The concept of a "permanent underclass" refers to a segment of the population that is structurally locked out of economic mobility — not temporarily poor, but effectively excluded from meaningful participation in the labor market and social life across generations. The idea is far older than it might seem. Its intellectual lineage stretches from Karl Marx's lumpenproletariat in the 1840s through Swedish economist Gunnar Myrdal's formal coinage of the term in 1962, to Charles Murray's incendiary 1984 book Losing Ground, and most recently to Silicon Valley's fears about artificial intelligence creating an entirely new stratum of economically irrelevant humans. This report traces the full arc of that intellectual journey, analyzes the major theoretical camps, and maps the competing frameworks — including ideas that directly challenge or propose alternatives to the underclass thesis.[^1][^2][^3]
Long before the term "underclass" existed, Western societies wrestled with the idea of a dangerous, irredeemable poor. In Victorian Britain, reformers and politicians worried about what they called the social residuum — people considered too morally degraded to be lifted by ordinary social programs. Shortly after came the concept of "unemployables," a group seen as constitutionally incapable of participation in industrial society. These ideas were deeply moralistic: poverty was framed not as a structural consequence of industrial capitalism but as a symptom of personal failure, vice, or hereditary defect.[^4][^5]
Henry Mayhew's monumental mid-19th century work London Labour and the London Poor offered one of the earliest systematic accounts of the world below the working class — describing the lives of casual laborers, street people, and prostitutes who existed entirely outside the formal economy. Mayhew did not moralize; he documented. But the reformers and eugenicists who followed him were not as restrained.[^2]
The early 20th century brought an even more dangerous intellectual development: the eugenics movement. ==The Eugenics Society in Britain actively promoted the concept of the "social problem group" in the 1930s — the idea that a hereditarily inferior subset of the population was responsible for crime, poverty, and degeneracy. After World War II, the same group was rebranded as "problem families" in the language of welfare policy. The framing changed from genetic to cultural, but the underlying logic — that a particular stratum was fundamentally separate and self-reproducing — remained intact.==[^4]
Karl Marx and Friedrich Engels coined the term Lumpenproletariat (German: "ragamuffin proletariat" or "rabble proletariat") in the 1840s, using it in The German Ideology and later in The Communist Manifesto. The concept refers to people entirely outside the regular labor system: beggars, criminals, vagabonds, prostitutes, and the chronically unemployable. Marx described this group as "the dangerous class, the social scum, that passively rotting mass thrown off by the lowest layers of old society".[^6][^7]
Critically, Marx did not romanticize the lumpenproletariat. Unlike the proletariat (whom he saw as the historical agent of revolution), the lumpenproletariat were, in his view, politically unreliable — easily exploited by reactionary and counter-revolutionary forces, often serving as hired muscle for capitalists against their fellow workers. They lacked class consciousness because they had no class position in the productive sense. This dismissal would be echoed and debated by sociologists for the next 150 years.[^8]
The lumpenproletariat and the later concept of "underclass" share striking structural similarities: both describe people excluded from regular wage labor, associated with crime and welfare, and seen as existing outside the social contract. The key difference is analytical framing — Marx saw this as a structural byproduct of capitalism's need for a "reserve army of labor," while conservative theorists like Murray would later frame it as a product of individual moral failure.[^9]
The specific term "underclass" was almost certainly invented by Swedish economist Gunnar Myrdal. In his 1962 book The Challenge to Affluence, Myrdal used the word — derived from the Swedish underklass — to describe a new social phenomenon emerging in the advanced capitalist economy of the United States. He defined the underclass as:[^10][^11]
> "A class of unemployed, unemployables, and underemployed who are more and more hopelessly set apart from the nation at large and do not share in its life, its ambitions, and its achievements."[^11]
Myrdal's framework was emphatically structural. He argued that economic development, while opening opportunities for many, simultaneously closed them for others. Automation, geographic shifts in industry, and the demands of a modern economy were creating a group that was permanently detached — not because they were lazy or morally deficient, but because the economic system no longer needed them. Importantly, Myrdal pointed out that the word underclass was absent from English usage at the time — most Americans considered their society classless, making the concept itself culturally difficult to accept.[^10]
Myrdal's structural analysis was groundbreaking, but it received far less attention than it deserved. The political winds of the 1970s and 1980s would ensure that the behavioral interpretation of the underclass — not the structural one — became dominant in public discourse.[^12]
While Myrdal focused on structural economic exclusion, the American anthropologist Oscar Lewis developed a parallel but importantly different framework: the "culture of poverty." Based on extensive fieldwork in Mexico City and Puerto Rico, Lewis argued that the poor were not merely victims of economic conditions but had developed a distinct subculture as an adaptation to chronic poverty.[^13][^14]
This subculture, Lewis argued, had its own logic, values, and behaviors — including fatalism, present-time orientation, distrust of institutions, and a sense of marginality. Crucially, once formed, this subculture transmitted itself intergenerationally. Children socialized into the culture of poverty internalized its values before they could fully understand the economic forces that had created them. Even if structural conditions improved, the culture could persist.[^15]
Lewis intended this as a sympathetic and radical argument — an indictment of capitalism for creating such conditions. He explicitly described the culture of poverty as "a reaction of the poor to their marginal position in a class-stratified, highly individuated, capitalistic society". But the theory was quickly stripped of this structural critique and weaponized politically. Critics noted that Lewis's framework easily lent itself to "blaming the victim" — treating poverty as a self-inflicted cultural failure rather than an economic and political outcome.[^14][^13]
In 1965, sociologist and future U.S. Senator Daniel Patrick Moynihan produced a government report titled The Negro Family: The Case for National Action, which argued that the deterioration of the Black family structure — particularly the rise of female-headed households in urban areas — was producing a self-perpetuating cycle of poverty that he called a "tangle of pathology". The report was immediately controversial. Critics argued it ignored the structural causes of family instability (discrimination, unemployment, the legacy of slavery) and shifted focus to Black cultural failure. Defenders said Moynihan was trying to galvanize liberal policy action. The debate over the Moynihan Report arguably never ended; it shaped underclass discourse for the next half-century.[^16]
In Britain, Conservative Secretary of State for Social Services Sir Keith Joseph delivered a landmark speech in June 1972 introducing the "cycle of deprivation" hypothesis. Joseph argued that social problems — poverty, poor parenting, school failure, criminality — were transmitted from parents to children not primarily through structural inequality but through inadequate socialization and parenting. Children raised in deprived families could not develop the skills and values needed for productive adult life.[^17][^18]
Joseph's framework was important for several reasons. It moved the debate toward intergenerational transmission — asking not just why individuals are poor but why entire families remain poor across generations. It also became a blueprint for later conservative welfare reform arguments in both Britain and the United States.[^18]
No figure did more to shape — and inflame — the underclass debate in the 1980s than Charles Murray, a political scientist at the Manhattan Institute. His 1984 book Losing Ground: American Social Policy, 1950–1980 argued that the welfare state itself was the primary cause of the underclass. Murray's core thesis was that social welfare programs created perverse incentives: they rewarded single motherhood, discouraged marriage, reduced the incentive to work, and thereby produced the very poverty they were meant to cure.[^19][^20]
Murray characterized the underclass by three behavioral markers:
The solution, ==Murray argued, was radical: abolish most welfare programs entirely. The hardship caused would be temporary; people would find work out of necessity. Losing Ground became the intellectual centerpiece of the Reagan administration's welfare reform rhetoric and was described as the "most heavily funded and widely-heralded piece of social research ever carried out under reactionary auspices".[^22][^20]==
==Critics were fierce. Economists like Christopher Jencks called the book "Social Darwinism" that provided "moral legitimacy for budget cuts" driven by ideology, not evidence. Others pointed out that welfare benefit levels had actually declined in real terms during the 1970s even as the behaviors Murray blamed on welfare had increased — a correlation that didn't fit his causal story==.[^23][^19]
Murray brought his ideas to the UK in November 1989 via a Sunday Times article, arguing that a similar "plague" of underclass pathology was spreading across the Atlantic. This triggered a major British debate that imported the American concept into very different sociological and policy contexts.[^21]
The most powerful structural counterweight to Murray came from sociologist William Julius Wilson, whose 1987 book The Truly Disadvantaged: The Inner City, the Underclass, and Public Policy rejected both conservative and liberal interpretations of urban poverty. Wilson agreed that inner-city communities showed alarming rates of crime, unemployment, single parenthood, and welfare dependency — but argued the causes were entirely different from what Murray claimed.[^24]
Wilson's structural argument centered on deindustrialization. During the 1970s, large manufacturing industries withdrew from American inner cities. This removed hundreds of thousands of entry-level jobs from communities that were already racially segregated and geographically isolated. At the same time, middle-class Black families — newly able to move to the suburbs after the Civil Rights Act — left urban neighborhoods, removing role models, social networks, churches, and community institutions. The combined effect was a community stripped of both economic opportunity and social resources.[^25][^26]
Wilson explicitly criticized Murray for ignoring these macro-economic shifts and for focusing on welfare incentives that were, in fact, declining. He also criticized liberal poverty researchers for avoiding the topic of race and class simultaneously, which had left the field open for conservatives to dominate the narrative.[^27][^25]
Wilson's work was enormously influential. It shifted serious academic debate away from behavioral explanations and toward structural analysis of labor markets, geography, and racial segregation.
Sociologist Douglas Massey added another structural layer that Wilson had underemphasized. In his landmark 1990 article "American Apartheid: Segregation and the Making of the Underclass," Massey demonstrated through simulation that residential racial segregation was a crucial missing variable in all prior underclass theories. When poverty rates rise in a racially segregated city, they don't spread evenly across the population — they concentrate with devastating intensity in Black neighborhoods. This concentration of poverty then creates self-reinforcing disadvantages: poor schools, high crime, deteriorating infrastructure, and social isolation. Massey concluded that policies to help minorities would fail without specifically attacking residential segregation.[^28][^29]
Sociologist Herbert Gans offered one of the most penetrating critiques of the underclass concept itself. In his 1994 essay "Positive Functions of the Undeserving Poor," Gans argued that the label "underclass" served ideological and political functions for more privileged Americans, not analytical ones. By labeling the poor as "undeserving," wealthier Americans:[^30]
Gans also argued that the behavioral definitions embedded in the "underclass" concept reflected mainstream middle-class values projected onto the poor, rather than empirical descriptions of how poor people actually behaved.[^32]
French-American sociologist Loïc Wacquant took the deconstruction furthest. In his 2022 book The Invention of the 'Underclass': A Study in the Politics of Knowledge, Wacquant traced how the term was transformed from Myrdal's careful structural concept into a racialized folk devil. His key findings:[^33]
Wacquant argued that the term operated as a "turnkey problematic" — a ready-made framework imposed on researchers by media, think tanks, and politics, compromising scientific autonomy.[^33]
Professor David Gordon, who led the largest-ever poverty research study in the UK, offered this blunt verdict:
> "These ideas are unsupported by any substantial body of evidence. Despite almost 150 years of scientific investigation, not a single study has ever found any large group of people with behaviors that could be ascribed to a culture or genetics of poverty."[^4]
The underclass concept, in its behavioral version, has consistently struggled to produce robust empirical evidence for intergenerational cultural transmission as a cause of poverty, as opposed to simply documenting that poverty clusters in families — which is equally explained by structural factors like wealth inheritance, school quality, and neighborhood effects.
British economist Guy Standing proposed a more modern and global analog to the underclass in his 2011 book The Precariat: The New Dangerous Class. Standing identified the emergence of a new global class defined not by poverty but by precarity — chronic economic insecurity and lack of stable identity:[^35]
The precariat differs from the traditional underclass in that it includes people who are working — but whose work provides no economic stability. It is rapidly growing due to outsourcing, digitization, and the decline of the public sector as a stable employer. Standing estimated that in some countries, the precariat constitutes a quarter of the adult population.[^36][^35]
French economist Thomas Piketty's landmark 2013 book Capital in the Twenty-First Century provided the mathematical scaffolding for understanding how class hierarchies become permanent. Piketty's central thesis: when the rate of return on capital (r) consistently exceeds the rate of economic growth (g), wealth concentrates in the hands of capital owners over time. Returns on capital compound faster than wages grow; the rich inherit and build on their advantage indefinitely while workers' share stagnates.[^37][^38][^39]
==Piketty's framework doesn't use the term "underclass" but provides the economic mechanism by which a permanent class hierarchy develops: those born with capital accumulate more capital; those born with only their labor — and no capital — fall further behind in relative terms. His proposed solution was a global progressive wealth tax==.[^39]
The underclass debate is ultimately a microcosm of the oldest argument in poverty research: Is poverty caused by culture (values, behavior, attitude) or structure (economic forces, discrimination, labor market dynamics)?
| Framework | Primary Cause of Poverty | Political Implications | Key Thinkers | |---|---|---|---| | Culture of poverty | Self-perpetuating values and behaviors | Behavioral interventions, reduced welfare | Oscar Lewis, Charles Murray, Keith Joseph | | Structural underclass | Economic deindustrialization, racial segregation | Jobs programs, anti-discrimination, urban policy | Gunnar Myrdal, William Julius Wilson, Douglas Massey | | Lumpenproletariat | Capitalism's reserve army of labor | Abolition of capitalism | Karl Marx, Friedrich Engels | | Precariat | Neoliberal labor flexibility, globalization | UBI, labor rights, reclaiming commons | Guy Standing | | Capital concentration | Returns to capital exceeding growth | Progressive wealth tax, redistribution | Thomas Piketty |
The ideological opposite of the permanent underclass theory is meritocracy — the belief that talent and effort determine outcomes, making poverty a temporary state for those who work hard enough. The term was coined by British sociologist Michael Young in his 1958 satirical novel The Rise of the Meritocracy, but in a deeply ironic twist, Young intended the book as a warning, not a celebration.[^40]
==Young's actual argument was that a society organized purely on merit would not produce mobility — it would produce a new, more rigid class system in which the elite could justify their privilege as earned, making them less sympathetic to the disadvantaged and more aggressive in defending inequality. A meritocracy identifies the winners as deserving and the losers as simply not good enough, which is arguably more psychologically brutal than an aristocracy, which at least makes no claim that the poor deserve their poverty.==[^41][^40]
Gary Becker's human capital theory provides the economic version of this counterargument: investment in education and skills development can overcome poverty by increasing individuals' productivity and wages. If the poor lack market-competitive skills, education and training are the solution. The theory underpinned decades of welfare-to-work policy.[^42]
The empirical problem: human capital theory assumes a relatively level playing field where skills are the binding constraint. But research consistently shows that social networks, family wealth, residential location, school quality, and racial discrimination are equally or more important than individual skill levels in determining economic outcomes.[^43][^44]
Sociologists Alejandro Portes and Min Zhou's segmented assimilation theory (1993) offers an important nuance: the permanent underclass is not just a domestic class but also a destination segment for certain immigrant groups. Their theory identifies three paths for immigrant integration:[^45]
Which path an immigrant family takes depends on their human capital, racial classification by the receiving society, and the nature of their social networks. This framework is significant because it shows the underclass is not an isolated community but a segment into which new groups can assimilate downward — the underclass as gravitational field rather than static population.[^47]
While American debate centered on the "underclass" (usually with racial and behavioral overtones), Europe developed the parallel concept of social exclusion. The key differences:[^48]
The French "solidarity paradigm" of social exclusion saw exclusion as a failure of social cohesion — a deficiency in the bonds that hold society together. This framework placed responsibility on the state and society to restore inclusion, rather than on individuals to reform their behavior.[^49]
One of the most widely discussed policy responses to the permanent underclass problem — particularly in its AI-era incarnation — is Universal Basic Income: an unconditional cash transfer to every citizen regardless of income or employment status. UBI proponents argue it would eliminate the poverty trap by ensuring no one falls below a livable floor, removing welfare stigma, and allowing workers to reject exploitative work.[^50][^51]
Critics raise significant concerns: UBI may be unaffordable; it may provide a rationale for eliminating targeted programs (Medicaid, housing vouchers, food stamps) that provide more value to the poor than an equivalent cash transfer; and it may not address the complex, non-financial dimensions of poverty. The debate is currently unresolved — pilot programs have shown positive effects on well-being but inconclusive effects on employment.[^52][^50]
In October 2025, writer Kyle Chayka published an article in The New Yorker titled "Will A.I. Trap You in the 'Permanent Underclass'?" that gave mainstream visibility to a conversation already running hot in Silicon Valley. The article observed that the term had been spreading as both a genuine fear and an online meme — phrases like "You have two years to create a podcast to escape the permanent underclass" becoming ironic badges of tech-industry anxiety.[^53][^54]
The fears are not entirely new in structure, but they are new in scale and speed. Anthropic CEO Dario Amodei, in a 20,000-word essay, warned explicitly that AI systems could leave less-skilled workers with "nowhere to go" — creating "an unemployed or very-low-wage underclass". His estimate: AI could eliminate up to 50% of entry-level white-collar jobs within five years, ascending from the bottom of the skill ladder to the top.[^55][^56][^57][^58]
What makes the AI-era underclass threat distinctive from previous waves of automation is the elimination of the retraining escape route. Historical pattern:[^58]
If AI displaces cognitive labor at scale, there is no obvious "next tier" to retrain into — especially given that 77% of new AI-related jobs require master's degrees, creating a skills barrier most displaced workers cannot clear.[^59]
==Philip Trammell and Dwarkesh Patel formalized the economic logic in their paper "Capital in the 22nd Century": if AI renders capital a true substitute for labor (not just a complement), then labor's share of income approaches zero. All returns flow to capital owners. Combined with Piketty's r > g dynamic, the result is permanent concentration of wealth in those who owned assets at the moment of AI transition. Without aggressive global capital taxation, "approximately everything will eventually belong to those who are wealthiest when the transition occurs, or their heirs".==[^60][^61]
Not all economists accept the AI permanent underclass as inevitable:
Baumol's Cost Disease reframing: Economist William Baumol observed in the 1960s that sectors resistant to automation (arts, education, human care) increase in relative economic value as productivity in automatable sectors rises. In an AI economy, the things AI cannot do — authentic human relationships, physical care, creative judgment, ethical reasoning — may become premium goods. The permanent underclass thesis assumes human labor becomes worthless; Baumol's insight suggests the opposite.[^62][^60]
Historical labor market resilience: Sam Altman and others cite the historical pattern of jobs evolving rather than disappearing: every 75 years, roughly half the economy's jobs change dramatically, yet employment persists. The question is whether this transition will happen faster than human adaptation can manage.[^63][^64]
Political response: The existence of democratic institutions and the scale of potential disruption creates strong political incentives for redistribution, UBI, or other compensatory mechanisms. Dario Amodei himself acknowledges that some form of income disbursement will likely emerge — pointing to China's dibao system as a precedent for large-scale unconditional cash transfer infrastructure.[^65]
The empirical picture is mixed but increasingly concerning:
One of the most important lessons from the history of the underclass concept is that naming creates reality. Loïc Wacquant's meticulous analysis shows how the same social phenomenon was described in structurally sympathetic terms by Myrdal, then reframed in morally condemnatory terms by journalists and conservative thinkers, with enormous consequences for public policy. The choice between calling someone a member of an "underclass" (emphasizing permanence and social separation) versus "someone experiencing poverty" (emphasizing temporariness and shared humanity) is never politically neutral.[^68][^12]
The same dynamic is playing out in the AI era. Calling displaced workers a "permanent underclass" naturalizes their exclusion — it suggests their position is a fixed feature of the social landscape rather than a policy choice that can be made differently. The meme-ification of the concept on social media has, paradoxically, both raised awareness and normalized the catastrophism it purports to critique.[^69]
The core intellectual divide that produced the underclass debate — structural causes vs. behavioral causes — was never really resolved. Empirically, the evidence favors structural explanations: deindustrialization, racial segregation, and lack of access to high-quality education and employment are better predictors of persistent poverty than cultural values or welfare dependency. But structural explanations produce policy implications that are harder to sell politically: you cannot fix deindustrialization with a speech about personal responsibility.[^26][^28]
The behavioral explanation persists in public discourse not because it is better supported by evidence but because it is politically more convenient — it places responsibility on the poor rather than on the economy, the housing market, or the racial caste system. This is what Herbert Gans meant when he argued the "underclass" label serves ideological functions: it legitimizes inequality by making it appear deserved.[^30][^31][^32]
The contemporary AI underclass debate has one feature that distinguishes it from all prior underclass debates: it is being articulated primarily by the people creating the technology that will do the displacing. Dario Amodei is not a conservative ideologue justifying welfare cuts; he is the CEO of an AI company warning about what his own product will do. This gives the structural argument unusual credibility across political lines.[^55]
If the AI permanent underclass materializes, it will be the first version of the concept that is genuinely technology-determined rather than policy-determined. Previous underclasses were created by the interaction of economic change and deliberate policy choices (welfare design, racial segregation, urban disinvestment). An AI underclass, in the extreme scenario, would be the product of a technological transition so rapid that policy has no time to compensate — a structural change so complete it forecloses the usual adaptation pathways.[^65][^58]
==The permanent underclass is not a single theory but a contested intellectual space where economics, sociology, political ideology, and moral philosophy collide. Its 160-year history moves from Victorian panic about the "social residuum" through Marx's lumpenproletariat, Myrdal's structural warning, Lewis's culture of poverty, Murray's welfare critique, Wilson's deindustrialization thesis, and into Silicon Valley's 2025 fear of AI-driven permanent exclusion. Each iteration reflects the anxieties and ideological commitments of its moment more than the clean findings of social science.==
What is consistent across all versions is a genuine observation: in any sufficiently unequal society, some people fall so far below the mainstream that normal mobility mechanisms stop working for them. Whether the cause is cultural, structural, racial, technological, or some combination, the effect — intergenerational poverty and social exclusion — is real and measurable. The debate is not about whether such a group exists but about why it exists and therefore what to do about it.
==The most intellectually honest answer is that it is caused by both structure and culture, but structure precedes culture: people adapt their values and behaviors to the economic conditions they face. Change the conditions — provide stable employment, desegregate housing and schools, invest in social infrastructure — and the cultural adaptations change with them. The permanent underclass is permanent only if the structures that produce it are left permanently in place.==
Marx and Enge...
Segmented assimilation theory, developed by sociologists Alejandro Portes and Min Zhou, presents ...