revolutionary reflections | Which side are you on? Work, class and the 99%

Discussions of Brexit and the general election have highlighted confusion about the nature of class and what we mean by working class or middle class. In an article based on a talk at Historical Materialism London in November, Bob Carter argues that clarity, rather than vague notions of ‘the many’ or ‘the 99%’, is vital to understanding our world and fighting to change it. A focus on exploitative relationships in what Marx called the ‘hidden abode of production’ is far more illuminating than arbitrary hierarchies of inequality. Carter illustrates the approach by applying ideas from Marx and Carchedi to the restructuring of workplace relationships for school teachers and others.

Screenshot from Man with a Movie Camera (1929)

After over 30 years of decline in trade union membership and density, a weakening of left-wing and radical organisations, the practical collapse of many social democratic parties and the rise of right-wing authoritarianism, there are welcome signs of a revival of discontent, radical political regeneration, and frustration and anger against austerity. These latter developments are evident in the support for Corbyn in Britain and Sanders in the USA. The developments, however, remain partial and uneven, with few signs of a revival in trade union militancy and increased workplace organisation. Notwithstanding the achievements of US teachers and the agility and invention of some small independent unions in Britain, the largely unsuccessful ‘business as usual’ approaches of most unions continues.

In this context, the ready and widespread adoption of vague notions of class expressed within phrases such as ‘for the many, not the few’, and ‘we are the 99%’, should cause Marxists some unease because they have consequences for class analysis and consciousness. Jane McAlevey[i], when talking of union organising, recounts practising semantics every day with organisers because how you talk about the union affects people’s expectations and practice. Should not the same be said for political organisation?

The central basis of the appeal of these vague slogans is a feeling of unfairness underpinned by economic inequality. The slogans are unashamedly populist, rejecting any clear definitions or complexities. The shift to concern with inequality is to be welcomed but it reflects a turning away from the idea of a working class organised in the workplace. As a basis for recognising class divisions and the dynamic of capitalism, the focus is woefully lacking. We need to go beyond sentiment and the arbitrary dividing lines and cut-off points that characterise both econometric and sociological conceptions of class divisions.

(Nearly) All in it together?

Econometric approaches to class were boosted by the reception given to Piketty’s Capital in the Twenty-First Century[ii]. As Husson’s review[iii] states:

it would be no exaggeration to say that the ‘Piketty group’ (including people such as Anthony Atkinson and Emmanuel Saez) has supplied a significant part of the arguments raised by recent social movements (the indignados, Occupy Wall Street, and such like) and even one of their watchwords: ‘We are the 99 percent!’

Husson argues that this approach is not without its problems: ‘Capital as a social relation is forgotten and the history of capitalism appears as an accounting mechanism’. Roberts[iv] makes the related point that by collapsing wealth and capital into the same category, Piketty fails to understand the dynamics of capitalism as a system.

Dorling’s account[v], which parallels Picketty’s, draws a particular conclusion: ‘statistics and evidence of a recent contraction of inequality within the 99 per cent’ show that ‘as the very richest become richer, the rest of us are becoming more equal’. Some 99 per cent of us are, in his words, increasingly ‘all in it together’. The analysis is based on the concentration of the economic gains from growth being captured by an extreme minority and so appears comfortable with the vast inequalities between the very poorest and the top twenty percent.

This failure to acknowledge significant social division extends to largely ignoring the nature of workplace social relations. Dorling claims, for instance, that headteachers are becoming ‘increasingly more like the majority they serve rather than the minority who are now much richer than them’. This judgement would be a strange one for many teachers given their day-to-day experiences of intolerable workplaces. Local Management of Schools and the Workforce Remodelling Agreement in England and Wales have undermined local authority control, fragmented cooperative relations between schools and encouraged the establishment of managerialism within schools, resulting in the conversion of layers of teachers into managers. The result has been strengthened Senior Management Teams, led by headteachers acting as accountable agents of state policy and who are liable to be dismissed if targets are not achieved. Under this regime, teachers are increasingly deskilled, bullied and stressed by the intensification of demands. Nor are these kinds of developments unique to teaching or restricted to the state sector. Beyond schools, Phil Taylor’s report for the Scottish TUC pointed to generally increasing work intensification, tighter supervision and harsher performance management. Any attempt at social transformation has to speak to the experience of workers and cannot ignore questions of management and control in every type of enterprise, be they shops, offices, schools or factories.

Marxism and Class

One would expect Marxist approaches to be particularly strong on class analysis and workplace relations. In contrast to social democratic perspectives that are largely restricted to seeking to constrain the free market, Marxists recognise the generation of inequality within the (still largely) ‘hidden abode of production’[vi]. Historically, therefore, orthodox Marxist perspectives emphasised exploitation, rather than economic inequality, concentrating on the ways in which surplus was pumped out of the labour under capitalism.

However, despite class being central to Marxism, class has presented problems for its theory. As the composition of the working class has changed, and particularly as advanced capitalist societies saw manual labour decline and non-manual labour expand, there has been pressure on Marxists to accommodate other approaches and to come to terms with sociology’s fascination with the middle class, which I discuss elsewhere[vii]. One response was to accept that white-collar and non-manual workers were somehow middle class – this being the standard sociological view since the 1950s. Marxist analysis was threatened by the seemingly inexorable growth of this group, making it attractive to revise Marx’s contention in the Communist Manifesto that ‘society as a whole is more and more splitting up into two great hostile camps, into two great classes directly facing each other’.

The late 1970s saw intense interest in the nature of class and a flurry of analyses identified a New Petty Bourgeoisie[viii], contradictory class locations[ix] and a Professional Managerial Class[x]. In the process of defining these groups, working class people were frequently implicitly defined and denigrated by representations that see them constituted through absences – they lack ownership, skills, education, organisational ability, or culture. While the debate has largely subsided, of these writers it is Wright that still has some contemporary purchase[xi].

Wright[xii] maintained that groups should not be forced into one class or another: ‘The concept of contradictory class relations within class relations . . . does not refer to the problem of pigeonholing people within an abstract typology’. For Wright, the polarisation of classes into the bourgeoisie and the proletariat was structured through three central processes – control over the physical means of production, control over labour power and control over investment and resource allocation. Capitalists controlled all three and workers were excluded from control. Control over some of the aspects and exclusion from others produced contradictory class locations. Nowhere in Wright was there an attempt to link class analysis with the production of surplus-value or surplus-labour, and workers achieved their status as workers only negatively, through the absence of dimensions of control, rather than through their positive performance of useful labour.

In Classes[xiii], Wright produced an auto-critique of his earlier work, concentrating particularly on the lack of sufficient weight given to relations of exploitation. Wright’s concept of exploitation, however, bore little resemblance to that of Marx. Exploitation was not based on the labour theory of value but on possession or non-possession of three exploitable assets – property, organizational position and skill. In all, Wright identified twelve class locations constructed from the amalgam of particular assets. In this model, it is possible for skilled workers to share in the exploitation of the unskilled and therefore not to be unambiguously working class. With a working class shorn of skilled and coordinating workers, it is not surprising that Wright[xiv] concludes that:

it is no longer axiomatic that the proletariat is unique, or perhaps universally the central, rival to the capitalist class for class power in capitalist society . . . there are other class forces that have the potential to pose an alternative to capitalism

These characterisations of workers as those denuded of skills and capacities reflected Weberian approaches to the sociology of class. They largely form the assumptions of social democracy, providing justification for social policies including state welfare systems that are essentially Fabian. For Marxists, however, they raise the question as to how, using these conceptions of class, will a socialism based on workers’ self-activity and emancipation ever to be possible? We need a theory of class that emphasises not the characteristics of individuals, but workers’ power and its potential further development. In other words, we must look to the workplace as the epi-centre of class analysis.

Guglielmo Carchedi and Class Theory

The contention here is that Carchedi’s work[xv] provided the most fruitful innovations in Marxist class theory. Following Marx, he maintained that under a pure model of capitalism there are two classes: capitalists and workers. Capitalists own the means of production (owners), do not produce value (non-labourers), appropriate surplus value (exploiters) of surplus labour (oppressors). Workers do not own the means of production (non-owners), produce surplus value (labourers) and have surplus value or surplus labour expropriated (exploited or oppressed). Capitalists in this model, although necessary to guarantee capitalist production, do not add use values and are not part of the labour process. Here Carchedi prefigures his later argument. Capitalists have real ownership of the means of production and, whether they operate in a productive or unproductive sphere of the economy, they are in an analogous position of exploiting or, where surplus labour rather than surplus value is produced, economically oppressing workers. This has become increasingly apparent as the managerial organisation and work processes of white-collar factories, such as government offices, schools, universities and call centres, come to resemble each other.

There are some wider clarifications and qualifications that should be noted here to avoid confusion. Carchedi’s main work on class was entitled On the Economic Identification of the New Middle Class. The fact that he specifies that it is an economic identification simultaneously indicates that there are other spheres of class determination. Indeed, he states ‘I will restrict my field of analysis to classes in relation to the economic structure. This limit should not be interpreted as an attempt to define classes in purely economic terms’. The centrality to capitalism of the social relations of production is neither a barrier to its wider influence on social organisation, including the family, nor does this approach limit the working class to those in paid work. Social reproduction both beyond the workplace and within (hospitals, education, and social welfare) are important arenas of contention and class struggle[xvi].

Moreover, that empirical reality is more complex than a simple two-class model is also recognised. Early capitalist production was characterised by employers gathering together traditional crafts, largely producing in the same ways under one roof. Marx termed this ‘the formal subsumption of labour’. In the course of capitalist development, products are increasingly the result of socialised, collective labour. As a result, the scope of productive labour expanded to create what Carchedi calls the collective worker. This Marx regarded as ‘the real subsumption of labour’. The prospect of continuing as a craft worker outside of direct employment was largely eradicated. Meanwhile, with the advent of the joint stock company, the individual capitalist withdrew from the productive sphere to be replaced by a managerial hierarchy. The old middle class of independent producers and providers of services increasingly became reconstituted as salaried managers and specialists, a new middle class, within capitalist organisations.

The work of management, as Marx noted, comprises two elements. Firstly, the supervision and control necessary to ensure the production of surplus value; Carchedi calls this the global function of capital. Secondly, the work of unity and coordination, necessary under any system of social production and hence part of the labour process. The varying configurations of these elements gives rise to complexity and to a new middle class that encompasses those carrying out a combination of control and coordination (as well as those who carry out only tasks of control but do not, in reality, own the means of production). Carchedi’s framework thus enables him to show the relationship between adding to use values, exploitation and the capitalist labour process. In the process, he also clarifies that unproductive labour is part of the collective worker equally engaged in a labour process with surplus labour extracted from them. Managers, while carrying out control and supervision, are not unproductive labourers but are non-productive and are paid from revenue rather than from variable capital.

Critics of Carchedi

The criticisms of Carchedi are not legion. Writers on class are far more likely to either simply dismiss him or ignore him altogether. Das[xvii], for instance, in a Marxist book of over 650 pages that has generally been warmly received[xviii], is weakened by having no pertinent references to Carchedi’s work on class. There were opportunities to do so: Das sees the centrality of exploitation to class formation. Despite noting with some justification that: ‘Unfortunately, whatever little work there is on the labour process aspect of class has fallen for the more fashionable culturalist concerns, almost banishing the element of material-political antagonism from the sphere of work’[xix], Das offers no systematic account of the capitalist labour process and its relationship to class.

Drawing on Marx, there is some recognition in Das’s book of the functions of capital, the despotic control of labour and the role of managers and overlookers, but general statements, that are at best ambiguous, obscure insights on the nature of relations at the point of production. Despite, for instance, noting that the capitalist appropriates surplus value from labour, Das comments:

that does not mean that capitalists do not work. They do work. But they do not have to work. Their income and property are disproportionately more than they would be if they were based merely in their own labour. (emphasis in original)

There is an absence of any recognition that supervision and control by capital, embedded in extended managerial hierarchies, comprise not unproductive labour but non-labour.

The approach leans towards a view that levels of autonomy and earnings in themselves take people out of the working class. Again, being working class is not associated with capacities and function of labour, but lower wages and increased supervision – implying that skills debar you from being part of the collective worker.

Other writers see in Carchedi’s approach a sterile structuralism[xx] or claim that such analytical perspectives on social classes necessarily end in ‘an aversion to empirical work and a tendency to create classes alfresco’[xxi]. Carchedi’s approach is entirely consistent with Marx’s method in Capital which centres on the major forces at work before making allowances for counter-tendencies. How capitalist social structures and practices are manifested in concrete conjunctions is determined by numerous influences on social relations, not by scholasticism. In short, it was a theoretical work, albeit with some illustrations, not a template. The following illustrates how it can be applied to understand contemporary class relations.

The theory in practice – contemporary workplace restructuring

As briefly indicated above, state initiatives have restructured the organisation and, simultaneously, the labour process and class relations in many areas of the public sector. Teaching in England and Wales can provide an illustration[xxii]. Through a combination of fragmentation, local management of schools, a national curriculum, standardised tests and targets producing league tables, and a widely-feared inspection regime, teachers’ work has been subject to deskilling, intensification and high levels of stress. A Workforce Remodelling Agreement (WRA) promised to relieve teachers of peripheral work by expanding the roles of teaching assistants and ancillary workers, thus allowing them to concentrate on teaching. Henceforth, they were to cease photocopying, mounting displays, collecting dinner money, and other roles that excused absence from the classroom and allowed informal interactions and conversations. They had been released from these duties in order to teach more: they were to do nothing else. In order to police the new boundaries, to pro-actively and systematically intervene in day-to-day practice, a new management regime was required. Achieving this new structure was engineered in part by new Management Allowances (MAs). Following the logic of the WRA, MAs for non-teaching responsibilities were removed and replaced by Teaching and Learning Responsibilities (TLRs). All schools were required to produce new staffing structures that gave additional payments, not for the miscellany of extra duties associated with MAs (for instance, timetabling or pastoral work), but only to a smaller, more highly remunerated group with responsibilities for the performance and achievements of students and their teachers. Heads of Department assumed titles such as Directors of Learning with new roles that made them accountable for the results of their departments and for carrying out new performance management procedures. Examined through the lens of the capitalist labour process this change reflected a shedding of tasks that were part of the collective worker and the assumption of more tasks of supervision and control that are not unproductive labour, but non-labour. With the development of more explicit, hierarchically organised line management, it is increasingly unrealistic to maintain, in line with Reid’s arguments[xxiii], that all teachers are simply workers and that control is solely the prerogative of ‘higher officials’.

Similar movements in the boundaries of control, and with them, changing relationships between the labour process and class, can be seen in developments within Her Majesty’s Revenue and Customs (HMRC)[xxiv]. With the introduction of Lean Production in British tax offices, Front Line Managers (FLMs) were required to work in a very different way. FLMs tended to be the most knowledgeable about tax work. Previously, they were responsible for allocating work, had substantial autonomy and frequently gave help and advice over tax issues that arose in the work process. They were therefore for most of the time contributing to outcomes and were part of the collective worker. These features, together with them being in the same union and bargaining unit as their teams, were reflected in the lack of hostility to them from the rest of the workforce. Indeed, in some tax offices, they were leading trade unionists, holding branch office and being amongst the most militant.

With the introduction of Lean Production, FLM roles changed markedly. They were required to monitor the work outputs of individuals by collecting hourly statistics, to report these and to mark up outputs on a whiteboard. Their input into decision-making on tax queries was (theoretically) made redundant by fragmentation of the work process and standard operating procedures supplemented by computerised information systems. In recording and reporting the performance of their subordinates, they simultaneously reported and made transparent their own effectiveness, an observation made of changes in the private sector by Nichols and Beynon as early as 1977[xxv]. In addition, they had also to administer disciplinary and sickness absence policies that became increasingly harsh and conflictual. With the defeat of opposition to the introduction of Lean, many of the FLMs took redundancy or retired and were replaced by staff with much less, or no experience of tax affairs, having either excelled in learning Lean mantras or having been transferred from the Excise section.

Remaining FLMs were threatened with discipline for raising objections to the irrationality and inefficiency of Lean. The result was an end to their specialist contribution, whether in the form of knowledge or material labour. Instead, FLMs were engaged in less challenging, de-skilled work, but were more accountable for the performance of workers, and hence perform more as agents of the state. They moved from doing unproductive work as part of the collective worker to performance of control and supervision and therefore non-productive labour. The idea that increased work intensity and de-skilling, divorced from the social relations in which they are embedded, in themselves constitute proletarianisation, falls a long way short of understanding the complexity of the capitalist labour process.


Marx and Carchedi were able to conceptualise the elements that, in the absence of complications and intervening factors, would produce the simplified class structure of capitalism – workers facing capitalists. When analysing real societies with all their historical variations and different insertions into the world economy, no such simplifications hold. Class is a relationship in process that is complicated by mediations. The process of capitalist production nevertheless produces workers and a working class that by its very nature struggles daily with the power of capital. Carchedi’s approach is not one that forces people into static categories. It is not a model or a template that can be laid across society to determine a class structure. Nor is it a structuralism in which people have no subjectivity.

Studying social relations in process and change can reveal both managers and workers recognising distinctions between who suffers or benefits from management initiatives and developments, and the difference between useful work and tasks that only arise because of the need to maximise the extraction of surplus value and surplus labour. The frontiers of work and control change in struggle[xxvi]. Workplace managers can be sympathetic or hostile in these struggles. It is, therefore, possible both to insist on a collective working class power, based on labour that produces use values and surplus, a power that is strengthened by the continued socialisation of production and thereby proletarianisation, and, at the same time, to simultaneously recognise the counter-currents, based upon strengthening control, supervision and oppression of workers on behalf of capital, functions that bring their own rewards, both subjective and material, in ways that makes a simple division between the 99% and the 1% simplistic, naïve and sometimes politically disarming, both in production and beyond.


[i] Jane McAlevey, No Shortcuts: Organizing for Power in the New Gilded Age (Oxford: Oxford University Press, 2016), pp.89-95

[ii] Thomas Piketty, Capital in the Twenty-First Century (Cambridge, MA: Harvard University Press, 2014).

[iii] Michel Husson, ‘Capital in the Twenty-First Century by Thomas Piketty: “Wealth of Data, Poverty of Theory”’, Historical Materialism, 23:1 (2015), 70–85.

[iv] Michael Roberts, Marx 200 – a review of Marx’s economics 200 years after his birth (, 2018).

[v] Danny Dorling, Inequality and the 1% (London: Verso, 2019), pp. 1,4,9.

[vi] Karl Marx, Capital, volume 1 (Harmondsworth: Pelican, 1976), p.279.

[vii] Bob Carter, Capitalism, Class Conflict and the New Middle Class (London: Routledge, 2015).

[viii] Nicos Poulantzas Classes in Contemporary Capitalism (London: New Left Books, 1975).

[ix] Eric Olin Wright, Class, Crisis and the State (London: New Left Books, 1978).

[x] Barbara Ehrenreich and John Ehrenreich ‘The Professional-Managerial Class’, in Between Labour and Capital, ed. Pat Walker (Sussex: Harvester Press, 1979), pp.5-45.


[xii] Wright, Class, Crisis and the State, p.62.

[xiii] Eric Olin Wright, Classes (London: Verso, 1985).

[xiv] Eric Olin Wright, Classes, p.89.

[xv] Guglielmo Carchedi, On the Economic Identification of the New Middle Class (London: Routledge, 1978).


[xvii] Raju J Das, Marxist Class Theory for a Skeptical World (Chicago: Haymarket Books, 2018).


[xix] Das, Marxist Class Theory for a Skeptical World, p.121

[xx] Chris Smith and Hugh Willmott, ‘The New Middle Class and the Labour Process’, in White-Collar Work: The Non-Manual Labour Process, eds. Chris Smith, David Knights and Hugh Willmott (Basingstoke: Macmillan, 1991), pp.13-34.

[xxi] Chris Smith and Paul Thompson, ‘Re-Evaluating the Labor Process Debate’, in Rethinking the Labor Process, eds. Mark Wardell, Thomas Steiger and Peter Meiksins (New York: State University of New York Press, 1999), p.219.

[xxii] For fuller accounts see Bob Carter, Howard Stevenson and Rowena Passy, Industrial Relations in Education: Transforming the School Workforce (London: Routledge, 2010); and Bob Carter and Howard Stevenson, ‘Teachers, workforce remodelling and the challenge to labour process analysis’, Work, Employment and Society, 26:3 (2012), 481-496.

[xxiii] Alan Reid, ‘Understanding teachers’ work: is there still a place for labour process theory?’, British Journal of Sociology of Education, 24:5 (2003), 559-573.

[xxiv] Bob Carter, Andrew Danford, Debra Howcroft, Helen Richardson, Andrew Smith and Phil Taylor ‘“They can’t be the buffer any longer”. Front-line managers and class relations under white-collar Lean Production’, Capital and Class 38:2 (2014), 323-343.

[xxv] Theo Nichols and Huw Beynon, Living with Capitalism: Class Relations and the Modern Factory (London: Routledge, 1977).

[xxvi] Carter Goodrich, The Frontier of Control: A Study in British Workshop Politics (London: Pluto, 1975).


Please enter your comment!
Please enter your name here