, whether it be legislature or whether it be a third party.”

When the final, brutal economic contraction of steel came in the early 1980s, it accelerated these processes further. Job loss sped up to the point of outright plant closure, and health care grew even faster, working countercyclically. It fed off the social decay from which the system was shielded by its public-private construction, as 17 percent unemployment by 1983 inflicted a massive negative health shock on the population and intensified the social demands on the health system, further speeding its growth across region now increasingly called “Rust Belt.” In 1981, Pittsburgh ranked 24th among U.S. cities for average length of hospital stays; of the 23 ahead of it, all but one were centers of concentrated industrial job loss. In 1980, Pittsburgh had 5 hospital beds per 1,000 people; by 1990, 7 per 1,000, and health care jobs were as numerous as metal jobs had been in 1960. Health care became, alongside punishment, our main institutional response to rising inequality.

This continuation of sectoral-specific inflation into the 1980s triggered alarm in Washington and led the Reagan administration and Congress to rewrite Medicare’s reimbursement formula in 1983, shifting from a retrospective cost-plus system to reimbursement rates fixed in advance by diagnosis — squeezing less acute admissions and long hospital stays. A decade of implementation caused the disintegration of the community hospitals, which could no longer function as general social shock absorbers. The academic medical hospitals, which were able to afford new investment in technologies and staff that allowed them to perform high-reimbursing complex procedures, prospered and bought up the collapsed community hospitals—eventually shutting many down. Nursing homes and home care agencies blossomed to absorb the market niche once held by community hospitals, extending the new low-wage labor market further.

The relationship between social inequality and rising demand for health care has never been interrupted, only pushed down different channels. Health care now accounts for one in seven jobs in the United States, and nearly one in five in Pittsburgh and similar cities. Nationwide, as sociologist Rachel Dwyer shows, the broader care economy (including education, domestic work, and so on) accounted for 56 percent of all low-wage job growth in the 1980s, 63 percent in the 1990s, and 74 percent in the 2000s.

The Next Shift was not originally intended as an intervention in health policy—although I hope I have said something meaningful there—but rather as a work of political economy and working-class history. I aimed to explore the question of what material forces configure social class when the factory is gone and the working class has been rendered dispersed and invisible, not just culturally but through real institutional mechanisms—as per UPMC’s claim in 2013 that there are no workers to be found there. Beginning from an intuition that we should not let deindustrialization be the end of the story of working-class formation led me paradoxically to the finding that the legacies of struggles within industrial capitalism, particularly long-lived historical institutionalizations of race and gender, continue to shape the terrain of class formation today and set the terms on which class conflict proceeds.

Those terms now entrap all of us. If the pandemic has made anything clear, it is that we all depend on care workers—a dependence which has caused their numbers to grow—yet they are individually disposable. As firms suppress labor costs, they generate impossible conditions for workers and harrowing experiences for patients. The growth of this workforce, our collective dependence upon the people who make it up, the outrages of their working conditions, and degradation of the care they are able to provide—all these are linked together by the historical exteriorization of social reproduction and the negotiation and allocation of health care through the private sector. For this reason, the crisis of care that we witness every day is both deeply historically rooted and, potentially, a lever of change for the millions of us whom the health care system touches, exploiting us and keeping us alive at once.

I’d like to thank the contributors to this symposium, and I’m looking forward to the discussion: the combined crises of our labor market and health care system could hardly be more pressing.