Migration strand abstracts


Immigrant family & fertility. Monday 12 September 16:45pm 

Migrant Total Fertility Rates: Time for a rethink?
Michael Murphy, Department of Social Policy, London School of Economics and Political Science   

Migrant Total Fertility Rates: Time for a rethink? Fertility of migrants is increasingly measured by the indicator of Total Fertility Rate (TFR) for migrant populations. We question whether this indicator can be usefully interpreted as a measure of relative fertility for such populations. The discussion is organised as follows. We summarise literature on the construction and interpretation of period and cohort fertility measures of family size. The conventional TFR on which it is based has some limitations as an indicator of the annual level of fertility. These are discussed and it is argued that these problems are more severe when migrant TFRs are produced. Migrants often have high fertility shortly after arrival, which inflates measures such as TFRs and invalidates comparisons of underlying migrant and native fertility. Substantial numbers of migrants may also distort the interpretation of national-level cohort estimates of fertility, especially of childlessness. We consider construction of migrant-specific TFRs. We embed discussion of migrant TFRs within a wider set of similar measures, especially total marital fertility rates (TMFR) that have addressed similar problems of interpretation. We discuss the construction of indicators of migrant fertility in relation to the uses of such measures as well as use of existing approaches. Some solutions to these problems are discussed. 


Ethnic differentials in the uptake of (in)formal childcare in Belgium and effects on parity progression
Karel Neels, Jonas Wood, University of Antwerp   

The positive association between fertility and female employment in OECD countries suggests that family policies have played an important role in reducing the ‘parent-worker’ conflict. The empirical literature, however, finds only small positive effects of family policies on fertility, but has typically failed to consider eligibility and uptake of family policies at the individual level, as well as population heterogeneity in the uptake and effect of these policies. Using longitudinal individual-level data from the 2001 Census and the National Register, we document ethnic and educational differentials in the uptake of formal childcare (kindergarten, daycare mothers) and informal childcare arrangements (family or household members) following a first birth in Belgium in 2001 and analyze the effect on subsequent parity progression in the period 2002-2005. Controlling for education and age at first birth, results show that both first and second generation migrant women from Southern Europe, Turkey and Morocco are more likely than Belgian women to have no care arrangement or only rely on informal care arrangements. Among Belgian women uptake of informal and particularly formal childcare arrangements is associated with higher second birth hazards, particularly among higher educated women and women with strong labour market attachments. This positive effect is less articulated among first and second generation migrant women, suggesting selective (non)uptake of family policies in migrant populations. 


What does the recent refugee crisis mean for the future of European demography? From policies and statistics to the voices of Syrian refugees
Ben Wilson 1, Ellie Ga 2, Dominik Hangartner 1, Dwan Kaoukji 3, 1The London School of Economics and Political Science, 2 Swedish Research Council, 3 BBC Media Action   

Europe’s recent refugee crisis is expected to have far-reaching implications, many of which are of interest to demographers. Yet at present, there is an almost total absence of demographic research on the crisis, its impact, or the refugees themselves. Similarly, there is a lack of research that considers the implications of the crisis for the future of European demography. This article responds to these issues with a mixed methods analysis of two new data sources: a survey of refugees who had recently arrived in Lesvos (Greece) at the beginning of December 2015, and a second data source based on eleven semi-structured interviews that were carried out with female Syrian refugees two weeks earlier. The results show that there is considerable variation in the characteristics and circumstances of male and female Syrian refugees, at least with respect to those who have chosen to come to Europe from Turkey via the Aegean Sea. In addition, the interviews with Syrian women demonstrate that there are a complex web of motivations surrounding their decisions to migrate and their plans for settling in Europe. The analysis concludes with a series of recommendations for future research, including a discussion of the role that demographers have in addressing the future needs of refugees and asylum seekers. 


Internal migration. Tuesday 13 September 11:00am 

Why do we move? An analysis of census microdata
Fran Darlington-Pollock 1, Nik Lomax 2, Paul Norman 2, 1 Queen Mary, University of London, 2 University of Leeds 

In this paper, we examine whether propensity to migrate has changed between 1991, 2001 and 2011, using the latest available census microdata. In the ‘age of migration’, exploring whether and how propensity to migrate varies according to different attributes is essential if we are to understand how best to meet the needs of (im)mobile groups. Migration is inherently selective: likelihood of migrating varies according to attributes such as age, sex, ethnicity, educational attainment, tenure, social class and health. Differences in the nature of the migration event, defined by distance moved, have also been determined. Using regression analysis, we model the likelihood of migrating (within Great Britain) as explained by the attributes listed above. We will show whether (a) the influence of these attributes on propensity to migrate (and distance moved) has changed over time, and (b) whether this varies between population subgroups stratified by age, sex and ethnicity. Our analysis reveals that the nature of the relationship between key determinants of migration, such as health or social class, and propensity to migrate has changed over time for different population subgroups. For example, younger migrants are more likely to be in good health than older migrants, with the inverse being true for older migrants. The age at which poor health rather than good health is associated with a higher likelihood of migration has increased. 


40 years of (not) moving: Identifying residential mobility trajectories with the British Cohort Study 1970
Nissa Finney 1, Lee Bentley 2, Naomi Tyrrell 3, 1 University of St Andrews, 2 University of Manchester, 3 Plymouth University   

Although theoretical understandings into migration and the life course have developed a great deal in recent decades, empirical studies devoted to this topic remain limited. This paper uses birth cohort data from the British Cohort Study 1970, collected for over 40 years, in order to track whole lifecourse patterns of residential mobility in Britain. This enables the following questions to be addressed: Can distinct residentially mobility trajectories be identified, from birth to mid-adulthood? How can the residential mobility trajectories be characterised? How are the residential mobility trajectories related to childhood context? How are the residential mobility trajectories associated with adult socio-economic and wellbeing outcomes? To investigate these questions, the paper develops a typology of residential movement from birth to middle-age, using sequence analysis techniques, and relates these to characteristics of both the parental household and the cohort member. Although a small number of studies have employed sequence analysis for investigating residential mobility, this study is original in its use of cohort data and focus from birth to mid-adulthood. Thus, the paper develops new insights into the way in which early and later life residential instability/mobility are related to one another, as well as to outcomes such as socio-economic position. 


New insights into migration and commuting derived from the Annual Survey of Hours and Earnings
Tony Champion 1, David Brown 2,1 Newcastle University, 2 Cornell University 

The literature on migration and commuting is very rich, but is lacking in two significant ways. One is that the primary thrust of these is cross-sectional, relating to a single point in time, as derived from separate censuses, rather than tracing individuals longitudinally. Secondly, the work on migration is focused almost exclusively on residential mobility, with very little attention being given to workplace mobility and thus providing very little intelligence about the changes in commuting journeys produced by changes of home or work address. The Annual Survey of Hours and Earnings (ASHE) has the potential for plugging these gaps, as it tracks a 1% sample of employees based on National Insurance number. ASHE has been running since 2002 on its current basis which includes geographical details of both home and workplace addresses, which can be recoded to a variety of geographies including rural-urban location and size of settlement. The paper goes on to demonstrate the value of ASHE in being able to identify frequency of moving home for this population, as well as whether a change of residence is accompanied by change of workplace in the same year or subsequently and, more generally, what effect any change of residence and/or workplace has on the length of the commuting trip. Policy relevance will also be discussed. 


Separation and Spatial Mobility: A Cross-National Comparison
Hill Kulu, 1, Julia Mikolai 1, Sergi Vidal 2, Christine Schnor 3, Didier Willaert 4, Fieke H. L. Visser 5, Clara H. Mulder 5, Michael J. Thomas 5, 1 University of Liverpool, 2 University of Queensland, 3 University of Leuven 4 Vrije Universiteit, Brussels, 5 University of Groningen 

This study investigates spatial mobility of separated individuals in five industrialised countries. While there is a large body of literature examining residential changes related to separation in selected individual countries, only a few studies have compared patterns across countries. Using longitudinal data and applying Poisson regression models we study the risk of a move of separated men and women in comparison with cohabiting and married individuals. We use time since separation to distinguish between moves due to separation and moves of separated individuals. Our analysis shows that separated men and women are significantly more likely to move than their cohabiting and married counterparts. The risk of a residential change is the highest shortly after separation and it decreases with duration since separation. The patterns are similar across countries, although the levels of spatial mobility are higher in Australia.


Migration, health and well-being. Tuesday 13 September 16:45pm 

Understanding age variations in the Migrant Mortality Advantage
Matthew Wallace 1, Michel Guillot 1, 2, Myriam Khlat 1, Irma Elo 2, Matthieu Solignac 2, 1 INED, 2 UPENN 

In the literature on migrant mortality, the most pervasive finding is that migrants tend to exhibit lower mortality than their host population, a “Migrant Mortality Advantage”. Typical explanations include selection effects, salmon bias, cultural effects, rapid health transition, and data artefacts. However, the relative contribution of these mechanisms remains unclear. One major limitation of the literature is that it largely ignores any age variations in mortality among migrants. Due to this lack of information, conclusions about the existence and scale of MMA are often made without any reference to age. This indicates that, relative to the host population, migrants exhibit a relative mortality risk that is constant over age. Age-adjusted rates may hide important fluctuations at ages where the scale of the MMA is larger or smaller than the age-adjusted rate. Likewise, explanations for MMA are often discussed with little reference to age when explanations such as selection effects may vary markedly with age. Using data from the UK, France and the US for the period 2000-2010, typical age patterns of relative migrant mortality are examined to understand the extent of age variation in the MMA. We observe extensive variation in the MMA by age and, in light of the theoretical framework, that while some explanations are consistent with observed age variations, others are not and can be dismissed as explanations of MMA. 


Ethnic mortality estimation – a method put to the test
Pia Wohland, Hull York Medical School, University of Hull 

The UK ethnic population composition is changing and diversity is increasing. By 2011, about 20% of the population in England and Wales defined themselves as not White British. Still, information on mortality for ethnic groups, an important population health indicator, is not routinely collected. This is even though numerous UK health studies found varying health outcomes for different ethnic groups. In the course of developing population projection for UK ethnic groups, we developed a method to estimate ethnic mortality by combining information on the spatial distribution of ethnic groups and overall mortality called the Geographical Distribution Method (GDM). With this method we found significant variations in mortality between groups, with the majority of ethnic minorities experiencing higher mortality compared to the White British majority. The question now is: how reliable are these GDM estimates? Even though other studies used similar methods to estimate ethnic mortality, to our knowledge no validation of results has ever been undertaken. In contrast to the UK, the US collects routinely information on ethnicity/race on the deaths certificate. From these data we can see that for example in 2011 life expectancy at birth varied considerably between racial/ethnics the gap between the Hispanic-the group with the highest life expectancy (LE) and Non-Hispanic black population-the group with the lowest LE was 6.5 years. US mortality data (deaths counts by ethnic/racial group) allow us to test and validate the GDM method. In this presentation we will discuss how US mortality data calculated with the GDM method compare to actual measured data. 


The role of deprivation transitions in explaining health inequalities in New Zealand Daniel J Exeter 1, Fran Darlington 2, Paul D Norman 2, Jinfeng Zhao 1, Grant Hanham 3, 1 School of Population Health, University of Auckland, New Zealand, 2 School of Geography, University of Leeds, UK, 3 Hanham Consulting Limited, Auckland, New Zealand    

‘Period-specific’ deprivation indices (e.g. NZDep2001, NZDep2013) are used to investigate health inequalities over time. However, this approach has limitations since the areas assigned to a given deprivation quintile in the first period may improve or worsen over time, so the deprivation quintiles in the second period may not be directly comparable. In parallel, residential mobilities of individuals over time are also ignored. Evidence suggests that movers are often socio-economically distinct from stayers but few studies have explored the extent to which transitions in deprivation circumstances associated with moving creates cumulative advantage/disadvantage on health outcomes. Using an encrypted national health identifier, we obtained a database cohort of 2.1 million adults aged ≥18 years between 2006 and 2013 linking routine national health datasets, to investigate whether the frequency and deprivation-related trajectory of moves an individual makes influences cardiovascular disease outcomes. We utilise a novel approach to measure the cumulative advantage/disadvantage exposure during deprivation transitions that occur during the residential mobility process. Logistic regression analyses are used to model the odds of patients having their first CVD event, controlling for socio-demographic factors and the deprivation transition index, compared to stayers. The policy implications of our research findings will also be discussed. 


Mobility of Young People During the Transition to Adulthood in Britain
Alina Pelikh, Hill Kulu   

This paper examines the life course trajectories of young people in England and Wales, who began their transition to adulthood in the era of neoliberalism with a special focus on moving trajectories. British pattern of the transition to adulthood is usually characterized by early transitions from school to work and heterogeneous household and family formation. The goal of the research is to look at the dynamics of cohort and gender changes in mobility among young people, controlling for both personal background characteristics (such as parental socio-economic status, region of birth) and interaction with other life domains, such as employment, education and partnership histories. For the analysis we use data from 18 waves of the British Household Panel Survey (BHPS), an annual survey consisting of a nationally representative sample of households recruited in 1991. We analyse the housing and moving trajectories by applying the techniques of multistate event history analysis, which is based on the set of competing risks models for repeated events (sequence of long- or short-distance moves). Preliminary results show that the youngest cohort postpones leaving parental home, but once they leave the parental nest, they show higher residential mobility than the two older cohorts. As for the gender gap in mobility, our results confirm an overall trend of females leaving parental home earlier than males and moving more often 2nd time, but by the higher order of moves these differences disappear. 


International migration. Wednesday 14 September 09:00pm 

The form and evolution of international migration flow networks
Guy J. Abel 1,2, Jack DeWaard 3, Zack W. Almquist 3, Jasmine Trang Ha 3, 1 Asian Demographic Research Institute, Shanghai University, Shanghai, China 2 Wittgenstein Centre (IIASA, VID/ÖAW, WU), Vienna Institute of Demography, Austrian Academy of Sciences, Vienna, Austria, 3 University of Minnesota, Minneapolis, MN   

Theoretical and empirical research on migration systems suggests convergence toward stable identifiable geographic structures comprised of networks of place-to-place migration flows over time. In this paper, we use new estimates of global bilateral international migration flows to provide an account of migration networks over the past half century. Using community detection methods, we identify international migration “communities,” or networks, over time, and subsequently examine the form and evolution of these networks with respect to their geographic concentrations and intensities. 


International migration into and out of the UK 1991-2014: trends by country of birth, citizenship and ethnicity for regions and local authorities
Philip Rees, Stephen Clark, University of Leeds 

Essential for ethnic projections for UK local authorities are assumptions about future international migration by ethnicity, informed by recent trends. Since 2014, official estimates are available for total immigration and emigration for local authorities in England. The key source for international migration is the International Passenger Survey (IPS) in which the sample of immigrants and emigrants is tiny (3 to 5 thousand). Adopting a strategy of “where angels fear to tread, fools rush in”, we have used the rich set of IPS tables published in November 2015 to create, using a microsimulation model (FMF), multi-dimensional arrays of immigrants and of emigrants, classified by country of birth, country of last or next residence, area of origin or destination within the UK (home country and English region), broad age and sex. We used the country of birth dimension of these arrays to link to ethnicity. Probabilities of ethnicity given country of birth were extracted from the 1991, 2001 and 2011 Censuses, interpolated for intervening years and applied to the arrays. The result is a rich picture of immigration and emigration by ethnicity, which we used to adjust local authority estimates. In the paper we describe the UK’s recent history of international migration both within the country and outside through the lens of ethnicity using time series indicators and graphs. Despite the large uncertainty ranges around the estimates, the trends look plausible and reveal the geographical patterns of ethnic immigration and emigration.  


A comparison of bilateral migration counts by time intervals
Guillermo Vinue, Guy Abel, Vienna Institute of Demography/Austrian Academy of Sciences   

Inconsistent data is a common and persistent problem in cross national comparisons in migration research. One source of inconsistency is the timing criteria used to define counts of migrants. Models to harmonise data based on differences in the timing criteria on the count of migrants has tended to be limited a single country (e.g. Courgeau (1973)) or synthetic data (e.g. Nowok and Willekens, (2010)). In this presentation we use questions from the Integrated Public Use Microdata Series (IPUMS) database on both the current region of residence at the number of years residing in current region from multiple countries and time periods. Using the individual data we are able to derive bilateral migrant count tables of region to region ows based on a range of timing definitions (e.g. 1 year, 2 years, ..., 5 years, ...). We propose a statistical methodology to explain the differences between counts of region to region migration over several used defined time intervals, controlling for region specific factors such as area and distance. The resulting model allows us to predict synthetic bilateral migrant ow counts over a range of timing criteria, where data is available for only a single timing criteria (such as 5-year definition). Some numerical and graphical results will be presented and some methodological issues will be discussed. We will also be able to establish relations between the changes in residences of the migrants. R statistics code will be freely available. 


Examining the link between climate, conflict and cross-border migration using gravity model
Raya Muttarak 1, Guy Abel 1,2, Michael Brottrager 3, Jesus Crespo Cuaresma 3, 1 Vienna Institute of Demography, Austrian Academy of Sceinces, Vienna, Austria, 2 Asian Demographic Research Institute, Shanghai University, Shanghai, China, 3 Vienna University of Economics and Business (WU), Vienna, Austria    

Establishing scientifically based links between climate change, conflict and migration is a complex task. This requires a panel data approach accounting for potential confounding factors in order to draw causal inferences on the relationship between climate and conflict. Likewise, the study of the relationship between climate change and migration needs to control for a range of other drivers of migration including social, political, economic and demographic factors. Given the complexity in establishing connections between climate change, conflict and migration, extant empirical studies on the subject are generally organized along the twin axes of climate change and conflict, and climate change and migration. To our knowledge, empirical literature that simultaneously investigates climate change, conflict and migration is scarce. To this end, this paper aims to systematically examine the connection between climate change and conflict and explore how the two factors interplay in influencing cross-border migration. Using gravity-type model and country-fixed effects, we attempt to draw a causal link between climate, conflict and migration. Cross-border migration is estimated based on the UNHCR global bilateral international refugee flows collected annually from 1951-2014. Climatic conditions are measured as rainfall variability, temperature anomalies and natural disaster events at a country-level. We control for demographic, social, economic and political characteristics of countries of origin and destination that drive conflict as well as serve as “push” and “pull” factors in determining migration. The preliminary results suggest higher outmigration flows from countries with greater number of violent conflicts and fatalities induced by the conflicts. The frequencies of natural disasters in the country of origin did not increase refugee flows but this could be because climate-induced migration often occurs within the country rather than cross-border. 


Effects of international migration on the age structure of young persons: local authorities in England 2001 to 2011
Paul Norman, School of Geography, University of Leeds   

This paper is mainly about the differences in fertility for UK and non-UK mothers since the number of young children in the population is largely due to births to mothers present in the population. First, this paper will demonstrate how the proportion of births in each of England’s local authorities by UK / non-UK mothers has changed during the 2001 to 2011 decade. Next, using new estimates of nativity as denominators, fertility rates by UK / non-UK mothers will be compared to demonstrate the impact of mothers who are immigrants. Finally, the change in the population aged 0-9 will be tracked for the decade to account for not only births but also for entries and exits by young persons (since, for example, children can arrive in the country having been born overseas). In terms of service demand, knowledge of these changes is essential for health services (e.g. midwifery re the impact of fertility rates) and education (e.g. if there is net migration gain at school age).