Alice Bloch (2004) 'Labour Market Participation and Conditions of Employment: A Comparison of Minority Ethnic Groups and Refugees in Britain'

Sociological Research Online, vol. 9, no. 2, <http://www.socresonline.org.uk/9/2/bloch.html>

Received: 3 Sep 2003     Accepted: 18 May 2004    Published: 31 May 2004


Abstract

This paper draws on data from the Labour Force Survey (LFS) and from a survey of 400 refugees in Britain in order to present an up to data comparison of the labour market experiences of minority ethnic groups and refugees. It will show that refugees experience lower rates of employment than their ethnic minority counterparts and that those refugees in employment are more likely to be in temporary and part-time work with poorer terms and conditions of employment and with lower wages. The reasons why refugees experience greater disadvantage in the labour market than others include structural barriers due to policies such as dispersal that can leave refugees isolated from social and community networks that provide information and advice and informal routes into employment but also leave refugees in areas with higher levels of unemployment. Migration patterns are also influential with refugees for the most part arriving more recently in Britain than people from minority ethnic groups. Refugees are also increasingly reliant on agents and smugglers to plan their route and destination and so asylum seekers can find themselves in countries where they have no social networks. Social networks and community organisations play an important role in the early stages of settlement. Finally, the circumstances of exile, attitudes to the country of origin and the insecurity of having temporary status in Britain all prevent economic activity.
Keywords: Dispersal; Employment; Job Seeking; Migration; Minority Ethnic Groups; Refugees; Self-employment; Social Networks; Terms And Conditions Of Employment

Introduction

1.1 This paper draws on data from the Labour Force Survey (LFS) and from a survey of 400 refugees [1] in Britain in order to present an up to date comparison of the labour market experiences of minority ethnic groups and refugees. It will show that refugees experience lower rates of employment and for those in employment, lower wages and poorer terms and conditions of employment than their minority ethnic counterparts.

1.2 The first part of the paper will examine the background to the research and will focus on the factors that have been found to affect labour market participation among minority ethnic groups and refugees. This includes structural barriers due to policies such as dispersal, migration patterns, group characteristics, regional variations in the economy, discrimination and English language proficiency. The paper will then provide a brief outline of the methods used and the sample characteristics. The third part of the paper will present the findings from the data on labour market participation, occupation and industry of employment, terms and conditions of employment, earnings and job search. Finally the conclusion will include a discussion in order to unravel the factors that result in refugees being more disadvantaged in the labour market than other minority ethnic groups in Britain.

Background

2.1 The disadvantage experienced by minority ethnic groups in the labour market compared to their white counterparts has been well documented. The average rate of unemployment among white men is six per cent compared to 13 per cent among minority ethnic men. A similar pattern is evident among women where five per cent of white women are unemployed compared to 13 per cent of minority ethnic women (Owen et al, 2000; Twomey, 2001). Research carried out by the Policy Studies Institute (Modood et al, 1997) shows that there are differences in the labour market experiences - both in terms of participation and earnings - between ethnic groups. Most advantaged are the Chinese, African Asian and white groups. Most disadvantaged are Pakistanis and Bangladeshis while Indian and Caribbean people are in the middle.

2.2 There is much less quantitative data available on the labour market experiences of refugees compared with minority ethnic groups. The limited information available suggests that refugees are more disadvantaged than their minority ethnic counterparts. Previous research has consistently shown the high levels of unemployment experienced by refugees. The Home Office, in a national study of 263 refugees, found that 36 per cent of refugees were unemployed (Carey-Wood et al, 1995). A study in Newham found unemployment levels among 180 refugees to be 46 per cent (Bloch, 2002a) while research by the London Research Centre (1998) placed unemployment rates at 51 per cent.

2.3 Findings of previous research show that those refugees who are employed tend to be male, more fluent in English, have lived in the UK longer and are more secure about their immigration status (Carey-Wood et al, 1995; Parkinson, 1998; Bloch, 2002a). Those who are working tend to be clustered in a few low-paid industries, such and catering, and their employment is characterised by poor terms and conditions and unsocial hours. In addition, refugees are often underemployed, in the sense of not utilising the skills and experiences that they bring with them on arrival to the UK (Bloch, 2002a; Girbash, 1991).

2.4 Refugees face structural barriers to employment. The 1996 Asylum and Immigration Act introduced penalty fines for any employers who employed people without the correct documentation. Some employers were reluctant to employ refugees in case they were in breach of the legislation or because they did not want the burden of checking documentation (Refugee Council, 1999). Since July 2002, which was after the fieldwork for this study was carried out, asylum seekers had their right to work withdrawn. Only those with a positive decision - that is those granted refugee status or Exceptional Leave to Remain (ELR) and more recently Humanitarian Protection (HP) - are eligible to work. A long-term adverse effect from an initial period of unemployment on employment outcomes for refugees has been found (Wooden, 1991; Valtonen, 1999) though the affects of the new policy in the UK are as yet unknown.

2.5 As for minority ethnic groups more generally there are differences in labour market participation between and within groups. The factors which are known to produce differential labour market experiences include discrimination, group characteristics, economic climate at the time of migration; regional labour market variations, social networks and community organisations, education, English language proficiency and aspirations for the migration.

2.6 Discrimination is a common experience, as identified by the Policy Studies Institute Fourth National Survey where 20 per cent of ethnic minority respondents said that they had been discriminated against in the labour market (Modood et al, 1997). More recently, the impact of religion on labour market outcomes has been highlighted. Brown (2000), for example, notes the differences between South Asian religious groups. More specifically Muslims, as a group, are less likely to be employed than other religious minorities (Brown 2000). Sikhs, Pakistani Muslims and Bangladeshi Muslims are under-represented in professional employment. Among those who are employed, there is disparity with Muslims earning less than other religious minorities (Lindley, 2002).

2.7 Group characteristics prior to migration are also influential for both refugees and minority ethnic groups more widely. Reasons for the economic success of East African Asians were that they were well educated on arrival to Britain, had transferable skills and could draw on resources and capital (Robinson, 1993). For other groups, notably Bangladeshi and Pakistani women, cultural influences are reflected in low levels of labour market participation and this is particularly the case among those who are married or living in a partnership (Owen, et al, 2000).

2.8 The prevailing economic climate at the time of migration has been found to be an important determinant of the labour market integration of refugees and asylum seekers. This is linked to the hyper cyclical nature of employment among minority ethnic groups in general (Leslie et al, 1998). A comparative study of Vietnamese refugees in Finland and Canada found that those who arrived in the country of asylum during an economic boom were successfully absorbed into the labour market. In contrast, those who arrived during a recession struggled to gain employment, not just in the short term but also longer term (Valtonen 1999).

2.9 Significant regional variation in labour market activity was found in research carried out by Ho and Henderson (1999). The reasons for variations included the local economic structures at the time of migration. The 1999 Immigration and Asylum Act introduced the compulsory dispersal of destitute asylum seekers dependent on the National Asylum Support Service (NASS) to eight regions around the country. Some asylum seekers have been dispersed to white homogeneous areas with high levels of unemployment while others are in more diverse regional centres. Certainly, prior to dispersal, the majority of refugees were located in London where community organisations and informal networks provided much of the settlement work including information and advice about employment and job seeking (Brophy et al, 1998; Bloch, 2002a). While it is known that such networks may have disadvantages in limiting language acquisition and restricting access to labour market information, they are a vital source of support for recent arrivals (Duke et al, 1999) and of facilitating integration into the labour market.

2.10 This is not to suggest that the migration experiences of refugees mirrors that of other minority ethnic groups because it is rarely based on a system of chain migration that characterised the migration patterns of established minority communities from former colonial countries (Castles and Miller, 2003). Moreover, the initial motivations of the migration differ as for refugees they are forced while others make a choice. Refugees can depend on agents or smugglers to organise their flight and have limited choice of their eventual destination (Robinson and Segrott, 2003). Once in the UK, before the introduction of dispersal, refugees headed for larger urban centres particularly London where there were established or emerging communities from whom they were able to draw on for support and information (Duke et al, 1999).

2.11 Levels of education and qualifications are important correlates of employment. Research carried out by Berthoud (1999) identified the lower levels of qualifications obtained by some minority ethnic groups in the UK as one factor influencing low rates of employment and earnings. For refugees, qualifications obtained overseas are usually not recognised and some are not transferable (Sargeant and Forna, 2001). Moreover, refugees who are forced to leave their homes in acute circumstances may not have the opportunity to gather documentary evidence of their qualifications or, if the infrastructure in the sending country has collapsed, they might not be able to obtain replica certificates and the like when in exile.

2.12 Proficiency in English will affect economic incorporation with rates of employment rising dramatically where there is fluency in English (Sargeant and Forna, 2001; Bloch, 2002b). English language skills vary by country of origin and gender, as do literacy skills in the first language. Where women have low levels of first language literacy, they can be disadvantaged in their acquisition of the English language (Bloch and Atfield, 2002).

2.13 For all new migrants attitudes and aspirations for the migration will affect settlement outcomes. Research with refugees has found that if people maintain the 'myth of return', that is aspirations for return migration, they are less likely to invest in employment and training opportunities because it implies a degree of permanency (Al- Rasheed, 1994). Equally, prolonged uncertainty due to the length of time it can take to determine asylum cases can make it difficult to get on with everyday life and to make plans for the future (Bloch, 2002a).

2.14 This paper will examine the labour market position of minority ethnic groups and refugees. It will explore some of the reasons for the disadvantages that refugees experience when compared to their minority ethnic counterparts. It will also note that, as with other minority ethnic groups in general, the experiences of refugees are heterogeneous.

Methods and Sample Characteristics

3.1 The data of minority ethnic respondents is combined data from the Spring, Summer and Autumn 2001 Labour Force Surveys giving a total minority ethnic sample of 15, 251. Included in the sample were men aged 18-64 and women aged 18-59 living in Britain. Data was weighted using the weighting variable provided with the LFS in order to provide population estimates. The sample was not broken down by length of residence or place of birth which means that the inter-generational differences are not explored. However, refugees are more recent arrivals than their ethnic minority counterparts. Table 1 shows that all refugee respondents were born outside the UK and only 20 per cent had been in the UK for 10 years or more. This is in contrast to ethnic minority groups where half (49.5 per cent) were born in the UK while a further 13.2 per cent had been resident since before 1970, 21.3 per cent arrived between 1970 and 1989 and 16 per cent from 1990 onwards (Owen, 2003).

3.2 The data for refugees is from a survey of 400 refugees, from five communities living in five localities in Britain[2]. The five refugee communities were from the Somali regions, Iraq, Kosova, Sri Lanka (Tamils) and Turkey. The communities represented a mix of experiences and migration patterns but where numbers were large enough to ensure that respondents could be located. The five regions in Britain where the fieldwork was carried out were London, Manchester, Leeds/Sheffield, Newcastle and Birmingham. In the absence of a sampling frame of refugees (see Robinson, 1998; Bloch, 1999) quotas were set for country of origin, length of residence, age, gender and region and multiple gate keepers were used to identify respondents for the study in order to maximise the extent to which the survey represented the diversity of the refugee experience (see Bloch 2002b for more detail on the survey). The fieldwork took place between February and May 2002.

3.3 The main topics covered in the questionnaire concerned qualifications, language and literacy skills, training, employment histories and job seeking strategies both in the UK and prior to coming to the UK. It also included questions on participation in English language classes, voluntary sector activity and social and community involvement among refugees in the UK. Finally, there were questions on socio-demographic characteristics. Included in the refugee survey were some questions from the LFS on employment and training to allow comparisons to be made between the two surveys.

3.4 The questionnaire used in the survey of refugees contained mostly structured questions although a small number of open-ended questions were included as follow-ups in areas where either more qualitative descriptive data was required or where the diversity of answers was such that it would have been difficult to provide comprehensive coding frames for pre-coded questions. This was especially the case with attitudinal topics or those that required a more detailed description of experiences. Using mostly closed response categories limits the extent to which the data allows for the exploration of the individual experience. However, it did enable the quantification of refugees' economic participation and in so doing allows for comparisons to be made both with other groups and between and within different refugee communities.

3.5 The questionnaires were translated into community languages using the dual strategies of back translation and the translating decentralising procedure (Brislin et al,1973). Back translation is a process whereby one bi-lingual, in this case one of the survey interviewers, translates the source language into the target language while another bi-lingual, the other interviewer from the same community, translates from the target language back to the source language without seeing the original. Every question is then examined and then where necessary, changes were made to the translations and/or the English language version. Using the translating decentralising procedure leaves the source and target language open to modification in order to achieve comparability. However, it was not possible to measure the impact that the translations may have had on the data collected and therefore the comparability and standardisation across linguistic groups.

3.6 A total of 80 interviews were carried out with each of the five refugee communities of which 40 were with men and 40 with women. Table 1 shows the sample of refugees by length of residence, age and region.

Table 1: Refugee sample by length of residence, age and region (frequencies)

 

Somalia

Turkey

Iraq

Sri Lanka

Kosova

Total %

Length of residence

 

 

 

 

 

 

Less than 1 year

6

7

3

4

0

5

1 year but less than 3 years

17

23

39

27

59

41

3 years but less than 5 years

14

14

8

9

15

15

5 years but less than 10 years

14

24

13

18

5

19

10 years or more

29

12

17

22

1

20

Total

80

80

80

80

80

400

 

 

 

 

 

 

 

Age

 

 

 

 

 

 

18- 24

10

7

13

9

21

15

25- 29

20

21

13

15

14

21

30- 34

14

19

14

18

13

20

35- 39

18

18

18

12

15

20

40- 49

12

12

20

11

14

17

50 or over

6

3

2

15

3

7

Total

80

80

80

80

80

400

 

 

 

 

 

 

 

Region

 

 

 

 

 

 

Manchester

15

0

3

4

10

8

Leeds/Sheffield

16

3

17

0

22

14.5

Birmingham

13

6

13

11

33

19

Newcastle

1

3

10

0

4

4.5

London

35

68

37

65

11

54

Total

80

80

80

80

80

400


3.7 In the refugee survey, 10 per cent of respondents had refugee status, 17 per cent were naturalised British citizens, 21 per cent had ELR, 22 per cent were asylum seekers on temporary admission, 29 per cent had Indefinite Leave to Remain (ILR) and 1 per cent were a citizen of a European Union country. More than half (57 per cent), therefore, had security of status (refugee, ILR or naturalised citizens).

Labour Market Participation

4.1 A comparison of the labour market participation of refugees and minority ethnic groups show the lack of activity among refugees. In total, 29 per cent of refugees were working which is much fewer than the 60 per cent of minority ethnic people. One of the differences in participation was between men and women. Table 2 shows that higher proportions of men were working than women.

Table 2: Labour market status of men and women by main minority ethnic groups and refugee communities (percentages)

Employed
All Men Women
Asian or Asian British 58 70 45
Of which:
Indian 67 76 58
Pakistani 46 64 27
Bangladeshi 41 59 18
Other 62 69 53
Black or Black British 64 68 60
Of which:
Caribbean 67 69 66
African 59 67 52
Other 66 71 61
Mixed 66 73 60
Of which:
White and Asian 66 72 56
White and Black Caribbean 64 75 55
White and Black African 80 91 74
Other 64 68 62
Chinese 57 66 49
Other 57 62 51
All 60 68 51
Refugees:
Somali 20 23 18
Turkish 31 40 23
Iraqi 24 45 3
Sri Lankan 45 63 28
Kosovan 23 38 8
All 29 42 16

4.2 With the exception of Tamils from Sri Lanka, who had higher rates of employment than Bangladeshi people and similar rates to Pakistani respondents, all the other refugee groups were much less likely to be working than their minority ethnic counterparts. The greater participation of Tamils in employment than other refugee communities has been found in other research (Bloch, 2002a). The reasons why Tamils are more likely to be employed than others are first, because Tamils arrived in Britain with greater fluency and literacy in English than the other refugee communities because English is taught in schools in Sri Lanka. Secondly, there is a long-standing Tamil community in Britain and Tamils made use of social and community networks, once within the country, in order to obtain employment.

4.3 Table 2 shows that around half of ethnic minority women (51 per cent) were working compared to 16 per cent of refugee women. Of those refugee women who were not employed, 22 per cent said they were looking for paid work compared with 12 per cent of ethnic minority women. There were differences between groups: among ethnic minority women people who described their ethnic group as Caribbean were most likely to be looking for work (33 per cent). Among refugees Tamil and Kosovan women the most likely to be looking for work (37 per cent and 35 per cent respectively).

4.4 Those refugees who said they were not looking for work were given an open-ended probe and asked 'why not'. The reason mentioned most for not looking for work among refugee women was because they were looking after the home and family (53 per cent). Turkish women were the most likely to say they were not looking for work due to family commitments (68 per cent) than others. The proportions among women from Sri Lanka was (62 per cent), Kosova (57 per cent), Iraq (53 per cent) and the Somali regions (23 per cent). The other main reasons given for not looking for work were included health, (14 per cent), a lack of English language skills and wanting to learn English first (both eight per cent). The proportion of ethnic minority women who were not looking for work was 58 per cent. There was a great deal of variation between groups reflecting cultural norms and expectations. Among Bangladeshi women, 83 per cent said that they were not looking for work because they were looking after the home and family compared with 73 per cent of Pakistani women, 49 per cent of Indian women, 46 per cent of African women and 39 per cent of Caribbean women.

4.5 Verifying the data found in other studies, the survey of refugees found that the main factor that affected the labour market participation among refugees was proficiency in English. However, an examination of Cramers V shows no strong associations between employment and any one variable.

Table 3: Proportion participating in the labour market by other characteristics

Percentages
Labour market participation Yes No Total
Spoken English: Cramer's V=.302[3]
Fluently 52 48 83
Fairly well 31 69 156
Slightly 14 86 133
Not at all 11 89 28
Whether arrived with a qualification: Cramer's V=.211
Yes 37 63 212
No 18 81 187
Whether obtained qualification in Britain: Cramer's V=.247
Yes 51 49 81
No 23 77 319
Length of residence: Cramer's V=.175
Less than three years 20 80 185
Three years but less than five 35 65 60
Five years or more 36 64 155
Whether sees Britain as home: Cramer's V=.054
Yes 30 70 260
No 25 75 140
Attitudes towards return migration: Cramer's V=.041
Yes 28 72 187
Maybe 34 66 41
No 28 72 170
Immigration status: Cramer's V= .069
'Secure status': Refugee, ILR, Naturalised British/ EU citizen 31 69 229
ELR 35 65 82
Temporary admission 25 75 89
Participation in training; Cramer's V=.252
Yes: past 67 33 30
Yes: current 41 59 17
No 25 75 353
Region: Cramer's V= .102
North East 17 83 18
North West 19 81 32
Midlands 26 74 76
London 30 70 216
Yorkshire and Humberside 35 65 58
Country of origin: Cramer's V=.201
Somalia 20 80 80
Turkey 31 69 80
Iraq 24 76 80
Sri Lanka 45 55 80
Kosova 22.5 77.5 80
Age: Cramer's V= .178
Less than 35
36 64 221
35 or over 20 80 179
Sex: Cramer's V= .288
Men 41.5 58.5 200
Women 15.5 84.5 200
Total number 114 286 400

4.6 Table 3 shows some regional diversity with more refugees working in the Leeds/Sheffield area than elsewhere although Cramers V shows a weak association. The regional diversity reflected, to some extent, the differentials found in local labour markets. Data from the Annual Local Area Labour Force Survey (2001-02) showed that the highest rates of employment in the five localities included in the refugee survey was in Leeds (77.8 per cent) followed by Sheffield (71.2 per cent), London (70.4 per cent), Newcastle (65.6 per cent), Birmingham (65.1 per cent) and Manchester (60.5 per cent) <http://www.statistics.gov.uk/cci/nugget.asp ?id=252 >. With the exception of Newcastle, the data correlates with the levels of employment among refugees that was highest in Leeds and Sheffield and lowest in Manchester. Thus, variations in local labour markets do seem to effect refugee employment as they do with other minority ethnic groups.

Self-employment

4.7 Self-employment was also an area where there were differences between refugees and minority ethnic groups. In total, 13 per cent of minority ethnic people who were working were self-employed while the proportion among refugees was 19 per cent. Table 4 shows that more men than women were self-employed.

Table 4: Whether employees or self employed by gender and minority ethnic groups and refugee communities (percentages)

Self- employed Employee
Men Women Me n Women
Asian or British Asian 21 8 79 92
Of which:
Indian 18 8 82 92
Pakistani 31 10 69 90
Bangladeshi 19 5 81 95
Other 14 9 86 91
Black or Black British 10 4 90 96
Of which:
Caribbean 11 3 89 97
African 9 6 91 94
Other 15 4 85 96
Mixed 14 6 86 94
Of which:
White and Asian 16 3 84 97
White and Black Caribbean 17 5 83 95
White and Black African 0 12 100 88
Other 13 7 87 93
Chinese 28 18 72 82
All 18 7 82 93
Somalia 0 14 100 86
Turkey 19 11 81 89
Iraq 39 0 61 100
Sri Lanka 12 0 88 100
Kosova 33 67 67 33
All 22 13 7 8 87

Base number refugees: 114

4.8 Iraqi men were the most likely to be self-employed followed by Kosovan, Pakistani and Chinese. No Somali men were self-employed. Refugee women were more likely to be self-employed than their minority ethnic counterparts. However, the numbers were small as only 31 refugee women were working of which four were self-employed.

4.9 People from minority ethnic groups are more likely to be self-employed that white people (Owen et al, 2000). One of the reasons for self-employment among migrants generally has been due to the difficulties in finding employment and so it has become an important route for absorption into the labour market. Research has shown this to be the case among East African Asians and Armenians (Malik, 1990; Robinson, 1993; Srinivasan, 1992).

4.10Metcalf et al (1997) note that the propensity for self-employment among Asians in Britain is a consequence of economics and culture. Self-employment offers an alternative to poor employment prospects and racism in the labour market for whole families. Secondly, self-employment provided standing within the family, which is culturally important. Self-employment is seen as something transitory for new migrants and acts as a springboard for future generations. The jobs carried out by self-employed refugees differed from Asians, including East African Asians who were from refugee backgrounds, in the study carried out by Metcalf et al (1997). The main differences were the greater diversity in the types of self-employment among South Asians when compared to refugees and their propensity to have set up businesses rather than being freelance. Part of the reason for this difference is that refugees face difficulties obtaining business loans due to their lack of collateral and lack of suitable loan guarantors.

4.11 Among South Asians, 42 per cent were in retail, 12 per cent in restaurants, 18 per cent in artisan-style services, six per cent were in clothing manufacture, nine per cent were cab drivers and the rest, 13 per cent, were in a small range of other services such as insurance sales and driving instructors. In contrast, more than half the refugees who were self- employed were working as interpreters or translators (55 per cent). The rest had catering or shop and garage businesses (18 per cent) or were self-employed mechanics (nine per cent) or drove a mini-cab (5 per cent) worked as an administrator (five per cent) and a legal representative (five per cent). This means that most self-employed refugees, unlike their minority ethnic counterparts, would not be in a position to offer employment to family members and friends and were working in areas that only required human capital. Moreover, no one who was self-employed had found his or her job through a family member. This was in contrast to the pattern of employment and job seeking prior to coming to Britain.

4.12 Before coming to Britain, 42 per cent of refugees had been working compared with 29 per cent in Britain. However, only 6.5 per cent described themselves as unemployed. More than one fifth had been students (21 per cent) and 27 per cent had been looking after the home and family. Of those who were working a third were self-employed compared to 19 per cent who were self-employed in Britain. Among those who were self-employed, a third were shop owners, more than one-fifth (22 per cent) were farmers and 15 per cent were trades people (builders, welders or carpenters). More than half (56 per cent) had found their job through a family member. Thus, before coming to Britain, refugees were self-employed in work that required more capital and in many cases had become involved in family businesses. Clearly this was not a route to employment in Britain due to the lack of collateral for loans and the relative lack of kinship networks.

Occupation and Industry

5.1 Research has shown that minority ethnic groups are occupationally segregated. Men from minority ethnic groups are over-represented in the service sector and under-represented in the agriculture, fishing, energy and water industries. Women are even more concentrated in a few industries than men (Owen et al, 2000). Table 5 shows the industry of employment by aggregated ethnic groups and gender.

Table 5: Percentage of employees in each industry[4] by ethnic group and gender

Asian/Asian British Black/Black British Mixed Chinese Other All
Men Wom Men Wom Men Wom Men Wom Men Wom Men Wom
Agriculture, hunting and forestry + + + 0 + + 0 0 1 0 + +
Mining, quarrying + 0 + 0 0 0 1 0 + 0 + 0
Manufacturing 21 13 15 6 15 6 10 5 10 7 17 9
Electricity, gas, water supply 1 1 + + 2 2 1 1 0 1 1 1
Construction 3 1 5 1 4 1 1 1 3 + 4 1
Wholesale and retail trade 17 18 15 13 15 14 10 11 20 10 17 15
Hotels and restaurants 10 3 4 6 7 4 36 17 10 7 9 5
Transport, storage and communications 13 6 14 5 14 7 4 1 9 5 12 5
Financial intermediation 4 7 4 5 6 3 11 4 6 6 5 6
Real estate, renting, business activity 13 11 18 11 10 12 13 11 11 17 14 12
Public admin and defence 4 9 6 12 7 9 4 3 6 5 5 9
Education 3 10 3 8 5 9 1 17 8 10 3 9
Health and social work 9 16 11 32 8 27 6 21 10 25 9 23
Other community, social and personal activities 3 3 5 4 6 5 2 8 4 6 4 4
Other 0 1 + 0 0 1 2 0 + 3 + 1
Total number 2076 1630 1021 1239 323 394 149 155 458 434 4027 3852

5.2 Table 5 shows that minority ethnic women are most likely to be employed in health and social work (23 per cent), both public sector areas of employment, while men are most often employed in manufacturing and wholesale and retail (both 17 per cent). There are, however, differences between communities. Most notable is the 36 per cent of Chinese men who work in hotels and restaurants. When the Asian and British Asian group is disaggregated it reveals that 46 per cent of Bangladeshi respondents work in hotels and restaurants and 23 per cent of Pakistani people are employed in manufacturing.

5.3 One of the very striking features of refugee employment, as Table 6 shows, is that those who are working are even more clustered in a few industries than their minority ethnic counterparts.

Table 6: Current and previous employment in Britain Frequencies

Job title Current job Previous job Total number Total %
Catering including waiter, fast food chef and delivery/kitchen porter 20 10 30 18
Interpreter/translator 17 0 17 10
Shop assistant/cashier 16 15 31 19
Administration/clerical 15 2 17 10
Factory worker 7 8 15 9
Advice provider 7 1 8 5
Community or social worker 4 3 7 4
Minicab/driver 3 3 6 4
Security guard/porter 3 1 4 2
Shop or garage owner/manager 3 0 3 2
Cleaner 2 1 3 2
Motor mechanic 2 0 2 1
Construction/building/decorating 2 2 4 2
Engineer 2 0 2 1
Nurse 2 1 3 2
Teacher 0 2 2 1
Carer 0 2 2 1
*Other 8 3 11 7
Total number 113** 54 167 100

**Missing: 1
*Other includes carpenter, IT trainer, service supervisor, receptionist, and legal representative
Source: Bloch, A. (2002b) p. 101.

5.4 Sixty per cent of refugees who were working were employed in just four types of work: catering, shop work, interpretation and translation and administration and clerical work. Before coming to Britain there was more diversity of employment among refugees. The jobs that refugees had done most often were: shop owner/shop keeper (12 per cent); teacher (10 per cent); office, administration and clerical (10 per cent); tradesperson (9 per cent); farmer (seven per cent); catering (7 per cent); shop assistant (7 per cent) and driver (taxi/truck) (6 per cent). However, there was also a range of employment including health care professionals, accountants, financial advisors, journalists, textile workers and engineers. Much of the work has comparable British equivalents though it is clear from Table 6 that refugees were not able to find employment in the same areas. This is due to a combination of factors that include work place English language proficiency, lack of UK work experience in relevant areas and the lack of recognition of overseas qualifications.

Terms and Conditions of Employment

6.1 A comparison of refugees and minority ethnic groups shows that refugees work under poorer conditions and have less choice over their employment situation then their minority ethnic counterparts. This section will examine the incidence of temporary work, part-time work, numbers of hours worked and entitlement to paid holiday.

6.2 A greater proportion of refugees were in temporary posts than people from minority ethnic groups: 25 per cent and 11 per cent respectively. The majority of refugees who had taken a temporary job took that type of employment because they could not find a permanent job (79 per cent) while among minority ethnic groups the proportion was 31 per cent.

6.2 More than half the refugees (54 per cent) who had a temporary post said their it was temporary because it was casual. The rest were in temporary work because is was seasonal (14 per cent), contract (14 per cent) or agency work (11 per cent). Of those minority ethnic people in temporary work, most were in temporary work because they had a contract (43 per cent) followed by working through an agency (26 per cent), casual (17 per cent), seasonal (three per cent). Clearly the profile differs between refugees and minority ethnic groups. Refugees are much more likely to have a casual job or seasonal work and less likely to have a contract than other ethnic minorities. It is not therefore surprising that refugees' temporary posts are for a shorter duration than their minority ethnic counterparts.

Figure 1. Length of time of temporary job: minority ethnic groups and refugees (percentages)
Base number refugees: 28
Base number ethnic minorities: 255

6.4 Figure 1, despite the small numbers of refugees, indicates that shorter duration of jobs compared to minority ethnic groups. More than half the refugees (54 per cent or 17 out of 28) were employed for less than a year while the same was true for 30 per cent of people from minority ethnic groups.

6.5 Refugees were also more likely to be working part-time than minority ethnic people mainly because they could not find a full-time post. Around one-third of refugees who were working were working part-time (35 per cent) compared with less than a quarter (23 per cent) people from minority ethnic groups. A comparison of men and women shows that nearly one- third (32 per cent) of ethnic minority women work part-time compared to 14 per cent of men. Among refugees the proportion of men working part-time was 31 per cent. This is comparable to the 35 per cent of women who also worked part-time. Figure 2 shows that more than half the refugees working part-time were doing so because they were unable to find a full-time job in contrast to minority ethnic groups where just under half did not want a full-time job.

Figure 2. Reasons for working part-time: minority ethnic groups and refugees (percentages)
Base number refugees: 38
Missing: 1

6.6 Fewer refugees were entitled to holiday pay (47 per cent) compared with people from minority ethnic groups (92 per cent). The higher incidence of self-employment, temporary and part-time work among refugees than among minority ethnic groups more generally helps to explain the differential.

Earnings

7.1 Levels of pay are known to vary between ethnic groups and within ethnic groups as there is a gender dimension to earnings (Owen et al, 2000). A comparison of the average hourly rate of pay commanded by refugees and minority ethnic groups in Table 7 shows that refugees' hourly earnings were only 79 per cent of that earned by other minority ethnic people. There was less difference between the earnings of refugee men and women than among minority ethnic men and women. Kosovan women and Bangladeshi men earn less per hour than others. Only one Iraqi woman was in employment and she was working as an engineer so the rate of pay of Iraqi women is not representative.

Table 7: Mean gross hourly pay (£) by minority ethnic group and refugee communities

Ethnic Group All Men Women
Indian 10.11 10.98 9.15
Pakistani 7.45 7.71 6.90
Bangladeshi 6.19 5.48 9.62
Caribbean 9.09 9.90 8.39
African 9.16 9.50 8.72
Chinese 9.12 10.36 8.19
All 9.26 9.65 8.79
Somalia 8.23 7.69 8.88
Turkey 7.30 7.29 7.31
Iraq 8.17 7.47 18.67
Sri Lanka 6.48 6.56 6.29
Kosova 7.53 7.95 5.42
All 7.29 7.26 7.39

Base number refugees: 105
Missing: 9

7.2 Regional variations can affect levels of earnings. Owen at al, (2000) found that average hourly rates for men in London were £11.81 compared to £9.04 for elsewhere. Among women in London pay was £9.64 and for those outside it was £7.27. Wages are therefore about 30 per cent higher for those in London than those outside of London. Among refugees, however, there was little difference in levels of pay by locality. Figure 3 shows that refugee women were earning more outside London than those in London while the reverse was true among men.

Figure 3. Mean gross hourly earnings among refugees by region and gender (£)
Base number: 105
Missing: 9

7.3 Qualifications had a greater impact on levels of pay for minority ethnic groups than refugees. The LFS data does not disaggregate those qualifications obtained in the UK with those obtained elsewhere so to enable comparisons of ethnic minorities with refugees, all qualifications were grouped together regardless of where they were obtained.

Table 8: Mean Gross pay per hour (£) by highest qualification

Minority ethnic groups Refugees
Degree or higher 13.71 9.04
A'level or equivalent 8.18 8.42
O'level/GCSE A-C or equivalent 7.67 6.09
None 5.67 6.30

Source: Bloch, A. (2002b), p. 99

7.4 Table 8 shows a linear relationship between levels of earnings and qualifications among minority ethnic groups. It shows that the higher the educational qualification the higher the rate of pay. There is also a very big leap in earnings for those who have attained a degree level qualification or higher. Refugees with a degree or higher only earn two-thirds of that earned by people from minority ethnic groups with the same qualifications. Also, there is a non-linear relationship between qualifications and earnings among refugees.

7.5 Having a degree or post-graduate level qualification from a British university increased the level of earnings among refugees. The average gross hourly rate of those with a degree or higher from a British university was £12.10 compared to an average rate of £8.23 among those who had obtained a degree from elsewhere. Thus studying at a high level in the UK helped to reduce the disparity between refugees and others though not eliminate it. However, the increase in pay associated with a degree from Britain suggests a lack of equivalence, by employers, of qualifications obtained elsewhere. In fact, only 16 per cent of those who arrived in Britain with a degree or higher had successfully got their qualification recognised.

7.6 One factor that did not interact with pay, in the refugee survey, was religion. The secondary analysis of the PSI fourth national survey found that Muslim respondents earned less than other religious groups (Lindley, 2002). Among refugees, the average hourly rate of pay was highest among Muslims (£7.89), lowest among non-Muslims (£6.48) while those with no religious affiliation held the middle ground (£7.19). However, like Brown (2000), the refugee survey found lower levels of economic activity among Muslims than others. While more than half (54 per cent) of those with no religious affiliation were working and 40 per cent of non-Muslims were working, only 21 per cent of Muslims were in paid work.

Job Seeking

8.1 Most refugees were either very satisfied or quite satisfied with their current job (76 per cent). Only nine per cent were quite dissatisfied or very dissatisfied and 14 per cent were neither satisfied nor dissatisfied. However, 31 per cent were looking for a new job to replace their present job and eight per cent were looking for an additional job. Among minority ethnic groups only eight per cent were looking for a new job, one per cent for an additional job while 91 per cent were not looking. The two reasons stated most often by refugees for looking for a new or additional job were because the pay in their current job was unsatisfactory (34 per cent) and because their present job might end (30 per cent) which reflects the casual nature of employment.

8.2 Methods of jobs seeking can be culturally specific (Marshall, 1989) and a comparison of the way in which refugees had found their last job before coming to Britain and their most recent job in Britain (i.e. those who were working or had worked in the past) shows the greater use of kinship networks in the country of origin compared to Britain.

Figure 4. Way in which refugees' had found most recent job in Britain and before coming to Britain (percentages)
Base before coming to Britain: 168
Base in Britain: 168
Other included: traineeship, political party and through other media.
Source: Bloch, A. (2002b), p.104

8.3 Figure 4 shows that in Britain refugees had found jobs through private agencies, community groups and from someone working in a place. None of these methods of job seeking had been used elsewhere.

8.4 Those who were not working were asked whether they were looking for work. The proportion of refugees looking for work was 37 per cent while among ethnic minorities it was 20 per cent. There was a large difference between men and women. Among ethnic minorities, a third of men were looking for work compared with 12 per cent of women. The proportions among refugees were 57 per cent and 23 per cent respectively. Those who were looking for work were asked what they had done in order to find a job. The responses in Figure 5 show the differences between refugees and other ethnic minorities.

Figure 5. Action taken to find a job: refugees and ethnic minorities (percentages)
Base number refugees: 149
* Ask community advisor was not included as a separate category in the LFS data.
** The refugee survey separated out friends and relatives and did not include colleagues and trade unions. The 70 per cent for refugees is the proportion that stated relatives or friends. When separated out, 66 per cent of refugees said they had asked friends and 48 per cent had asked relatives about work.
Source: Bloch, A. (2002b), p. 112

8.5 Refugees were more likely that minority ethnic people to use informal contacts and community advisors. Ethnic minorities were more likely than refugees to study the situation vacant listing in papers or journals, to answer advertisements or register with a private agency. There was little difference between refugees and minority ethnic groups either in the use of statutory job search provision or in the outcomes of using statutory services.

Conclusion

9.1 This paper has shown the relative disadvantage that refugees experience both in participating in the labour market and in their poor terms and conditions of employment when compared with minority ethnic groups more generally. Refugees have lower levels of employment, are more clustered in a few industries and occupations, work for less pay and with poorer terms and conditions of employment than their minority ethnic counterparts. Though refugees experience some of the same barriers as other minority ethnic people the data shows that they are clearly disadvantaged in the labour market and that the additional barriers that they face act as a real impediment to economic participation and to obtaining jobs that are commensurate with their skills and qualifications. The conclusion highlights the key factors that differentiate refugees from minority ethnic groups and in so doing affect economic activity.

9.2 First, are migration patterns to the UK with refugees arriving more recently than ethnic minority people. Different migration trajectories affect language, knowledge of UK employment structures and job seeking as well as access to social networks and sources of information all of which can further disadvantage refugees in the labour market.

9.3 Secondly, reasons for migration and attitudes to the country of origin can also be influential in migration outcomes. Historically there has been a strong and influential chain migration element to migration in the post-WWII period among Britain's ethnic minorities (see Castles and Miller, 2003) based on kinship networks that facilitate migration and help with accommodation and jobs for new arrivals by those already in the UK. Refugee migration, however, is often unplanned as some flee under acute circumstances (Kunz, 1981). Increasingly agents are playing a role in channelling asylum seekers to particular countries and sometimes this is without the asylum seeker making a choice about their destination (Robinson and Segrott, 2003). This means that asylum seekers can find themselves in countries where they have no pre-existing contacts and do not speak the language and so negotiating a new system and finding employment can be more difficult than for those who come to Britain with already established networks in place.

9.4 Thirdly, some refugees experience difficulties adjusting to a new society and may have mental and physical health problems from torture and imprisonment (Coate and Kamasa, 1997). In this study, more than one-fifth of refugees (21 per cent) said that they were not looking for work due to health problems. Women can also feel additional anxiety and stress due to separation from family members (Manderson et al, 1998). Certainly refugee women have disproportionately low levels of economic participation.

9.5 Fourthly, the circumstances of exile and attitudes towards the country of origin can affect settlement in the country of asylum (Kunz, 1981). Although just under two-thirds of refugees (65 per cent) said that they saw Britain as home, nearly half (47 per cent) said that they would like to return to their country of origin to live and a further 10 per cent said that they might like to return. Those in paid employment were slightly more likely to see Britain as home than those who were not (see Table 3) though being in paid employment did not influence whether or not refugees wanted to return to their country of origin. One-third of those who didn't want to return to their country of origin said that it was because they had now made their life in the UK, a quarter said that it was because of their children's future, one-fifth said that they did not want to return because it was unsafe and 10 per cent said it was because their family was in the UK. Economic issues were not mentioned as a factor influencing attitudes to return migration. However, those who did not want to return home tended to be doing higher status work including interpreting and translating than those who wanted to return home who were more likely to be working in catering or as shop assistants. Research with migrants carrying out irregular employment in Southern Europe noted that workers did not want to do work commensurate with their skills and qualifications as it indicated a level of commitment to the migration that they were not prepared to make (Reyneri, 2001). The situation of refugees and their employment profile might in some cases reflect this thesis.

9.6 The introduction of dispersal adds another element to the refugee experience that differentiates them from ethnic minorities. Earlier the variation in levels of employment by region was shown (see Table 3). Though policy has changed and now excludes all asylum seekers from permission to work, at the time of this survey dispersal meant that those asylum seekers who had been part of the dispersal programme were unable to move to seek employment.

9.7 Dispersal means that there is less access to social and community networks as they are less established in these newer areas of settlement. This in term can affect employment as these networks have, in previous research, been found important especially among those who were not fluent in English and were not therefore able to seek employment through statutory and formal mechanisms (Bloch, 2002a). In this research refugees in London were more likely to have found their current job through informal contacts (friends, relatives, community group or from someone already working at the organisation) than those in the regions (58 per cent and 37 per cent respectively). Clearly informal routes to employment were important in areas where these networks exist and the lack of networks might impede refugee employment.

9.7 The sixth factor differentiating refugees from others is immigration status. Uncertainty about immigration status affects the ability to make a long-term commitment to the country of asylum (Bloch, 2002a) and we have already seen that those with secure status are more likely to be working than those with ELR or asylum seekers on temporary admission. The asylum determination process can be a lengthy one. In this study 20 per cent of those on temporary admission had been n Britain for three years but less than five and a further 10 per cent had been in Britain for five years or more. This uncertainty can have an adverse affect on settlement.

9.9 Finally, refugees do experience structural barriers and the removal of the work concession in July 2002 that excludes asylum seekers from applying for permission to work may well have a longer-term affect on participation due to the initial period of unemployment. It is also likely to result in deskilling so that those with professional qualifications are less likely to ultimately engage in work commensurate with their skills (Stewart, 2003).

9.10 Some of the barriers to the labour market are as a result of being a refugee, such as migration patterns, while others are more structural and are due to policy such as dispersal and excluding asylum seekers from the labour market. This is within the current migration policy context of 'managed migration' which grants permits for entry under schemes such as the Highly Skilled Migrant Programme and the Seasonal Agricultural Workers' Scheme to attract workers with skills that are already present in the UK among refugees. Given the obvious disparity in employment outcomes between refugees and minority ethnic people and the skills and experience that refugees bring with them, perhaps working towards decreasing the structural barriers to employment faced by refugees might be a good starting point for helping their economic adaptation, for decreasing their obvious disadvantage in Britain and for utilising skills that are already present.


Notes

1 The term refuge will be used to describe all forced migrants (that is refugees, people with Exceptional Leave to Remain, Indefinite Leave to Remain, asylum seekers on temporary admission and naturalised British and EU citizens who came to Britain initially as forced migrants), unless a distinction is specified. On some occasions the distinction is important due to the hierarchical set of rights associated with different statuses with asylum seekers having the fewest rights because they and are excluded from employment services and the labour market. For more about differential rights see for example Zetter and Pearl, 1999 and Sales, 2002.

2 The research was commissioned by the Department for Work and Pensions

3English language standard was self-assessed.

4 Using Standard Industrial Classification 1992 (SIC1992)


Acknowledgements

Thanks to Liza Schuster and Jo Bacon for commenting on drafts of the paper and to the three anonymous referees for their helpful comments. Thanks also to Dave Wilkinson for the LFS data.

References

AL-RASHEED, M. (1994) 'The myth of return: Iraqi Arab and Assyrian refugees in London', Journal of Refugee Studies Vol. 7, No. 2/3, pp. 199-219.

BERTHOUD, R. (1999) Young Caribbean men in the labour market - a comparison with other ethnic groups, York: YPS for Joseph Rowntree Foundation. BERTHOUD, R. (2000) 'Ethnic employment penalties in Britain', Journal of Ethnic and Migration Studies, Vol. 26, No. 3, pp. 389-416.

BLOCH, A. (1999) 'Carrying out a survey of refugees: some methodological considerations and guidelines', Journal of Refugee Studies, Vol. 12, No. 4, pp. 367-84.

BLOCH, A. (2002a) The Migration and Settlement of Refugees in Britain, Basingstoke: Palgrave.

BLOCH, A. (2002b) Refugees' Opportunities and Barriers in Employment and Training, Research Report 179, Leeds: Department for Work and Pensions http://www.dwp.gov.uk/asd/asd5/rrep179.asp

BLOCH, A. AND ATFIELD, G. (2002) The Professional Capacity of Nationals from the Somali Regions in Britain, London: Refugee Action and the International Organisation for Migration, unpublished report.

BRISLIN, R., LONNER, W. AND THORNDIKE, R. (1973) Cross- Cultural Research Methods, New York: Wiley-Interscience.

BROWN, M.S. (2000)'Religion and economic activity in the South Asian population, Ethnic and Racial Studies, Vol. 23, No. 6, pp. 1035-1061.

BROPHY, M., MAXEY, K., ABRAHAM, T., AYBEK, T., BIRD, P. and STANIKZAI, F. (1998) Refugee Education, Training and Employment in Inner London, London: FOCUS TEC.

CAREY-WOOD, J., DUKE, K., KARN, V. AND MARSHALL, T. (1995) The settlement of refugees in Britain, London: HMSO.

CASTLES, S. AND MILLER, M. (2003) The Age of Migration, Basingstoke: Palgrave, 3rd Edition.

COATE, J. and KAMASA, K. (1997) Refugees and Asylum Seekers in Haringey, London: London Borough of Haringey.

DUKE, K, SALES, R. AND GREGORY, J. (1999) 'Refugee resettlement in Europe', Chapter 6 in Author, A. and Levy, C. (Editors.) Refugees, Citizenship and Social Policy in Europe, Basingstoke: Macmillan.

GIRBASH, C. (1991) Manchester Vietnamese Employment and Training Survey, Manchester, The Centre for Employment Research, Manchester Polytechnic.

HO, S.Y. AND HENDERSON, J. (1999) 'Locality and the variability of ethnic employment in Britain', Journal of Ethnic and Migration Studies Vol. 25, No. 2, pp. 323-333.

KUNZ, E.F. (1981) 'Exile and resettlement: Refugee theory', International Migration Review, Vol.15, No.1, pp. 42-51.

LESLIE, D., DRINKWATER, S. AND O'LEARY, N. (1998) 'Unemployment and earnings among Britain's ethnic minorities: some signs for optimism', Journal of Ethnic and Migration Studies, Vol. 24, No. 3, pp. 489-506.

LINDLEY, J. (2002) 'Race or religion? The impact of religion on the employment and earnings of Britain's ethnic communities', Ethnic and Racial Studies, Vol. 28, No. 3, pp. 427-442.

LONDON RESEARCH CENTRE (1998) Refugees: Breaking down the barriers to work, London: London Research Centre and Peabody Trust.

MALIK, F. (1990) A Survey of the Armenian Community in London, London: London Research Centre.

MANDERSON, L., KELAHER, M., and MCMANUS, K. (1998) 'A women without a man is a woman at risk', Journal of Refugee Studies, Vol. 11, No. 3, pp. 267- 283.

MARSHALL, T. (1989) Cultural Aspects of Job Hunting, London: Refugee Council.

METCALF, H., MODOOD, T. AND VIRDEE, S. (1997)Asian Self- employment: The interaction of culture and economics, London: Policy Studies Institute.

MODOOD, T., BERTHOUD, R., LAKEY, J., NAZROO, J., SMITH, P., VIRDEE, S. AND BEISHON, S. (1997) Ethnic Minorities in Britain - diversity and disadvantage, London: PSI.

OWEN, D., REZA, B., GREEN, A., MAGUIRE, M. AND PITCHER, J. (2000) 'Patterns of labour market participation in ethnic minority groups', Labour Market Trends, November, pp. 505-510.

OWEN, D. (2003) 'The demographic characteristics of people from ethnic minority groups in Britain', Chapter 3 in Mason, D. (Editor) Explaining Ethnic Difference: Changing patterns of diversity, Bristol: Policy Press.

PARKINSON, R. (1998) Survey of Refugee Needs in Hackney for the Hackney Refugee Training Consortium, Research Consultancy Training: Cardiff.

ROBINSON, V. (1993) 'Marching into the Middle Classes? The Long-term Resettlement of East African Asians in the UK', Journal of Refugee Studies Vol. 6, No. 3, pp. 230-247.

ROBINSON, V. (1998) 'The importance of information in the resettlement of refugees in the UK', Journal of Refugee Studies Vol. 11, No. 2, pp. 146- 160.

ROBINSON, V. and SEGROTT, J. (2003) Understanding the Decision Making of Asylum Seekers, Research Study 243, London: Home Office.

SALES, R. (2002) 'The deserving and the undeserving? Refugees, asylum seekers and welfare in Britain' Critical Social Policy, Vo. 22, No. 3, pp 456-478.

SARGEANT, G. AND FORNA, A. (2001) A Poor Reception: Refugees and asylum seekers: welfare or work, London: The Industrial Society.

SRINIVASAN, S. (1992) 'The class position of the Asian petty bourgeoisie', New Community, Vol. 19, No. 1, pp. 61-74. SLY, F., THAIR, T. AND RISDON, A. (1999) Trends in the labour market participation of ethnic groups', Labour Market Trends, pp. 631-639.

STEWART, E. (2003) 'A bitter pill to swallow: Obstacles facing refugee and overseas doctors in the UK', New Issues in Refugee Research, Working Paper 96, Geneva: UNHCR.

TWOMEY, B. (2001) 'Labour market participation of ethnic groups', Labour Market Trends, January, pp. 29-42.

VALTONEN, K. (1999) 'The societal participation of Vietnamese refugees: case studies in Finland and Canada', Journal of Ethnic and Migration Studies Vol. 25, No. 3, pp. 469-491.

WOODEN, M (1991) 'The experiences of refugees in the Australian labour market', International Migration Review, Vol. 25, No. 3, pp. 514-534.

ZETTER, R. and PEARL, M. (1999) 'Sheltering on the margins: Social housing provision and the impact of restrictionism on asylum seekers and refugees in the UK', Policy Studies, Vol. 20, No. 4, pp. 235-254.