Introduction

In 1995 Sir David Attenborough released a series, ‘The Private Life of Plants’, which used new filming techniques to dramatically speed up the passage of time. The results were sensational. Peaceful forest glades were revealed as whirling, chaotic worlds of dynamic change and perpetual competition. Plants formed friendships, went to war over territory and ran desperate races for access to precious light. The plant world is ever-changing, but because the change is slow, it usually escapes our notice. Social change is often like this too. Societies are in constant flux. The jobs people do, the education they have, where they come from, where they go to, how they think of themselves and what they value – none of these things are constant. Yet because such change is often slow, it goes unnoticed until a watershed moment draws attention to it. The EU Referendum was such a moment, when a shock result forced people to pay closer attention to changes long underway all around them. What people saw in the wake of the ‘Leave’ victory was a land suddenly divided, at odds with itself and locked into intractable conflicts: Brexitland. But the divides Brexit exposed were not new. They had been building in the electorate for years.

Two demographic shifts have been gradually reshaping British society for many decades – educational expansion and ethnic diversification. Just one generation ago a majority of the British electorate were white voters with few or no educational qualifications. University was the preserve of a privileged minority,Footnote ¹ and ethnic minority communities were still small and concentrated in the largest cities.Footnote ² Educational expansion and mass migration have since driven a slow but relentless transformation of the electorate, with the youngest generations dramatically more highly qualified and ethnically diverse than the oldest.Footnote ³ In this chapter, we examine these demographic shifts and explain how their relationship to an influential worldview – ethnocentrism – gives them disruptive political power.

Ethnocentrism is the ‘view of things in which one’s own group is the centre of everything’.Footnote ⁴ The presence or absence of ethnocentric views plays a major role in determining on which side of the Brexitland divide people fall. On one side of this divide are the voters we call ‘identity conservatives’ – white voters with lower levels of formal education who most frequently hold ethnocentric worldviews, making them more strongly attached to in-group identities like national identity and more threatened by out-groups such as migrants and minorities. On the other side are two ‘identity liberal’ groups – university graduates and ethnic minorities – who for different reasons reject ethnocentrism. The conflict between these groups runs right through the heart of the electorate, and the activation of this conflict is a major source of the political upheavals and volatility of the past decade. The new political context we call ‘Brexitland’ is one in which identity conflicts between the formerly dominant but now declining identity conservative group and the growing but not yet dominant identity liberal groups have become a central structuring feature of British politics. The vision of Britain each side embraces is one its opponents reject. This conflict in worldviews, once mobilised, produces polarised politics – it is hard to compromise with those whose values you abhor. But it also produces highly dynamic and competitive politics, because at the present time neither identity conservatives nor identity liberals are large enough groups to prevail consistently and set the terms of debate.

The polarisation of identity politics is also exacerbated by the way white school leavers, graduates and ethnic minorities are clustered in certain age groups and areas. Both ethnic and educational change are generationally structured. The British electorate today contains older cohorts dominated by ethnocentric white voters, and younger cohorts where university graduates and ethnic minorities predominate. Social norms and social experience are generationally structured,Footnote ⁵ with views about what is ‘normal’ strongly influenced by experiences in early adulthood, leaving the generations deeply divided in their experiences and their values. Many of those who grew up in a more ethnically homogeneous, socially conservative Britain have a profoundly different view of what Britain is and ought to be than members of the youngest generations, who have grown up in a much more ethnically diverse and socially liberal country.

Geography has a similar polarising effect.Footnote ⁶ People on both sides of the identity politics divide live in distinct locations through choice and circumstance. The migration of university graduates for study and work concentrates them in the prosperous towns and big cities where university campuses and job opportunities are found. Ethnic minorities are also concentrated in larger cities, a result of past and present migration patterns.Footnote ⁷ By contrast, white voters with low education levels move less often, and are becoming concentrated in more ethnically homogeneous and less economically successful rural and small-town areas.Footnote ⁸ These growing ethnic and educational differences between big cities and small towns are further exacerbated by a growing age gap. While the recruitment of students and young graduates is making large towns and cities younger, smaller towns are rapidly ageing as these groups move away while older residents remain.Footnote ⁹ These trends magnify identity conflicts by increasing social segregation and reducing the level of contact and common experience between people on either side of the identity politics divide. Graduates mainly live among other graduates, in ethnically diverse places that accord with, and reinforce, their belief in a dynamic and diverse Britain. White voters with low education levels also live around similar people, in ageing and declining places which accord with, and reinforce, their sense of marginalisation and stagnation.

In this chapter we set out the demographic changes which are driving the emergence of identity politics: educational expansion and rising ethnic diversity. We then introduce the concept of ethnocentrism, the tendency to see the social world in terms of groups and group conflict; and illustrate how both educational levels and ethnicity are closely linked to this worldview. We then show how the generational and geographical polarisation in education levels and ethnic diversity serve to deepen the divides between people and places, and to magnify the political impact of the new identity politics conflicts. But while demographic change is inevitable, demography is not political destiny, as political parties have a vital role to play in determining how these new divides are mobilised into political competition.

Education: university expansion and the rise of the graduate class

Britain is in the middle of a historic transformation from a society of school leavers to a society of university graduates. This is part of an international trend evident in most developed democracies since the Second World War,Footnote ¹⁰ and which is fast spreading to other countries as they become more prosperous.Footnote ¹¹ Higher education is widely seen as economically and socially beneficial, and there has been a near-universal tendency for wealthier nations to invest in expanding access to it. This process was initially slow in Britain, as governments held on to an elitist model where only a small minority attended university, but then made a late and dramatic switch to mass higher education, sending university attendance rates sharply upwards. The legacy of this tortoise to hare transformation is a particularly stark generational divide in education levels.Footnote ¹² The first wave of British university expansion occurred in the 1960s, but this was modest and left university attendance at around 10 per cent of each cohort. Universities then grew slowly for the next two decades, with attendance rates in the 1970s and 1980s still in the 13–15 per cent range.Footnote ¹³

The second wave of university expansion was the product of reforms by the early 1990s Conservative government, in particular the 1992 Education Act, which upgraded a large set of educational institutions to degree-awarding university status. The effects were substantial: university attendance rates more than doubled from 15 per cent in 1988 to 33 per cent in 1994, and have continued to rise since. The domestic undergraduate population in 2000 was six times larger than in 1960, and by the mid-2010s university attendance rates at eighteen were approaching 40 per cent. The expansion of higher education since the early 1990s means that the British electorate is currently divided between generations born before the 1970s, who grew up with an elitist higher education system, and those born since, who have grown up with ever-expanding mass higher education provision and consumption – though access remains skewed towards the wealthy and the middle class.Footnote ¹⁴

University expansion is one part of a broader trend of increasing access to education. The proportion of students staying on beyond the compulsory school leaving age (fifteen until 1973, sixteen since) rose from 20 per cent in the early 1960s cohorts to 40 per cent in the early 1980s and over 70 per cent among the 2000s cohort. The cumulative effect of these repeated waves of educational expansion is a slow but steady rise in the overall formal education levels of the electorate, as younger cohorts brought through a dramatically expanded education system gradually replace older cohorts where most students left school at the earliest opportunity (see Figure 2.1). The changes year-on-year are small, but their cumulative effect is striking. When Margaret Thatcher won her third election victory in 1987, seven voters in ten had left formal education at sixteen or earlier, and university graduates (8 per cent) were outnumbered five to one by voters with no formal qualifications at all (42 per cent). By the time Tony Blair won his third election victory in 2005, the graduate share had more than doubled, but graduates were still heavily outnumbered by unqualified voters. When Theresa May faced the electorate in 2017, nearly a quarter of voters had a university degree, triple the share of Thatcher’s time, and graduates substantially outnumbered the unqualified. The year 2010 was an important turning point in this process: graduates were heavily outnumbered by the unqualified in every general election held before this point, but outnumber the unqualified in every election held after it.

Figure 2.1 Share of respondents who report having no formal qualifications and who report having an undergraduate degree or more

Source: British Social Attitudes surveys 1985–2016.

Racial diversity and immigration: the rise of multiracial Britain

The second great demographic change of the post-war era is Britain’s transformation into a racially diverse society. While Britain has long incorporated multiple distinct national cultures within a single state,Footnote ¹⁵ and has longstanding and politically distinctive religious minorities,Footnote ¹⁶ the rise in ethnic, racial and religious diversityFootnote ¹⁷ seen since the onset of mass migration from the Commonwealth in the 1950s has been different in scale and scope to what came before.Footnote ¹⁸ Britain’s shift from a nearly all-white society to a racially diverse one has occurred, like the transformation of education levels, within a single lifetime, as Figure 2.2 illustrates. A pensioner turning seventy-five in 2019 spent their childhood in a society where less than one in a hundred people was born outside Europe or belonged to an ethnic minority, while a youngster turning eighteen in the same year has only ever known an ethnically diverse Britain with large well-established ethnic and religious minority communities, a country where around one person in five belongs to an ethnic group other than ‘white British’ and one in seven residents was born in another country.

Figure 2.2 Ethnic minority population of the UK, 1951–2011

Sources: Census (1991–2011 ethnic minority population figures and 1951–2011 total population figures); Owen (Reference Owen1995) (1951–1981 ethnic minority population estimates).

This transformation is more than a matter of raw numbers. The nature of diversity has changed as minority communities have grown and become more established. In the 1950s Britain of current pensioners’ youth, residents from ethnic minority groups were almost all first-generation migrants who had usually come to Britain as adults. Race and migration were conflated, racial diversity was an imported phenomenon not, outside a few districts in Britain’s largest cities,Footnote ¹⁹ a home-grown aspect of mass British culture. As time passed, a ‘second-generation’ ethnic minority population emerged, born and raised in Britain, and with no memory of their parents’ countries of birth. As these British-born minority communities grew and settled across a wider range of neighbourhoods, racial diversity became more embedded in everyday British social life. Black and Asian people born in Britain were less willing to accept discrimination and disadvantage;Footnote ²⁰ and the experience of being treated differently to other native-born fellow citizens shaped, and continues to shape, their social identities and political priorities.Footnote ²¹ Rising diversity has thus led to new debates about the meaning of British identity, with British-born ethnic minorities favouring multicultural understandings of Britishness which recognise and include them, while older white voters still hold to an understanding of Britishness framed by the homogeneous pre-migration society in which they grew up. The idea of reversing the process of ethnic change through state-sponsored repatriation schemes remained popular with many white voters for several decades after mass migration began. Yet the absurdity of mass repatriation in a country with a large, rapidly growing British-born ethnic minority population was already obvious in the 1970s. As the West Midlands-born black British comedian Lenny Henry told television audiences at the time: ‘Enoch Powell wants to give us £1,000 to go home. Suits me. It only costs me 50p to get to Dudley.’Footnote ²²

Debates over identity and diversity are now about much more than immigration and will continue to evolve as the British population changes. The 2011 Census revealed that the ethnic minority population in England and Wales was evenly split, with 48 per cent born in Britain and 52 per cent born abroad. Many of the British-born 48 per cent will be third- or even fourth-generation British-born. And the boundaries between ethnic minorities and the white majority are being further blurred as majority and minority communities intermarry. The fastest growing ethnic group in Britain is those reporting a mixed racial heritage, a group for whom a bright line separation between majority white British and ethnic minority identities makes little sense.Footnote ²³ The 2011 Census found that six per cent of children under ten had mixed heritage, seven times the share among fifty-somethings. This made the mixed heritage group larger than any other single minority ethnic group among Britain’s youngest residents – and continued rises in mixed marriages will ensure further growth in the years to come.Footnote ²⁴ Among the youngest cohorts, ethnic identities have become knitted together at the most intimate level, as their family heritage binds them to multiple communities.Footnote ²⁵

At the same time as established British-born minority communities have come of age and found their voice in society, new waves of migration have continued to bring new settlers to Britain, both from the original migrant countries and increasingly from elsewhere. Migration rates rose sharply from the late 1990s onwards, a shift big enough to constitute a ‘second wave’ of post-war migration. The scale and diversity of this new wave of migration is illustrated in Figure 2.3, which shows estimates of the total foreign-born population by broad region of origin. The overall migrant population nearly doubled between 2005 and 2017, rising from 5.3 million to 9.4 million. While there was substantial growth in all migrant populations, this varied a lot between regions of origin. Growth was slowest in the ‘Old’ EU migrant population (rising from 1.2 million to 1.7 million) and the diverse population from the rest of the world (up from 1.8 million to 2.4 million). The population of migrant residents born in the Asian subcontinent grew faster, nearly doubling from 1.7 million to 2.9 million. And the fastest growth of all came in the population of migrants from the new EU member states such as Poland, Romania and Lithuania – this population exploded from less than 200,000 in 2005 to nearly 2 million by 2017, a dramatic development with major political consequences, as we shall see in later chapters.

Figure 2.3 Migrant populations resident in Britain by broad region of origin, 2005–17

Source: Office for National Statistics.

The second wave of migration to Britain has been more regionally and ethnically diverse, but it is also distinct in another politically consequential way – the new migrants are much less likely to have voting rights than those who arrived earlier. The first wave of post-war migrants to Britain were for the most part Commonwealth and Irish citizens who possessed full political rights, including voting rights in general elections, from the day they arrived in Britain. These migrants rapidly became an important electoral constituency for politicians to court in many locations, and migrants’ electoral power helped to counter-balance, at least in part, the influence of ethnocentric white voters threatened by their arrival.Footnote ²⁶ Commonwealth migrant communities could use the power of the ballot box as one route to secure and protect their rights and status. Their British-born descendants, the vast majority of whom hold British citizenship, can do likewise. Things are rather different for the new migrants who have arrived since the late 1990s. A much larger share of these migrants come from non-Commonwealth countries, in particular the EU. They do not have general election voting rights on arrival in Britain, but can secure such rights only by becoming British citizens, which they are less prone to do than earlier waves of immigrants.Footnote ²⁷ This marginalises the new migrants in electoral politics and skews the political debate towards those threatened by their arrival.

While the profile of British society has changed dramatically over the past few decades, these changes would not matter politically unless education or ethnic identity had an important effect on voters’ values and political choices. Education levels and ethnic identity are associated with a number of differences in outlook and values, the most important of which concern ethnocentrism – the tendency to see the world in terms of groups and group conflict. It is the presence or absence of this perspective as a prevailing influence on how voters understand the world that is central to understanding how conflicts over identity arise from demographic shifts in education levels and ethnic diversity. Understanding what ethnocentrism is, why it matters to people and how to measure it, is therefore the task we turn to next.

Ethnocentrism: how educational and ethnic divides translate into political conflicts

The idea that structural changes can lead to the emergence of new conflicts between groups with different values and priorities is not a new one. There is a large academic literature mapping out how social change can drive political change in this way, the most prominent example being the work of Ronald Inglehart.Footnote ²⁸ In a series of ambitious comparative studies, Inglehart developed a model of mass social change, with rising prosperity driving a gradual shift away from ‘materialist’ values focused on securing the basic essentials in life and towards a set of ‘post-materialist’ values focused on individual rights, self-actualisation and liberty. Inglehart argued that this is a generational process: the values individuals hold are formed in the ‘socialisation’ period of young adulthood and shape their political priorities for the rest of their lives. As a result, the political shift away from economic priorities towards social and humanitarian priorities lags several decades behind the economic shift from poverty to prosperity – generational change is slow, and older generations who grew up before prosperity arrived stick around in the electorate for a long time. Related arguments have built upon this influential account, proposing a similar generationally structured transformation in religious attitudes and behaviours,Footnote ²⁹ in gender norms and gender roles,Footnote ³⁰ and in the emergence of democratic values and institutions.Footnote ³¹ Similar generational value-change arguments have also been used more narrowly, to explain differences in political values and identities,Footnote ³² and shifts in these over time, with different generations retaining a lasting concern with the political problems prevalent in their youth,Footnote ³³ and lasting attachments to the valuesFootnote ³⁴ and political partiesFootnote ³⁵ which were dominant when they came of age.

These accounts all share a core argument. The social, economic and political conditions people experience during their youth have a lasting impact on their politics, so changes in those conditions are followed by a much slower, generationally structured change in political allegiances and priorities. Social and economic change may be rapid, but the political change it produces comes later, and more slowly. Our account shares many elements of this story. We also focus on generationally structured social transformations – educational expansion and rising ethnic diversity – and, as in these earlier accounts, the political changes we examine are in part the product of the mobilisation of lasting differences between the worldviews of different generations, worldviews shaped by their experiences in youth. We concentrate on one aspect of this broader story – the division between those who embrace an ethnocentric worldview, with groups and group conflict at its heart, and those who reject and oppose such a view of the world. This division forms the focus of our story for two reasons.

First, ethnocentrism and identity conflicts have proven explanatory power in a range of contexts and are becoming more important as developed societies grow more diverse. The political power of group conflicts has long been evident to researchers working on American politics, where ‘the color line’ has been one of the most powerful of political and social divides,Footnote ³⁶ and where race and racial attitudes are still among the strongest predictors of political choice many decades after the Civil Rights movements ended the formal segregation and disenfranchisement of African Americans.Footnote ³⁷ More recently, conflicts over immigration and its impact on American identity have come to the fore, adding another layer of identity conflict to the longstanding ‘scar of race’.Footnote ³⁸ In Europe, a second large body of research has shown how mass immigration and the growth of Muslim minorities have sparked the emergence of new radical right parties which mobilise support from ethnocentric voters threatened by these developments, while views about immigration and diversity also exercise a growing influence on voters’ choices between mainstream parties.Footnote ³⁹

There is a broad consensus in these two large research communities on a number of key points. There are deep and growing divides in white majority populations centred on identity attachments and views of out-groups. These divides have proved uniquely capable of shifting white vote choices and disrupting political alignments,Footnote ⁴⁰ especially when voters perceive particular migrant or minority groups as threatening.Footnote ⁴¹ Conversely, ethnic minorities’ political choices are strongly influenced by their experience of white hostility and discrimination, giving them a lasting attachment to the (usually left or liberal) parties which have fought for their political and social rights, and a lasting hostility to centre-right and radical right parties which have mobilised ethnocentric sentiments in the majority electorate.Footnote ⁴²

The second reason we focus on ethnocentrism is that it provides an intuitive framework for understanding many of Britain’s recent political upheavals. As we show in Chapter 5, it was the activation of ethnocentric sentiments among identity conservatives that pushed immigration to the top of the political agenda in the 2000s, and it was ethnocentric voters threatened by immigration who turned against the New Labour government and then later the Conservative-led Coalition government, opening the door for UKIP in the 2010s. Meanwhile, as we discuss in Chapter 7, identity liberals – graduates and ethnic minorities – were becoming an ever more central part of the Labour Party electoral coalition, shifting the centre-left electoral coalition away from poorer, economically left wing but ethnocentric white school leavers and towards better off, economically moderate but identity liberal university graduates and ethnic minorities. These emerging identity conflicts then moved to centre stage in the defining political drama of the past decade – the EU Referendum – with ethnocentric narratives of an in-group ‘taking back control’ from hostile and threatening out-groups defining the campaign to leave the EU (see Chapter 8). Identity conflicts also played a major role in the re-alignment of Scottish politics – with arguments about ‘us versus them’ used to drive the rise of the SNP, which then channelled such sentiments in a very different direction in their bid to secure Scottish independence (see Chapter 9).

The first person to name the tendency to see the world through the lens of groups ‘ethnocentrism’ was the sociologist William Graham Sumner in the early twentieth century. Sumner argued this tendency to attach to social groups and to denigrate rival groups was universal in human social life and explained a diverse range of otherwise puzzling behaviour.Footnote ⁴³ A large body of research across several disciplines has confirmed this core intuition – people everywhere are indeed remarkably prone to identifying with social groups and turning even arbitrary and explicitly meaningless group contests into emotive battles of ‘us versus them’.Footnote ⁴⁴ But this tendency, like many aspects of personality and worldview, varies across populations in systematic and predictable ways. Political scientists and social psychologists have revealed ethnocentrism to be a stable personality orientation leading some people to be chronically prone to seeing social life in terms of attachments to in-groups and hostility to out-groups,Footnote ⁴⁵ while others seldom think about social problems in terms of group conflicts. As Donald Kinder and Cindy Kam, who have studied the phenomenon and its political effects in the United States, put it, ethnocentrism is ‘a readiness to reduce society to us and them … a readiness to reduce society to us versus them’.Footnote ⁴⁶

The origins of ethnocentrism, and why it varies between people, have been much debated, with different researchers emphasising different aspects of the complex mix of psychological and sociological forces which encourage or discourage the group conflict habit. Ethnocentrism may be an aspect of the ‘authoritarian personality’ – a tendency to value and insist upon conformity, order and authority, and to find diversity, ambiguity and uncertainty threatening.Footnote ⁴⁷ Or it may be one of a basket of tools employed by people with a ‘closed’ personality disposition,Footnote ⁴⁸ who find a complex and unstable social world more threatening and harder to deal with than those with a more ‘open’ disposition.Footnote ⁴⁹ Group competition can also stimulate ethnocentric thinking. People may attach to a group who share a set of goals and compete for economic resources or political influence to secure these goals. Doing so will often put them into competition with other groups seeking the same scarce resources for different goals, and the resulting conflict over resources and influence encourages ethnocentric thinking – sometimes people see an issue as a matter of ‘us versus them’ because it is, indeed, a competition between groups.Footnote ⁵⁰ Inequalities and hierarchies between groups can have a similar effect. When a group feels its privileges or resources are under threat from a competing group, or when a group feels its members do not get fair treatment or a fair share, then all group members will be more prone to see the world in terms of ‘us versus them’.Footnote ⁵¹

There are a number of key features to ethnocentrism which make it a valuable tool for understanding Britain’s recent political disruptions. Ethnocentrism is a stable personality orientation, one that varies between individuals, and can be activated among those who hold it when they perceive threats to the in-groups they care about, particularly threats from opposed and disliked out-groups. Ethnocentric voters are sensitive to threats and will mobilise politically against them, leading ethnocentrism to become a stronger predictor of political preferences and choices when such threats are present. It is the stable links between ethnocentrism and demographic features of the electorate which enable it to translate demographic divides into political conflicts; and it is the capacity of ethnocentric ‘us versus them’ thinking to be activated by threats that gives ethnocentrism its disruptive ‘flash’ potential, with large and rapid shifts in voters’ priorities and behaviour occurring when a perceived threat emerges. Both aspects are crucial to understanding the current political context, which features both long-term structural change and rapid, volatile changes in voter behaviour.

The ethnocentric worldview is stable over time. Those who express a stronger ‘us against them’ worldview at one point in time reliably express similar views if you ask them again years or even decades later. Political scientists Donald Kinder and Cindy Kam reviewed a range of different studies which went back to the same people repeatedly over many years and found that in all these studies ethnocentric attitudes were stable over time. Ethnocentrism behaved in this respect like many other political values, being shaped most in the impressionable years of youth, and becoming harder to shift thereafter.Footnote ⁵² Studies of both in-group attachments such as national identity,Footnote ⁵³ and out-group hostilities such as racial prejudice,Footnote ⁵⁴ have arrived at similar conclusions.Footnote ⁵⁵

This brings us to the second critical feature of ethnocentrism: it varies by education level and ethnic identity. Educational expansion and rising ethnic diversity have therefore opened up new divides between ethnocentric voters and those who reject their group-centred worldview. Education strongly predicts levels of both in-group attachment and out-group hostility in the white majority, as we shall see in the next chapter. The relationship between ethnocentrism and ethnic identity is a little more complex. Ethnic minorities are also prone to express hostility to out-groups – for example, large numbers express hostility to a close relative marrying someone from a different ethnic or religious groupFootnote ⁵⁶ – but they also tend to reject the group identities favoured by ethnocentric white voters, such as attachment to an ethnically defined nation. This is a consequence of the structural position of ethnic minorities, who are frequently the targets of ethnocentric hostility from the white majority. Ethnic minorities therefore understandably tend to reject the forms of identity and group attachment most attractive to ethnocentric white voters, as they are usually excluded from these and often face hostility from those who most strongly express them. While ethnic minorities are just as prone to ethnocentric thinking as the white majority, this tendency is less likely to find political expression among minority groups, who instead tend to align with white graduate ‘conviction liberals’ in an ‘identity liberal’ alliance of groups opposed to ethnocentric white ‘identity conservatives’, as we show in more detail in the next part of this chapter and in Chapter 3.

The stability of ethnocentric worldviews, and their lasting association with educational and ethnic differences, raises a paradox we need to resolve. How can stable attitudes, and very slow demographic change, account for rapid and dramatic political shifts such as those seen in Britain over the past decade? The third key feature of ethnocentrism resolves this paradox: the political influence of ethnocentric attitudes depends on the political context. Ethnocentrism becomes activated when ethnocentric voters perceive a threat, leading ethnocentric voters to rally around their threatened in-groups and mobilise hostility to the out-groups seen as posing the threat. Growing identity divides can remain latent in the electorate until ethnocentric voters perceive a threat, at which point change can happen very rapidly.Footnote ⁵⁷ The most influential examination of this threat–activation dynamic comes from Karen Stenner, who works in a tradition with long historical roots, which proposes that the tendency of hostility towards out-groups is a result of an authoritarian personality.Footnote ⁵⁸ In a series of studies Stenner demonstrates how authoritarian voters who value social order and conformity behave no differently to others when they feel secure, but their attitudes and behaviour are transformed once they perceive a threat to the things they value.Footnote ⁵⁹ Authoritarians become dramatically more intolerant towards out-groups, and they mobilise to eliminate the threat and restore the security and conformity they crave. A seemingly stable political situation can be rapidly transformed by surges in intolerance, demands for strong leadership and action against threatening out-groups.Footnote ⁶⁰

Immigration, for example, is an issue with a strong tendency to trigger such dynamic threat responses – immigrants are by definition people seeking to cross a group boundary and join a new group. Ethnocentric voters are prone to perceive immigrants as a threat to the national in-group and will mobilise to defend their national in-group when migration is a salient issue. Other issues such as national security, terrorism and war inherently involve group conflict and threats from out-groups, and all such issues tend to activate authoritarian personalities.Footnote ⁶¹ Stenner has argued that such disruptive surges in authoritarian sentiment are a structural feature of politics, and liberal democracies need to develop better mechanisms to channel and respond to them.Footnote ⁶²

Such threat-mobilisation dynamics have been observed for a long time. A striking early example comes from the research of British sociologist Margaret Stacey, who examined the social and political upheaval caused by an influx of immigrants into the small Oxfordshire market town of Banbury in the 1930s.Footnote ⁶³ Stacey found ethnocentric activation in response to threat, with Banburian locals showing the same pattern of hostile mobilisation against threatening newcomers that we see in the political conflicts over immigration now playing out on a national stage. The arrival of new people who spoke differently and looked different, the subsequent pressures on existing provision and, even more tellingly, the appearance of new and alien shops on Banbury High Street triggered ethnocentric hostilities among native Banburians, who told Stacey they felt like strangers in their own town. Yet the twist in this tale is that the immigrants who settled in Banbury in the 1930s were all fellow Englishmen and women coming to work in a newly opened factory.Footnote ⁶⁴ Yet because Banburians saw the issue in terms of a locally defined ‘us’ facing a threat from alien outsiders, their latent ethnocentric tendencies were activated, and they mobilised to defend themselves from this threat to local identity and traditions.

For insular 1930s Banburians, migrant workers coming from a few counties away were already alien enough to be seen as a threatening ‘them’. In the more mobile and globalised societies of today, in-groups and out-groups tend to be more broadly defined, but the pattern of dynamic ethnocentric response to threat remains the same. As British society has undergone two massive demographic changes, three new groups of voters have emerged: one associated with higher levels of formal education and the distinct values associated with it; the second a product of growing racial and ethnic diversification; and the third arising from a formerly dominant segment of the white majority reacting defensively to decline. We now present a thumbnail sketch of these groups, and the tensions between them, to provide a summary illustration of how tensions between these groups shape identity conflicts in current British politics. In the next chapter we will provide a more extensive account of these groups’ identities and values, and of the political arguments which mobilise them against each other.

From demographic change to political conflict: conviction liberals, necessity liberals and identity conservatives

Identity polarisation: generations and geography

Both educational expansion and ethnic transformation are structured by generation, and the oldest British cohorts have dramatically different educational and ethnic profiles to the youngest. Both identity liberal and identity conservative voters also cluster together and live apart from the other group. Graduates and ethnic minorities congregate in big cities, while white school leavers concentrate in smaller towns and rural areas. These are polarising tendencies – identity liberals and identity conservatives increasingly live and socialise among people from their side of the identity politics divide, and apart from those on the other side. Such geographical segregation also has the potential to increase the electoral impact of identity conflicts, as the British electoral system is built around competition for control of small, geographically defined constituencies. Growing numbers of these seats are dominated either by identity liberal or identity conservative voters, giving the MPs representing them a strong electoral incentive to represent the locally dominant viewpoint, and thus helping to mobilise identity conflicts into Parliamentary politics when they arise (see Chapter 5).

The generational polarisation of identity politics is illustrated in Figure 2.4, which shows the changing proportions of under forties and over seventies who belong to the core identity conservative and identity liberal demographics. In 1986, the core identity conservative group of white school leavers was a dominant majority among all age groups, though it was smaller among the youngest cohorts. The group has steadily declined since, but the retreat has been much more rapid in the youngest age groups, who have grown up since the rise of mass higher education. The share of British residents under forty who are white school leavers fell from nearly two-thirds in 1986 to less than a third in 2016. Among those over seventy – a large and high-turnout group in Britain’s ageing society – this decline was much slower. Nearly two-thirds of the oldest generations were still white school leavers in 2016. Identity conservatives are a shrinking minority in the youngest cohorts but continue to define majority opinion in the oldest cohorts. The generational rise of identity liberals mirrors the generational decline in identity conservatives. Graduates and ethnic minorities were a small minority of all cohorts in 1986, but have grown rapidly since, with the growth strongly concentrated in the youngest generations. By 1996, identity liberals made up a fifth of those under forty, by 2006 they were up to over a third, and by 2016 they had risen to nearly 40 per cent of the youngest generations, substantially outnumbering identity conservatives. As we moved into the 2010s, identity liberals were the dominant group defining mainstream opinion in the youngest generations, while remaining a minority group outnumbered by identity conservatives in all the older cohorts.

Figure 2.4 Share of under forties and over seventies belonging to the core identity conservative and identity liberal demographic groups, 1986–2016

Source: British Social Attitudes surveys, 1986–2016.

Very large generational divides have thus opened up between the youngest cohorts, where identity liberals are now dominant, and the oldest cohorts, where identity conservatives still set the tone. Such divides will be with us for a long time, because generational replacement is a very slow process. Ethnocentric older cohorts dominated by white school leavers will remain in the electorate for decades, providing a large and persistent constituency for ethnocentric politics. Conversely, while university graduates and ethnic minorities are far more numerous among the cohorts growing up since university expansion and the second wave of mass migration, it will be many years before generational replacement enables them to achieve the kind of overall electoral dominance that white identity conservatives enjoyed just a few decades ago. Both the educational and ethnic trends driving this generational replacement process are accelerating – the children attending school in Britain today are the most ethnically diverse in history, and university attendance rates continue to break records each year. As a result, the differences in the demographic composition of young and old cohorts will continue to rise in coming years, increasing the potential for generational conflict.

These differences in composition are consequential because our social lives are structured by generation – we go to school and university with people the same age, then join workplaces organised into age-structured hierarchies, with those from the same generation starting at roughly the same time and moving up the workplace hierarchies together.Footnote ⁷⁷ Peers born within a few years of each other are therefore heavily over-represented within friendship groups and social networks, and the mix of people in a generation has a big influence on our everyday social experiences. When a cohort is more ethnically diverse, diversity will seem more ‘normal’ to its white members. When a cohort has a higher share of graduates, university education will be seen as a more ‘normal’ aspiration, and the political values found among graduates will be seen as more ‘normal’ too. Generations dominated by identity liberals show, for example, much more liberal attitudes towards intermarriage between different ethnic groups, and have a strong social norm stigmatising expressions of opposition to such marriages.Footnote ⁷⁸

A similar polarising trend is evident when we look at where identity liberals and identity conservatives live. Britain’s ethnic minorities have always been unevenly geographically distributed, reflecting the legacies of early urban settlement and chain migration. Ethnic minority communities are concentrated in larger urban areas – in particular, the largest English cities such as London, Birmingham and Manchester.Footnote ⁷⁹ While levels of ethnic diversity are rising all over the country, driven by immigration, ethnic minority population growth and the internal migration of ethnic minorities,Footnote ⁸⁰ the pattern of growth in minority communities has been highly uneven. Figure 2.5 illustrates this, plotting the change in local ethnic diversity between 2001 and 2011 against the starting level of local ethnic diversity.Footnote ⁸¹ The pattern is clear: the more diverse a place was in 2001, the larger the increase in ethnic diversity it has experienced since. The ethnic minority share in the least diverse places rose on average by two percentage points in the following decade, while in the most diverse places the increase was 15 percentage points – more than seven times as large.

Figure 2.5 Local authority change in ethnic diversity 2001–11 by starting levels of ethnic diversity in 2001, England and Wales

Source: Census 2001 and 2011.

With diversity rising most where it was already high, and rising least where it began low, local identities and experiences of diversity are diverging. Britain’s ethnically diverse cities, led by London, are becoming ‘superdiverse’ places with a multitude of internally diverse migrant and ethnic minority communities,Footnote ⁸² where there is no locally dominant group and the white British are just one ethnic community among many. At the other end of the spectrum, a large part of the British population – and the bulk of older ethnocentric white voters – live in rural, small town and urban contexts which are still nearly mono-ethnic, with 95 per cent or more of the population identifying as white British, and with only modest growth in their small minority communities. The everyday experiences of diversity are diverging between communities, just as the formative experiences of diversity are diverging between generations.

Britain’s towns and cities are also diverging by education levels. Mass university education now produces two major waves of internal migration in Britain every year – the first as new students move to start university, the second when new graduates move again in search of work. Both moves tend to increase the segregation of communities by education level.Footnote ⁸³ Britain’s universities are nearly all located in large towns and cities, while the students they educate come from all over the country. Each autumn, young people depart en masse from the smaller towns and rural areas where they grew up and flow into Britain’s big cities and university towns to register for their degree studies. A few years later, when these same young people flow out of the university campuses, it is again the biggest cities that benefit, and the smaller communities that lose out. Most graduates either stay in the city where they have studied or move on to other large cities – particularly London – where graduate job opportunities are best.

Mass higher education is thus experienced by smaller communities as a massive loss of youth and potential – the higher a young student flies, the more likely they are to leave, and the less likely they are to return. The expansion of university education has ramped up the scale of this process, and as access rates approach 50 per cent Britain’s universities now annually suck the ‘best and brightest’ from every community in the country into the nation’s large urban areas, while those who leave school after GCSEs and A-levels typically remain where they are. Growing geographical segregation increases the disruptive political potential of identity conflicts in several ways. The first is a matter of composition.Footnote ⁸⁴ As Britain’s big cities and smaller communities have diverged, the common ground between different communities has shrunk and the differences between their values and priorities have grown. Where we live, like the generation we grow up in, also has a powerful and lasting influence on the social networks we form, so geographical segregation increases the contact we have with like-minded groups, while reducing everyday contact with people whose experiences and views are different. This entrenches the views that individuals encounter more frequently and marginalises opinions that, though frequently held outside the immediate peer group, or local area, are seldom encountered within it.Footnote ⁸⁵

The political impact of geographical segregation is also magnified by Britain’s electoral system. Members of Parliament are elected to represent a single area and all the voters within it, so when places become more polarised by education and ethnicity, this increases the influence of distinctive local attitudes and priorities of MPs. Legislators representing ethnically diverse and graduate-heavy seats have an electoral incentive to faithfully represent the identity liberal worldview of their constituents, while MPs in seats where white voters with low education levels are locally dominant have a similar incentive to represent the ethnocentric outlook of their local voters, even if, as we show in Chapter 5, such views are often far away from such MPs’ personal values.

From demographic change to political change: is demography destiny?

Is demography destiny? Can we predict the impact these dramatic ongoing demographic changes will have on our political system? Our answer is a resounding ‘No’, for two reasons. First, previous claims that demographic change will produce inevitable political shifts have come to grief, because they have underestimated the capacity of political actors to respond and adapt to demographic change. Four prominent examples from recent British political history illustrate this. In the wake of Labour’s fourth successive election defeat in 1959, two leading researchers asked ‘Must Labour lose?’Footnote ⁸⁶ The authors argued the likely answer was ‘Yes’, but the Labour Party begged to differ, winning four of the next five elections. British election researchers reflecting on Labour’s repeated successes in the 1970s then argued that demographic and generational change would further cement Labour’s dominance in the contests to come.Footnote ⁸⁷ Mrs Thatcher put paid to that notion. Researchers in the 1990s, writing in the wake of four Conservative election victories, once again raised the existential question, asking whether ‘Labour’s Last Chance’ had passed, with the social changes unleashed over many years of Conservative rule portending terminal decline for the opposition.Footnote ⁸⁸ As in 1960, this diagnosis arrived just a few years before its refutation, in the form of a new leader, a new strategy and renewed electoral success. Finally, researchers and party strategists argued in the early 2010s that the Conservative Party would struggle to secure a general election majority unless it could improve its appeal to Britain’s rapidly growing ethnic minority communities.Footnote ⁸⁹ A few years after these reports were published, the Conservatives under David Cameron won their first House of Commons majority in over two decades, despite failing to improve their appeal to ethnic minority voters.Footnote ⁹⁰ In all of these cases, the parties confounded predictions of doom by finding ways to adapt their appeal and shift their electoral coalitions to address the effects of demographic change. There is no inevitable read across from demographic change to political change because parties are not passive observers, but active agents who react to change and shape its political meaning.

This argument applies with particular strength to identity politics, because of the dynamic nature of identity conflicts, with voters’ preferences shifting rapidly in response to the rise and fall of perceived threats. This process provides many openings for elite actors looking to influence the political impact of identity conflicts. Parties and leaders can seek to activate or de-activate identity attachments by framing political narratives in ways that emphasise divisions between groups or focus on the values and identities that unify otherwise diverse people. As we shall see, such choices have played a large role in how parties have approached identity issues to date, and the choices they make have lasting consequences on the political form identity conflicts take. Different choices may open up new paths to resolving such conflicts in future.

For both of these reasons, we argue not that demography is destiny, but rather that demographic shifts change the electoral resources available to parties and open up the potential for new political conflicts to emerge. Demographic change creates new opportunities and new risks for parties, but does not determine the choices they make in the face of new challenges, or the consequences of these choices. Different reactions to the same trends can channel demographic change towards divergent political outcomes, as we illustrate in our analysis of the very different ways demographic divides and identity conflicts have been mobilised in Scotland (see Chapter 9). It is the choices parties have made in response to new risks and opportunities, and the mobilisation of voters in response to these choices, which have charted the road to Brexit, and beyond into the Brexitland politics of today, with the country divided as never before by identity conflicts. The story we aim to tell in the coming chapters encompasses both the changing opportunities for identity politics which have arisen from demographic change, and how the choices made by different governments have shaped the form of identity politics that has emerged. In this first section, we now move to a more in-depth review of the identity attachments and identity conflicts that define Brexitland (Chapter 3), before examining how these divides were first mobilised a generation ago during the first wave of post-war immigration (Chapter 4).

Introduction

Long-term demographic changes have driven the emergence of an identity politics divide, with two growing groups of identity liberals on one side and a declining, formerly dominant, group of identity conservatives on the other side. In this chapter we lay out in more detail the values and attitudes underpinning this divide and explain why identity conflicts between these groups are often intense and difficult to resolve. Two reactions to rising ethnic diversity have pulled the white majority in different directions: a process of gradual overall accommodation to change has been offset by rising polarisation within the white population. As we showed in Chapter 2, successive cohorts of British voters are growing up in steadily more diverse social contexts. Each generation of white voters expresses higher levels of comfort with diversity and ethnic minorities than its predecessor. A multicultural Britain is becoming a part of ‘normal’ social life, but this happens slowly because each generation’s sense of ‘normal’ is informed by the conditions when they grew up, and older generations whose norms were informed by earlier social contexts stick around in the electorate for a long time.

Both ethnic diversification and rising social acceptance of it are set to continue. As we noted in the preceding chapter, the fastest growing ethnic minority group in Britain today is the mixed ethnicity group: children with parents from different ethnic groups. Not only does this point to rising acceptance of diversity in the most intimate of social spheres, it also means the youngest cohort just arriving in the electorate features a growing group for whom ethnic diversity is a daily experience around the childhood dinner table as well as in the playground, on campus or in the office. At the time of writing (2020), the segment of the white British population who grew up before mass migration began in the 1950s, and who therefore express the strongest opposition to diversity, is elderly and declining. Conversely, the youngest voters currently joining the electorate express greater acceptance of diversity in all walks of social life than every older cohort. Within two decades, there will be virtually no voters left with any direct memory of Britain before the onset of mass migration.

While diversity is becoming more accepted overall, there are new arguments arising around the terms of this accommodation, over how and where to draw the lines between in-groups and out-groups, and what forms of group-based judgements are socially acceptable. The overall drift towards more inclusive attitudes has been accompanied by rising polarisation within the electorate. Older generations in general, and in particular white school leavers, are less exposed to diversity in their everyday lives, and often see ethnic change as threating to their understandings of British identity and culture. They favour government action to slow down or reverse this process of change. Younger generations and liberal graduates see diversity and ethnic change as both inevitable and laudable, and want to see the government focus instead on stronger action to combat the discrimination and disadvantage faced by ethnic minority groups. There are fundamental disagreements not only about the substance of policymaking in response to rising diversity, but also about how to talk about diversity. What one side sees as legitimate expressions of anxiety about the speed of change and attachment to traditional identities is criticised by the other as illegitimate expressions of prejudice. This race card politics – with fundamental disagreements over where the line is drawn in discussions of groups and group attachments – is a growing obstacle to compromise and dialogue to resolve the new political conflicts of Brexitland.

This chapter will first expand on how we measure the differences between the three identity camps, focusing on their relative tendency towards ethnocentric ‘us’ and ‘them’ thinking. Then we will tackle the question of why these differences are so hard to bridge and introduce the social norms employed by each distinct identity camp to defend their position and attack the legitimacy of their opponents. Such norms polarise discussions by denying the legitimacy of the opponent’s concerns, making compromise and even basic engagement in a meaningful debate more difficult. Ultimately, it is this lack of mutual recognition and dialogue that generates the intense and polarised disputes which characterise Brexitland identity conflicts.

Identity conservatives: ethnocentrism as a political agenda

Ethnocentrism has two central aspects: attachment to in-groups and negative attitudes towards out-groups. To track the evolution of ethnocentric attitudes and identity politics over the long run, we have sought out measures which have been asked reasonably often on long-running political and social surveys. For in-group attachments, we draw particularly on measures of national identity. The nation is one of the most salient in-group identities for voters. Who does, and does not, belong to the nation is a central question in debates over diversity and immigration in Britain, as it has been in other countries experiencing mass immigration and ethnic change. National identity is also regularly asked about in surveys, so we have relatively rich data to draw upon in examining its effects. However, the nation is unlikely to be the only identity important to ethnocentric voters, and readers should bear in mind that other forms of group attachments which are not captured in the data sources available to us are also likely to matter to ethnocentric voters. The measures we use include belief in British superiority to other nations, the protection of Britain’s culture and economy from foreign influence, and the notion that Britain should put its national interests before international cooperation. Later, we also make use of preferences for English or Scottish over British national identity, because both these forms of nationalism have strong and politically consequential ethnocentric elements.Footnote ¹

Out-group hostility comes in many forms, including negative stereotypes, feelings of threat, negative emotions and discriminatory behaviour amongst others, but as we wish to track the evolution of British politics over a long period, we are once again forced to focus our analysis on what is regularly available in existing data sources. The most frequently available measures of out-group hostility, and thus the ones we focus on, are ‘social distance’ measures capturing opposition to social contact with minority groups, self-rated racial prejudice and various measures of hostility to immigrants as an out-group.

Both in-group attachments and out-group hostility show a strong generational pattern,Footnote ² as illustrated in Figure 3.1. Older generations consistently express stronger support for an ethnically exclusive national identity, and more opposition to ethnic minority in-laws. When we have measures asked repeatedly over many years, we find these attitudes are generally stable over time within generations, while showing large and persistent differences between generations.Footnote ³ There are also deep and enduring divides by education level and ethnicity in ethnocentrism, as Figure 3.2 illustrates. White school leavers are much more likely, for example, to agree that birth and ancestry are very important markers of ‘being British’ and to agree that those who do not share British culture and traditions can never be ‘truly British’. By contrast, large majorities of white graduates and ethnic minorities reject birth and ancestry as markers of Britishness, and both groups are also more likely to reject the argument that the national in-group should exclude those who do not share British culture and traditions. However, while the link between demographics and identity attachments is strong, it is not perfect: there is a substantial minority of low-qualification whites who reject ethnocentric conceptions of the nation, and a substantial minority of graduates and ethnic minorities who express at least some support for them. The same patterns obtain for hostility to minority and migrant out-groups.Footnote ⁴

Figure 3.1 Share of people in different generations expressing ethnocentric views (percentages)

Source: British Social Attitudes, 2013.

Figure 3.2 Ethnocentric national identity among white school leavers, white graduates and ethnic minorities (percentages)

Source: British Social Attitudes, 2013.

Ethnocentric voters also have a distinctive political agenda encompassing a range of issues where groups and group conflict are salient. We illustrate this in Table 3.1, where we show the differences in the views of those who score highest and lowest on measures of ethnocentrism. Across a range of issues, including immigration, equal opportunities, views of the EU and views of devolution and constitutional reform, ethnocentric voters consistently favour stances which protect or enhance the position of their in-group, while opposing policies which protect or enhance the position of out-groups. Ethnocentric voters hold negative views of the EU, seeing it as a threatening out-group which constrains the sovereignty of their national in-groups. Ethnocentric voters are also strongly prone to negative views of immigrants and tend to oppose policies which support and protect ethnic minorities. On any issue framed as a conflict between in-groups and out-groups, ethnocentric voters will reliably line up behind the policies seen as best defending ‘us’ against ‘them’. The main exception is constitutional preferences – while ethnocentric Scottish voters show a stronger tendency to see the UK system as biased against them, and to favour reforming it or leaving it altogether, ethnocentric English voters do not (yet) express similar resentments about United Kingdom political institutions.Footnote ⁵ This could change in the future, if ethnocentric English voters come to see the other nations of the UK, or the UK’s overarching political institutions, as opponents frustrating the preferences of their in-group.

Table 3.1 Ethnocentrism and views on political issues involving group conflict

Issue	Agreement with statement, high ethnocentrism	Agreement with statement, low ethnocentrism	Difference
Immigration attitudes
Immigration should be reduced ‘a lot’	83	27	56
Migration is bad for the economy	73	17	56
Migration undermines British culture	69	16	53
Asylum seekers should not be allowed to stay	46	10	36
Britain would lose its identity if more Muslims came*	88	50	38
Britain would lose its identity if more Eastern Europeans came	82	48	34
Equal opportunities/multiculturalism
Oppose government assistance to support ethnic minority customs and traditions	72	43	29
Ethnic minorities should blend into society, not maintain customs and traditions	70	50	20
Government takes better care of ethnic minorities than the white majority*	65	30	35
Euroscepticism
Would vote to leave the EU in a referendum	62	22	40
Little or no benefit to UK from EU membership	54	23	31
UK should not follow EU decisions it disagrees with	76	40	36
Devolution/constitutional arrangements
Support for English Parliament (England)	22	19	3
Support for English independence (England)	20	13	7
England benefits more than Scotland from UK (Scotland)	37	22	15
Support for Scottish independence (Scotland)	30	20	10

Sources: British Social Attitudes, 2013; Scottish Social Attitudes, 2013 (Scottish devolution items), items marked * from British Election Study, 2010. Ethnocentrism measured using ethnic nationalism, except on questions marked with * where self-rated prejudice is used due to data constraints.

While ethnocentric attitudes inform a coherent worldview and political agenda focused on conflict between in-groups and out-groups, the lines of this conflict are not drawn in the same way by all ethnocentric voters. Ethnocentrism is a tendency to divide the world into ‘us’ and ‘them’, but the nature of the boundaries used to separate ‘us’ from ‘them’ and the political and social issues seen as ‘us versus them’ conflicts vary between individuals and evolve over time. The political context people grow up with informs where these lines are drawn. Over time, new minority groups who are initially seen as alien and threatening come to be accepted as part of a broader ‘us’, with their cultural and racial differences recognised but no longer seen as a threat.Footnote ⁶ Even within the most ethnocentric demographic groups, there is substantial potential for tension between older and younger generations who draw the lines between ‘us’ and ‘them’ differently, with older cohorts rejecting groups that younger cohorts accept. Figure 3.3 illustrates this, showing how the share of both white university graduates and white school leavers who accept the idea of an ethnic minority in-law rises steadily among younger generations who have grown up in a more diverse Britain.

Figure 3.3 Share of white graduates and school leavers accepting the idea of an ethnic minority in-law

Source: Ford (Reference Ford2008).

Why identity conflicts are polarising: a clash of social norms

The conflicts between identity conservatives and identity liberals flow from their very different ethnocentric tendencies, the status of the majority in-group and the problems faced by minority out-groups. Yet the mere existence of such differences does not explain the polarised nature of political arguments over race, immigration and other identity issues. Such debates often evoke very strong emotions because they involve strong normative claims. In fact, paradoxically one of the most polarising aspects of identity politics conflicts stems from a point of very broad agreement. There is a general social consensus that racism is a personal failing and a social evil, so those judged racist, or even accused of racism, face a substantial social stigma. Disagreement as to what constitutes racist, and thus unacceptable, behaviour therefore becomes very heated because all involved recognise that the stakes are high. The result is an emotive tug of war between identity liberals seeking to apply more expansive definitions of racism in order to expunge prejudice from society, and identity conservatives pushing back against such definitions, which they feel inhibit free expression of legitimate views and group attachments, and stigmatise them unfairly.

To illustrate how widespread social norms sanctioning racism are, we designed an experiment to compare the punishments people suggested for racist behaviour in an everyday social situation with those they would impose for other forms of social transgression. We asked respondents in a nationally representative survey to imagine that they were a shop manager, and that they need to decide what sort of disciplinary action, if any, to take against an employee who was rude to customers in various ways. We randomly varied the kind of rude behaviour they had to judge, with a total of four options tested: criticising mothers for not controlling their children; suggesting customers were too poor to buy the shop’s goods; criticising customers for using poor English; and making racist comments towards a customer. Our design was therefore testing whether there was a distinctive stigma attached to racism, in comparison with other sources of rudeness, resulting in a stronger punitive response. We offered our respondents a range of reactions of differing severity, including dismissing the employee immediately, giving them a warning, and doing nothing.

The results, presented in Figure 3.4, confirm that racist behaviour is indeed taken more seriously than other forms of social transgression. While the majority of our hypothetical managers were not keen to tolerate rudeness on any grounds, they were more likely to dish out warnings rather than dismiss employees in most of our scenarios. Only 2 per cent would dismiss an employee for being rude about a customer’s parenting. Surprisingly, given Britain’s class divisions and the salience of debates over food banks and poverty in recent years, only 4 per cent would dismiss an employee for ridiculing a customer’s poverty. As we expected, the sanctioning of rudeness based on race is much more severe, reflecting the strength of anti-racism norms. Thirty-five per cent of respondents recommended firing the employee for making racist comments, twice as many as would react to rudeness based on language skills. This sharp rise in support for the most punitive action was mirrored by a sharp decline in those saying they would take no disciplinary action at all – while many people were happy to let rudeness about naughty children slide, virtually no one was willing to let racist comments or rudeness about language skills pass without taking action.Footnote ²⁴

Figure 3.4 Share of respondents who would dismiss an employee for different forms of rudeness to a customer (percentages)

Source: YouGov survey commissioned by the authors, March 2018.

The vast majority of imaginary shopkeepers, fully 97 per cent, saw racism as a transgression they should punish, an impressively broad social consensus, though people varied in how severe they thought the punishment should be. As Figure 3.5 illustrates, the strength of the sanction people apply is closely related to their own anti-prejudice norms. Those who expressed the strongest motivation to police their own prejudices also applied the toughest sanctions to others: nearly half of those at the top of the anti-prejudice scale would dismiss the racially prejudiced employee. Meanwhile, those who did not express a strong need to control their own prejudices were also more forgiving of others’ transgressions – less than a third of those at the bottom of the anti-prejudice scale would dismiss the racially offensive employee, while roughly a quarter would take no action at all.

Figure 3.5 Predicted probability of dismissing the employee and of taking no action by levels of motivation to control prejudice

Source: YouGov survey commissioned by the authors, March 2018.

Anti-racism norms are thus a potent force in society – people are more willing to take strong action against racist behaviour than against other forms of anti-social behaviour, and very few people were willing to ignore racism entirely. A perceived violation of anti-racism norms can therefore do real harm to someone’s social position or even their livelihood. To be called a racist in today’s Britain is a powerful stigma that can shut people off from the respect and support of peers and colleagues and put reputation and employment at risk. Identity conservatives are particularly aware of the power of the label ‘racist’, which they often resent as a means used by liberals to shut down discussion and marginalise their concerns.Footnote ²⁵ This is why the common use of the phrase ‘I’m not a racist, but …’, while widely perceived as being a prelude to saying something that is in fact racist, actually reflects invocation of a shared recognition that racism is unacceptable. It is an attempt to reach out to those on the other side of the conversation, seeking common ground and trying to neutralise an anticipated hostile response to views the speaker worries may be seen as contentious. Yet it usually has the opposite effect, alerting the other side to the fact that views they reject will be expressed. The roots of this social tension lie in an ongoing struggle to settle the boundaries separating beliefs and behaviour that should be stigmatised as racist from more benign expressions of group attachment and judgements about others. The result is a tug of war between identity liberals, who seek to broaden definitions of prejudice and strengthen the social norms stigmatising its expression, and identity conservatives seeking to defend what they see as legitimate expressions of group preferences and group attachment.

Many examples of this tug of war can be found in the political debate over immigration. Identity liberals frequently frame anxieties over migration inflows or calls for greater control as either direct expressions of prejudice, or efforts to legitimate racist or xenophobic motivations. A former Labour immigration minister attacked UKIP campaign posters in 2014 criticising the economic effects of immigration, featuring white workers in hard hats, as ‘racist’,Footnote ²⁶ while in 2010 Labour Prime Minister Gordon Brown dismissed a voter who expressed similar anxieties about migrants from the EU as a ‘bigoted woman’.Footnote ²⁷ When Labour opposition leader Ed Miliband made a pledge to introduce immigration controls, complete with branded merchandise, this was met with a torrent of negative commentary from identity liberals, who attacked the pledge as ‘pandering to racism’,Footnote ²⁸ and years later continued to reference ‘Ed’s racist mug’ as an example of Labour appeasing prejudice.Footnote ²⁹ Actions labelled as racist by identity liberal commentators and political activists have included: expressing anxiety about or opposition to immigrationFootnote ³⁰; displaying symbols of English national identity such as the George Cross flagFootnote ³¹; voting for UKIP in 2014 and 2015Footnote ³²; and campaigning or voting for Brexit in 2016.Footnote ³³ In all of these cases, and more, the strategy employed by identity liberal campaigners has been to expand the definition of racism to include these actions directly or to achieve the same goal indirectly by ascribing racist motives to ambiguous behaviours.

Identity conservatives show the opposite tendency – looking to defend expressions of in-group attachment or hostility towards certain out-groups as expressions of ‘legitimate concerns’Footnote ³⁴ and to exclude them from the unacceptable label of racism. A few examples from across the political spectrum illustrate this approach. Nigel Farage, leader of UKIP and a prominent figure in the Leave.EU campaign, defended claiming that many families would not want Romanian neighbours by asserting this highlighted ‘real concerns’ driven by Romanian migrants’ involvement in organised crime.Footnote ³⁵ He similarly defended the deployment in the EU Referendum campaign of a poster depicting thousands of Middle Eastern refugees on the Croatia–Slovenia border in 2015, under the slogan ‘Breaking Point’, by claiming the poster was ‘the truth’ and ‘an example of what is wrong inside the European Union’.Footnote ³⁶ Boris Johnson, Prime Minister at the time of writing, defended an August 2018 article calling it ‘weird and bullying’ for Muslim women to wear face-covering veils, then went on to compare women who did so with ‘letterboxes’ or ‘bank robbers’, as an example of politicians ‘speaking directly’ about voters’ concerns, something he said the public wanted to see.Footnote ³⁷

The defence of identity conservative stances as ‘legitimate’ is often combined with criticism of overly expansive identity liberal definitions of racism, a pattern of argument found frequently among Labour politicians seeking to defend efforts to reconnect with identity conservative voters against accusations of pandering to racism from their own identity liberal activists. In 2016, Labour candidate for Mayor of Greater Manchester Andy Burnham accused his party’s campaigners of ‘avoiding people’s eyes and shuffling away’ when voters raised concerns about immigration and attacked the tendency to label and stigmatise those who raised such concerns:

The emotionally heated character of such political arguments once again reflects the high stakes in this debate. There are political risks to drawing the line too narrowly or too broadly. An overly restrictive definition of racism will underestimate its prevalence and seriousness as a social problem, weaken the case for political and social action to tackle it, and could legitimate behaviour which harms the lives and interests of minority groups. Yet there are also risks to an overly expansive definition of prejudice. If the term ‘racist’ is applied too broadly by one side in political debates, it may lose its sting – being seen less as a fundamental and universal taboo and more as a tool of political rhetoric. As the definition of racism and racist behaviour is broadened, it can become diluted, weakening the social consensus for action against prejudice. At the extreme of this would be a situation where radical identity liberals posit that all white people are racist, perhaps irredeemably so, with more identity conservative white voters responding by treating prejudice as a personality quirk to be tolerated, rather than a social injustice in need of urgent correction.Footnote ³⁹ Identity conservatives have also liberalised over time, with younger generations expressing much more comfort with racial diversity than their parents, and it can therefore be alienating for them to be criticised as bigots by their identity liberal peers. They may feel, with some justification, that the goalposts are being moved, as despite adopting more tolerant views, they are still criticised as narrow-minded by identity liberals.

Identity conservative voters stung by such criticism often look to contest the expanding application of anti-racism norms by laying the counter-charge of ‘political correctness’ – arguing that some complaints of racism reflect overly draconian and inflexible social rules on speech and behaviour. Many people on this side of the Brexitland divide often think enough has now been said and done on the issue of racial inequalities (despite empirical evidence that these persistFootnote ⁴⁰) and see the zeal of anti-racist campaigners as excessive and oppressive, a view expressed through phrases such as ‘political correctness gone mad’. Identity liberals see this invocation of ‘political correctness’ as a defensive deflection, used to belittle or dismiss claims about prejudice, and undermine support for policies aiming to combat it. Both sides of the identity politics divide have developed rhetorical tools to undermine their opponents’ arguments, while seeking to impose their own framing on identity conflicts. Identity liberal activists attempt to police the terms of identity conflicts by deploying a more expansive definition of prejudice and stigmatising their opponents’ objections as intolerance. Identity conservatives counteract this by deploying a narrower definition of prejudice and saying it is instead ‘politically correct’ identity liberals who are being intolerant, by seeking to marginalise those with different views.

The charge of political correctness does not carry the same sting as an accusation of racism, but it may nonetheless be influential. We tested this possibility in a second survey experiment fielded to a representative sample of voters. We presented respondents with a question about the value of diversity in London, but half were also given a statement dismissing positive views of diversity as ‘the politically correct thing to say nowadays’. By comparing those randomly assigned to see the ‘political correctness’ treatment with those who did not, we could test whether flagging up ‘political correctness’ as a counter-argument would persuade people to be more openly critical of London’s diversity. As Figure 3.6 illustrates, this was indeed what happened. The share saying London benefitted from diversity drops 9 per cent, while the share saying diversity either made no difference or made London worse off rose. The balance of support in the population shifts from overall rating diversity as beneficial to London to being evenly split between supporters and critics. Making ‘political correctness’ salient thus seems to achieve for identity conservatives some of what making anti-racism norms salient achieves for identity liberals – shifting opinion about which views are legitimate to express. The positions people are willing to adopt in identity conflicts depend on how the conflicts are framed. Each side has cards to play in this – voters adopt more liberal positions if they think anti-racism norms are truly at stake, but some will also adopt more conservative positions if the invocation of ‘political correctness’ calls the relevance of these anti-racism norms into question.

Figure 3.6 Views about the benefits of diversity, with and without ‘political correctness’ counter-argument (percentages): Is London better off, or worse off because of ethnic diversity?

Source: YouGov, 2015.

When we look at who is most responsive to the political correctness prompt, we find an initially surprising pattern. Those who express the strongest motivation to control prejudice, and were in the previous experiment keenest to fire a racist employee, are also most prone to express neutral or negative views more frequently when praise of diversity is framed as ‘political correctness’.Footnote ⁴¹ This fits with the idea that those strongly motivated to follow anti-prejudice norms are therefore highly sensitive to the framing of arguments about diversity, and will also respond to cues which indicate that the expression of inclusive attitudes is not normatively required. Those motivated to control prejudice are thus responsive to the standards set by wider society, and will shift their stance in response to both liberal and conservative shifts in the framing of identity arguments. This raises the stakes further in the constant dialogue over the social acceptability of attitudes about groups and group attachments. The support for ethnic minorities’ rights and policy responses to ethnic inequality will often depend on which of the two social norms prevails: anti-racism or anti-political correctness.

Conclusion

This chapter has unpacked the vexing question of how group attachments and views of racial and ethnic diversity define and divide identity conservatives and identity liberals. We have outlined the ethnocentric attitudes that form the basis of the identity conservative worldview: attachment to narrowly drawn in-groups and strong suspicions of outsiders. We have also shown how the worldview of identity liberals goes beyond simply lacking such ethnocentric predispositions, and involves embracing diversity and a focus on protecting the rights of racial and religious minorities. We have examined why it is so hard for these two groups to settle their differences, or even to engage in a constructive dialogue over identity issues. The high normative stakes of identity politics arguments make them inherently polarising. As a result, differences in worldviews can quickly become questions of legitimacy and moral worth, with both sides incentivised to police and penalise perceived infractions of social norms, and to contest the definitions and rhetoric employed by their opponents.

There is a very broad political and social consensus that racial prejudice is wrong, and that those who hold racially prejudiced views should be stigmatised. As such, the stigma attached to bigotry is powerful: accusations of racism can and do end political careers, and the perception that a party or policy is racist can erode support for it.Footnote ⁴² Yet while there is broad agreement that racism is a social evil and racists should be punished, there is no such consensus over how to define prejudice or how to sanction those who express it. The result is a tug of war between identity liberals, who seek to broaden definitions of prejudice and strengthen the social stigma attached to it, and identity conservatives, who seek to defend what they see as legitimate group preferences and criticise the excessive extension of anti-prejudice norms as itself an instance of intolerance. Any expression of group attachment or group judgement can become part of this ‘race card politics’ – with identity liberals seeking to stigmatise such expressions as unacceptable prejudices, while identity conservatives defend them as a legitimate expression of group identities and anxieties. Identity conservatives also frame these debates in emotive and polarising terms, attacking identity liberals as intolerant ‘politically correct’ zealots who use politicised accusations of racism to stifle the expression of legitimate viewpoints and stigmatise those who hold them.

This polarisation of debates over identity matters. When both sides are interested in policing the behaviour of their own social group, and seeking to apply their own normative framings to identity debates, people become more focused on the symbols and rhetoric used in the debate and cease to consider the substance of the issues they are debating. The emotional heat that identity conflicts generate makes coalition-building across identity divides is more difficult. Having laid out the general dividing lines between identity conservatives and identity liberals, we now turn to consider how the conflict between them first became activated by the issue of immigration. The origins of more recent identity conflicts over immigration and other identity politics issues, and of the political parties’ reputations on these issues, lie in the heated political arguments which took place in the 1960s and 1970s, during the first wave of post-war migration. This is the story we tell in the next chapter.

Introduction

Political conflicts triggered by native hostility to newcomers are nothing new in British politics. As we saw in Chapter 2 with the defensive reactions of 1930s Banburians to migrants moving from elsewhere in Britain to work in their town, the activation of ethnocentrism does not even require immigrants to cross national borders. But while any influx of outsiders can trigger ethnocentric reactions, the deepest divides and most lasting conflicts have come over international immigration and the rising ethnic and racial diversity that successive waves of it have generated. This chapter tells the story of why this is so and how it came to be. We examine the first wave of sustained non-white migration to Britain from the 1950s to the 1970s, showing how conflicts over this migration became mobilised into politics. The choices taken during this wave of migration set up an identity politics alignment in the electorate, and this alignment in turn has shaped more recent identity politics conflicts over immigration. The dilemma facing this earlier generation of politicians will be familiar to those following the contemporary migration debate – policymakers agreed a liberal policy regime and unwittingly triggered an influx of migrants, then faced pressure to restrict this inflow when it activated ethnocentric hostility among the white majority, while at the same time a pressing need emerged to protect the new migrant communities from this ethnocentric hostility. The choices politicians made in response to these conflicting demands had a lasting impact, aligning identity conservative voters with the Conservative Party, which came to be seen as more willing to control the ‘threat’ from immigration, and aligning identity liberal voters with the Labour Party, which came to be seen as more willing to protect the rights of migrants and minorities.

There are three parts to the story, which parallel and foreshadow events in the decade leading up to the EU Referendum of 2016. The first is a large and persistent elite–mass gap on immigration, which led more liberal and cosmopolitan political elites to introduce reforms granting extensive migration rights to a large population in order to improve Britain’s international position, while underestimating the scale and intensity of public hostility this would trigger. In the first wave of immigration, the goal was to secure Britain’s place at the heart of a post-Imperial community of nations – the Commonwealth – and open borders between Commonwealth members was seen either as a valuable goal in itself,Footnote ¹ or as an acceptable price to pay to secure lasting political influence within this community. More than fifty years later, another identity liberal-dominated political elite came to very similar conclusions when considering whether to fully open Britain’s borders to migrants from the post-Communist countries acceding to the EU. In both cases, the unintended consequence of these decisions was a surge in migration as far more people opted to exercise newly granted free movement rights than political elites had anticipated, activating ethnocentric hostilities in the native electorate who perceived the new migrants as a threatening out-group.

The second parallel between the two waves is that the ethnocentric sentiments activated by migration were successfully mobilised by political actors arguing for more radical migration restriction policies. As public opposition to immigration grew, the policy response from elites constrained by a commitment to an open borders principle and unwilling to alienate migrant-sending countries was piecemeal and slow in coming. Substantial migration continued for a number of years, and public concern remained high, but without an effective mainstream political outlet, until a new political actor mobilised ethnocentric voters behind more radical proposals and transformed the political situation. The appearance of a credible electoral threat broke the logjam, pushing one of the main political parties to embrace more radical migration restrictions, breaking with their earlier commitment to uphold open borders principles. In the 2010s, this was the story of UKIP’s rise, as identity conservatives frustrated with successive governments’ inability to control migration turned to the radical right and eventually forced the Conservatives to offer an option to exit Britain’s open border arrangements with the EU via a referendum on Brexit. The story played out in a similar way in the 1960s as identity conservative voters, frustrated with repeated governments’ unwillingness to control Commonwealth migration, turned to Enoch Powell’s radical right insurgency, which eventually forced the Conservatives to concede radical reforms which effectively ended the migration rights of most Commonwealth citizens.

The final similarity between the two periods is that both also involved a substantial counter-mobilisation by identity liberals opposed to the rise of radical right actors and seeking to protect migrant minorities from ethnocentric intolerance. In the first wave, committed identity liberals within the Labour Party were pivotal in pushing through the first race relations legislation – writing anti-racism norms into British law, and laying the groundwork for a longer-run project of re-imagining Britain as a multicultural society where minority cultures are celebrated and minority rights protected. This, too, is already finding its echo in the Brexitland political cycle, with a shift towards pro-migration attitudes since 2016,Footnote ² and the emergence of activist groups devoted to protecting the rights of EU migrants and fighting the oppressive ‘hostile environment’ rules applied to migrants by the Home Office since the mid-2010s. While these movements have not, as yet, had the kind of lasting legislative impact that the proponents of race relations legislation had in the 1960s, they have already shifted the balance of power on migration by activating and politically mobilising pro-migration sentiments among the much larger contemporary identity liberal electorate.

The origins of the first wave: the entanglement of citizenship and Empire

The story of the first wave begins with the British Nationality Act (BNA 1948) of 1948, one of the most liberal pieces of citizenship and migration legislation passed by a Western democracy. The BNA 1948 defined British citizenship for the first timeFootnote ³ and did so in very expansive terms. A common citizenship with identical rights was conferred on all residents of Britain and of the current and former territories of the British Empire, including the vast and populous Indian subcontinent.Footnote ⁴ Eight hundred million people across the globe acquired full British citizenship rights, including the right to settle and work in Britain, and to participate in British mainland politics from the moment they arrived.Footnote ⁵

Given this remarkable openness, it is rather surprising that facilitating mass migration was not a goal, or even an expected effect, of the BNA 1948 legislation. Instead, its Parliamentary authors aimed to cement Britain’s political status at the heart of an open and integrated Commonwealth of former imperial states. While close links with the former Empire were seen as essential to Britain’s future prosperity and influence, mass immigration was not expected to be part of that equation, nor were all parts of the former Empire seen as equally important. The emphasis of the political elite was on maintaining close relations with the white colonial settler societies of the ‘Old Commonwealth’ – Canada, Australia and New Zealand.Footnote ⁶ In the decades prior to the BNA, the primary circulation of people within the Empire had been between Britain and these countries, and the BNA 1948 aimed to protect this system by confirming unrestricted rights to migrate to and from Britain and the Commonwealth. It was ‘a fundamentally backward-looking document reaffirming the status quo as it had existed for decades’.Footnote ⁷ The desire was to preserve economic and political connections between Britain and the diverse global network of territories it had developed over centuries under the aegis of Empire in a new post-Imperial era of independent Commonwealth states.

It was not possible to preserve this right for the white settler states of the ‘Old Commonwealth’ while excluding the black and Asian majority Commonwealth states without writing an explicit ‘colour bar’ into the legislation, something identity liberal politicians, crafting legislation just years after a world war against a racist dictatorship, were unwilling to consider. British legislators therefore conferred a single, undifferentiated set of citizenship rights on all residents of Imperial and Commonwealth territories. The policymakers who thus opened up the opportunity to migrate to Britain to hundreds of millions of people in Caribbean, Asian and African territories did not, however, give much consideration to what might happen if large numbers chose to exercise this right. The issue of migration to Britain from the current and former Imperial colonies was not mentioned once in the extensive committee and Parliamentary debates on the BNA.Footnote ⁸ Yet, as labour shortages developed in Britain’s post-war economy, rapidly expanding numbers of black and Asian Commonwealth citizens began to exercise their rights, moving to Britain in search of better work and higher incomes. The first inflows came from the West Indies, beginning with the arrival of the famous Empire Windrush with hundreds of Jamaican migrants seeking work, just months after the passage of the 1948 Act.Footnote ⁹ As the 1950s progressed, the numbers grew and migration diversified, with flows from the West Indies augmented by arrivals from India and Pakistan.Footnote ¹⁰

Public opposition to migration in the first wave

As Commonwealth migration flows increased, ethnocentric sentiments in the electorate were activated and strong public opposition began to manifest itself. Polling is sparse in this period, but the evidence available underscores that public opposition to ‘coloured’ migration, as it was then called, was intense and widespread from the outset (see Figure 4.1). Close to 90 per cent of poll respondents supported strong restrictions on Commonwealth or ‘coloured’ migration, and around 70 per cent expressed approval of the first restrictive legislation passed by the Conservatives in 1962. The share of the public who supported the BNA 1948 policy of full Commonwealth migration rights typically sat at around 10 per cent, while substantial parts of the public were supportive of very restrictive measures such as banning family reunion migrationFootnote ¹¹ or state-sponsored repatriation of settled migrants.Footnote ¹² This opposition was, from the outset, racially discriminatory – the overwhelming focus of public attention and hostility was migration from the West Indies and the Indian subcontinent – indeed ‘coloured’ migration was the issue pollsters typically asked about, rather than Commonwealth migration in general.Footnote ¹³

Figure 4.1 Opposition to immigration and support for migration restrictions, 1961–6 (percentages)

Source: Archive of historical immigration polling compiled by Professor Will Jennings.

One unusual survey conducted during this period provides a stark illustration of the discriminatory nature of public opposition to immigration. In 1967, Gallup pollsters asked the public identical questions about the benefits and harms from Commonwealth and Irish migration. Flows of migrants from Ireland were at this point as substantial as settlement from the entire Commonwealth combined,Footnote ¹⁴ and, unlike Commonwealth migration, Irish migration remained unrestricted at the time of the survey, so if public concern was driven by the actual pressures generated by migration then opposition to Irish migrants should be as high as, if not higher than, opposition to Commonwealth migrants. Yet, as Figure 4.2 reveals, Commonwealth migrants attracted much stronger public opposition than Irish migrants. Three in five voters felt Britain had been harmed by the settlement of Commonwealth migrants, while only a fifth felt that way about the Irish migrant population. Racially different Commonwealth migrants activated ethnocentric hostility in a way that white Irish migrants did not.

Figure 4.2 ‘Do you think on the whole this country has benefitted or been harmed through immigrants coming to settle here from the Commonwealth/Ireland?’

Sources: Gallup (1967); Professor Jennings historical polling archive.

This strength of the hostile reaction to Commonwealth migration reflects the demographics of the 1960s British electorate, which was dominated in this period by identity conservatives. White school leavers – the core identity conservative demographic group – formed a large majority of the population. These voters were consistently much more likely to express ethnocentric hostility to ‘coloured’ migrants, and to support policies which would halt migration to Britain or repatriate already settled migrants, as Table 4.1 below illustrates by showing education divides in various immigration questions. University graduates, who were at a time a tiny minority, were much less likely to express hostility to Commonwealth migrants, and large majorities of graduates opposed all the draconian migration restriction policies proposed during this period. The migrants of the first wave faced more intense and widespread racially motivated hostility because Britain in the 1960s and 1970s was a society more dominated by the ethnocentric demographic groups most prone to such hostility.

The elite–mass divide on immigration in the first wave

Political arguments about migration in the first wave, like those today, were seldom a matter of narrow economic costs and benefits, but were a clash of outlooks between more cosmopolitan political elites concerned to maintain Britain’s status in the international community and a more ethnocentric electorate opposed to the settlement of outsiders they found threatening. For Britain’s post-Imperial elite, the people living in Britain’s Commonwealth were part of an ideologically constructed ‘us’ stretching across the former Imperial territories, a community of interest defined by a common history. Britain was at the centre of a global network, so Britain should have a globalised form of citizenship which crossed continents, knitting together all those with a political and historical bond to the country.

The British public did not share this view – their sense of ‘us’ was much more narrowly defined, racially and territorially. ‘Us’ for the British public of the 1950s and 1960s was white British people born and resident in Britain. Migrants from the Caribbean and south Asia were not part of any in-group they recognised, and they saw no reason why people from thousands of miles away should have an unrestricted right to join their national and local communities. This ethnocentric opposition to Commonwealth migrants was for the most part not softened by the economic reality of post-war labour shortages,Footnote ¹⁵ or the major contribution Commonwealth citizens made to the war effort. This illustrates how immigration debates are chronically prone to activate ethnocentric concerns about groups and group conflict, which cannot be resolved with technocratic claims about economic or foreign policy benefits. In this way, too, arguments during the first wave of migration resembled, and influenced, those during the second wave.

Debates over Commonwealth migration exposed deep divides between the identity liberal minority and the majority not only over the issue of who to let in and on what terms, but also over whether migration was a political priority at all. While liberal university graduates, and much of the policymaking elite, saw the arrival of relatively modest numbers of black and Asian migrants as a trivial matter, large parts of the electorate – in particular identity conservatives – reacted with intense hostility to Commonwealth immigration from the moment it began. While most identity liberal politicians, many of whom had fought Nazi racism in the Second World War, abhorred the use of race or ethnicity to judge migrants, many of their ethnocentric constituents felt just as strongly that racial and ethnic differences were a legitimate basis for restricting migration.

These tensions between identity liberals and conservatives over immigration in the first-wave period divided the parties internally, driving a wedge between the political elites and electorates of both Labour and the Conservatives. Both parties’ ruling elites tended to have stronger attachments to the Commonwealth, an intense and widely shared social norm sanctioning racial prejudice and discrimination, and a tendency to see migration and open borders pragmatically in terms of political and economic benefits. The support bases of both parties differed in all these regards – there was little attachment to Empire or Commonwealth amongst the mass electorate, whose primary loyalty was to a narrowly drawn sense of national identity defined by ancestry and birth. Social norms sanctioning expressions of racism were weak or absent in this period – as seen, for example, in the widespread and explicit use of discrimination in rental housing (‘no dogs, no blacks, no Irish’), and in popular culture – 1970s television sitcoms regularly featured racial stereotypes and insults, which viewers would see as outrageous and unacceptable just a decade or two later.Footnote ¹⁶ The mass electorate, and the mass membership of both parties,Footnote ¹⁷ expressed a strong preference for white over non-white migration, with many wanting the latter completely stopped or reversed.

The first evidence of the political power of such ethnocentric sentiments came in the 1964 general election campaign in Smethwick, where the Labour Shadow Foreign Secretary Patrick Gordon Walker – a prominent campaigner against migration restrictions – lost his seat to an obscure Conservative candidate following a racially charged campaign, featuring leaflets using the slogan ‘If you want a nigger for a neighbour, vote Labour’. As the defeated Labour MP left Smethwick town hall after the count, Tory supporters yelled after him: ‘Where are your niggers now, Walker?’ and ‘Take your niggers away!’Footnote ¹⁸ While there was growing evidence of the disruptive power of this activated ethnocentric hostility, the leadership of both parties remained reluctant to respond to it. Early politicians who explicitly mobilised such concerns, such as Smethwick winner Peter Griffith, were ostracised by their fellow MPs and shunned by the parties’ leadership figures.Footnote ¹⁹ While many MPs in both parties were privately worried about rising public hostility to black and Asian migrants, openly articulating or sympathising with such ethnocentric sentiments was taboo.

The political activation of ethnocentrism: Enoch Powell and ‘Rivers of Blood’

The dam finally broke when, for the first time, a prominent member of the Conservative Party elite – Shadow Cabinet member Enoch Powell – broke the taboo and articulated in full the identity-based hostility to migration widely shared in the electorate in the infamous ‘Rivers of Blood’ speech in April 1968, which used emotive rhetoric and lurid imagery to attack liberal Commonwealth immigration policies:

Powell was fully aware that such an open and visceral violation of anti-racism social norms would provoke a strong reaction from his colleagues: ‘I can already hear the chorus of execration … how dare he say such a thing? How dare I stir up trouble and inflame feelings …?’ However, he defended his stance by arguing that the growing opposition of white ethnocentric voters to migration was both legitimate and too important to ignore: ‘The answer is that I do not have the right not to do so … I simply do not have the right to shrug my shoulders and think about something else. What [my constituent] is saying, thousands and hundreds of thousands are saying and thinking …’Footnote ²¹

The chorus of execration Powell anticipated was indeed swift to arrive. Conservative leader Edward Heath repudiated Powell’s position, sacked him from the Shadow Cabinet, and never spoke to him again.Footnote ²² This move was overwhelmingly supported by his senior Shadow Cabinet colleagues – four of whom threatened to resign themselves unless Powell was dismissed. Heath cited the ‘racialist tone’ of Powell’s speech as the reason for his sacking, which he called ‘unacceptable from one of the leaders of the Conservative Party’ and ‘liable to exacerbate racial tensions’. The Times, newspaper of record for the British ruling class, denounced Powell’s speech as ‘evil’, calling it ‘the first time that a serious British politician has appealed to racial hatred in this direct way in our post-war history’.Footnote ²³

The public response was quite different – two-thirds of voters said Heath was wrong to sack Powell, and over 75 per cent said they agreed with his views on immigration.Footnote ²⁴ Powell received thousands of letters in support,Footnote ²⁵ and overnight became the most widely known Conservative politician after Prime Minister Heath himself.Footnote ²⁶ Polling in the months before the speech already showed a large majority believed controls on immigration were not strict enough, while a substantial minority backed a ‘total ban on coloured immigration’. Support for both policies rose in the wake of Powell’s intervention (see Figure 4.3). There was also lower, but still widespread, public support for Powell’s more controversial and draconian proposals – including banning family reunion and the repatriation of settled migrants.

Figure 4.3 Support for Powellite positions on immigration before and after ‘Rivers of Blood’ speech in April 1968

Source: Jennings, Opinion polls database, 2018.

Enoch Powell transformed the debate over immigration, mobilising ethnocentric identity conservatives and driving a wedge between liberal anti-racist political elites and the mass electorate. Powell’s interventions were also the point at which a lasting divide emerged in the parties’ reputations on immigration. Before Powell, the two main parties were seen by voters as rather similar on the issue. New migration restrictions in 1968 had been introduced by the Labour government, and the Conservative leadership – which had distanced themselves from Powell’s stances – showed little initial interest in further action. But after ‘Rivers of Blood’, it was Powell who made the running in the migration debate. Powell continued to sit as a Conservative MP, and his association with the Conservatives led voters to see them as a party favouring strict and racialised immigration control. Powell’s strident and hostile language also forced a stronger response from Labour’s leadership, pushing the party into stronger defences of ethnic minorities’ rights. After Powell’s interventions, voters saw a clear divide between the Conservatives as the party of migration restriction and opposition to diversity, and Labour as the party of liberal migration policy and multiculturalism.Footnote ²⁷

We can trace the emergence of this divide in the British Election Study (BES) surveys. In 1964 and 1966, a majority of respondents, when asked which party was more likely to stop immigration, said ‘neither’, suggesting most voters had noticed the cross-party identity liberal consensus against strict immigration control. This changed in the wake of ‘Rivers of Blood’. Nearly six out of ten respondents to the 1970 British Election Study saw the Conservatives as more likely to halt immigration, compared with just 4 per cent who named Labour.Footnote ²⁸ Indeed, thanks to Powell, voters in 1970 saw the Conservatives’ immigration policy as a great deal more restrictive than it actually was. This is illustrated in Figure 4.4, which shows the share of voters whose perceptions about the parties’ immigration policies were more restrictive than reality, accurate or less restrictive than reality. Some 58 per cent of voters in 1970 inaccurately claimed the Conservatives’ policy was either to totally halt further immigration (36 per cent) or to repatriate settled migrants (22 per cent), whereas only a fifth correctly identified that Conservative policy was to ‘allow immediate families and a few skilled workers’. As the share of voters who personally supported such restrictive policies was higher still (50 per cent backed a total halt, 20 per cent backed repatriation), these misperceptions, driven by Powell’s rhetoric, were electorally valuable, enabling the Conservatives to attract support from ethnocentric identity conservatives. Indeed, the party was able to have its cake and eat it on immigration: Powell’s widely reported polemics on the issue signalled a restrictive stance on migration, but by holding him at arm’s length the party leadership could avoid fully committing itself to a strongly anti-immigration stance that would violate elite anti-racism norms and jeopardise relations with the Commonwealth. The balancing act worked,Footnote ²⁹ bringing big gains for the Conservatives among the most ethnocentric and pro-Powell voters,Footnote ³⁰ despite the party leadership’s official disapproval of Powell and continued opposition to his most draconian proposals.

Figure 4.4 Voters’ perceptions of Conservative and Labour immigration policies, 1970

Source: British Election Study 1970. The study offered respondents four options on migration policy: repatriation of settled migrants and stopping all existing migration were coded ‘more restrictive than reality’ for both parties; allow immediate family and a few skilled workers was coded as ‘accurate’ for both parties as it is the closest analogue to the proposals both made; while allowing new workers and families free entry were coded as ‘less restrictive than reality’ for both parties as both intended to keep in place the strict quotas on labour migration imposed since the mid-1960s, while the Conservatives proposed further restrictions on top of this.

While the Conservative leadership was initially uncomfortable with Powell’s high profile and fiery rhetoric, they were unable to resist the pressure for migration restriction produced by his campaigns. The party pledged to introduce new immigration restrictions during the 1970 election campaign, and Heath fulfilled this pledge within a year of taking office with the passage of the 1971 Immigration Act (IA 1971). This maintained the concept of Commonwealth citizenship, but stripped it of practical meaning by creating two classes of Commonwealth citizens: ‘patrial’ citizens, with an unrestricted right to abode in Britain; and ‘non-patrial’ citizens, who had no such automatic right. Patriality was awarded to all those with a British-born parent or grandparent, a provision that, in effect, introduced a ‘colour bar’ while avoiding explicit recognition of race in immigration policy. Patriality ensured continued access to Britain for most white Commonwealth citizens (who typically had at least one British-born grandparent), while excluding most non-white Commonwealth citizens. However, to head off identity liberal criticism of racial discrimination, the Act also awarded ‘patrial’ status automatically to all Commonwealth citizens who had lived in Britain for more than five years, along with their families. Most already settled Commonwealth migrants of all ethnic origins therefore retained full residence rights, though in practice no effort was made to formally document such rights and thus secure them against future challenge, storing up problems which would emerge decades later in the form of the ‘Windrush’ scandal.Footnote ³¹

Heath’s 1971 legislation attempted to address the anxieties of identity conservative voters, and ensure their continuing support for the Conservatives, by stripping most black and Asian Commonwealth citizens of their migration rights. However, the legislation failed in its political goal before it was even implemented. On 7 August 1972, Idi Amin’s regime in Uganda issued a decree giving tens of thousands of south Asian residents with British Commonwealth citizenship just ninety days to leave the country.Footnote ³² While the previous Labour government had abandoned Kenyan Asian Commonwealth citizens when they found themselves in a similar position four years earlier, leaving them stateless, Edward Heath opted to uphold the anti-racist and pro-Commonwealth principles espoused by the pre-Powell political elite, in defiance of public opinion and despite vehement opposition from Powell himself. Heath pledged to fully honour the passport rights of Ugandan Asians to settle in Britain. He defended the choice as a matter of principle: ‘[We have] no choice but to stand by Britain’s obligation …’Footnote ³³ A massive airlift was organised to safely remove Ugandan Asian citizens with British passports, and the Heath government pursued an intensive diplomatic effort to ensure those without such passports would find a safe haven in other countries ahead of Amin’s deadline.Footnote ³⁴

Nearly 30,000 Ugandan Asian refugees were admitted to Britain in a matter of weeks. The unexpected and rapid arrival of large numbers of ethnic Indian refugee migrants was a scenario likely to provoke an intensely hostile reaction from ethnocentric white voters, and Enoch Powell wasted no time in looking to once again mobilise such sentiments, this time against his own party. Powell led the political campaign against the Ugandan Asians, repeatedly attacking his own government’s policy – for example, accusing the Attorney General of ‘prostituting his office’, for supporting their claims.Footnote ³⁵ Heath, like his Labour predecessors, discovered that Powell’s speeches were more important than Westminster legislation in driving media headlines and public perceptions on immigration. Although Heath had passed the restrictive IA 1971 just a year earlier, the Ugandan Asians crisis and Powell’s renewed anti-immigration campaign turned ethnocentric voters against the Conservatives, who were now seen as being ‘soft’ on immigration. Heath gained no credit for his restrictive reforms from ethnocentric voters, who instead were now being mobilised against him by one of his own backbenchers.

The consequences of this backlash are clear in the 1974 British Election Study. As we have seen, most voters in 1970 thought Conservative immigration policy was more restrictive than it actually was, thanks to Powell’s anti-immigration polemics. Now, with Powell campaigning against his own party for being too soft on immigration, public sentiment swung the other way, as Figure 4.5 illustrates. The share of voters who thought the Conservatives favoured the strictest migration control policies – repatriation or a total halt to immigration – fell from 58 per cent in 1970 to 36 per cent in 1974. Conversely, the share who thought Heath’s Tories favoured large-scale new migration or uncontrolled entry of migrants – options never entertained by the Heath government – tripled from 8 per cent to 24 per cent. Meanwhile, the share of the electorate believing Labour favoured liberal migration policies fell, despite the party’s strong support for Heath’s stance on the Ugandan Asians and its opposition to the IA 1971, particularly the discriminatory ancestry rules.Footnote ³⁶ The Conservatives’ electoral advantage on immigration restriction largely disappeared, even as the Conservatives’ immigration policy shifted in a restrictive direction. What Powell gave, Powell could take away. With the electorate again overwhelmingly in favour of strong restrictions on migration (73 per cent favoured repatriation or a total halt to migration in 1974), the loss of this restrictive reputation was costly. The Conservatives’ 1970 lead among the most ethnocentric voters disappeared in 1974. While immigration was far from the only issue on the agenda in the turbulent mid-1970s, it is quite possible that Edward Heath’s principled act of generosity to the Ugandan Asians, and Enoch Powell’s fiery criticism of this generosity, contributed to the Conservatives’ narrow defeat in the two elections of 1974.

Figure 4.5 Prevalence of misperceptions about Conservative and Labour migration policies in 1970 and 1974 (percentages)

Source: British Election Studies, 1970 and 1974. See note to Figure 4.4 for details of coding.

After Powell: the consolidation of an identity politics divide

With the Conservatives no longer perceived as committed to immigration control after the Ugandan Asian crisis, space opened up for new parties to exploit ethnocentric sentiments. In another parallel with the politics of the second wave of immigration, the radical right surged in the late 1970s by attracting identity conservative voters who had lost faith in the government’s ability to control immigration. The extreme right and openly racist National Front, emerging as Britain’s fourth largest party in the mid-1970s,Footnote ³⁷ foreshadowed the later turn to the BNPFootnote ³⁸ and UKIP following a similar loss of public faith in the government of the late 2000s.Footnote ³⁹ In both cases, Conservative leaders sought to win back migration sceptics with new and stronger promises of control – David Cameron’s ‘tens of thousands’ pledge and Theresa May’s ‘red line’ on EU free movement echoing the earlier bid by new Conservative opposition leader Margaret Thatcher to win back ethnocentric voters in the late 1970s.

Thatcher, elected as leader in 1975, was influenced both by Enoch Powell’sFootnote ⁴⁰ views and by the public reaction to them. She recognised from the outset the disruptive political power of immigration, and the rewards the issue could provide to politicians able to articulate and mobilise ethnocentric anxieties. Her reflections on the issue in her memoirs are worth quoting at length:

Thatcher’s thoughts bring together several aspects of the identity politics conflict over immigration. She, like Powell before her, was aware of the strong anti-racism norm among the political elite but, again like Powell, she did not believe such norms were shared by most voters. She attacked those propagating such norms as hypocritical – demanding acceptance of migrants while living in areas unaffected by their arrival – and defended the hostile sentiments expressed by those who she argued had to live with the disruptive consequences of migration. This account, written in 1995 and reflecting on political disputes from decades earlier, seeks both to legitimate the political mobilisation of ethnocentric sentiments and to undermine those who sought to stigmatise such mobilisation as a violation of anti-racism norms. It could easily have been delivered by a UKIP politician or Brexit campaigner defending ethnocentric opposition to mass migration twenty years later.

Mrs Thatcher’s sympathy with the ethnocentric sentiments of identity conservative voters was also evident when she was leader of the opposition. In a widely reported interview with Granada’s ‘World in Action’ programme in January 1978, Thatcher expressed sympathy with British voters who were ‘rather afraid that this country might be rather swamped by people with a different culture’.Footnote ⁴² This had an immediate impact on public perceptions, just as Powell’s interventions had ten years earlier. The share of voters who regarded the Conservatives as most likely to stop immigration, which had languished at around 30–35 per cent since the Ugandan Asians crisis, jumped to over 50 per cent after the ‘World in Action’ interview, and remained at this level through to the 1979 election.Footnote ⁴³ In the British Election Study conducted after the 1979 election this figure rose to over 60 per cent, including two-thirds of Conservative identifiers and more than half of Labour identifiers – the highest figures recorded since the studies commenced in 1964. Thatcher, like Powell, thus proved able to shift public views of the Conservatives’ immigration stance by employing strong restrictionist rhetoric, aligning herself with the concerns of electorally dominant identity conservatives. Unlike Powell, however, either due to her more careful tone or her more elevated position as Conservative leader, she was able to activate such ethnocentric sentiments without suffering a significant political cost from violating elite anti-racism norms.Footnote ⁴⁴

While Thatcher made no specific policy commitments in her 1978 intervention, once elected she moved quickly to enact fundamental reform of British citizenship and immigration policy, abandoning the framework established in the 1948 BNA and replacing it with a new conception of citizenship and national identity based on heritage. The 1981 British Nationality Act (BNA 1981) not only severed Britain’s citizenship links with its former colonies, it also ended an even longer-standing citizenship principle – the ius soli principle under which, since 1914, children born on British territory had an automatic right to British citizenship. Following the BNA 1981, children born in Britain to non-citizen residents have to register to obtain British citizenship, and can acquire citizenship only if they can meet residence or parental citizenship conditions.Footnote ⁴⁵ The provision, like the IA 1971, stored up problems for later, as there remained a widespread assumption that people born in Britain automatically acquired British citizenship,Footnote ⁴⁶ and many migrant families therefore made no effort to compile the documentation needed to secure their children’s citizenship rights later on. Children born to non-citizen parents after BNA 1981 came into force, who have lived in Britain their whole lives, have often been shocked to discover once they turned eighteen that the British state regarded them as migrants who could be subject to Home Office control and exclusion from some public services.Footnote ⁴⁷

The radical reforms of the BNA 1981, and the subsequent sustained drop in migration to Britain,Footnote ⁴⁸ cemented the links between ethnocentric attitudes and Conservative support which Mrs Thatcher had re-forged.Footnote ⁴⁹ As Figure 4.6 illustrates, those expressing ethnocentric attitudes were consistently more likely to also express a Conservative partisan identity throughout the Thatcher–Major governments of the 1980s and 1990s, a link which holds even after adjusting for the other demographic and attitudinal differences between ethnocentric voters and others.Footnote ⁵⁰ Conversely, those expressing racial prejudice or other ethnocentric attitudes, such as opposition to immigration, were consistently less likely to hold Labour partisanship throughout this period. Meanwhile, the rapidly growing ethnic minority electorate showed a strong and persistent alignment to Labour, reflecting the campaigns against them and their parents by Powell and Thatcher, along with Labour’s passage of race relations legislation. This powerful and lasting alignment began to tilt previously competitive or Conservative-leaning seats in England’s largest cities decisively towards Labour as ethnic minority populations grew.Footnote ⁵¹

Figure 4.6 Conservative advantage and Labour disadvantage in party identification among ethnocentric voters

Source: British Social Attitudes, 1983–1996. Measure of ethnocentrism is ‘Do you think of yourself as prejudiced against people of other races?’ The same pattern is found with other measures of ethnocentrism (see the Online Appendix: www.cambridge.org/Brexitland).

Mobilisation on the left: entrenching anti-racism norms

Both Conservative and Labour governments passed restrictive immigration reforms during the first wave, some with black and Asian migration as the clear focus of policymakers’ attention. Unhappiness with this approach was widespread among identity liberal elites and resulted in a counter-mobilisation of those most strongly committed to anti-racist norms. Identity liberals successfully pressed Labour into passing a series of Acts writing anti-racist norms into law, through pioneering race relations legislation. This legislation steadily expanded to cover more areas of life, including housing, employment and public services, and the anti-racism political debate moved from simply outlawing racially motivated discrimination to enabling fast-growing ethnic minority groups to retain many of their customs and accommodating their religious requirements. As the political and policy debate moved towards cultural recognition and providing ethnic groups with special exemptions from usual legislation (such as wearing motorcycle helmets by SikhsFootnote ⁵²), Britain became one of the European leaders in implementing multiculturalism policies.Footnote ⁵³ Unlike many continental counterparts, but similar to the United States, British law was early to recognise indirect, as well as direct, discrimination (1976 Race Relations Act). Crucially, the choice of dealing with discrimination through the civil, and not the criminal, justice processes has meant that the burden of proof in discrimination cases has been lighter, and made it easier to raise a complaint. This choice was directly modelled on the US system of legislation,Footnote ⁵⁴ and the early entrenchment of anti-racism norms in law has played an important role shaping the subsequent path of policy- and law-making in the area of race in Britain.Footnote ⁵⁵

The emergence of a distinct identity liberal political coalition was also catalysed by events in the decades between the two waves of immigration. One flashpoint was the racially motivated murder of teenager Stephen Lawrence in 1993. This case was subject to a formal inquiry headed by Sir William Macpherson, instigated in 1997 by the Labour government following a sustained and broad-based public and media campaign mobilising anti-racism norms.Footnote ⁵⁶ The Macpherson report received wide press and public attention and a generally sympathetic response. It was influential in prompting further legal protectionsFootnote ⁵⁷ against what Macpherson called ‘institutional racism’,Footnote ⁵⁸ and a sustained campaign, unusually led by the socially conservative tabloid press, to prosecute and jail Lawrence’s suspected murderers.Footnote ⁵⁹ Often dubbed the ‘murder that changed Britain’, the tragic death of Stephen Lawrence not only further entrenched legal protections against discrimination, but also catalysed the spread of anti-racism norms through society and politics. From mainstream news media revealing and scrutinising institutional racism in the police, Home Office and political parties, to parties across the political spectrum naming tackling racism as a key policy priority,Footnote ⁶⁰ the case highlighted the potential for counter-mobilisation of anti-racism norms in response to extreme expressions of ethnocentric hostility. Although, as we showed in the previous chapter, the public do not always rally behind these anti-racism norms, by the 1990s they were no longer a preserve of the political elites, as they had been during the first wave of migration.

Conclusion: why did liberal immigration policies persist for so long?

Given the large majorities opposed to black and Asian migration throughout the first wave, the puzzle posed by this period is not that conflicts over immigration arose, but that relatively liberal rules governing Commonwealth migration were maintained for so long. British policymakers in 1948 granted a huge portion of the world’s population rights to reside and work in Britain. It took them fourteen years to begin restricting these rights, and thirty-five years to completely curtail them. Vestiges of this liberal citizenship regime remain even today – citizens of Commonwealth countries (and Ireland) retain the right to vote and stand in British general elections from the day they arrive in Britain. This is a right very few new migrants enjoy in other developed democracies, and a right not shared by other migrants to Britain, who cannot vote in general elections without first obtaining British citizenship.

Three overlapping factors explain this persistence. The first is legal and institutional path dependence. The 1948 choice to grant Commonwealth residents citizenship rights acted as a constraint on political elites who were reluctant to ignore or unilaterally abandon the obligations the citizenship regime created. Policymakers’ refusal to write discrimination into law by creating different and unequal classes of citizens was a key factor delaying the introduction of Commonwealth migration restrictions in the 1950s, and the obligations of the state to all citizens were invoked by Edward Heath to justify his decision to assist the Ugandan Asians in 1973. We see similar path dependence emerging again some fifty years later following the EU A8 enlargement of 2004. The decision to align Britain with a large international structure – this time the EU – once again led the government of the day to forgo immigration restrictions and triggered an unforeseen influx of migrants that was opposed by ethnocentric voters. Once the decision was made, political elites again felt themselves bound by legal and normative obligations to respect citizenship rights, this time the free movement rights of EU citizens, limiting their ability to respond to rising public concern (see Chapters 6 and 7).

The second factor was foreign policy. Britain’s political elites regarded close political, economic and diplomatic links with the Commonwealth as a key policy goal. With the sun finally setting on the British Empire, policymakers saw integration and cooperation across a post-Imperial ‘Anglosphere’ as the best way to renew Britain’s place in the world. Immigration control was therefore resisted by political elites as likely to cause frictions with Britain’s Commonwealth partners in the short run and weaken the bonds between Britain and its former colonies in the long run. Their successors forty years later were similarly reluctant to make a major push for reform to migration in the EU because, once again, this conflicted with central foreign policy goals. In both periods, political elites saw domestic disquiet over immigration as an acceptable price to pay for maintaining close connections with key international partners.

Finally, elite opposition to immigration control was not just a matter of policy constraint or expediency. It also reflected deeply held and widely shared elite social norms sanctioning racial prejudice and discrimination. The political classes of the 1950s–1970s included many men who had personally fought in a global war against a racist, genocidal dictatorship. Veterans of that war abhorred the racism they associated with their Nazi adversaries, and those perceived to be mobilising similar dark forces in domestic politics were ostracised. Thus, Peter Griffiths, who ran an openly racist constituency campaign in 1964, was dubbed a ‘Parliamentary leper’ by the prime minister following his election, and never served in ministerial office. Enoch Powell, the first senior politician to mobilise ethnocentric opposition to migration, correctly anticipated that it would end his career in the Conservative leadership, but perhaps did not anticipate that it would also end several longstanding friendships with Conservative colleagues. Even Margaret Thatcher, the archetypal dominant and domineering prime minister, was successfully discouraged from public interventions on migration by the normative objections of Cabinet allies.

While such social norms did not, in the end, prevent politicians in either party from introducing and then extending racially discriminatory migration controls, they were still consequential. Anti-racism norms acted as a brake on the political mobilisation of hostile public sentiment by the mainstream parties, and slowed and diluted the policy responses to this ethnocentric sentiment. Normative concerns also motivated Labour politicians to balance anti-immigration legislation with equalities legislation, entrenching anti-racism norms in law. And such norms acted as a sign of things to come. Racial equality was already a core personal value for university graduates in this period and anti-racism norms would therefore only grow in political significance as university expansion dramatically increased the share of graduates in the electorate.

The BNA 1981, which came into force in 1983, brought the political story of the first wave of immigration to a close. This was a complex story of grand imperial ideals and unintended consequences, of noble stands and messy compromises, as governments of both parties wrestled with an issue that divided them internally, and where their political and ethical instincts often strongly diverged from the strongly anti-migration stance of the electorate. Yet the political legacy of these conflicts was simpler: a clear and lasting divide in the parties’ reputations on immigration and diversity. The Conservatives, thanks to the strident and long-remembered stances of Powell and Thatcher in particular, became seen as the opponents of ethnic diversity and supporters of tight immigration control. The Labour Party, despite inconsistent and sometimes unprincipled positions on immigration, emerged as the party of identity liberals, migrants and minorities – in part, thanks to their passage of race relations legislation, but also simply by being the main opposition to the party of Powell and Thatcher, and thus the only viable alternative for those threatened by Conservative mobilisation of white ethnocentric hostility. This partisan alignment over race and ethnocentrism was still in place when immigration once again began to disrupt politics in the 2010s and, as we shall see, it played an important role in shaping the political impact of these new disruptions.