The Great British Class Survey and the Return of Class Today

In the first quarter of the twenty-first century, inequality is firmly back in the public eye. In 2014 the World Economic Forum highlighted income disparity as one of the principal risks to economic and political security today. International non-governmental organizations such as Oxfam have drawn attention to the cycles of advantage transmitted across generations and pointed to the unequal opportunities that reinforce privilege. Longstanding problems such as poverty appear to be getting worse even though we live in more affluent times, and have, increasingly, been juxtaposed with the burgeoning fortunes of the super-rich.

This book shows, focusing on the British case, how these spiralling levels of inequality are remaking social classes today. It is one thing to point to growing economic inequalities, but we also need to know how people themselves understand these divisions. Do they overlap with wider social, cultural and political fractures? Can we identify distinctive social classes who share common lifestyles, identities, social networks and political orientations as well as levels of income and wealth? Only in this case can we talk about ‘class for itself’ (i.e. class as composed of people who are class conscious), rather than ‘class in itself’ (i.e. class as a social group), to use Karl Marx’s idiom. In this book we argue that classes are indeed being fundamentally remade. Away from longstanding differences between middle and working class, we have moved towards a class order which is more hierarchical in differentiating the top (which we call ‘the wealth elite’) from the bottom (which we call ‘the precariat’ which consists of people who struggle to get by on a daily basis), but which is more fuzzy and complex in its middle layers. We will show in this book how social classes arise from the concentration of three distinctive kinds of capital: economic capital (your wealth and income); cultural capital (your tastes, interests and activities), and social capital (your social networks, friendships and associations). Each of these is fully introduced and explained in a later chapter. Understanding class as based on these three capitals allows us to understand how growing economic inequality is also associated with growing class inequality between the top and the bottom. We can also avoid the tendency to see class as a throwback. to the old industrial era of blue-collar factory workers, coal miners and farm labourers, pitted against mill owners, professionals and managers. Our focus on economic, cultural and social capital offers our alternative to previous sociological analysis focusing on classes as groupings of occupations and which do not adequately illuminate the wider cultural and political significance of class. We can also restate the importance of class, given that critical commentators such as Owen Jones or Danny Dorling rarely use the concept of class directly,1 while economists, who in recent decades have done most to demonstrate the growth of inequality, tend not to use the concept of social class at all, which they see as too crude to capture contemporary economic divisions.

Our perspective allows us to see how expanding economic inequalities are associated with class divisions more broadly. We can identify today’s major social fractures, especially at the top and bottom, and understand the difficulties of social mobility into the most advantaged positions, the intensification of geographical divisions, and the growing power of elite universities.

Introducing the Great British Class Survey

The current explosion of interest in questions of class came home to us in 2013 when we published findings from the BBC’s Great British Class Survey, which was publicized by the media and provoked astonishing interest across the globe. When the BBC persuaded us to help them design their web survey, we truly did not know how much interest it would generate. Who could be bothered to give up twenty minutes of their time to answer a battery of arcane questions on their leisure interests, cultural tastes, social networks and economic situation? We were therefore gratified when within a few weeks the survey elicited over 161,000 responses – to become the largest survey of social class ever conducted in Britain.

Having downloaded the first wave of data from the BBC in April 2011, we spent nearly two years trying to make sense of its patterns. After several false starts, we elaborated a new sociological model in April 2013 which proclaimed the existence of seven new classes, which we discuss further in Chapter 5.2 The BBC promoted our findings with an impressive set of visual infographics. These were then hooked up to an interactive ‘Class Calculator’, whereby people could spend less than a minute tapping in replies to a few questions about their income, savings and house value, their cultural interests and their social networks, and were then told which of our ‘new’ classes they fitted into.

The results were staggering. Within a week seven million people – roughly one in five of the British adult population – clicked on the Class Calculator to find out which ‘new’ class they were in.3 Social media buzzed with debate and a tidal wave of popular and academic comment overwhelmed us. During our careers we have written extensively about social class for mainly academic audiences, but had never experienced popular interest on this scale. There was extensive blogging and media interest in the arguments on a level rarely – if ever – seen for sociological research.

This is a prime message of this book – that social class is now a very powerful force in the popular imagination once again. People in Britain are aware of, interested in and also upset about class. During the media storm associated with the launch of the Great British Class Survey, we heard of train passengers chatting about which class they were now in and schoolchildren talking in the playground about class. There were some very odd incidents. Demand for theatre tickets in London increased by an average of 191 per cent in the week after the GBCS launch. Louise Mullock, spokesperson for Seatwave, remarked: ‘We recorded a near universal increase in ticket demand which we were at a loss to explain, until we realized that it corresponded directly with the BBC’s Class Calculator becoming public.’4 It seemed that large numbers of people responded to the Class Calculator’s questions treating theatre attendance as an indicator of cultural capital by deciding to get out more.

Scientific experiments are normally expected to stand back from the research they are conducting in order to provide distanced and ‘objective’ results, for instance using randomized control tests when comparing which medical interventions are effective. However, in the case of the GBCS, we could not do this. Interests in class are themselves so highly loaded that if we try to stand back, then we miss the energies, intensities, but also the hostility and insecurity that are bound up with class. Indeed, this is a fundamental argument of our book. We like to think of ourselves as living in a democratic society where individuals are supposed to have equal rights. Yet we also know that people’s economic fortunes can be strikingly different. Symbolically, class is a lightning conductor for the anxieties this discrepancy between economic realities and our beliefs provokes.

This issue is crisply demonstrated in the differential take-up of the GBCS web survey itself. It turns out that those who are interested in doing a twenty-minute web survey are far from being typical of the population as a whole. The map (Figure 0.1) shows the distribution of where people participated in the GBCS in numbers greater or lesser than we would have expected, given the population of the area. (We have taken that underlying population from the 2011 census.) So, if there was a perfect equality of participation across the entire country, all local areas would match the figure we expected, based on the census. However, that clearly wasn’t the case and there were big disparities in levels of participation that we actually observed at the local level. So this map displays those disparities as a percentage of what we would have expected to see. In the white-shaded areas, levels of participation were half (or even lower) than what we would have expected, given the local population base, while in darker-shaded areas, many more people completed the web survey than we could have predicted, up to 50 per cent higher or more. Thus, the map reveals the geographical distribution of those who did the GBCS, comparing them to the population in different unitary authorities in Great Britain (including the Greater London boroughs) and in Northern Ireland by county.5 We can note at the outset that Northern Ireland has markedly lower response rates in comparison to other regions of the UK. It seems likely that considerable proportions of the Northern Irish people, notably those of a Republican hue, may not have been attracted to a survey entitled the ‘Great British Class Survey’. In this sense national identity may have trumped class and this may well also explain the lower response rates in swathes of western Scotland and even into the north of England.

Figure 0.1
Figure 0.1

The Geography of Participation in the GBCS in Relation to Population

At the other end of the scale, areas overcompensating in terms of response to the survey (indicated in darker shades), were heavily clustered in the west of London and to a large extent in the south-east of England more generally. These areas are the BBC’s heartland. However, the geography is complicated. Edinburgh also had a very high response rate – and Glasgow residents were not far behind. Looking at a more refined spatial scale, university towns and cities were also over-represented, with Oxford, Cambridge and York all standing out from their hinterlands with particular clarity and Aberdeen, Brighton, Exeter, Canterbury, Aberystwyth and Norwich all being in darker shades.

The take-up of the GBCS is itself testimony to marked geographical divisions and poses a fundamental question as to the reasons behind such stark differences in why some people are interested in class, and others are not. It is the more affluent who seem more interested in the topic of class, even though they might also be sceptical of the survey results. Even the most basic indicator of the geography of GBCS respondents is itself indicative of the profound power of a class divide.

The plot thickens when we consider what the occupations were of people who were most likely to do the GBCS. Tables 0.1 and 0.2 demonstrate major differences. A huge 4.1 per cent of all those replying to the GBCS were chief executive officers (CEOs), which turns out to be twenty times more than we would expect, given the total number of CEOs in the labour force. We also see a dramatic over-representation of business and related finance professionals, and also all kinds of scientists, researchers and professionals. Experts, of all kinds, were drawn in droves to the GBCS.

Occupation % of GBCS Amount Over-Represented
CEOs 4.1 × 20.4
Business, research and administrative professionals 0.9 × 5.8
Business and related finance professionals 1.5 × 13.7
Natural and social scientists 1.2 × 8.3
Physical scientists 0.4 × 5.9
Barristers and judges 0.4 × 4.8
Actuaries, economists and statisticians 0.5 × 4.8
Engineering professionals 1.3 × 4.8
Journalists 1.2 × 4.5
Table 0.1

Over-represented Occupations in the GBCS

Contrast this, however, with a list of those who did not hasten to the BBC’s Lab UK website to complete the GBCS, as displayed in Table 0.2. This opposite group consist of people in largely unskilled manual occupations. Out of the 161,000 respondents, not a single cleaner or worker in the elementary (basic) services or plastics processing answered. There were also very few glaziers, fork-lift truck drivers or the like. In comparing these tables, the stark power of a class divide is clear.

Occupation % of GBCS Amount Under-Represented
Elementary service occupations
Elementary cleaning occupations
Plastics process operatives
Forklift-truck drivers 0.01 × 0.3
Glaziers, window fabricators 0.01 × 0.3
Roofers, tilers and slaters 0.02 × 0.4
Rubber process operatives 0.02 × 0.4
Vehicle / paint technicians 0.02 × 0.4
Packers, bottlers, canners and fillers 0.02 × 0.4
Table 0.2

Under-represented Occupations in the GBCS

These are very simple findings, but they display, in miniature, the power of class divisions in Britain, which divide an affluent, educated, professional and managerial group living in west London and the Home Counties on the one hand, from a group of manual workers, more likely to be living outside the south-east of England on the other. Such is the imprint of class that it is marked in the very conduct of the research tool designed to reveal its significance.

But the issue is even more complicated than this, for the GBCS is not only about class, it is also about being ‘British’. We have already seen that Northern Irish and Scottish residents were less likely to respond to the GBCS than those in England. There was also a high skew in which ethnic groups, and the kind of members of different ethnic groups, who did the GBCS. Table 0.3 shows that the proportion of black and Asian ethnic minorities is considerably fewer than we would expect, if the sample is representative of the British population. Why was this the case? It might be that these ethnic groups are more likely to be in the lower levels of the class structure, members of which we have already seen are unlikely to respond to the GBCS. It is also possible that the connotations of ‘Great British’ might be taken to be ‘white British’, and hence of little relevance to ethnic minorities. Among those ethnic minorities who did do the GBCS, a high proportion were students or university graduates, indicating again that the GBCS is skewed towards the proportionately better educated from within these groups. Comparing Asians with white people, for instance, will largely turn out to be a comparison of the well-educated members of both groups. Thus, we can’t simply assume there is only a class skew at work in the GBCS, as the biases in the data run in several different ways and we need to be attentive to all of these when we report findings.

Ethnic Group % of the Ethnic Group who Undertook the GBCS % of the Population who Undertook the GBCS (2011 Census, England and Wales) % of Each Group's Graduates who Undertook the GBCS
White 90.14 85.97 63.4
Black 0.90 3.32 59.9
Asian 2.18 6.81 68.1
Chinese 0.94 0.70 71.9
Mixed 1.89 2.18 62.2
Others 3.96 0.10 65.0
Table 0.3

Ethnic Skew in the GBCS

The GBCS certainly can’t be used to report nationally representative findings. But in a sense, this is actually a very important point, for as we see it, this skew is highly revealing. We would be failing to recognize the power of class in contemporary Britain if we did not understand how people’s engagement with forms of knowledge and expertise is implicated in class itself. This is a point which we will return to time and again as this book proceeds.

Nevertheless, this observation also poses fundamental challenges for our analytical strategy. Clearly, if we simply reported the findings from the GBCS, we would be relying on the voices of the most affluent and well educated, and could hardly do justice to class divisions, and their intersection with ethnicity, gender, age and suchlike. And indeed the problems of relying on a survey with such an intense sample skew have been amply discussed by commentators on the GBCS in previous publications.7 This is why in this book we have used numerous other sources to allow us to redress this bias.

Firstly, once the extent of the sample skew came to light the BBC agreed to conduct a small nationally representative survey of 1,026 people, conducted by the market-research firm GfK, asking identical questions to those in the GBCS, allows us to provide benchmarks to correct for the biases of the GBCS6. We use this national sample to provide representative patterns where this is appropriate for our arguments, and use the larger GBCS when we want to mine down into more detailed analyses. We will make it clear which of the samples we are using.

Secondly, given the immense media and public interest in the GBCS, we decided to follow up with fifty additional qualitative interviews in order to find out more about what people think about class, in ways which go beyond survey response tick boxes. We directed some of these interviews at those kinds of people who did not tend to do the GBCS, namely those at the bottom of the class structure, in order to find alternative means of accessing groups who were not predisposed to complete the survey. We use these qualitative interviews and ethnographic vignettes to flesh out our account and bring out issues which are left hanging by the survey itself. We also use these interviews to zoom in on members of the elite class, who are of particular interest to us.

Thirdly, and very importantly, in reporting the results of the GBCS we also draw on extensive sociological research that underpins our long-term engagement with these issues. The authors of this book have extensive direct experience with numerous other projects examining the remaking of social class divisions and we will deepen and bring out the connections with other research as we proceed. A particular reference should be made to the Cultural Capital and Social Exclusion project which conducted a comprehensive national survey, created focus groups and undertook qualitative interviews to assess the nature of cultural engagement in Britain (and the findings of which were published in Bennett et al., 2009).8 This pioneered many of the questions on cultural capital that we used in our GBCS project and offers much additional research bearing on the topics of the book.

The Great British Class Survey is, then, an experiment in the true sense of the word. Both the data itself, as well as the public interest in the research, give us an unparalleled opportunity to think about the new kinds of class divisions that exist in contemporary Britain. But in understanding the lessons from this study we need to stand back from the data to look at the wider dynamics of class which inform us whether people are likely to have completed the questionnaire itself.

This book provides our interpretation of the significance of class today. We do not attempt to summarize swathes of literature. Rather, drawing on and distilling the extensive research we have conducted in recent decades, we try to offer a provocative account, which we hope will generate discussion and reflection inside and outside academic circles. This book is thus not simply a report on our work with the GBCS, but represents a much wider engagement with sociological research on class.

Our analysis is based on the British case, on which we are experts. We do not claim that the British experience is typical or that it lays down a course which other nations will follow. Far from it. However, we do think that the issues discussed here are unlikely to be confined to this country alone and will have resonances around the planet, even while we fully acknowledge that as one of the wealthier nations in the world, the relationship between wealth elites and poorer classes will be very different in other countries. And the British case is symbolically important too. Over the past century, reflections on the nature of class relations in Britain have generated pivotal arguments on the changing nature of citizenship, welfare, poverty, cultural snobbery, political radicalism and reform across the globe. In all these cases, by recognizing the peculiarities of the British case, it is also possible for those in other locations to identify how class relations work in other contexts. If our book provokes reflections from other national contexts, then we will have had a positive influence.

The topic of class is far from being a dispassionate one. There are bitterly contested views about what classes are, how to measure and analyse them, and their overall significance for society. And we are far from being neutral in these debates. We have been at the forefront of a group of British sociologists who have insisted over recent years that class remains fundamental to sociological analysis. We have also championed the thinking of the French sociologist Pierre Bourdieu as offering the most perceptive approach to unravelling the complexities of class today, in a programme of work which has been labelled ‘cultural class analysis’. In this book, we do not debate with other perspectives directly, so as not to be distracted from our main aim.

Our book has a logical form and is best read in sequence. In the first part we explain how existing thinking about class continues to focus on the divide between the middle and working classes, which we will show is outdated. Part Two argues for our multidimensional approach to class, in which we unravel how economic, cultural and social capital each contribute to inequality, and in Chapter 5 we show how we can link these different kinds of capital together to develop a new approach to class which shows there is no neat dividing line between working and middle class, but a more hierarchical class order, in which the wealth elite stands clearly above the rest. Part Three then shows how social class impacts on social mobility, education and geography. Part Four zooms in on our two extreme classes – the wealth elite and the precariat – to demonstrate the profound class divide that now exists. We draw out the political implications of our arguments in the conclusion.

  • 1. Danny Dorling, Injustice (Bristol: 2010), and Owen Jones, The Establishment, and How They Get Away with It (London: 2014).

  • 2. Mike Savage, Fiona Devine, Niall Cunningham, Mark Taylor, Yaojun Li, Johannes Hjellbrekke, Brigitte Le Roux, Sam Friedman and Andrew Miles, ‘A New Model of Social Class? Findings from the BBC’s Great British Class Survey Experiment’, Sociology, 47(2), 2013, 219–50.

  • 3. This number had risen to nearly nine million by the end of 2014, with over ten thousand people continuing to click on the Class Calculator each week. See further, Fiona Devine and Helene Snee, ‘Doing the Great British Class Survey’, Sociological Review, 63(2), 2015, 240–58.

  • 4. See

  • 5. Spatial units with fewer than a hundred GBCS respondents have been suppressed.

  • 6. However, this was a particular problem for the analysis of ethnicity, since the numbers of ethnic minorities in the nationally representative sample gathered by the market-research company GfK (see also Appendix: The Great British Class Survey, section 2), were too few to allow inferences to be readily drawn. Readers should therefore be aware of this limitation and the implication that ethnicity cannot be satisfactorily analysed using either the data-set the GBCS provides or the nationally representative sample’s data-set garnered by GfK.

  • 7. See Colin Mills, ‘The Great British Class Fiasco: A Comment on Savage et al.’, Sociology, 48(3), 2014, 437–44, as well as the response by Mike Savage, Fiona Devine, Niall Cunningham, Sam Friedman, Daniel Laurison, Andrew Miles, Helene Snee and Mark Taylor, ‘On Social Class, Anno 2014’, Sociology (forthcoming).

  • 8. See Tony Bennett, Mike Savage, Elizabeth Silva, Alan Warde, Modesto Gayo-Cal and David Wright, Culture, Class, Distinction (Abingdon: 2009).

Next chapter

Contesting Class Boundaries

Chapter 1