## Yann Algan, Alberto Bisin, Alan Manning, and Thierry Verdier

Print publication date: 2012

Print ISBN-13: 9780199660094

Published to Oxford Scholarship Online: January 2013

DOI: 10.1093/acprof:oso/9780199660094.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use. date: 29 January 2020

# Cultural Integration in the United Kingdom

Chapter:
(p.260) 8 Cultural Integration in the United Kingdom
Source:
Cultural Integration of Immigrants in Europe
Publisher:
Oxford University Press
DOI:10.1093/acprof:oso/9780199660094.003.0008

# Abstract and Keywords

This chapter compares a relatively wide range of outcomes for the main ethnic minorities in the UK with the outcomes for white natives. The chapter also compares the outcomes for the foreign and UK born. The indicators we look at are fertility, marriage, and divorce rates, interethnic marriage, spousal age gaps, the gender gap in education, employment rates, national identity, religiosity, and language use. The chapter finds substantial heterogeneity across ethnic minority communities but also evidence that in almost all dimensions and for all groups, the UK-born minorities are closer to white natives than the foreign born.

# 8.1 Introduction

The UK has had a much longer history of large-scale immigration than many other European countries. For a long time there was a certain smug satisfaction that its generally tolerant and accommodating approach to cultural diversity had been relatively successful, although there is no doubt that problems of racism persisted. But this self-satisfaction has, in many quarters, now turned to alarm that some immigrant groups are not following the stereotypical immigrant path of economic and cultural integration into mainstream society. But, while views on this topic are often very strongly held, the evidence base is often weaker than one would like. That is what we seek to address in this chapter.

The plan of the chapter is as follows. The next section summarizes very briefly the history of immigration into the UK since 1945, the policy towards integration and the voluminous existing literature on the economic and social circumstances of ethnic minorities in Britain. The third section provides details about the data used in our analysis and presents some descriptive statistics as background for our findings in subsequent sections. The fourth section studies fertility, the fifth marriage and divorce, the sixth the gender gap in educational attainment, the seventh female employment, and the eighth values like national identity, religiosity, and language.

All in all, we find considerable heterogeneity across ethnic minority communities along the outcomes considered. However, we also find (p.261) evidence of a marked change in all these areas and this change is always in the direction of the behaviour of the indigenous British.

# 8.2 A brief history of immigration and integration policy in the UK since 1945

## 8.2.1 Immigration

Compared to many other European countries, the UK began to experience sizeable immigration much earlier, starting fairly soon after 1945. In the 1950s immigrants from the Caribbean and in the 1960s from the Indian sub-continent arrived, primarily as workers to help alleviate labour shortages. As the economy worsened in the 1970s, there were fewer economic migrants, though there was a steady trickle through family reunification and the 72,000 Ugandan Asians expelled by Idi Amin. However, as the economy improved again in the 1990s, there was a return of economic migration, with sizeable inflows from Eastern Europe (especially after the enlargement of the EU) and from Sub-Saharan Africa. In addition, the 1990s saw sizeable inflows of refugees. The proportion of immigrants in the UK population is now at its highest level since 1945 and the immigrant population is very diverse—for a summary of the ethnic minority population see Peach (1996).

## 8.2.2 Integration policy

By European standards, the UK began to wrestle with the question of how best to incorporate immigrant populations into society very early. What emerged as the dominant idea (essentially a form of ‘multiculturalism’) is well-summarized by the following quotation from the Home Secretary Roy Jenkins in 1966:

I do not regard [integration] as meaning the loss, by immigrants, of their own national characteristics and culture. I do not think that we need in this country a ‘melting pot’, which will turn everybody out in a common mould, as one of a series of carbon copies of someone’s misplaced vision of the stereotyped Englishman…I define integration, therefore, not as a flattening process of assimilation but as equal opportunity, accompanied by cultural diversity, in an atmosphere of mutual tolerance.

This led to early (by European standards), legislation against discrimination (the first law being the 1965 Race relations Act) and a generally sympathetic attitude to allowing cultural and religious exemptions to laws and practices, for example allowing Sikh motorcyclists to wear (p.262) turbans instead of helmets and Muslim policewomen to wear the hijab on duty. There was a belief that if natives were hospitable to immigrants, the minorities would, in return, come to feel part of the wider community—just one big happy family. The reality was often far from this rosy picture, as there were riots in many British cities in the early 1980s and various organizations, notably the police, have been widely criticized for institutional racism.

But more recently there has been a feeling that this strategy of multiculturalism has failed to create a common core of values, primarily because it offered minorities more than it asked from them in return and that some communities chose not to integrate into the wider society. Events like the London bombings of 2005 have shocked people into thinking something has gone badly wrong. For example, the chairman of the Commission for Racial Equality (a non-departmental public body aimed to tackle racial discrimination and promote social equality, currently merged into the new Equality and Human Rights Commission) argued in a TV interview that multiculturalism was leading to segregation, saying that ‘too many public authorities particularly [are] taking diversity to a point where they [are] saying, “actually we’re going to reward you for being different, we are going to give you a community centre only if you are Pakistani or African Caribbean and so on, but we’re not going to encourage you to be part of the community of our town”’. The reaction has included not just a wringing of hands but also substantive changes to policy—immigrants becoming citizens now have to pass a test on language, culture, and history designed to mould their values into those deemed appropriate.

## 8.2.3 Existing literature on immigrants and ethnic minorities

There is a vast amount of research on the ways in which the economic and social circumstances of ethnic minorities in Britain differs from that of the indigenous white population.2 The earliest papers on economic outcomes (most commonly measured as earnings, employment, and unemployment) were probably Chiswick (1980) and Stewart (1983). Since then, there have been many studies, considering diversity in the ethnic minority experience (see Blackaby et al., 1997; Modood et al., 1997; Clark and Drinkwater, 2007; Elliott and Lindley, 2008 inter alia), the difference between first and second-generation immigrants (e.g. (p.263) Blackaby et al., 2002, 2005), the importance of language fluency (Leslie and Lindley, 2001; Lindley, 2002a; Dustmann and Fabbri, 2003), rates of integration (Bell, 1997; Clark and Lindley, 2006), the role of religion as opposed to ethnicity (Lindley, 2002b), and differences in time-use (Zaiceva and Zimmermann, 2007). These studies have given us excellent snapshots of the position of different ethnic minorities. In particular, earnings and employment penalties are typically found to be largest for the Pakistanis and Bangladeshis who are among the most economically disadvantaged groups in British society.

But, there is much less in the way of research into how this is changing over time. This is probably due to the fact that many ethnic minority populations in Britain are of relatively recent origin so that, until very recently, it has been hard to say anything very precise about trends. But there are a number of recent studies that do explicitly address the question of changes over time. Lindley et al. (2006) investigate how women’s employment rates among ethnic minorities have been changing, paying particular attention to the changing role of education. Clark and Drinkwater (2007) compare data from the 1991 and 2001 censuses, looking at the way in which employment and unemployment rates have changed for different ethnic minorities. They find little change in the gap between the employment rates for Pakistanis and Bangladeshis on the one hand and whites on the other. Similar persistence in employment disadvantage is found in Berthoud and Blekesaune (2007) using General Household Survey data from 1974 to 2003 and in Dustmann and Theodoropoulos (2010), who, however, report more pronounced inter-generational improvements on educational achievement for ethnic minorities compared to white natives. Georgiadis and Manning (2011) look at how the behaviour of ethnic minority communities is changing over time, taking an approach somewhat similar to that used here but with a narrower range of variables.

The main contribution of our study is that we complement and extend the existing literature in two ways: (1) we present evidence on the differences between white natives and each of the main UK ethnic minorities for a wide range of outcomes, some of which haven’t been considered by other studies and (2) we document patterns of change in the behaviour of ethnic minorities over time by comparing the outcomes for the foreign and UK born, with the evidence suggesting convergence of behaviour of all ethnic minorities towards that of white natives.

# (p.264) 8.3 Data and background

The main data used in this chapter comes from the Labour Force Survey (LFS) for the years 2000–2008 inclusive. This is the main UK household survey for the collection of information on economic activity. It is an address-based household sample, with each household being interviewed for five successive quarters and one-fifth being replaced each quarter. The LFS contains information on country of birth but no information on country of parental birth for the UK born. This means that we cannot, strictly speaking, identify first-generation Britons, that is, the children of immigrants. This is different from every other chapter in this book. The standard practice in UK research, which we follow here, is to use self-defined ethnicity as a measure of being a first-generation Briton. Therefore, the analysis of the descendants of immigrants is restricted to ethnic minorities. For the sample period under analysis it is reasonable to assume that almost all of the non-white UK born have at least one immigrant parent, though this assumption will become less true in future years.3

Table 8.1 reports the sample proportions for natives, first-generation immigrants, and second-generation immigrants for the UK, using the current standard classification of ethnicity in UK surveys.4 First-generation immigrants represent around 8.6 per cent of the sample, of which half (49.4 per cent) are of white ethnicity, 11 per cent are from India, 7.6 per cent Black African, and 6.5 per cent from Pakistan. The share of second-generation immigrants (those who are UK born but their ethnicity is not White British) in the sample is 6.6 per cent, of which the largest groups are ‘other white’ (27.4 per cent), Indian (14 per cent), Pakistani (13.2 per cent), and Black Caribbean (10.8 per cent).

The differences in the fraction of the ethnic minority communities who are UK born largely reflect the fact that they arrived in the UK at different times. Black Caribbean immigration into the UK began earliest (in the 1950s), followed by Pakistanis and Indians,5 who began to arrive in (p.265)

Table 8.1 Ethnicity and place of birth composition of the Labour Force Survey 2000–2008.

Ethnic origin

Foreign born

UK born

White native

0

84.8

Other

8.6

6.6

of which (%)

Other white

49.4

27.4

Black Caribbean

4

10.8

Black African

7.6

6.5

Other Black

0.3

1.5

Black mixed

0.6

7.6

Indian

11

14

Pakistani

6.5

13.2

2.9

4.4

Other Asian

4.8

2.4

Chinese

2.8

1.7

Other mixed

1.1

6.1

Other

8.7

4.1

Note: Data source is the Labour Force Survey (LFS) 2000–2008. Proportions are computed using individual sampling weights. The other white category also includes foreign-born white British.

large numbers in the 1960s. The Bangladeshi and Chinese communities are more recent, so have the lowest proportion of UK born among adults.

In the analysis that follows, we exclude some ethnic groups because the sample sizes are too small or because the groups are too heterogeneous for analysis to be reliable. We exclude the two mixed categories (that are mostly UK born), and the four ‘other’ categories (other white, other Asian, other black and other) as they are very heterogeneous. This leaves us with seven groups for our analysis—white natives, Indian, Pakistani, Black African, Black Caribbean, Bangladeshi, and Chinese.

# 8.4 Specifications

As in other chapters, the specification we are estimating is the following:

$Display mathematics$

Note that this specification assumes that birth cohort and other regressors (typically age and education) have the same effect on the outcome for all ethnicity and nativity groups. We do have evidence from other research (Georgiadis and Manning, 2009) that this is not true but we want to have a consistent specification across all country chapters. (p.266) Because white natives are the vast majority of all samples, the coefficients on birth cohort and other regressors are going to be mostly influenced by the white native coefficients. The coefficients on the ethnicity and nativity dummies will then be close to what one would get from an Oaxaca decomposition assuming white native coefficients.

# 8.5 Fertility and marriage

## 8.5.1 Fertility and age at first child

In this section we consider the two outcomes related to fertility—the number of children and age at first birth. The LFS is not ideal for investigating fertility, as it does not ask retrospective questions about the number of children a woman has had. The best we can do is to exploit its household-based structure to see the number of dependent children who are living with a woman. As children will tend to leave the family home at some point, older women will be seen to be living with fewer children just because their children are older. So, we restrict our sample to women aged between 18 and 40, to capture the youngest ages at which women are likely to have children and an age when few women’s children will have left home. To capture the fact that, for many women in this age group, fertility will not be completed fertility, we include a polynomial in the age of the woman as explanatory variables (these coefficients are not reported). We also control for education (which has a negative effect on fertility). We include dummy variables for each of our ethnic groups, interacted with whether the individual is UK or foreign born. The results in Table 8.2 are reported relative to white UK-born women.

All ethnic minority groups, with the exception of the Chinese, have higher fertility rates than white native women. But it is also striking that, for all ethnic groups, fertility rates are lower among the UK born compared to the foreign born. For example, foreign-born Pakistani women have 0.83 more children than white natives but UK-born Pakistani women have 0.45 more. For Bangladeshis, the foreign born have 0.98 more children, but the gap falls to 0.31 for the UK born. For Black Africans, the foreign born have 0.4 more children but the gap falls to 0.18 for the UK born. For Indians, fertility among the UK born is not significantly different from the white natives.

Table 8.3 now considers age at first birth, which we compute by taking the current age of the woman minus the age of their oldest child in the household. There are similar problems with this measure as with our measure of number of children but it probably gives the right (p.267)

Table 8.2 OLS estimates of the number of dependent children for females by ethnicity and place of birth.

Ethnicity

Foreign born

UK born

White native

Reference

Indian

0.216***

0.043

(0.023)

(0.023)

Pakistani

0.833***

0.447***

(0.034)

(0.033)

0.984***

0.309***

(0.045)

(0.065)

Black Caribbean

0.013

0.020

(0.056)

(0.027)

Black African

0.399***

0.178***

(0.029)

(0.056)

Chinese

−0.166***

−0.192***

(0.035)

(0.062)

R2

0.211

Observations

541,234

Note: Data source is the Labour Force Survey (LFS) 2000–2008, the sample is all women aged between 18 and 40 inclusive. Dependent children are all children below 16. Controls include age, age squared, and education, clustered standard errors at the individual level in parentheses,

*** significant at 1%,

** significant at 5%.

Table 8.3 Estimates of the age of the mother at first birth by ethnicity and place of birth.

Ethnicity

Foreign born

UK born

White native

Reference

Indian

−1.95***

−0.32

(0.15)

(0.17)

Pakistani

−2.3***

−1.88***

(0.16)

(0.17)

−4.2***

−1.93***

(0.18)

(0.4)

Black Caribbean

−1.01**

−1.01***

(0.4)

(0.2)

Black African

−1.58***

−1.28***

(0.17)

(0.41)

Chinese

1.53***

1.47**

(0.33)

(0.7)

R2

0.211

Observations

539,278

Note: Data source is the Labour Force Survey (LFS) 2000–2008, the sample is all women aged between 18 and 40 inclusive. A censored regression model is estimated, controls include age, age squared, and education, clustered standard errors at the individual level in parentheses,

*** significant at 1%,

** significant at 5%.

(p.268)

Table 8.4 Estimates of the probability of marriage/cohabitation for women by ethnicity and place of birth.

Ethnicity

Foreign born

UK born

White native

Reference

Indian

0.330***

−0.113***

(0.027)

(0.017)

Pakistani

0.412***

0.106***

(0.020)

(0.020)

0.437***

0.045

(0.023)

(0.038)

Black Caribbean

−0.110**

−0.229***

(0.047)

(0.017)

Black African

−0.011

−0.248***

(0.025)

(0.028)

Chinese

−0.004

−0.117

(0.039)

(0.061)

R2

0.113

Observations

128,294

Note: Data source is the Labour Force Survey (LFS) 2000–2008, the sample is all women aged between 18 and 40 inclusive. A linear probability model is estimated, controls include age, age squared, and education, clustered standard errors at the individual level in parentheses,

*** significant at 1%,

** significant at 5%.

impression.6 Table 8.3 reports the estimates and one sees a similar pattern to that seen in Table 8.2. With the exception of the Chinese, ethnic minority women are younger at first birth than white native women, but the difference is smaller for the UK born. On both these measures, fertility seems to be moving towards the white native pattern.

## 8.5.2 Marriage and divorce rates

We next consider marriage patterns (see Berthoud, 2005, for an existing analysis). In Table 8.4 we report estimates for the probability for currently being married or cohabiting. Our sample is women aged between 18 and 40 so our models can be thought of as estimating the difference in marriage rates across women in these age groups. We control, as before, for age and education. For the foreign born, Table 8.4 shows that all those from South Asian communities are very much more likely to be married than white native women. However, this gap falls dramatically for the UK born, even becoming negative for UK-born Indians and only remaining significantly positive for Pakistanis. Black Caribbean (p.269)

Table 8.5 Estimates of the probability of divorce/separation for women by ethnicity and place of birth.

Ethnicity

Foreign born

UK born

White native

Reference

Indian

−0.051***

0.013

(0.004)

(0.010)

Pakistani

−0.014**

0.071***

(0.007)

(0.012)

−0.025**

−0.0001

(0.010)

(0.026)

Black Caribbean

0.167***

0.129***

(0.015)

(0.016)

Black African

0.209***

0.113***

(0.010)

(0.028)

Chinese

−0.007

−0.021

(0.010)

(0.025)

R2

0.019

Observations

916,963

Note: Data source is the Labour Force Survey (LFS) 2000–2008, the sample is all non-single women aged between 18 and 40 inclusive. A linear probability model is estimated, controls include age, age squared, and education, clustered standard errors at the individual level in parentheses,

*** significant at 1%,

** significant at 5%.

women are less likely to be married than white women, especially for the UK born, where the differential is 23 per cent. One also notes that UK-born Black African women have much lower marriage rates than white natives.

These differences in marriage rates may indicate different propensities to marry (or cohabit) in the first place, or differences in divorce and separation rates. To investigate the latter, Table 8.5 considers the fraction of ever-married women who are currently divorced or separated. As the married category includes those who have divorced and remarried, this will be an under-estimate of those who have ever divorced but probably gives the right picture. For the foreign born, those from the South Asian communities are significantly less likely to be divorced, whereas the Chinese are as likely to be divorced as white natives. But Black Caribbeans and Black Africans are significantly more likely to be divorced, by more than 15 percentage points for both ethnic groups. However, among the UK born significant differences in divorce/separation rate appear for the Pakistanis, Black Caribbeans, and Black Africans, who are more likely than white women to be divorced. The observation for Pakistanis is particularly interesting but does chime with some who have written that the practice of taking a spouse from Pakistan—which remains very common—results in marriages that do not last.

(p.270)

Table 8.6 Proportion of exogamous individuals by gender, ethnicity, and place of birth.

Ethnicity

Men

Women

White native spouse

Non-white native spouse

Total

White native spouse

Non-white native spouse

Total

White native

0

3.6

3.6

0

3

3

Foreign born

Indian

5.4

4.6

10

4.7

4.7

9.4

Pakistani

3.4

4

7.4

2.3

3.8

6.1

1.7

3.4

5.1

1

3.4

4.4

Black Caribbean

22.8

10.1

32.9

14.7

9.7

24.4

Black African

10

12.8

22.8

9

8

17

Chinese

9.8

6.2

16

28.6

8.8

37.4

UK born

Indian

14.4

5.8

20.2

15.7

7.2

22.9

Pakistani

7.8

7.3

15.1

3.7

6.8

10.5

8.3

20.8

29.1

6.7

1.7

7.4

Black Caribbean

46.7

15.8

62.5

30.4

14

44.4

Black African

22.5

16

38.5

10.6

19

29.6

Chinese

53.6

13.1

66.7

70.4

7.6

78

Note: Data source is the Labour Force Survey (LFS) 2000–2008. Proportions are computed using individual sampling weights.

These marriage patterns suggest, in line with other evidence, that the South Asian communities are relatively conservative in their marital patterns, with women typically marrying relatively young and divorce being relatively rare, while Black Caribbeans have been much less conservative. One would also see this pattern if one differentiated between cohabitation and marriage—cohabitation would be rare among South Asians and much more common among Black Caribbeans. However, as for fertility, there is a clear indication that differences in behaviour are falling.

## 8.5.3 Inter-ethnic marriage

Perhaps the most striking way in which communities can converge culturally is by marrying outside their own ethnic group (see also Coleman, 1994). Table 8.6 reports the fraction of each community that is married to someone of a different ethnicity. We also report the fraction of individuals who are married to white natives. We report exogamy rates separately for men and women as there are some interesting differences.

(p.271)

Table 8.7 Estimates of probability of exogamy by ethnicity and gender.

Ethnicity

Men

Women

Foreign born

UK born

Foreign born

UK born

White native

Reference

Reference

Indian

0.0105*

0.0826***

0.0191***

0.117***

(0.00438)

(0.0124)

(0.00477)

(0.0127)

Pakistani

−0.000780

0.0500***

−0.000671

0.0177

(0.00530)

(0.0128)

(0.00573)

(0.00982)

−0.0169*

0.165**

−0.0198**

0.00777

(0.00721)

(0.0555)

(0.00723)

(0.0243)

Black Caribbean

0.362***

0.556***

0.253***

0.372***

(0.0206)

(0.0203)

(0.0203)

(0.0221)

Black African

0.103***

0.255***

0.0598***

0.149***

(0.0105)

(0.0376)

(0.00944)

(0.0326)

Chinese

0.0595***

0.545***

0.253***

0.634***

(0.0138)

(0.0591)

(0.0173)

(0.0535)

R2

0.053

0.052

Observations

834,571

817,757

Note: Data source is the Labour Force Survey (LFS) 2000–2008, the sample is all married individuals. A probit model is estimated, estimates presented are marginal effects, controls include education, age, and age squared, clustered standard errors at the individual level in parentheses,

*** significant at 1%,

** significant at 5%.

Exogamy rates are lowest for white natives (about 3 per cent), but vary very considerably across ethnic minority communities. Among the foreign born, exogamy rates are lowest for the South Asian communities, but extremely high among the Black groups and the Chinese. For all groups, exogamy rates are much higher among the UK born. Among the South Asians, there is some indication that exogamy with white natives is higher for Indians than the Pakistanis and Bangladeshis (where religion may be more of an obstacle). Exogamy rates for some groups are extremely high—78 per cent of UK-born Chinese women are exogamous, as are 66.7 per cent of UK-born Chinese men, and 62.5 per cent of UK-born Black Caribbean men.

There is a lot of information in Table 8.6 which also does not control for age and education. Table 8.7 reports regression estimates—the patterns are very similar to those reported for Table 8.6.

## 8.5.4 Spousal age gaps

Table 8.8a reports estimates for the age gap between wives and their husbands, which could perhaps be interpreted as a measure of gender relations, with a larger age gap reflecting greater gender inequality. For the foreign born, all ethnic minority groups have a significantly (p.272)

Table 8.8a OLS estimates of the age gap between husband and wife for all individuals by ethnicity and place of birth.

Ethnicity

Foreign born

UK born

White native

Reference

Indian

1.413***

−0.221

(0.078)

(0.133)

Pakistani

1.628***

−0.149

(0.123)

(0.142)

4.230***

1.894***

(0.203)

(0.401)

Black Caribbean

1.703***

−0.762***

(0.328)

(0.202)

Black African

3.188***

0.247

(0.186)

(0.346)

Chinese

1.294***

0.596

(0.209)

(0.492)

R2

0.023

Observations

759,733

Note: Data source is the Labour Force Survey (LFS) 2000–2008, the sample is all non-single women aged between 18 and 40 inclusive. Controls include age, age squared, and education, clustered standard errors at the individual level in parentheses,

*** significant at 1%,

** significant at 5%.

Table 8.8b OLS estimates of the age gap between husband and wife for endogamous and exogamous individuals by ethnicity and place of birth.

Ethnicity

Endogamous

Exogamous

Foreign born

UK born

Foreign born

UK born

White native

Reference

Reference

Indian

1.430***

−0.374***

1.423***

0.471

(0.078)

(0.122)

(0.354)

(0.432)

Pakistani

1.560***

−0.275

2.886***

0.494

(0.127)

(0.148)

(0.582)

(0.516)

4.281***

2.019***

4.128***

−1.304

(0.207)

(0.400)

(1.223)

(1.460)

Black Caribbean

2.333***

−0.468

0.994

−0.832**

(0.403)

(0.252)

(0.630)

(0.344)

Black African

2.841***

0.435

5.415***

−0.347

(0.185)

(0.322)

(0.680)

(0.912)

Chinese

0.993***

0.229

1.761***

0.612

(0.228)

(0.548)

(0.426)

(0.577)

R2

0.022

0.016

Observations

751,708

724,190

Note: Data source is the Labour Force Survey (LFS) 2000–2008, the sample is all married individuals. Controls include age, age squared, and education, clustered standard errors at the individual level in parentheses,

*** significant at 1%,

** significant at 5%.

(p.273) greater spousal age gap than white native women, it being largest among Bangladeshis (4.2 years) and Black Africans (3.2 years), and smallest among the Chinese (1.3 years). However, it is striking that, among the UK born it is only for the Bangladeshis who have a significantly different spousal age gap and even that is much reduced. UK-born Black Caribbeans have a significantly lower spousal age gap than white native women.

One possibility is that this is driven by the higher rates of exogamy among the UK born that we saw in Table 8.6. However, Table 8.8b shows that this is generally not the case. In Table 8.8b we estimate separate spousal age gap equations for endogamous and exogamous groups with the reference group, in both cases, being all white native women. Although there are some significant differences in spousal age gaps between exogamous and endogamous couples (though sample sizes are small for the exogamous group), it is clear that the declining gaps are present among endogamous couples.

# 8.6 Educational attainment and the gender gap in education

It is of very considerable interest how the level of education of ethnic minorities compares with that of natives (see Briggs et al., 2005; Modood, 2005, for other research on the educational attainment of ethnic minorities). The gender gap in education is also a good way of looking for evidence of gender equality. Table 8.9 shows the average age left full-

Table 8.9 Average age left continuous full-time education and proportion left full-time education by the age of thirteen for men and women by ethnicity.

Ethnicity

Average age left full-time education

Proportion of people who left continuous full-time education by the age of thirteen

Men

Women

Men

Women

White native

17.2

17.1

0.19

0.15

Indian

19.6

18.6

2.14

4.46

Pakistani

18.3

16.2

4.68

15.2

17.5

15.7

8.4

19

Black Caribbean

17.2

17.5

1.5

0.7

Black African

20.6

19

1.8

5.8

Chinese

20

19.4

3.7

4

Note: Data source is the Labour Force Survey (LFS) 2000–2008. Proportions are computed using individual sampling weights.

(p.274) time education for different ethnic groups. This measure of education is not ideal as a given age left full-time education may reflect very different types and quality of education, especially when comparing the foreign and UK born. But, unfortunately, the UK LFS does not adequately code foreign qualifications so the measure we use here is the best available.

Table 8.9 shows that, for men, it is white natives who, on average, left full-time education at the youngest age. The Black Africans, Chinese, and Indians are the best-educated. Among women, Pakistanis and Bangladeshis have lower levels of education than white natives, clearly indicating a gender gap in education for these groups. A smaller gender gap is found among the Indians and Chinese.

However, these figures on average age left full-time education hide a lot of variation. The last two columns look at the fraction of communities who left full-time education by the age of 13, that is, who have a very low level of education. This should be impossible for those born and brought up in the UK and one sees essentially zero rates among white natives. The fractions are higher for all ethnic minority communities—and very high for some groups. Most strikingly, 19 per cent of Bangladeshi women and 15.2 per cent of Pakistani women have completed education by the age of 13, so levels of education are low for these groups.

Table 8.10 OLS estimates of the gender gap in age left continuous full-time education by ethnicity, place of birth, and birth cohort.

Indian

Pakistani

Black Caribbean

Black African

Chinese

UK born

Born before 1970

0.760***

1.277***

3.198

−0.181

0.497

−1.089**

(0.203)

(0.331)

(1.713)

(0.102)

(0.364)

(0.549)

Born after 1970

0.546***

0.875***

0.951

−0.139

−0.065

0.337

(0.156)

(0.207)

(0.572)

(0.203)

(0.393)

(0.456)

Foreign born

Born before 1970

1.382***

2.907***

2.422***

−0.294**

2.084***

0.799***

(0.099)

(0.171)

(0.300)

(0.120)

(0.166)

(0.250)

Born after 1970

0.903***

1.719***

1.457***

−0.452

1.278***

0.814**

(0.162)

(0.207)

(0.247)

(0.325)

(0.199)

(0.369)

R2

0.075

0.116

0.100

0.034

0.037

0.073

Observations

38,202

21,614

7,145

19,709

18,163

7,245

Note: Data source is the Labour Force Survey (LFS) 2000–2008, the sample is all individuals aged 26 and above. Clustered standard errors at the individual level in parentheses,

*** significant at 1%,

** significant at 5%.

(p.275) Table 8.10 reports regression estimates for the gender gap in education. In these regressions we also interact the ethnicity and foreign-born dummies with a cohort dummy for whether the respondent was born before or after 1970. The reported coefficients are gender gaps. For all the foreign-born groups except the Black Caribbeans there is a significantly larger gender gap in education than among white natives. However, this gender gap is smaller for later birth cohorts. Among the UK born, the gender gaps are lower and, within this group, lower for those born after 1970 (although small sample sizes make it hard to draw precise conclusions on this).

To summarize: the gender gap in educational attainment is larger among Pakistani and Banglasdeshi communities than for the other main ethnic minorities. In large part, this is the result of enormous past differences in the educational attainment of men and women in the countries of origin. But there is marked change, driven in part by changes among both the UK and foreign born,7 and in part because of the change in the share of the communities who are UK born. Our conclusions here are consistent with those of more qualitative studies (e.g. Ahmad et al., 2003) who conclude that cultures often portrayed as opposed to the education and employment of women seem to be producing growing cohorts of highly motivated young women.

# 8.7 Female employment

We now turn to an analysis of female employment. For white natives the last 60 years have seen a large growth in female employment rates, though there is some evidence that the growth is now slowing or even stopping. But many of the ethnic minorities come from cultures in which female employment is lower, so female employment is an interesting indicator of cultural change. It may, of course, also reflect economic opportunities.

Table 8.11 reports female employment rates, by ethnicity, place of birth and—because it is so important—marital status and the presence of dependent children. The first row reports employment rates for all women. These are highest for white natives, though Black Caribbeans are only slightly behind. However, the exceedingly low rates for Pakistani (25 per cent) and Bangladeshi (17 per cent) women are quite (p.276)

Table 8.11 Female employment rates by ethnicity, place of birth, marital status, and presence of dependent children.

White Native

Indian

Pakistani

Black Caribbean

Black African

Chinese

All Women

0.74

0.64

0.25

0.17

0.71

0.59

0.63

UK born

All

0.74

0.75

0.45

0.46

0.73

0.73

0.77

Single

0.7

0.81

0.72

0.7

0.67

0.7

0.78

Married

0.78

0.88

0.72

0.8

0.91

0.92

0.82

Married with dependent children

0.74

0.68

0.33

0.31

0.76

0.71

0.73

Foreign born

All

0.61

0.18

0.14

0.7

0.56

0.6

Single

0.76

0.3

0.54

0.66

0.54

0.64

Married

0.61

0.25

0.22

0.72

0.72

0.6

Married—with dependent children under 16

0.6

0.15

0.11

0.7

0.53

0.57

Note: Data source is the Labour Force Survey (LFS) 2000–2008, the sample is all women aged between 25 and 59 inclusive.

Table 8.12 Estimates of employment probability for women by ethnicity and place of birth.

Ethnicity

Foreign born

UK born

White native

Reference

Indian

−0.185***

−0.0421**

(0.00815)

(0.0129)

Pakistani

−0.555***

−0.322***

(0.00848)

(0.0167)

−0.587***

−0.298***

(0.0122)

(0.0416)

Black Caribbean

−0.0543***

−0.0619***

(0.0134)

(0.0115)

Black African

−0.255***

−0.0959***

(0.0102)

(0.0256)

Chinese

−0.210***

−0.0343

(0.0161)

(0.0373)

R2

0.053

Observations

948,814

Note: Data source is the Labour Force Survey (LFS) 2000–2008, the sample is all women aged between 25 and 59 inclusive. A probit model is estimated, estimates presented are marginal effects, controls include education, age, and age squared, clustered standard errors at the individual level in parentheses,

*** significant at 1%,

** significant at 5%.

(p.277) striking. This is well known (see, for example, Cabinet Office, 2003, Berthoud and Blekesaune (2007) and Clark and Drinkwater (2007)). The Equalities Review went so far as to say that the gap in employment rates between Pakistani/Bangladeshi and white women would never be eliminated (Cabinet Office, 2007). However, one can also see in Table 8.11 that there is a very large difference in employment rates between the foreign and UK born. For example, UK-born Pakistani women have an employment rate of 45 per cent—still low, but much higher than the rate of 18 per cent for foreign-born Pakistani women. For Bangladeshi women, the figures are 46 per cent and 14 per cent, respectively. There is also some indication that UK-born women from these communities are no longer stopping employment on marriage but waiting until they have children.

The employment rates of Table 8.11 do not control for age or education. Table 8.12 reports estimates from specifications that control for education. For the foreign born, women from all ethnic minorities are significantly less likely to be in employment than white native women, with the largest gaps being for Pakistanis and Bangladeshis. But, these gaps are much reduced among the UK born.

Again, we see evidence of convergence in behaviour. The quantitative conclusions we have drawn here mesh well with the more qualitative studies of Ahmad et al. (2003) and Aston et al. (2007).

# 8.8 Values and beliefs

## 8.8.1 National identity

Since spring 2001 the LFS has asked about the national identity of respondents (though not in Northern Ireland), a question motivated by concern that some immigrant groups did not think of themselves as British. The specific question asked is ‘What do you consider your national identity to be? Please choose as many or as few as apply’. There are six possible responses: British, English, Scottish, Welsh, Irish, and Other. The order in which these responses are listed depends on the country of residence so English is the first option in England, Scottish in Scotland and Welsh in Wales. For the purposes of this chapter we group British, English, Scottish and Welsh into a single ‘British’ category and we will use the term British to refer to any of these answers in what follows.

Table 8.13 reports estimates from a probit equation—the coefficients are differences from white natives. In line with Manning and Roy (2010), and Georgiadis and Manning (2009) we find that all ethnic minorities are less likely to report a British national identity than white natives but (p.278)

Table 8.13 Estimates of the probability of reporting British national identity by ethnicity and place of birth.

Ethnicity

Foreign born

UK born

White native

Reference

Indian

−0.406***

−0.0819***

(0.00582)

(0.00528)

Pakistani

−0.340***

−0.0692***

(0.00707)

(0.00555)

−0.346***

−0.0497***

(0.0111)

(0.0110)

Black Caribbean

−0.374***

−0.0764***

(0.0108)

(0.00512)

Black African

−0.629***

−0.0802***

(0.00725)

(0.01000)

Chinese

−0.580***

−0.0962***

(0.0119)

(0.0156)

R2

0.474

Observations

1,944,169

Note: Data source is the Labour Force Survey (LFS) 2001–2008, the sample is all individuals aged 16 and above. A probit model is estimated, estimates presented are marginal effects, controls include education, age, and age squared, clustered standard errors at the individual level in parentheses,

*** significant at 1%,

** significant at 5%.

the gap is very much smaller for the UK born (in the 5–10 percentage point range) than among the foreign born (where it is in the 30–60 percentage point range). In line with other studies, it is worth pointing out that the Muslim groups (the Pakistanis and Bangladeshis) whose loyalty to Britain is often questioned are the ethnic minorities who are most likely to report a British national identity.

## 8.8.2 Religion

Since 2002, the Labour Force Survey has collected data on religion and Table 8.14 documents the proportions of different ethnicities describing themselves as of different religions. We also report the fraction with no religion and the fraction who report that they are practising their religion. The groups from the Indian sub-continent and Black Africans remain very religious, as very few report having no religion compared, for example, to the 56.6 per cent share of individuals reporting no religion among the Chinese. The most religious are the Pakistanis and Bangladeshis who are overwhelmingly Muslim.8 These groups are also much more likely to be practising their religion.

(p.279)

Table 8.14 Reported religion and whether practising religion by ethnicity.

Religion

White Native

Indian

Pakistani

Black Caribbean

Black African

Chinese

Christian

82.5

7.9

1.3

0.24

84

76.7

25.4

Buddhist

0.14

0.3

0.02

0.05

0.2

0.1

14.5

Hindu

0.01

46.1

0.18

1.07

0.24

0.27

0.17

Jewish

0.43

0.1

0.02

0.03

0.12

0.04

0

Muslim

0.07

13.3

96.5

97

0.62

18

0.2

Sikh

0.01

27.6

0.11

0.1

0.01

0.04

0

Other religion

0.7

1.95

1.22

0.28

2.3

0.93

3

No religion

16.2

2.6

0.66

1.21

12.5

3.9

56.6

% practising religion

17

64

80

82

33

65

6

Note: Data source is the Labour Force Survey (LFS) 2002–2008 for religious denomination and LFS 2002–03 for whether practising religion, the sample is all individuals aged 16 and above.

Table 8.15 reports estimates for the differences in the proportions who are practising their religion. In line with Table 8.14, all the South Asian and Black groups are more religious than white natives, with the Pakistanis and Bangladeshis standing out as being the most religious. There is evidence of less religiosity among the UK born than the foreign born, although the decline is noticeably less marked for Pakistanis.

What this suggests is that, while there is some evidence of a trend towards lower rates of religiosity among all the ethnic minorities studied here, the trend is less marked for Pakistanis than for other groups. This is perhaps consistent with the evidence in Bisin et al. (2007) that Muslims are more serious about their faith than adherents to other religions, although the Muslim Bangladeshis do show a marked decline in religiosity for the UK born.

## 8.8.3 Language

If one has problems with the English language, it is likely to be very hard to assimilate into British culture and one is very likely to remain economically disadvantaged. The LFS asks9 whether English is the first language at home and, if some other language other than English, Welsh, Gaelic, or Ullans is spoken, whether the respondent has language difficulties with work and education. We code an individual as reporting (p.280)

Table 8.15 Estimates of the probability of whether practising religion by ethnicity and place of birth.

Ethnicity

Foreign born

UK born

White native

Reference

Indian

0.604***

0.448***

(0.0130)

(0.0200)

Pakistani

0.748***

0.679***

(0.0130)

(0.0197)

0.805***

0.609***

(0.0121)

(0.0559)

Black Caribbean

0.266***

0.161***

(0.0201)

(0.0147)

Black African

0.627***

0.389***

(0.0140)

(0.0376)

Chinese

−0.0592***

−0.0286

(0.00860)

(0.0247)

R2

0.087

Observations

315,866

Note: Data source is the Labour Force Survey (LFS) 2002–2003, the sample is all individuals aged 16 and above. A probit model is estimated, estimates presented are marginal effects, controls include a dummy for whether the individual is female, education, age, and age squared, clustered standard errors at the individual level in parentheses,

*** significant at 1%,

** significant at 5%.

language difficulties if they report problems with either work or education.

In our analysis we assume that no white natives have language problems.10 We also assume the same for Black Caribbeans, the vast majority of whom come from English-speaking islands. Table 8.16 reports rates of using English at home for the other groups. Only 11 per cent of foreign-born Bangladeshis and 19 per cent of foreign-born Pakistanis use English at home, compared to 30 per cent of foreign-born Indians and Chinese and 47 per cent of Black Africans. For all ethnic minorities the proportions rise very markedly for the UK born, though a sizeable minority continue to use a language other than English at home.

Table 8.17 presents estimates of the proportions reporting difficulties with English. Among the foreign born, 22 per cent of Bangladeshis, 16 per cent of Pakistanis, 15 per cent of Chinese, and 10 per cent of Indians and Black Africans report difficulties. In many ways these differences reflect differences in educational attainment reported earlier in the chapter in Table 8.9. Among the UK born these proportions become (p.281)

Table 8.16 Estimates of the proportion with English as the first language at home by ethnicity and place of birth.

Ethnicity

Foreign born

UK born

Indian

0.294***

0.620***

(0.0109)

(0.0184)

Pakistani

0.189***

0.509***

(0.0112)

(0.0216)

0.114***

0.546***

(0.0133)

(0.0517)

Black African

0.466***

0.865***

(0.0151)

(0.0236)

Chinese

0.284***

0.784***

(0.0216)

(0.0374)

R2

0.48

Observations

8,257

Note: Data source is the Labour Force Survey (LFS) 2002 second quarter, 2003 second quarter, and 2006 third quarter, the sample is all individuals aged 16 or above who are not white natives or Black Caribbeans, as all individuals with either ethnicity speak English at home. A linear probability model is estimated, the constant term is not included in estimation, controls include education, age, and age squared, clustered standard errors at the individual level in parentheses,

*** significant at 1%,

** significant at 5%.

Table 8.17 Estimates of the proportion reporting that English language difficulties are causing problems in finding a job or in education.

Ethnicity

Foreign born

UK born

Indian

0.0912***

0.0191**

(0.00653)

(0.00726)

Pakistani

0.165***

0.0160

(0.0105)

(0.00884)

0.220***

0.0263

(0.0175)

(0.0187)

Black African

0.101***

0.0157**

(0.00901)

(0.00579)

Chinese

0.155***

0.0561**

(0.0155)

(0.0182)

R2

0.14

Observations

8,257

Note: Data source is the Labour Force Survey (LFS) 2002 second quarter, 2003 second quarter, and 2006 third quarter, the sample is all individuals aged 16 and above who are not white natives or Black Caribbeans, as all individuals with either ethnicity speak English at home. A linear probability model is estimated, the constant term is not included in estimation, controls include education, age, and age squared, clustered standard errors at the individual level in parentheses,

*** significant at 1%,

** significant at 5%.

(p.282) dramatically smaller, even though, as Table 8.16 shows, a large fraction do not use English as a first language at home. However, this perhaps comes as no surprise given that education is in English.

# 8.9 Conclusion

This chapter has compared the behaviours of the largest ethnic minorities in Britain with white natives across a wide, though not exhaustive, range of indicators. In all these dimensions there are significant differences across ethnic minorities some of which are well-established in the literature, as, for example, the strikingly low employment rates for Pakistani and Bangladeshi women but there are other differences less well-documented. An example of the latter is the finding that the Muslim minorities (the Pakistanis and Bangladeshis) are more likely to report a British National identity compared to other ethnic minority communities both among the foreign and UK born. Moreover, another striking common pattern that emerges is the extent to which differences in behaviours between ethnic minorities and white natives tend to be less pronounced for the UK than the foreign born. This indicates a general pattern of cultural integration, something perhaps not surprising to those who study the topic but not the impression one might gain from public discourse on the subject. The rate of cultural integration is faster for some variables than others—it is probably religion that shows the slowest rate. This has the implication that within religions, behaviours are changing so that what it means to be a good Christian or Muslim or Hindu is changing over time.

It is an important question whether, in future years, this process of convergence will continue until behaviours are the same or whether permanent differences will remain. Statistical analysis of data inevitably can only tell us about the past. But it is clear that there are very powerful forces that are acting to change the behaviour of immigrant communities once they are in the UK and it is not unreasonable to guess that these will continue into the future.

References

Bibliography references:

Adsera, A. and Chiswick, B. (2007) Are There Gender and Country of Origin Differences in Immigrant Labor Market Outcomes Across European Destinations? Journal of Population Economics, 20, 495–526.

(p.283) Ahmad, F., Lissenburgh, S., and Modood, T. (2003) South Asian Women and Employment in Britain: The Interaction of Gender and Ethnicity. London, Policy Studies Institute.

Aston, J., Hooker, H., Page, R., and Willison, R. (2007) Pakistani and Bangladeshi Women’s Attitudes to Work and Family. Department for Work and Pensions Research Report, No. 458.

Bell, B.D. (1997) The Performance of Immigrants in the United Kingdom: Evidence from the GHS. Economic Journal, 107, 333–344.

Berthoud, R. (2005) Family formation in multi-cultural Britain: diversity and change. In: G. Lowry, T. Modood, and S. Teles (eds) Ethnicity, Social Mobility and Public Policy. Cambridge, Cambridge University Press.

Berthoud, R. and Blekesaune, M. (2007) Persistent Employment Disadvantage. Department for Work and Pensions research Report No. 416.

Bisin, A., Patacchini, E., Verdier, T., and Zenou, Y. (2007) Are Muslim Immigrants Different in Terms of Cultural Integration? CEPR Discussion Papers 6453.

Blackaby, D.H., Drinkwater, S., Leslie, D.G., and Murphy, P. (1997) ‘A Picture of Male and Female Unemployment Among Britain’s Ethnic Minorities. Scottish Journal of Political Economy, 44, 182–197.

Blackaby, D.H., Leslie, D. G., Murphy, P.D., and O’Leary, N.C. (2002) White/Ethnic Minority Earnings and Employment Differentials in Britain: Evidence from the LFS. Oxford Economic Papers, 54, 270–297.

Blackaby, D.H., Leslie, D.G., Murphy, P.D., and O’Leary, N.C. (2005) Born in Britain: How are Native Ethnic Minorities Faring in the British Labour Market. Economic Letters, 88, 370–5.

Briggs, A., Burgess, S., and Wilson, D. (2005) The Dynamics of School Attainment of England’s Ethnic Minorities. CMPO Working Paper 05/130, Bristol University.

Cabinet Office (2003) Ethnic Minorities and the Labour Market: Final Report. London, Cabinet Office.

Cabinet Office (2007) Fairness and freedom: The Final report of the Equalities Review. http://webarchive.nationalarchives.gov.uk/20100807034701/http:/archive. cabinetoffice.gov.uk/equalitiesreview/upload/assets/www.theequalitiesreview.org.uk/equality_review.pdf.

Chiswick, B.R. (1980) The Earnings of White and Coloured Male Immigrants in Britain. Economica, 47, 81–87.

Clark, K. and Drinkwater, S. (2007) Ethnic minorities in the Labour Market: Dynamics and Diversity. York, Joseph Rowntree Foundation.

Clark, K. and Lindley, J. (2006) Immigrant Labour Market Assimilation and Arrival Effects: Evidence from the UK Labour Force Survey. IZA Discussion Paper No. 2228.

Coleman, D. (1994) Trends in Fertility and Intermarriage Among Immigrant Populations in Western Europe as Measures of Integration. Journal of Biosocial Science, 26, 107–136.

Dustmann, C. and Fabbri, F. (2003) Language Proficiency and Labour Market Performance of Immigrants in the UK. Economic Journal, 113, 695–717.

Dustmann, C. and Theodoropoulos, N. (2010) Ethnic Minority Immigrants and their Children in Britain. Oxford Economic Papers, 62, 209–233.

(p.284) Elliott, R.J.R. and Lindley, J.K. (2008) Immigrant Wage Differentials, Ethnicity and Occupational Segregation. Journal of the Royal Statistical Society, Series A, 171, 645–671.

Georgiadis, A. and Manning, A. (2009) One Nation Under a Groove? Identity and Multiculturalism in Britain. CEP Discussion Paper No. 944.

Georgiadis, A. and Manning, A. (2011) Change and Continuity Among Minority Communities in Britain. Journal of Population Economics, 24, 541–568.

Leslie, D. and Lindley, J.K. (2001) The Impact of Language Ability on Employment and Earnings of Britain’s Ethnic Communities. Economica, 68, 587–606.

Lindley, J.K. (2002a) The English Language Fluency and Earnings of Ethnic Minorities in Britain. Scottish Journal of Political Economy, 49 (4), 467–487.

Lindley, J.K. (2002b) Race or religion? The Impact of Religion on the Employment and Earnings of Britain’s Ethnic Communities. Journal of Ethnic and Migration Studies, 28(3), 427–442.

Lindley, J.K., Dale, A., and Dex, S. (2006) Ethnic Differences in Women’s Employment: The Changing Role of Qualifications. Oxford Economic Papers, 58, 351–378.

Manning, A. and Sanchari, R. (2010) Culture Clash or Culture Club? National Identity in Britain. Economic Journal Features, 120, F72–F100.

Modood, T. (2005) The Educational Attainments of Ethnic Minorities in Britain. In: G.C. Loury, T. Modood, and S.M. Teles (eds) Ethnicity, Social Mobility and Public Policy: Comparing the US and UK. Cambridge, Cambridge University Press.

Modood, T., Berthoud, R., Lakey, J., et al. (1997) Ethnic Minorities in Britain Diversity and Disadvantage. London, Policy Studies Institute.

Peach, C. (1996) Ethnicity in the 1991 Census, Volume 2: the Ethnic Minority Populations of Great Britain. London, HMSO.

Stewart, M.B. (1983) Racial Discrimination and Occupational Attainment in Britain. Economic Journal, 93, 521–541.

Zaiceva, A. and Zimmermann, K.F. (2007) Children, Kitchen, Church: Does Ethnicity Matter? IZA Discussion Paper No. 3070.

## Notes:

(1) The authors would like to thank Andrew Clark (PSE) for his helpful comments.

(2) There is also an enormous literature, which we do not seek to summarize here, on other countries—see Adsera and Chiswick (2007) for an interesting comparison of European countries.

(3) Information on parental country of birth can be identified in the LFS for individuals who live in the same household as their parents. This is the case only for 40 per cent of adults (aged 16 and above), UK-born non-whites in the LFS 2000–2008 inclusive. However, among individuals in the latter group with information on parental country of birth, 80 per cent have at least one parent born outside the UK.

(4) There are 15 categories after 2001 and 13 beforehand, the extra two groups being two extra mixed ethnicity categories. Table 8.1 reports the 13 categories of the earlier classification.

(5) This is the case for adults only, whereas if one also considers children then Bangladeshis have the third highest proportion of UK-born.

(6) One could use a censored regression model for those women who, when observed, have not given birth.

(7) Changes among the foreign born might be the result of the changes in the source countries discussed above, but another factor that might be important is the changing selection of immigrants into the UK.

(8) It is hard to know from this data whether the non-Muslims have converted or were brought up that way, as there are small religious minorities in both countries.

(9) This is only for one quarter every three years, so sample sizes are much reduced for the analysis that follows.

(10) In doing this we ignore the fact that a non-negligible fraction of the white native population do have literacy problems. However, we have little choice as the LFS does not ask the language difficulty question to those who report using English, Welsh, Gaelic, or Ullans at home.