European Immigrants in the UK Before and After the 2004 Enlargement

In progress European Immigrants in the UK Before and After the 2004 Enlargement Simonetta Longhi (1) and Magdalena Rokicka (1,2) (1) Institute for Social and Economic Research, University of Essex (2) IBE- Educational Research Institute Warsaw Part of the project Migrant Diversity and Regional Disparity in Europe. Financial support from NORFACE research programme on Migration in Europe - Social, Economic, Cultural and Policy Dynamics is acknowledged

2004 European Enlargement Free movement of people/workers among EU15 countries Enlargement in 2004 to 8(+2) Eastern European (low wage) countries Most of the EU15 made temporary exception to free movement from EU8 countries Only Ireland, Sweden and the UK allowed free movement from the day of accession What happened in the UK?

Dark green: EU15 Lighter green: EU8 (2004 enlargement) Lightest green: enlargement 2007 Source: http://en.wikipedia.org/wiki/ File:EU2004-2007.svg

2004 European Enlargement Free movement of people/workers among EU15 countries Enlargement in 2004 to 8(+2) Eastern European (low wage) countries Most of the EU15 made temporary exception to free movement from EU8 countries Only Ireland, Sweden and the UK allowed free movement from the day of accession What happened in the UK?

Empirical Research Questions Are EU8 immigrants arrived after 2004 different from those arrived before? Are EU8 immigrants different from EU15 immigrants? How do EU8 and EU15 immigrants compare with British people? How do they compare with people who remained in the country of origin? (Ireland, Germany, Poland)

How do EU8 immigrants arrived after 2004 compare - to those arrived before 2004? - to other EU15 immigrants? - to British people?

UK Labour Force Survey Household dataset focusing on labour market and job characteristics of individuals living at private addresses in the UK Immigrants are captured in the dataset, but does not focus on immigrants Questions on country of birth, nationality, date of arrival in the UK We use data 1997-2010 (plan to extend to 2013)

2011 Stock of EU Immigrants in the UK Num ber of Im m igrants 0 1 00 00 0 2 00 00 0 3 00 00 0 4 00 00 0 0 N um ber of Im m igrants 5 00 00 0 1 00 00 00 1 50 000 00 1998q1 1999q1 2000q1 2001q1 2002q1 2003q1 2004q1 2005q1 2006q1 2007q1 2008q1 2009q1 2010q1 2011q1 1996 1997 1998 1999 2000 2001 2002 2003 0 2004 2005 2006 2007 2008 2009 2010 quarter year Tot EU15 Immigrants Tot EU8 Immigrants Tot EU15 Immigrants Tot EU8 Immigrants 1997q1

Evolution of European Immigration 1997 Of all immigrants: more than 45% from Ireland; 20% from Germany Among recent immigrants: 25% from Ireland; 25% from Germany; less than 20% from France 2004 Of all immigrants: more than 35% from Ireland; 20% from Germany Among recent immigrants: almost 25% from Poland; less than 20% from Germany; about 10% from Lituania 2010 Of all immigrants: more than 25% from Poland; more than 20% from Ireland; about 15% from Germany Among recent immigrants: 25% from Poland; more than 10% (each) from Lithuania, Latvia, Hungary

Models y * it ' ' = X i j + TI β 1 i 2 β + ε j ij * y it Models analyse differences in ( ) : Region of residence (12 areas) Participation (active, student, other inactive) Employment (employed, self-employed, unemployed) Job characteristics (temporary + part-time + second job dummies) Industry (8 industries) Wages Aim: analyse differences between (white) British, EU15, and EU8 immigrants, and between immigrants arrived before and after 2004 Expect differences pre-post 2004 for EU8 but not for EU15 TI i : dummies for EU15, EU8, EU15 arrived after 2004, EU8 arrived after 2004 Other controls: (sex, age), marital status, whether dependent children in household, dummies for qualification levels, (dummies for part-time, temporary job) dummies for quarter and year of the interview

Region of Residence Reference: (1) (2) (3) (4) (5) (6) London North East Yorkshire and the Humber East Midlands East Anglia Rest of South East South West EU15-0.041* -0.035* -0.021* 0.002 0.059* 0.001 EU8-0.039* -0.019 + -0.017 + 0.016* 0.078* -0.026* EU15 arriving on or after 2004-0.011-0.015 0.011 0.024* -0.015-0.051* EU8 arriving on or after 2004-0.009 0.021 + 0.042* 0.002-0.105* 0.006 (7) (8) (9) (10) (11) West Midlands North West Wales Scotland Northern Ireland EU15-0.018* -0.029* -0.022* -0.037* 0.017* EU8-0.040* -0.070* -0.044* -0.026* -0.041* EU15 arriving on or after 2004-0.040* 0.005 0.007 0.040* 0.008 EU8 arriving on or after 2004 0.028* 0.049* 0.027* 0.029* 0.067* Observations 789,916 Log likelihood -1,875,153 Marginal effects of a multinomial probit model; + Significant at 5%, * Significant at 1%. Other explanatory variables: dummies for women; married; whether dependent children; dummies for medium, low, other qualification; dummies for period of arrival (before 1992; after 2007); quarters and year of the survey

Region of Residence Reference: (1) (2) (3) (4) (5) (6) London North East Yorkshire and the Humber East Midlands East Anglia Rest of South East South West EU15-0.041* -0.035* -0.021* 0.002 0.059* 0.001 EU8-0.039* -0.019 + -0.017 + 0.016* 0.078* -0.026* EU15 arriving on or after 2004-0.011-0.015 0.011 0.024* -0.015-0.051* EU8 arriving on or after 2004-0.009 0.021 + 0.042* 0.002-0.105* 0.006 (7) (8) (9) (10) (11) West Midlands North West Wales Scotland Northern Ireland EU15-0.018* -0.029* -0.022* -0.037* 0.017* EU8-0.040* -0.070* -0.044* -0.026* -0.041* EU15 arriving on or after 2004-0.040* 0.005 0.007 0.040* 0.008 EU8 arriving on or after 2004 0.028* 0.049* 0.027* 0.029* 0.067* Observations 789,916 Log likelihood -1,875,153 On average EU15 and EU8 immigrants concentrate in London and its surrounding area (East Anglia and the Rest of the South East) Marginal effects of a multinomial probit model; + Significant at 5%, * Significant at 1%. Other explanatory variables: dummies for women; married; whether dependent children; dummies for medium, low, other qualification; dummies for period of arrival (before 1992; after 2007); quarters and year of the survey

Region of Residence Reference: (1) (2) (3) (4) (5) (6) London North East Yorkshire and the Humber East Midlands East Anglia Rest of South East South West EU15-0.041* -0.035* -0.021* 0.002 0.059* 0.001 EU8-0.039* -0.019 + -0.017 + 0.016* 0.078* -0.026* EU15 arriving on or after 2004-0.011-0.015 0.011 0.024* -0.015-0.051* EU8 arriving on or after 2004-0.009 0.021 + 0.042* 0.002-0.105* 0.006 (7) (8) (9) (10) (11) West Midlands North West Wales Scotland Northern Ireland EU15-0.018* -0.029* -0.022* -0.037* 0.017* EU8-0.040* -0.070* -0.044* -0.026* -0.041* EU15 arriving on or after 2004-0.040* 0.005 0.007 0.040* 0.008 EU8 arriving on or after 2004 0.028* 0.049* 0.027* 0.029* 0.067* EU8 immigrants arrived after May 2004 (enlargement) are more dispersed across UK regions and less likely to be in London (and in the Rest of the South East) than all other immigrants and white British people. They are more likely to locate in the Midlands, North West, Yorkshire, Scotland and Wales compared to those arrived before 2004 Observations 789,916 Log likelihood -1,875,153 Marginal effects of a multinomial probit model; + Significant at 5%, * Significant at 1%. Other explanatory variables: dummies for women; married; whether dependent children; dummies for medium, low, other qualification; dummies for period of arrival (before 1992; after 2007); quarters and year of the survey

Labour Force Participation (1) (2) Reference: Active Men Women Student Other inactive Student Other inactive EU15 0.034* 0.023 0.018* 0.017 EU8 0.016 + 0.048* 0.001 0.032 EU15 arriving on or after 2004 0.005-0.030 0.006 0.027 EU8 arriving on or after 2004-0.028* -0.105* -0.039* -0.071* Observations 397,468 387,151 Log likelihood -165873-231752 Immigrants are slightly more likely than British people to be students. Those immigrants from EU8 countries arrived after the enlargement, however, are comparatively less likely to be inactive than British people: they may come to the UK for work-related rather than study-related reasons. Marginal effects of a multinomial probit model; + Significant at 5%, * Significant at 1%. Other explanatory variables: age; years in the UK (age for natives); dummies for women; married; whether dependent children; dummies for medium, low, other qualification; dummies for period of arrival (before 1992; after 2007); dummies for regions, quarters and year of the survey

Employment Status (1) (2) Reference: In paid employment Men Women Self-employed Unemployed Self-employed Unemployed EU15-0.010 0.000 0.005-0.022* EU8 0.118* 0.008 0.030* -0.025* EU15 arriving on or after 2004-0.061 + 0.010-0.007 0.000 EU8 arriving on or after 2004-0.185* -0.036* -0.054* -0.011 Observations 328,017 282,285 Log likelihood -203548-116769 EU8 immigrants are more likely to be self-employed than British people, but those arrived after the enlargement are less likely to be self-employed than those arrived before. They are also less likely to be unemployed. Maybe self-employment was a forced choice for those arrived before 2004, maybe lack of entrepreneurial spirit or have higher motivation to take paid employment Maybe new EU8 immigrants are more likely to be temporary migrants, less willing to invest in self-employment and more likely to accept lower quality jobs (i.e. temporary and have comparatively lower wages) Marginal effects of a multinomial probit model; + Significant at 5%, * Significant at 1%. Other explanatory variables: age; years in the UK (age for natives); dummies for women; married; whether dependent children; dummies for medium, low, other qualification; dummies for period of arrival (before 1992; after 2007); dummies for regions, quarters and year of the survey

Industry (I) Men (1) (2) (3) (4) (5) (6) Ref: Agriculture, energy Manufacturing Construction Distribution hotels restaurants Financial services Transport Public sector EU15 0.046 + -0.112* -0.006 0.035* 0.040* 0.026 EU8 0.048 0.059* -0.040-0.009-0.032-0.022 EU15 arriving on or after 2004 0.031-0.052 0.003-0.001 0.027-0.013 EU8 arriving on or after 2004 0.140* -0.131* 0.016 0.069* -0.022* -0.069* Observations 293,949 Log likelihood -522109 Women (1) (2) (3) (4) (5) Ref: Agriculture energy manufacturing Construction Distribution hotels restaurants Financial services Transport Public sector EU15 0.019 0.014 0.006 0.007-0.049 + EU8 0.035 + 0.105* -0.008-0.007-0.128* EU15 arriving on or after 2004 0.032 0.056 + -0.007-0.030-0.031 EU8 arriving on or after 2004 0.103* -0.045 + 0.030* -0.028-0.070 + Observations 252,045 Log likelihood -336479 Marginal effects of a multinomial probit model; + Significant at 5%, * Significant at 1%. Other explanatory variables: age; years in the UK (age for natives); dummies for women; married; whether dependent children; dummies for medium, low, other qualification; dummies for period of arrival (before 1992; after 2007); dummies for regions, quarters and year of the survey

Industry (II) Compared to British men EU15 more likely to work in manufacturing, financial services and transport, and less likely to work in construction EU8 more likely than British men to work in construction EU15xAfter No clear differences from those arrived before EU8xAfter more likely to work in manufacturing and financial services and less likely to work in construction, transport and the public sector Compared to British women EU15 less likely to work in the public sector EU8 less likely to work in the public sector and more likely to work in construction and in distribution hotels and restaurants EU15xAfter No clear differences from those arrived before EU8xAfter more likely to work in construction, financial services, but less likely to work in distribution hotels and restaurants, and in the public sector

Job Characteristics (I) (1) (2) (3) Temporary job Part-time job Second job Men Women Men Women Men Women EU15 0.017 + -0.020* -0.005 0.003-0.001 0.001 EU8 0.022-0.032* 0.002-0.009-0.008-0.007 EU15 arriving on or after 2004 0.001-0.028* 0.003-0.015-0.033 + 0.039 + EU8 arriving on or after 2004 0.012-0.004-0.041* 0.152* -0.011 0.018 Observations 255,124 249,312 355,891 327,835 308,629 268,948 Log likelihood -46590-53734 -98793-203836 -42449-56392 Immigrant women are less likely to have a temporary job than British women. EU8 immigrant men arrived after the enlargement are less likely than British men and all other immigrants to hold a part-time job, while EU8 immigrant women are more likely to hold a part-time job. Similar results for second jobs for EU15 immigrants arrived after the enlargement Marginal effects of a multinomial probit model; + Significant at 5%, * Significant at 1%. Other explanatory variables: age; years in the UK (age for natives); dummies for women; married; whether dependent children; dummies for medium, low, other qualification; dummies for period of arrival (before 1992; after 2007); dummies for regions, quarters and year of the survey

Job Characteristics (Wages) (1) Men (2) Women EU15 0.134*** 0.016 EU8 0.021-0.233*** EU15 arriving on or after 2004-0.001-0.035 EU8 arriving on or after 2004-0.186*** 0.034 R2 0.412 0.416 Observations 179,166 186,867 Immigrants from EU15 countries earn higher hourly wages than British people; no difference between EU8 immigrants arrived before the enlargement and British people; those arrived after the enlargement suffer a wage penalty (maybe related to their higher probability of having a paid job) EU8 women earn less than British women, but no additional wage penalty for those who arrived to the UK after the enlargement. Language proficiency? Temporary immigrants? + Significant at 5%, * Significant at 1%. Other explanatory variables: age; years in the UK (age for natives); years of tenure in the job; dummies for women; married; whether dependent children; part-time; temporary job; dummies for medium, low, other qualification; dummies for period of arrival (before 1992; after 2007); dummies for regions, quarters and year of the survey

How do immigrants compare to those who remain in their country? (those who did not migrate) The case of Ireland, Germany, and Poland

EU Labour Force Survey Similar to UK LFS (household dataset focusing on labour market characteristics) Collected in all EU countries; harmonised across countries less detailed but reasonably comparable across countries Data availability (years of data collection) differs across countries Focus on Ireland, Germany and Poland LFS: the largest immigrant groups (enough observations in the UK LFS for comparison with Irish, German and Polish LFS)

Models y * ' it = X i 1 j γ + ε ij Same models as previous analysis but simpler; estimated separately for types of immigrants and countries of origin * Models analyse differences in ( y it ) : Employment (employed, self-employed, unemployed) Job characteristics (temporary + part-time + second job dummies) Other controls: (sex, age), marital status, dummies for qualification levels, dummies for (quarter and) year of the interview

Summary of Descriptive Stats With the exception of Irish people, migrants are younger and less likely to be married than the average (working age) population in the sending country Immigrants from Ireland and Germany have on average higher levels of education than those remaining in their country of origin; most Polish immigrants have levels of education which are not easily translated into British qualification ( other qualifications ) For Irish and German people, the distribution across activity levels is similar for immigrants and for those remaining in the country of origin. Polish people who migrated to the UK are comparatively more likely to be in paid employment and less likely to be self-employed, unemployed or inactive, compared to Polish people living in Poland

Summary of Model Results (2009) Irish and Polish people arrived in the UK before the enlargement have a larger probability of being self-employed than similar people in Ireland and Poland. If arrived after the enlargement, both Irish and Polish immigrants to the UK are less likely to be self-employed than those arrived before and than similar people who remained in the country of origin. German men are less likely to be self-employed in the UK than similar men remaining in Germany. The opposite seems to be true for German women. No differences across arrival cohorts Immigrants seem much less likely to be unemployed than similar people who did not migrate people may migrate when they have a good chance of securing a job abroad. Immigrants are also less likely to have a temporary or parttime job than similar people remaining in the country of origin.

Preliminary Conclusions There are clear differences between EU15 and EU8 immigrants and between EU8 immigrants arrived before and after the 2004 enlargement (in terms of region of residence, labour force participation and job characteristics) There are differences between immigrants and individual with similar observed characteristics who remain in their country of origin (but need to clarify how much of the difference may be due to differences in each country s labour market) Changes in immigration rules may have caused changes in the quality of immigrants arriving to the UK more work needed on this issue