The Rise and Fall of the Industrial Working Class in America, 1880-2010

This page has argued that Trump is a symptom of the breakdown of elite-mass relations in the United States. He consummated a movement that has been underway for at least two decades: the migration of the white working class to the GOP and that of the college-educated to the Democratic Party. Both are still underway. And it will likely get worse before it gets better. What is driving the class realignment is the confrontation of class ideologies. In the grip of Boasian antiracism, American elites have become increasingly convinced that the white working class is racist. Antiracism functions as a discourse of moral self-congratulation for the winners of the hourglass economy. It complements the neoliberal ‘script of the self’ of the middle class that fetishizes individual achievement over everything else. This is reflected in loose ties to family, place and community — you must be prepared to abandon all for an opportunity halfway across the world. It is also reflected in middle class child rearing practices, beautifully documented by Barbara Jensen in her authoritative Reading Classes: On Culture and Classism in America (Cornell Univ. Press: 2012).

Meanwhile, the white working class did not get the memo on ‘the century of the self’. In the cultural world of the white working class, meaning is made not through the worship of individual achievement, but through tangled web of relationships with one’s situated community of family and friends. A sense of place features prominently in working class culture, leading social scientists like Kathy Cramer to mistake working class ideology for ‘rural consciousness’. Beyond family and place, the meaning making of the white working class centers on an ideology of hard work — you get up in the in the morning and do the job you need to do to put food on the table for your family. This is a central myth of almost totemic significance. Antiracist scholars often mistake ‘hard working Americans’ as code for white people; suggesting that this Republican soundbite resonates with the white working class because it is racist. They are largely mistaken.

In her brilliant work, The Dignity of Working: Men Morality and the Boundaries of Race, Class, and Immigration (Harvard Univ Press: 2002), the doyen of cultural sociology, Michele Lamont, explains the centrality of hard work in working class ideology as making a virtue of a necessity. That may be so, but the notion, shared by Marxian and non-Marxian scholars alike, that manual work, difficult, dirty, and risky as it may be, is valued exclusively for its extrinsic rewards, ie for provision, is mistaken. As Jeff Torlina argued forcefully in Working Class : Challenging Myths about Blue-Collar Labor (Lynne Rienner: 2014) and Matthew Crawford had in his extraordinary book, Shopclass as Soulcraft: An Inquiry into the Value of Work (Random House: 2010), manual work is intrinsically rewarding in ways that are only invisible to those who have never worked with their own hands.

Manual work, particularly, skilled manual work is highly valued for the routine and intermittent opportunities for solving concrete problems that it affords, for the tangible fruit of one’s labor; the reward that comes from making things of value and quality; for the teamwork and camaraderie of those laboring side by side in shared ecology of attention; and, indeed, valued for the intermittent difficulties that have to be surmounted, often with backbreaking work, since they are, in fact, opportunities for proving one’s worth in the presence of one’s equals. Only the class ideology of the elite has allowed the perpetuation of the Marxian myth of worker’s alienation from their work. As Torlina argues forcefully, workers are far from alienated from their work. It is instead a source of considerable intrinsic rewards and satisfaction that is again anchored in a situated community of skilled practice that Crawford has documented so lovingly.

When the Industrial Revolution got underway in earnest — in America; not England — in the closing decades of the nineteenth century, what obtained was, Gregory Clark has argued, also an ‘industrious revolution‘ (Journal of Economic History: 1998). What made factories productive was, above all, factory discipline. The tempo of work on the factory floor was unlike anything encountered in farming, herding, hunting or gathering; approximated only by intermittent mega-building projects with slave labor earlier in human history. In order to sustain the tempo of factory work, what was required was nothing short of the creation of a whole new script of the self. Lamont’s ‘disciplined self’ emerged not merely from the basic necessity of putting food on the table — although that was a prerequisite to get men to work indoors practically nonstop from sun up sun down for the first time in human history — but from the lived history of the new world of factory work. The construction process of the industrial working class, of the disciplined self, was a long and arduous process. Indeed, it was not until the mid-20th century that it would become central to the meaning-making of the working class. This industrial working class that emerged as the dominant strata in American society at midcentury is the subject of the present investigation.

In what follows, we shall not attempt a historical sociology of the industrial working class in America. Nor shall we attempt to historicize the construction process of the disciplined self. We shall leave both these tasks to future work. Instead, in what follows, we shall document the rise and fall of the industrial working class. We have argued before that what matters, when we think about class, is not so much the mode of production, but the mode of reproduction. We are interested here in the historically specific culture of the American working class and its mode of reproduction. The unit of analysis is not the individual who will be disciplined into a cog in the machine of industrial capitalism. Rather the unit of reproduction is the family. For it is families that reproduce class culture. Class is passed on at your parents dinner table.

What emerged at midcentury was a distinctive form of family life that can be approximately described by the male breadwinner-female homemaker ideal type. This emerged in America from the trials of the early twentieth century. It reached its zenith after world war II. And even though it would soon go into decline, it continues to exercise an almost totemic significance in the American imaginary.

All data below is from the IPUMS database of the Census Bureau. We extract data on 18-54 year old non-Hispanic white males from every census between 1880-2010. Our dataset consists of 15.8 million observations. It took some time to automate the process that allowed me to construct the following graphs. But it was all worth it, for the story they allow me to tell is worth telling. Keep in mind that we are looking at data spanning 140 years. Movements in these series should interpreted as structural change rather than cyclical fluctuations.

The rise of industrial capitalism was part of a broader process of modernization that is reflected in the hockey stick graphs for stature, life expectancy, income, and so on; all of whom show the hockey stick running roughly from 1890-1960. We begin with the dual of these hockey stick variables. Figure 1 displays the decline of the classic traditional occupations — farmhands and domestic servants. The recency of American modernity is evident from the fact that, as late as 1950, 3.5 percent of the working class was engaged in these traditional occupations. By 1960, it had fallen to 1.6 percent; by 1970 to 1.1 percent. And it continued its slow decline thereafter. In 2010, 0.6 percent of our non-Hispanic white male breadwinners were engaged in these occupations.


The rise and fall of the industrial working class is captured by the next figure. We define industrial workers as those classified as craftsmen and operators in the 1950 occupation classification (codes 500-595 and 600-690 respectively). This includes all the skilled manual workers both within and without the industrial sector proper — machinists, blacksmiths, electricians, plumbers, tailors, painters, bakers, and so on. We shall take this to be our occupational definition of the industrial working class. The share of this class rose from 26 percent in 1880 to 36 percent is 1920; when it suddenly and inexplicably stalled. It remained mired at that level through the Great Depression. It was only after World War II that the share of industrial workers would explode upwards to 43.6 percent in 1950; peaking at 44.4 percent in 1960. It stayed high at 43.6 percent in 1970 before it begins to decline dramatically. By 2010, the share had fallen back to 28.5 percent, below the 29.3 percent it had attained in 1900. What this graph captures is the full world historical cycle of the rise and fall of the industrial working class in America.


What happened? Economic growth resumed after the stagflation of the 1970s; as did industrial production. What happened is labor saving innovations, including automation, foreign competition, and the offshoring of many production processes. The deindustrialization of America was matched by other industrial nations, including the Soviet Union, as Stephen Kotkin has documented. There were exceptions, but postindustrial rustbelts are part of a much bigger story that cannot be confined to a neat story of American decline under the onslaught of the neoliberal counterrevolution. That is not to say that there is nothing to the narrative of barbarians at the gate. But what let them through door was the weakness of the great industrial firms who had underwritten the Treaty of Detroit on which had rested the corporatist-liberal synthesis of midcentury American political economy. The profits of the industrial firms would be restored, without a restoration of underlying productivity, through a process of oligopolization supervised by Wall Street, and reproduced ever since through the deployment of overcapacity to deter entry.

The next figure graphs the structural transformations of the American occupation structure over 1880-2010. We have put them on the same scale to facilitate comparison. The size of the industrial workforce that build America would later be matched by the managerial and professional elite who have cornered all the gains since the neoliberal counterrevolution. Their share of the workforce grew relentlessly decade after decade from 14.5 percent in 1940 to 38.7 percent in 2000, when it seems to have finally plateaued. It came down modestly to 38.5 percent in 2010. These occupations are, as we shall see, only open to the college-educated minority, whereas the others are largely run by the high school-educated majority. The other two categories are numerically much smaller. Clerical workers, low-status office workers of all sorts, reached roughly 14 percent by 1930 and have stayed at that level ever since. The last category, of unskilled laborers and service workers, janitors, cooks, waiters, and attendants of all sorts, rose from 12 percent in 1880 to 16 percent in 1940, before declining to 9 percent. This reflects not an absolute decline in numbers, but a relative decline as the share of the professionals and managers continued its rise and that of industrial workers reached its zenith. Since that it has tended upwards, as industrial jobs were lost and workers were forced into unskilled service occupations amidst much self-congratulation about the new economy. In 2010, unskilled service workers accounted for 16.6 percent of the workforce, an all time high. occupational_structure.png

What happened to the men who would have been skilled manual workers at midcentury once industrial decline accelerated after the 1970s? A few may have been absorbed in the unskilled service occupation; a few may have even climbed up the ladder to become managers and professionals — these are the class migrants who would do the most authoritative works on the white working class. (We shall have occasion to visit their work later.) But many simply abandoned the labor force for a life of disability or indolence, as the next figure shows. We decompose this set by age groups to interpret the pattern. The dramatic increase in 18-24 year olds who exited the labor market may be due to those who have gone on to study, although here too the sharp spike in 2010 suggests discouragement. The exit of older workers, particularly for the 35-54 age group, cannot be explained away by appealing to greater investment in education. In fact, the greater rise in 35-54 year olds relative to 25-34 year olds is probably due to the greater ease and acceptance of going on disability for older workers. Almost 10 percent of the former and 8 percent of the latter were no longer looking for work by 2000.


Let’s move on to those who stayed in the workforce. In order to understand the structural changes underway, we now condition on the respondent sporting a high school diploma but not a college degree. (The education data is unavailable before 1940.) These are the men who would have worked in the skilled manual occupations had the midcentury occupational structure survived. So where did they end up? The next figure shows that their share of clerical jobs did not rise at all, and there was only a slight uptick in those migrating to the professional and managerial class. The rest were absorbed into the unskilled service occupations, whose share more than doubled from about 8 percent in 1960 to more than 20 percent in 2010. So there were effective two options available to high school graduates facing increasingly scarce skilled manual jobs. Either exit the labor market for a life of indolence and dependence, or accept a poorly paid, unskilled job in the service occupations. Most took the latter option, no doubt in order to put food on the table for their families.


What about the college-educated middle class? They maintained their share of the professional and managerial occupations. Although some with college degree too ended increasingly in the unskilled service occupations, as the next figure shows.


How did these transformations of the occupational structure affect family formation and family reproduction? Before we get to these graphs, we must note that cultural changes that began soon after the world struggle at midcentury and gathered pace after the 1960s, was already undermining the male breadwinner-female homemaker model of the midcentury working class family. In his magisterial monograph, that inspired me to dig into IPUMS data, Labor’s Love Lost: The Rise and Fall of the Working-Class Family in America (Russell Sage: 2014), Andrew Cherlin argued, contra Murray’s argument in Coming Apart: The State of White America, 1960–2010 (Crown Forum: 2012), that the sexual revolution, and late-20th century cultural changes more generally, affected both working class and middle class families. Yet, family stability and family reproduction declined in both but the decline was arrested in the middle class whereas it wasn’t in the working class because of the continued deterioration of economic prospects for the high school educated in the modern hourglass economy. As we shall see, that is exactly right.

The next two figures show the probability of marriage by education and occupation classes. The latter allows us to extend the series back to 1880. But the pattern that emerges from both is consistent. The decline of marriage has been much more pronounced in the working class by either definition. probMgivenEdu.png


The pattern also holds if we condition on both education and occupation. Whereas the decline in family formation rates has slowed in the professional-managerial class; they have continued to fall alarmingly even for those high school graduates who do skilled manual work.


The next figures document the pattern of family reproduction. We again use multiple definitions of class. The next figure uses education classes. family_reproduction_by_education.png

As before, using occupational classes allows us to extend the graphs back to 1880. Here, blue collar refers to skilled manual workers, and white collar refers to professional and managerial occupations — as in the classic midcentury definition of these terms. We can see that the midcentury pattern was truly exceptional for both the middle class and the working class.


The next figure uses both occupation and education to define classes. By industrial working class we mean breadwinners with high school diplomas who perform skilled manual work; by middle class we mean college-educated breadwinners who fill professional or managerial roles.


No matter how you slice and dice it, the decline of the working class family is much more pronounced than the decline of the middle class family. And the movement shows no sign of abating. The next figure shows that at least some of the decline in the reproduction of families with a high school-educated male breadwinner can be traced to structural change in the American economy. The top-left panel displays the probability of getting skilled or unskilled work conditional on having a high school diploma but not a college degree. The top-right displays the probability of family formation conditional on the occupation of a high school graduate. The bottom-left panel displays the probability of family reproduction conditional on occupation and high school diploma. The bottom-right shows the divergence of family reproduction rates conditional on educational class. The patterns make it clear that changes in the occupational structure can explain the divergence in the rates of the reproduction of the two educational classes that constitute the bulk of American society. For while family reproduction rates have declined for both skilled and unskilled occupations, the levels are much lower for the unskilled occupations. Of course, the story is even sorrier for those who have exit the labor market for a life of dependence and indolence.


We have shown that structural changes in the American economy have strongly conditioned class reproduction in America. Not only is the midcentury ideal of the male breadwinner-female homemaker family no longer available to most of the working class, even basic family reproduction is now beyond the means of your average working class man. But it gets worse still. The next figure shows that the dispersion of earnings of families with male breadwinners, as captured by the interquartile range, has increased dramatically since the 1970s.


This development is just as, if not even more, pronounced for working class families than middle class families. fyiqr_highschool.png


And this remains true even if we condition on our high school graduate finding work in one of the skilled manual occupations.



In future work, we shall investigate the historical sociology of the white working class. We have seen that transformations in the occupational structure of the American economy strongly conditioned patterns of family formation and reproduction. Specifically, the modern hourglass economy has undermined the economic foundations of the working class family, and left the survivors with ever meager resources to deal with the catastrophic consequences. How did white working class communities cope with the trauma? What ripple effects did it have on working class ideology? on the meaning making of the working class? on boundary work? We know from the work of cultural sociologists that they doubled down on the script of the ‘disciplined self’ to manage the painful transition from the rewarding life of skilled industrial work to the indignities of service occupations and a life of precarity. Is that why boundary work became more pronounced against the underclass? against ‘hard living’ whites as well as racialized minorities? And how did these shifts in discourse and reality come to weigh on elite-mass relations in the United States? What precisely was the recipe for the clash of class ideologies now manifest? How exactly did we get here?





6 thoughts on “The Rise and Fall of the Industrial Working Class in America, 1880-2010

  1. Very nice work! Overall, strongly suggestive of a radical loss of social solidarity post-1960 (a sociological concept, Asabiyyah, often traced to Ibn Khuldun) that has been incorporated into econometric socio-historical models by Peter Turchin (cf. “Ages of Discord,” highly recommended).

    Regarding Marx, he got the forecast right, that capitalism leads to concentration of wealth and immiseration of the working class, but history suggests he got the solution wrong. Adam Smith suggested capitalism wouldn’t work without moral sentiments (basically, empathy), esp. from the elites, which seems to have disappeared in USA.

  2. Actually, since income, and wealth, and educational opportunity inequality and its attendant self-justifying Boasian antiracialism seem to be a Policy Tensor obsession, I will add my own surmise that the best way to reestablish elite-populist rapport would be to aggressively steepen the progressivity of the income tax to the point where top household incomes might be, say, at most 50X the bottom quartile cutoff (roughly $30K to $1.5m). This could be done in a revenue-neutral way.

    For the societal benefits of more optimal inequality see Wilkinson’s TED talk at or his book “The Spirit Level.”

    1. I read The Spirit Level with great interest, of course. I am not persuaded that, at least as a strategy for restoring elite-mass relations, aggressively taxing high incomes is a winning strategy. The reason is that vertical income and wealth polarization greatly pisses off the college-educated middle class, particularly the progressives in Brooklyn, but it works against the grain of the meaning-making of the working class. Basically, the ‘billionaires should go extinct’ is boundary work against financial elites that is simply not congruent with the boundary work of the white working class — which is directed not against financial elites but against the social and cultural elites who run the country and control the media. Any strategy that does not aim to root out the contempt with which the middle class holds the white working class, is superficial and unlikely to work as advertised. This is why I have become so skeptical of the prospects for the progressive candidates, who have basically gone all-in on the diagnosis you suggest — the masses are pissed off at elites for cornering all the gains; ergo, the solution is high taxes for the rich, and freebees for the masses. Well, is that what they are really pissed off about? Have you bothered to listen to them? Even a cursory stroll through the literature on the cultural sociology of the working class (begin with Lamont), or that on classism (begin with Barbara Jensen and Joan C. Williams), should be enough to convince you that this causal diagram (‘inequality -> instability’) is profoundly mistaken. That’s NOT what the revolt of the working class is about. But that being said, Bernie could still prevail against Trump, largely because people trust him. That is a valuable political resource that Clinton simply did not have.

      1. Tensor:


        1. In an evolutionary adaptationalist human tribe type way, I have come to believe after reading Wilkinson that our innate preference is for much less inequality than we have in USA now. So I take this as axiomatic, subject to optimization and the usual epigenetic nature/nurture flexiblity. Turchin also holds to this.

        BTW, my tax-heavy solution would only about replicate the actual income distribution in Japan, where there is still lots of solidarity. Our top marginal tax rates (70%+) were in the 50s and 60s when USA still had post-WWII solidarity going for it. WWII (~1942) was a great income leveling event; see Piketty and Saez data on this. (But do we really need a common enemy?)

        2. I agree that the working class (of which I have some in-law relatives) work hard and well to support their families, without the precious egotism of upper middle class professionals. So I don’t suggest freebies for the masses, nor do they want them, but less taxes. I do believe the elites need to feel some pain tax-wise, however. And we need to kill the Fed, which subsidizes the banks and other big money players while permitting the little people to be repeatedly shafted and expropriated through fraudulent conveyance of bad debt and various forms of fraud.

        3. ‘inequality -> instability’ is the lesson of history, T-man. Most people don’t get it. We are due for depression, revolution, world war, hyperinflation, or some combination thereof, without a miraculous gain in competence and cooperation in bad debt renegotiation, internationally, and fairer distribution of product generally. This would involve the banks charging off some of the bad debt they’re still carrying in hopes of buying up the rest of the world (eg, Greece).

        4. So the working class is well aware that they’re being shafted economically, and do indeed want redress on that (more progressive taxation, no more bank bailouts, level playing field, etc.); and that on top of that they know the elites are blaming them for their problems with the elites’ Boasian antiracialism, ie, by accusing the working class of being racist and stupid, and possibly religious to boot. That’s a double whammy of obnoxiousness and rent-graft flowing down to the working class. If you look at IQ by occupation, the bands are wide, with a lot of overlap. There are a lot of high-IQ people working trades because they couldn’t afford college.

        I propose making profit-sharing mandatory for all corporations, as it is with the best tech start-ups, where everyone get the benefits of capital ownership. The Japanese corporations do this, I believe.

        As Yves at says, the neoliberal synthesis can be summed up as:

        Rule 1. Because markets.
        Rule 2. Go die

        Change needed.

        Keep up the good work!

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s