Dreams and Nightmares of the World’s Middle Classes

Göran Therborn

The world has been getting contradictory messages about its class structure. According to one authoritative account, it has reached a ‘global tipping point’—‘half the world is now middle class or wealthier’. This was based on figures marshalled by Homi Kharas, a former World Bank chief economist now at Brookings. More excitably, the Economist has hailed the ‘relentless rise’ of a ‘burgeoning bourgeoisie’ and trumpeted the arrival of a middle-class world. Yet serious scholarship also assures us of the opposite: according to Peter Temin, emeritus professor of economics at mit, we should be concerned about ‘the vanishing middle class’.footnote¹ Readers could be forgiven for feeling bewildered. What is going on in economics—and in the economic sociology of the real world? This contribution will examine the varying definitions of ‘middle class’ in play and the contrasting trajectories analysed by development economists, sociologists and financial journalists across the different sectors of the world economy. It will go on to outline a rather different future for the world’s middle classes than either of the extremes suggested here. But first, a few historical and conceptual considerations may be in order, for the concept of the ‘middle class’ has long given rise to debate.

The term ‘middle class’ entered the English language two centuries ago—‘sometime between 1790 and 1830’, according to Eric Hobsbawm—as a rising industrial society overtook the ‘military’ order of monarchy and aristocracy.footnote² The nineteenth century saw intensive discussion over where this new society was headed and the place of the middle class within it. The liberal argument was that the task of government should, and would, fall to the middle class, ‘the most wise and the most virtuous part of the community’, as James Mill put it.footnote³ Had this already been accomplished? For Tocqueville, writing in 1855, the reign of the middle class had been realized not only in the United States but also in France, where the July Revolution of 1830 marked its ‘definitive’ and ‘complete’ triumph.footnote⁴ Would the emergent middle-class society lead on to a new and stable political order? In the later decades of the nineteenth century, this was increasingly questioned. Novel ‘isms’ appeared: mobilizing ideas, first and foremost socialism, which theorized ‘middle-class society’ as capitalism, doomed to be overthrown by the expanded ranks of the industrial working class.

Middle class and bourgeoisie

Most strikingly, nineteenth-century discussions featured a conceptual variety notably absent from current treatments of the ‘middle class’. This derived from the flourishing of a number of national languages, each expressing a particular history of class formation and conflict. In Western Europe, there were three major concepts circling around a similar social phenomenon, each from a different angle: the English ‘middle class’ was complemented by the German Bürgertum and French bourgeoisie.footnote⁵ Both originated in medieval urban law, denoting a category of urban residents with special civic and political rights. After the French Revolution, the ‘bourgeoisie’ grew increasingly synonymous with both the English ‘middle class’ and the classe(s) moyenne(s). But it also took on two distinct connotations. One was culturally pejorative: as Flaubert put it, ‘Hatred of the bourgeois is the beginning of all virtue.’footnote⁶ Second, from the 1870s a clear distinction arose between the bourgeoisie and the ‘middle’ or ‘new’ social strata. The bourgeoisie were the big capital owners: bankers and industrialists, the new peak of the social pyramid—that is to say, the upper class.footnote⁷ The middle class—the German Mittelstand; the petite bourgeoisie or couches moyennes in French—was something different. In the Communist Manifesto Marx and Engels paid handsome tribute to the ‘revolutionary’ historical role played by the bourgeoisie, now seen as the embodiment of capital and the sworn enemy of the working class.

Another noteworthy difference: work was a crucial attribute and value of the nineteenth-century middle class, the thing that separated it from the rent-consuming nobility. ‘Work is the burgher’s ornament’, wrote Friedrich Schiller in a famous ballad. ‘Blessed is he who has found his work, let him ask no other blessedness’, added Thomas Carlyle in Past and Present.footnote⁸ In today’s discussions, the middle class is overwhelmingly defined in terms of consumption, or rather consumer capacity, measured in dollars (as corrected by international purchasing power parities); occasionally it is specified by some middling location on the national ladder of income distribution—but never by reference to its work. This is all the more remarkable, since contemporary American usage typically deploys the term as a euphemism for the working class.

What are the implications of this mutation of middle-class discourse, from work to consumption? The Economist’s enthusiastic hailing of another ‘two billion bourgeois’ offers a clue.footnote⁹ Like the entry of ‘capitalism’ into the vocabulary of business executives, it is a celebration of victory and power. So long as socialism was seen as a danger, terms like ‘capitalism’ and ‘the bourgeoisie’ were banished to the margins; the acceptable terms were ‘market economy’ and ‘business’. As we shall see, the discursive change connotes an important shift in social hegemony. But first we should examine the conditions that gave rise to the twenty-first century’s new thinking on the middle class.

Contra Mill and Tocqueville, the nineteenth century did not usher in a middle-class world, for the twentieth century was above all defined by the working class. Although social democracy and communism were born in Europe, working-class socialism became a world model, conspicuous in the Chinese and the Vietnamese revolutions, with their repercussions throughout East and Southeast Asia; in revolutionary Mexico and Fidelista Cuba; in the large progressive movements of Latin America—Peronist Argentina and Vargas’s Brazil, not to mention the pt of more recent times—and in anti-colonial struggles, from Nehru’s Congress via Arab socialism to the anc of South Africa. The labour movement was a major force in the achievement of universal suffrage and the welfare state. It was the main ally—although seldom an exemplary one—of the feminist and anti-imperialist movements. The middle classes were largely in hibernation through these twentieth-century periods of revolution and reform; they gained political salience at times of rising fascism and authoritarianism. But the motor force of working-class reform peaked in the years around 1980 and then rapidly declined.

The end of the working-class century had an economic basis in the accelerating deindustrialization and financialization of the capitalist core; more obliquely, a sociological factor was the social dissolution stemming from the 1968 cultural movement. Yet this did not immediately herald a new middle-class dawn. Western neoliberalism was allergic to any kind of class discourse, and Eastern European anti-Communists preferred to refer to themselves as ‘civil society’, although in power they would claim middle-class credentials.footnote¹⁰ If, as Hobsbawm thought, the idea of the middle class was born in the West, it was reborn in the East and the South.footnote¹¹ In the 1980s the middle class was ‘discovered’ in conservative East Asia, as an outcome of the rapid economic growth of the ‘four little dragons’: Taiwan, South Korea, Singapore, Hong Kong.footnote¹² The middle classes were emerging as a significant political force in the region, playing central roles in the broad popular movements that put an end to military dictatorships in Seoul and Taipei.

In China, the concept travelled a rockier path to acceptance. In the 1980s, Chinese academic interest in the middle class was in part inspired by the American neo-Marxism of Erik Olin Wright and his colleagues. After 1989, following Tiananmen, government orthodoxy hit back. A prominent line-following sociologist put it this way: socialist China could not allow a ‘middle class’ to appear, since this could ‘overturn our socialist system’. Whereas middle-class theory in the West ‘exists to cover up the issue of class conflict’, in socialist societies it ‘divides the proletariat, separating businesspeople and intellectuals from the proletariat, creating a subversive force.’ After a period of silence, however, the middle-class discussion reopened, and from 2001 the debate was won decisively by the argument that ‘in any society, the middle class is the most important force in maintaining social stability’—a cushion between the upper and the lower classes, the bearer of moderate and conservative ideologies, and the crux of a broad and stable consumer market.footnote¹³ To many Chinese scholars in the 2000s the middle class also became an egalitarian ideal, the key to an ‘olive-shaped’ social structure.footnote¹⁴Vietnam’s post-Communist conceptual change was encapsulated by the Deputy Prime Minister, Hoàng Trung Hải. ‘The young middle-class population will be the driving force in Asia’, Hải declared, referring to ‘the arrival of a billion more middle-class consumers’. Thirty years earlier, his predecessor would have referred to the working class as the ‘driving force’.footnote¹⁵

1. southern dreams

The new middle-class dream in the Global South had, first and foremost, an Asian setting. It was woven and promoted by figures in the World Bank’s orbit, seconded by business consultancies and investment bankers. It appeared at the start of the millennium, the belle époque of outsourced global capitalism. As noted above, the Eastern rediscovery of the middle class in the 1980s was made by sociologists, concerned with changing occupational structures and class formations, and interested in their social and political implications. The new triumphalism, by contrast, was almost exclusively about consumption. ‘Middle class’ meant anybody who had some money to spend. By the same token, it soon came to mean being non-poor, as defined by official national poverty lines.footnote¹⁶ The notion of a middle class that began just above the poorest 20 per cent of the population—who, in poor countries, tend to be very poor indeed—was bolstered in 2000 by an influential paper from William Easterly, a tough-minded Hayekian then at the World Bank. In ‘The Middle-Class Consensus and Economic Development’, Easterly argued that the inequality represented by the (low) income share of the three middle quintiles of the population—which he dubbed the ‘middle class’, without any supporting arguments—was a hindrance to development.footnote¹⁷ Middle-class expansion therefore became synonymous with the decline of poverty—a conceptual linkage that connected the concerns of development economists about poverty reduction with the interests of business consultants hunting for new markets.

Rising Asia

Indeed, it was us business consultants and bankers who first started to big-up the Asian middle-class dream. In 2007, McKinsey predicted that middle-class Indian consumers would grow from 50 million to 583 million by 2025. The following year, Goldman Sachs foresaw global inequality tumbling thanks to the ‘exploding world middle class’.footnote¹⁸ The Economist was tail-ending this boosterism with its claim of ‘two billion more bourgeois’. The first major quantitative overview of the ‘bulging’ middle class appeared in January 2009, authored by World Bank economist Martin Ravallion. It defined ‘middle class’ as living on $2–$13 a day, the upper line chosen as roughly equivalent to the 2005 us poverty line at 2005 purchasing power parities—in other words, the rising ‘bourgeois’ were economically equivalent to the American poor. On Ravallion’s reckoning, the global middle class had swelled by more than 800 million between 1990 and 2005.

A closer look, however, revealed that 622 million of these were in ‘developing East Asia’, which basically meant China. But if the $2–$13 Chinese middle class had ‘exploded’ from 15 to 62 per cent of the population, the changes in the other world regions were comparatively modest. In South Asia, those with $2–$13 to spend per day had risen from 17 to 26 per cent of the population; in Africa, from 23 to 26 per cent; in the Middle East and North Africa, from 76 to 79 per cent; in Latin America, from 63 to 66 per cent—though in each case, the new ‘bulge’ was mainly concentrated at ‘just above $2 a day’. Meanwhile in Eastern Europe and Central Asia, the middle class had actually declined slightly, from 76 to 73 per cent.footnote¹⁹

Seen through this narrow $2–$13 lens, the development of the Asian ‘middle class’ in the 1990s and 2000s was indeed impressive, eliciting an avalanche of congratulatory literature. The most significant contribution was a 2010 report, ‘The Rise of Asia’s Middle Class’, from the Asian Development Bank, an inter-state body based in Manila. Published in the depths of the recession caused by the 2008 North Atlantic financial crash, the adb report’s press release predicted that ‘developing Asia’s rapidly expanding middle class is likely to assume the traditional role of the us and Europe as primary global consumers and help rebalance the global economy.’ It claimed that Asian consumers would spend 43 per cent of worldwide consumption by 2030.footnote²⁰ The subtext: the Asian middle class will save the world, or at least the world capitalist economy. According to the adb, the middle class of ‘developing Asia’—that is, excluding Japan—had grown from 569 million to 1.9 billion between 1990 and 2008, or from 21 to 56 per cent of the population. This middle class was now defined as earning $2–$20 a day, the upper limit set at about the poverty line in Italy. The poor, those below $2 a day, had correspondingly shrunk from 79 to 43 per cent. Most of this change was concentrated in China—but not all; the Indian middle class had expanded from 29 to 38 per cent between 1993 and 2005. The figures were based on surveys, and not very robust. Calculations from the national accounts gave a somewhat different picture but with the same impressive tendency, the poor decreasing from 69 to 17 per cent across ‘developing Asia’, and the ‘middle class’ rising from 31 to 82 per cent.footnote²¹Fascination with the Asian middle class has not led to any agreement on the actual size of the animal. A Chinese stock-taking some years ago found scholarly estimates of the prc’s middle class that ranged from 4 to 33 per cent, far short of the adb’s boosterish estimate of 89 per cent.footnote²² The Indian ‘middle class’ may comprise anything between 10 and 64 per cent of the population. A recent study found that 50 per cent of the population fell within the $2–$10 bracket in 2011–12; interestingly, it puts the ‘rise’ a decade later than the adb, not in the 1990s but in the period 2004–12. The lion’s share of this was attributable to people moving from less than $2 a day to between $2 and $6 a day.footnote²³ In fact, the most interesting discussion in Asia about the new middle class is probably happening in India, where a heterogeneous intellectual public is debating not just its size and growth but its socio-political meaning, in relation to a national political project of ‘changing India’. For Leela Fernandes, it represents ‘the political construction of a social group that operates as a proponent of economic liberalization’. For Dipankar Gupta, on the other hand, the term ‘middle class’ looks ‘sickly’ in India precisely because there is no project attached to it: instead, ‘we are besotted with consumption statistics.’footnote²⁴

The question posed by the 2010 adb report—whether Asia’s new middle-class consumers can compensate for falling growth in us middle-class spending power—was also addressed by the World Bank/Brookings economist Homi Kharas. Using a more intercontinental definition of middle-class consumption, $10–$100 a day, Kharas dispensed with descriptions of the class related to democracy, entrepreneurship or ‘contributing to human capital and saving’—that is, the entire gamut of traditional middle-class characteristics—insisting instead that ‘what makes the middle class special focuses on consumption’. He predicted a rise in the global middle class from 1.8 billion in 2009 to 4.9 billion by 2030, as the world economy pivoted towards Asia, which was expected to account for 85 per cent of the increase, driven primarily by the Indian middle classes (which ‘could overtake China by 2020’) and by China.footnote²⁵ The rise of a huge Asian consumer market is clearly part of an ongoing shift of the global economy. But ‘middle class’ and ‘poverty’ are not blank signifiers which can arbitrarily be put to any use. The Tokyo-based economist John West has drawn attention to the distortions created by the conscious or sub-conscious transference of historical Western connotations of ‘middle classness’ to contemporary Eastern nations, giving rise to what he calls ‘Asia’s mythical middle-class society’.footnote²⁶

African hopes

The African Development Bank followed its Asian counterpart in 2011 with an upbeat report on the Dynamics of the Middle Class in Africa: ‘the middle class is widely acknowledged to be Africa’s future’. Since it was ‘associated with better governance, economic growth and poverty reduction’, fostering its development ‘should be of primary interest to policy makers’.footnote²⁷ Using the same $2–$10 definition, the report claimed that the African middle class (including North Africa) had increased to 34 per cent of the continent’s population by 2010, after stagnating around 28 per cent between 1980 and 2000. At 327 million, it was now ‘roughly the size of the middle class in India or China’ (a stretch, given that the adb had claimed a Chinese middle class of 845 million, more than 80 per cent of the total population of Africa in 2010). The ranks of this new African bourgeoisie, as the Economist put it, had grown by 122 million since 2000; of these, some 93 million lived on $2–$4 a day. The African Development Bank called this sub tier a ‘floating class’, vulnerable to slipping back into poverty. Another 23 million, with $4–$10 dollars a day, were members of the ‘lower middle class’. Finally, an ‘upper middle’ at the $10–$20 mark (the lowest rung of Kharas’s definition of the global middle class) had actually declined in terms of population share since 1980, falling from 15 to 13 per cent.

Other studies of the African middle class have been more sober. Henning Melber, editor of one of the best collections, notes with bafflement the impact on African studies of the consumptionist approach set out above, initiated by ‘a handful of economists’. But Melber also acknowledges the popular attraction of middle-class identity, drawing on a study of the Black township of Soweto in Johannesburg, where two-thirds of respondents considered themselves middle class in a population where 7 per cent worked in middle-class occupations, 25 per cent were wage workers, 23 per cent were unemployed, 21 per cent were casual labourers and the rest were pensioners or students.footnote²⁸ By around 2015, the hype about a new African middle class had dissipated. From London, the Financial Times reported that foreign corporations were downscaling on the continent for want of middle-class consumers. The Economist also took note, now describing the African middle class as ‘few and far between’. Both presented starkly reduced estimates of its size: 15 million across eleven of the continent’s larger national economies, reported the ft based on a survey by Standard Bank, or just 6 per cent of the population according to the Economist, citing the Pew Center.footnote²⁹

Latin American circumspection

Latin American interest in the middle class also intensified in 2010, but took a very different form. The hype was absent, and socio-economic and sociological perspectives were more prominent than $2 head counts. The lower level of excitement had its background in the hemisphere’s undistinguished ranking in middle-class growth tables: a rise of just 3 percentage points between 1990 and 2005 according to Ravallion, as cited above. From 2010, three major reports appeared. The Middle Class in Latin America, a sociological analysis combining an occupational class approach and income distribution, was published by cepal, the un Economic Commission for Latin America—a rough equivalent of the Asian Development Bank and an important actor in analytical and policy discussions in the region. Meanwhile a socio-economic assessment came from the Americas desk of the oecd Development Centre in its annual Latin American Economic Outlook 2011, subtitled ‘How Middle Class is Latin America?’. Three years later the World Bank put out a weighty contribution, Economic Mobility and the Rise of the Latin American Middle Class. These institutional responses used different class definitions and painted three different pictures of Latin America.

The cepal study proceeded from the question, ‘What are we talking about when we talk about the middle class?’, and drew a map of social stratification in which the ‘middle strata’—the term used in preference to ‘middle class’—could be located. The middle strata were defined in terms of occupation (white collar) and income (four times the urban poverty line). While the study identified a substantial growth in the size of these social layers, overwhelmingly due to an increase of the lower middle, its conclusions emphasized social heterogeneity and inter-country variation, illustrated by five concluding monographs on contrasted countries.footnote³⁰

The contribution from the oecd Development Centre was policy oriented and aimed to identify the conditions of middle-class support. The guiding assumption was that if these sectors had ‘stable employment and reasonably robust incomes’, they would provide ‘a solid foundation for economic progress’, while if they had ‘precarious incomes and unstable employment’, then ‘their political preferences may veer toward populist platforms not necessarily conducive to good economic management.’footnote³¹ These ‘middle sectors’ were defined by their very middleness: that is, households with an income of 50–150 per cent of median income, an unmotivated expansion of the more common 75–125 per cent range once proposed by the distinguished economist Lester Thurow. The effect was inflationary: the oecd’s ‘middle sectors’ contained more contract-less ‘informal’ workers than formal employees.footnote³² The report also provided a comparison of Italian and Latin American trends. While the ‘middle sectors’ comprise over 60 per cent of Italy’s population, in Uruguay and Mexico they make up about 50 per cent, in Chile and Brazil around 45 per cent, in Argentina 40 per cent, and in Colombia and Bolivia just over a third. With regard to policy prospects, the oecd report ends on a cautiously optimistic note about the middle sectors and the potential for positive changes in income distribution, social protection and opportunity creation.

The World Bank product on Latin America was an elaborate investigation with two main themes, income mobility and the rising volume of the middle class. It starts with a fanfare: the continent is ‘a middle-income region on the way to becoming a middle-class region’. But the music becomes more subdued along the way: the region is not yet a ‘middle-class society’ where ‘most people earn a sufficiently high income to consume, live and behave [sic] like middle-class citizens’. In fact, ‘vulnerability to poverty remains a serious concern for the majority, and social policies will continue to play an important role for the foreseeable future.’ Nevertheless, the World Bank predicts a great future for the Latin American middle class: by 2030 it will have grown from a mere 30 per cent to a convincing 40 per cent of the continent’s population.footnote³³Drawing on the ‘vulnerability approach’ proposed by two World Bank colleagues,footnote³⁴ the report launched yet another definition of the ‘middle class’, now based on economic security—once again ignoring the historical connotations of the term. On this basis, the middle class are those with a less than 10 per cent probability of falling into poverty within five years. In some Latin American countries this would translate into a household per capita income of $10 a day; in others, it would not. The authors pragmatically settle for $10 a day as their lower bound. Without providing any substantiating rationale, they also add an upper bound of $50 a day. On this basis, they declared that the Latin American middle class had doubled between 1992 and 2009, rising from 15.5 per cent of the population to nearly 30 per cent.footnote³⁵In sum, the highest-flying Southern middle-class dreams have been Asian, concentrated in China and India, although they have encompassed all of ‘developing Asia’ apart from the western war zones. In this twenty-first-century imaginary, the Global South is surfing an ever-rising middle-class wave, the most important social change of the age. At its most far-reaching, the middle-class dream was linked to a shift in the centre of global economic gravity, from North America and Europe to Asia. If there was no consensus on the shape and content of this class, nor on the pace of its growth, it was widely agreed that it would mean more money and more consumption. The future understandably looks more modest viewed from Abidjan or Santiago de Chile, the headquarters of the African Development Bank and cepal respectively. In both Africa and Latin America, the middle-class dream has more often had some contact with the realities of social structure. But the Southern dream persists; the latest prediction from Homi Kharas suggests that by 2030 the ‘middle class’ will be dominant, with 63 per cent of the world’s population.footnote³⁶

2. northern nightmares

While the middle classes were touted as ‘rising’, ‘expanding’ and ‘exploding’ in the South, they were found to be shrinking in the North. ‘We have observed’, concluded leading inequality scholars Anthony Atkinson and Andrea Brandolini, ‘a downsizing of the middle class from the mid-1980s to the mid-2000s’. In a study of fifteen oecd countries, the ‘middle 60 per cent’ lost income shares ‘to the benefit of the richest fifth’ in all except Denmark, and in ten countries the middle class actually shrank.footnote³⁷ In 2011, Francis Fukuyama wondered aloud: ‘But what if the further development of technology and globalization undermines the middle class and makes it impossible for more than a minority of citizens in an advanced society to achieve middle-class status?’ Indeed, ‘there are already abundant signs that such a phase of development has begun.’ Fukuyama then raised an even bigger fear: ‘Can liberal democracy survive the decline of the middle class?’footnote³⁸The oecd has been worried about increasing inequality in the rich countries since its 2008 Growing Unequal? report, but it took a further decade before it focused on the difficulties of the middle class. In 2018 it put out an overview of darkening middle-class views on social mobility, socio-economic status in comparison with their parents, and future prospects.footnote³⁹ There followed in 2019 a wider-ranging study, Under Pressure: The Squeezed Middle Class—no question mark added—using a band of 75–200 per cent of median disposable income as its definition of the middle class. The size of this population in oecd—i.e., rich-world—countries shrank on average from 64 to 61 per cent between the mid-1980s and the mid-2010s. Its distance from the rich has widened, the income of the wealthiest 10 per cent increasing by a third more than that of the middle class. In addition, middle-class income share has fallen more than its population share, dropping 5 percentage points; the upshot has been rising debt—20 per cent of middle-class households now spend more than they earn. Sweden is outstanding in squeezing its middle class: population share has fallen by 7 percentage points, income share by 11 points. The corresponding figures for the us are 4 and 9 points; the British middle class has stayed put as a proportion of the population while losing 5 points of its income share.footnote⁴⁰The only positive development in the North has been the increased entry of the over-65s into the ranks of the middle-income class, except in the us. For the rest, Under Pressure paints a grim picture, concluding that ‘many middle-class households consider our socio-economic system unfair’ because they haven’t benefited from it as much as high-income groups. Furthermore, a ‘middle-class lifestyle is increasingly expensive, notably when it comes to housing, a good education and health care’. Labour-market prospects for many in the middle class are uncertain: one in six middle-income workers are in jobs ‘at high risk of automation’. Under Pressure is not apocalyptic, in contrast to the stream of national lamentations we shall look at below, but it tersely comments that, for many, ‘the middle-class dream is increasingly only a dream’.footnote⁴¹ What has gone wrong?

The Northern middle-class downturn started in the us in the late 1970s. It was brought to public notice through the work of a few sharp-eyed observers in the course of the mid-80s, though their findings were at first denied by mainstream opinion leaders. In 1986 a Federal Reserve economist, Katherine Bradbury, published a paper on ‘The Shrinking Middle Class’, which found a decline of 5 percentage points in the proportion of families with incomes of $20,000–$50,000 during 1973–84, of which 4 points were due to downward mobility.footnote⁴² In their excellent book, The Great U-Turn, Bennett Harrison and Barry Bluestone situated this fall in the context of historical developments within American capitalism: falling profits due to foreign competition leading to deindustrialization, corporate restructuring and financialization, hollowing out and polarizing the us labour market. ‘Does all this portend an end to the middle class in America?’, the authors asked. But these were pre-apocalyptic times, and they replied in the negative: ‘the middle class in America is resilient. Workers struggle to maintain their wages against the force of deindustrialization’.footnote⁴³

Following the crash of 2008, even the official tone darkened. A White House task force set up by the Obama Administration to study the problem used bland and cautious language, defining ‘middle class’ with an ideological emphasis on ‘aspirations’—to home ownership; college education for their children; health and retirement security; family vacations. The principal finding was that, with the cost of health care, college and housing rising faster than incomes, it had become ‘more difficult’ now for many Americans ‘to achieve middle-class status’.footnote⁴⁴ As the decade progressed, the tone became more apocalyptic. In 2017, mit economist Peter Temin marshalled evidence to show that the American middle class—now defined as those with 67–200 per cent of us median income—was vanishing; its income share had dropped from 63 per cent in 1970 to 43 per cent in 2014. This hollowing middle was leaving the us a ‘dual economy’, in the sense of Arthur Lewis’s analysis of Third World capitalism, with an fte sector (finance, technology, electronics) comprising about 20 per cent of the population and setting the rules for the economy, while a low-wage sector harboured the remaining 80 per cent.footnote⁴⁵

The implications of capitalism’s new turn for the European middle classes were discovered relatively late.footnote⁴⁶ Only in the past decade have middle-class nightmares come to haunt European writers, following the financial crash of 2008. In the uk, the director of a ‘radical centre’ think-tank anguished in Broke: Who Killed the Middle Classes? about their ‘impoverishment’ and ‘corrosion’, warning that ‘the destruction wreaked on the working classes is now in store for them’—and wondering if Marx would belatedly be proved right: capitalism might yet culminate in the proletarianization of the middle classes. In Germany, journalist Daniel Goffart announced ‘the end of the middle stratum’—here, those on 70–150 per cent of median income—which had declined from 48 to 41 per cent of the population between 1991 and 2015, with a further threat to employment looming from digitalization. In France, the social geographer Christophe Guilluy announced ‘the end of the Western middle class’, a concept which for him was ‘above all cultural’, its disappearance measured ‘by the loss of a status’ embodying the European or American way of life; the ‘popular categories, workers and employees’ had now been downgraded ‘from desirables to deplorables’.footnote⁴⁷

3. the middle classes by daylight

In this brief survey of the recent literature we have met a bewildering variety of groupings labelled ‘middle class’. Clearly, as Ferreira and his World Bank colleagues observe in their study of Latin America, ‘defining the middle class is no trivial matter’.footnote⁴⁸ While definitions as such are neither correct nor incorrect, they may be illuminating or obfuscating, in line with historical usage or arbitrarily idiosyncratic; when clothed in everyday language, novel definitions may have misleading connotations. In other words, the concepts underlying these dreams and nightmares need to be scrutinized by daylight. The Southern middle-class dreamworld is premised on a zero-sum linkage between the middle class and poverty; the rise of one is the flipside of the other’s decline. As we have noted, this is literally a bankers’ and business consultants’ outlook—Goldman Sachs, McKinsey, the corporate-oriented development banks, the World Bank—and it frames the world in a very peculiar way: wide-angled at times, it is only capable of seeing a world of trade and consumption; there are no producers, no working class and no social relations.

In bankers’ eyes, ‘middle class’ and ‘poverty’ are defined solely by the dollar sign. The relational and the relative have been absolutized and turned in upon themselves. Middle class is an intrinsically relative concept, denoting a stratum between at least two others. Poverty refers to having fewer resources in relation to relevant others, as indicated by the fact that rich and poor countries draw different poverty lines. In this sense, poverty is also relative. This discourse has an incentive, economic and political, to drain these terms of sociological meaning, as this allows one to be inflated and the other diminished. Yet using an everyday-language concept like ‘middle class’ with an idiosyncratic technical definition can mislead the reader, and to treat such a historically established and socio-politically charged concept in this way is either unwise or dishonest. The frequent counts of the Southern ‘middle class’—or, more cautiously, ‘middle sectors’—as those on $2–$4 a day include street vendors, day labourers and other contract-less and rights-less workers. Sixty per cent of the oecd’s Latin American ‘middle sectors’ are in the informal economy.footnote⁴⁹ It apparently requires an exceptional mind to discern workers in the middle-class haze of the Southern dreamworld.

In 2008, a study by Abhijit Banerjee and Esther Duflo posed the question, ‘What is middle class about the world’s middle classes?’ Looking at households with a per capita expenditure of $2–$10 and drawing on extensive Third World research, they found that ‘while there are many petty entrepreneurs among the middle class, most of them do not seem to be capitalists in waiting. They run businesses, but for the most part only because they are still relatively poor and every little bit helps.’ Why is this important? ‘It leads us to the idea of a “good job”’—an idea that economists have often resisted, ‘on the grounds that good jobs may be expensive jobs.’ But, Banerjee and Duflo conclude, ‘nothing seems more middle class than the fact of having a steady well-paying job’.footnote⁵⁰The logic of the Southern dreamworld suggests that the expansion of the ‘middle classes’ means that poverty is about to disappear, just as it supposedly has in large parts of the North. According to the World Bank, poverty in Europe—measured as less than $3.20 a day—is non-existent: at this level, the poverty rate stands at zero in France, Germany and the uk; in Sweden, it is a mere 1 per cent. By contrast, looking at broader indicators, the Eurostat economists more convincingly see 22 per cent of the eu population ‘at risk of poverty and social exclusion’.footnote⁵¹ For poverty is a social concept, not a biological one, nor a quantity of money below a certain line. As such it is intrinsically relational, referring to a disposal of resources below the median, regardless of whether it is defined as ‘absolute’—below some monetary level—or as ‘relative’, below some percentage of the population.

If the Southern rise of the middle class looks less rosy in the clear light of day, its Northern apocalypse looks less calamitous. Since the mid-80s, the oecd middle classes—defined at 75–200 per cent of median income—have on average declined from 64 to 61 per cent of the population, while their share of national income has decreased by 5 percentage points. Sweden and the us were the epicentres of decline, with middle-class income shares falling by 11 and 9 percentage points, respectively—although despite this, it’s interesting that Sweden has not yet been a subject for ‘middle-class nightmare’ discourse. In France, Ireland and Denmark, on the other hand, the economic middle has (slightly) increased in size over this period.footnote⁵² The problems encountered by Northern youth and young adults in accessing higher education and accommodation are real enough, in rich-world countries with high university fees and marketized housing. But the literature concentrating on the middle class fails to see the systematic inequalities produced by contemporary post-industrial capitalism. Its discourse is the nightmare of a class attempting to isolate itself from these dynamics. But in what directions do they tend?

4. converging roads to inequality

To paraphrase Oscar Wilde on England and America, we may say that the Global South and the Global North are divided by a common class. However, the evidence suggests these ‘middle classes’ are converging on the highway of twenty-first-century capitalist inequality. The Southerners are coming from poverty and the Northerners from relative comfort, but it seems likely that they are going to find each other, struggling and striving, abandoned by an ever-richer bourgeoisie, and with uncertain relations to the popular classes of workers, precariat and unemployed. And though they are nationally divided, they live under the same climate cloud (and face similar viral risks). Certain trends are apparent, even if we restrict our examination to income share.

Since the upper limit for the Southern middle class in Ravallion’s World Bank overview was the us poverty line, the recent trajectory and social horizons of the us poor may well indicate something about the future of the ‘rising’ Southern middle classes. The poorest Americans, around 20 per cent of the us population, are roughly equivalent to the Southern ‘middle classes’.footnote⁵³ As Table 1 shows, their experience since 1980 has been of falling further behind. What Thomas Piketty’s team at the World Inequality Lab call the ‘middle forty per cent’—the core and upper ranks of the American middle class—have also lost ground to the rich, or what may properly be called the bourgeoisie. The us development is extreme, but not sui generis. Between 1985 and 2017, the British ‘middle forty’ lost 4 percentage points of its income share, while the top ten increased its share by 5 points. In Germany, the top ten appropriated 8 extra percentage points of national income, while the ‘middle’ lost 1 point, and in France the ‘middle’ lost 2 points and the top ten gained 3 points.footnote⁵⁴

The Northern experience, then, suggests that the next stage up from poverty is the experience of widening inequality, which for those at the losing end represents another kind of poverty—the consciousness of having only meagre resources with which to get by in life—tacitly recognized as such by the ruling authorities of the North. Will the Southern middle classes meet a similar fate? It’s worth recalling that the Global North experienced a period of ‘inclusive growth’—that is, growth with decreasing inequality—in the 1945–80 period, the time of labour-movement influence. The Southern middle-class dreamers are deliberately erasing the memory of those times, but the question should be asked: is any equalization in sight in the South? A thorough answer would require another essay. As Table 2 shows, however, ongoing distributive tendencies in China and India point to a non-lagged convergence on the road of mounting inequality. In other words, yesterday’s Southern dreams are likely to turn into nightmares similar to those of the North.

In China and India, heartlands of the ‘rising middle class’, even the ‘middle forty’ is losing ground: the income growth rate for the bottom 50 per cent of the population is less than half that of the whole population. In India growth for the ‘middle forty’ has been just half of the national average. Neoliberal India has become the us of the Global South—and, like the us, has a clear historical U-curve of economic inequality. The income share of the Indian top one per cent is back to its level in the colonial 1930s.footnote⁵⁵ The virtual exclusion of the bottom half of the us population from sharing the proceeds of economic growth over the past thirty years tells us something important about capitalist democracy.

The probable future in store for the Southern middle classes under the current world system is highlighted in Table 3.footnote⁵⁶ It should be kept in mind that the Northern figures refer to disposable income after taxes and transfers: in other words, they include the remaining, if eroded, effects of the 1945–80 equalization, a period such as the South has not yet entered, and which under present conditions perhaps never will.

Developments in China and India are crucial, but they cannot be assumed to hold for the whole of the South. Empirical data are still lacking for many big countries in Asia and Africa, but the figures that do exist indicate some diversity. In Brazil, under the pt governments, the income of the bottom half of the population grew faster than that of the nation; but in absolute terms, the top 10 per cent captured 58 per cent of total income growth, and the bottom half, 16 per cent.footnote⁵⁷ Inequality has been rampant in post-apartheid South Africa, with the bottom half and the upper-middle class (the 50th to 90th percentiles) each losing about 10 percentage points of national income share, to the benefit of the top decile. In Nigeria, the bottom 90 per cent are also losing out heavily to the richest 10 per cent. In Turkey, Thailand and Malaysia, on the other hand, some economic equalization has occurred. Egyptian income distribution has changed less over the past three decades, according to the World Inequality Database, but there has been a growing concentration of income at the top.footnote⁵⁸Most important for future trends: nowhere in the South is there any evidence of a sustained egalitarian drive. There was such a tendency in Latin America in the first decade of the century, but it has been halted—by right-wing politics, above all, but also recently in Mexico by covid-19.footnote⁵⁹ New rounds in the battle between equality and privilege lie around the corner, in Argentina and Chile. But for now, growing inequality remains the most likely outcome.

5. a political outlook

Those who hope that the rising middle classes will bring about a good society—given what the oecd calls ‘their intolerance of corruption, and their trust in others’—need to take account of the ambitious young Modi supporters described by Snigdha Poonam in Dreamers: the middle-class aspirants who run sophisticated online and call-centre scams from small-town India, churning out clickbait, selling fake jobs and diplomas or extorting money from elderly Americans by threatening them with the Internal Revenue Service.footnote⁶⁰The middle-class discourses should rather be read symptomatically, as an expression of larger processes of development. In the North, the key point is that the prevailing middle-class literature is essentially a critique, though often oblique, of the ongoing increase of inequality. It is not a narrative of a middle class threatened from below by trade unions or state handouts to the poor. It is about a class being abandoned, left behind by a previously admired economic leadership and lifestyle model. In other words, it is an objectively progressive discourse, despite its occasional apocalyptic self-pity. It may indicate a sizeable potential base for progressive taxation. As the oecd’s Under Pressure shows, the ‘squeeze’ on the Northern middle class mainly affects the young and young middle-age generation born after 1975–80.footnote⁶¹ This was the generation that rallied behind the surprisingly successful campaigns of Corbyn and Sanders.

The field of work is another meeting-ground for the left, the labour movement and the salaried middle class. There is a growing contradiction between, on the one hand, the middle-class professionalism of teachers, health personnel, public-service employees and civil servants, and on the other, the increasingly invasive capitalist-managerial notion of labour for profit. The latter is, and should be, an affront to every true professional, who takes pride in learnt expertise and pleasure in the intrinsic value of her work. The incipient digital revolution is likely to hit the professions hard, as well as the mass of white-collar employees. A widespread middle-class environmentalism is already clashing with the accumulation drive of real-estate developers, ruthless extraction corporations and producers of pollution. The Southern growth of the ‘middle class’, however defined, is part of a rapid, large-scale social change which will never create an industry-centred—and thereby a socio-politically industrial working-class centred—society similar to that previously found in the North. Industrial and manufacturing employment has already started to decline in Asia and Latin America, and is most unlikely to grow beyond current Asian levels in Africa.footnote⁶² The social structuring of forces for equality and social justice will be different this time around.

It is already clear that the covid-19 pandemic is a Great Unequalizer, virally as well as economically, with ferocious discrimination within as well as between classes, between men and women, generations and ethnic groups. What this means for the dreams and nightmares we have been examining is an accelerated convergence of the Northern and the Southern middle classes on the bleak path of inequality. Their common abandonment by big digital capital, headed by Amazon and Microsoft, has been magnified many times over. Most of the Northern middle class of small business and freelance ‘entrepreneurs’ have been economic losers in the Coronavirus crisis. Even more so are the Southern informal workers on $2–$6 dollars a day, supposedly incorporated into the middle-class conglomerate, but now likely to fall back into raw poverty. The World Bank and cepal have already signalled that the much-touted decline of extreme poverty in the world will be reversed.footnote⁶³ By contrast, in North and South alike, upper middle-class managers, bureaucrats and professionals have had a relatively good crisis, keeping their salaries and safely working from home.

The 2020 pandemic has therefore divided the middle class, while the gap between its upper ranks and the real bourgeoisie is widening further, due to the billions of dollars of pandemic ‘stimulus’ snaffled by the latter.footnote⁶⁴ Middle-class aspirations are being thwarted by a surge in youth unemployment, in both North and South. The ‘forward march’ of the Southern middle class, by whatever definition, has halted. Northern nightmares, on the other hand, are likely to continue. The frenetic preoccupation with consumption in mainstream middle-class discourse might appear frivolous in the shadow of the Coronavirus, and under the darkening clouds of climate change.

Further important questions—the processes of contemporary middle-class formation, social development and political potential—lie beyond the scope of this paper. For now, what conclusions may be drawn? First, the world can only be understood through its differences and inequalities, taking a 360-degree perspective. Failing that, the world looks very different depending on one’s vantage point; a vista from the North may look upside down from the South, and vice versa. Second, the middle class has a discursive centrality in the early twenty-first century, corresponding to that of the working class a century before. It should be read symptomatically, as an indicator of profound social change, as well as critically, as an ideology of consumer capitalism. Third, prevailing middle-class discourse is deeply—if not always deliberately—ideological, inflating out of all proportion a nebulous entity with strong political connotations—the middle class—and portraying a world of consumers without producers. Fourth, this discourse is also deceptive in absolutizing both the middle class and poverty. Poverty is always relative, the losing end of the prevailing level of unequal resource distribution; and the middle has to be in the middle of something. Finally, the emergent middle classes of the South are heading into the maelstrom of capitalist inequality, where they look set to converge with the hard-hit middle classes of the North. The covid-19 pandemic is currently shattering the middle-class dream in the South and accelerating the inegalitarian tendencies analysed above. Where this will lead is still an open question.

Back to issue