Mixed Race Studies

A Longitudinal Study of Migration Propensities for Mixed Ethnic Unions in England and Wales

The Institute for the Study of Labor (IZA)
Bonn, Germany
Discussion Paper No. 6394
February 2012
21 pages

Zhiqiang Feng, Research Fellow
University of St. Andrews

Maarten van Ham, Professor of Urban Renewal
Delft University of Technology and IZA

Paul Boyle, Professor of Geography and Sustainable Development
University of St. Andrews

Gillian M. Raab, Research Fellow
University of St. Andrews

Most studies investigating residential segregation of ethnic minorities ignore the fact that the majority of adults live in couples. In recent years there has been a growth in the number of mixed ethnic unions that involve a minority member and a white member. To our knowledge, hardly any research has been undertaken to explicitly examine whether the ethnic mix within households has an impact on the residential choices of households in terms of the ethnic mix of destination neighbourhoods. Our study addresses this research gap and examines the tendencies of migration among mixed ethnic unions in comparison with their co-ethnic peers. We used data from the Longitudinal Study for England and Wales. Our statistical analysis supports the spatial assimilation theory: ethnic minorities move towards less deprived areas and to a lesser extent also towards less ethnically concentrated areas. However, the types of destination neighbourhood of minority people living in mixed ethnic unions varied greatly with the ethnicity of the ethnic minority partner.

INTRODUCTION

Residential integration is regarded as a measure of structural assimilation of ethnic minority populations and has drawn long-standing interest from academic studies (Park and Burgess 1969; Lieberson 1963; Massey 1985; Allen and Turner 1996). Residential integration is not only an indicator of the degree of ethnic assimilation, but also further enhances social and cultural integration. Conversely, ethnic segregation is deemed to hinder social interaction with majority populations, and to marginalise ethnic minority populations. Hence the British government has increasingly promoted community cohesion and residential integration.

While a body of research has examined aggregate levels of residential segregation of ethnic minority groups and the cross-sectional residential locations of ethnic minority populations at the individual level, few studies have examined the determinants of the actual residential migration of ethnic minorities in relation to characteristics of neighbourhoods of origin and destination (Finney and Simpson 2008). Little is known about how ethnic minority people move between neighbourhoods with different levels of concentration of their own groups and with different levels of deprivation. Most existing studies of ethnic segregation ignore the fact that the majority of adults live in couples. In recent years there has been a growth in the number of mixed ethnic families that involve a minority member and a white member (Feng et al, 2010). However, to our knowledge, almost no research has been undertaken to explicitly examine whether the ethnic mix within households has an impact on tendencies of residential migration between different types of neighbourhood. In the US, a few studies which examined the residential locations (but not mobility) of ethnic populations, have taken the ethnic mix within households into account. Ellis et al. (2006) used cross-sectional data in the US and came to the conclusion that mixed-ethnic households are less likely to live in minority ethnic neighbourhoods. White and Sassler (2000) also used US census data and found that Latinos and blacks who married a white spouse were more likely to reside in higher status neighbourhoods, while in contrast the marriage of a white person to a non-white person seemed to result in them residing in a lower-status neighbourhood than they might otherwise have done. Although Ellis et al (2006) argued that their results are more likely due to mixed-ethnic couples choosing to live in mixed-ethnic neighbourhoods, rather than mixed neighbourhoods ‘creating’ these couples, it is difficult with cross-sectional data to come to any firm conclusion about this. The same is true for the study by White and Sassler (2000) due to the use of cross-sectional data. In their review of geographies of mixed ethnic unions, Wright et al (2003) called for further research on migration of mixed ethnic unions in a longitudinal perspective.

With this study we fill this gap, and use longitudinal data from the Office for National Statistics Longitudinal Study (ONS LS), to explore whether minority people in mixed ethnic unions were more likely to move to areas which are less concentrated in their own group than ethnic minorities living in mono ethnic unions. In our analyses we also take the level of deprivation of neighbourhoods into account…

…In the past decades Britain has witnessed a growing ethnic diversity in populations. In England, for example, the percentage of ethnic minorities has risen from 4.6 % to 8.6 % between 1981 and 2001 (Rees and Butt 2004). It is estimated that nearly a million people report themselves as having a mixed-ethnic identity in Britain today (CRE 2006). Along with the trend in diversity the number of marriages and partnerships between people of different ethnic groups is also on the rise (Aspinal 2003; Coleman 1985; 2004; Voas 2009; Song 2010). The one per cent census sample from the Office for National Statistics (ONS) Longitudinal Study (LS) reveals that the total number of mixed ethnic unions reached 5,139 in 2001 in England and Wales, a 46 Per cent increase from 1991 (Feng et al. 2010)…

…The ONS LS was a unique and very rich dataset. However, we acknowledge that the data has some limitations. We did not have information on migration between two censuses. Some couples might move more than once between 1991 and 2001. The British Household Panel Survey (BHPS) is a panel dataset which provides annual information for sample couples. However, the number of mixed ethnic unions in the BHPS is too few for a meaningful statistical analysis. The other limitation is the self reported ethnicity can change over time. It is not a big problem for South Asians as they reported their ethnic identity very consistently over time. But the consistency was not high for Black Others who were part of the Black group in our analysis (Platt et al 2005). Therefore our results here should be treated with caution…

Read the entire paper here.