Inheritance and International Migration: The Influence of Tenure and Timing Effects in Europe

Transcription

1 Inheritance and International Migration: The Influence of Tenure and Timing Effects in Europe Albert Sabater Centre for Housing Research, School of Geography & Geosciences University of St Andrews Paper to be presented at the 25th European Network for Housing Research Conference Workgroup: 1st Choice WS-16 Minority Ethnic Groups and Housing Universitat Rovira i Virgili June 19-22, 2013 Abstract It is widely acknowledged that intergenerational transfers are a major source of social bonds. In Europe the setting up of welfare systems, including pension systems, and the growth of homeownership has increased the number of people who inherit assets. Whilst abundant literature has dealt with the description and explanation of intergenerational transfers, little is known about the patterns of downward financial flows to younger generations via inheritance among the ageing migrant population in Europe. Aim: Of importance here is to assess the differences in the chances of leaving an inheritance between older native-born and foreign-born populations in Europe. This is of particular importance given the diverse character of the ageing process and the homeownership gap between natives and immigrants in Europe. This paper makes three contributions. First, it provides evidence that immigrant groups are generally less likely to leave inheritance than their native counterparts, although differences are also found among immigrant groups and countries of residence. Second, it highlights the importance of homeownership (outright and with mortgage) which is, by far, the most powerful predictor for leaving an inheritance for both natives and foreign-born populations. Finally, it indicates that age at migration and length of residence (timing effects) are also two robust predictors for leaving an inheritance, thus signalling the consistency of theoretical predictions of the assimilation-integration model. The findings are policy relevant and suggest the need to tackle intergenerational inequality of the increasingly diverse ageing population in Europe, so as to avoid unequal socioeconomic positions across generations. Methods & Data: Various binary logistic regressions are carried out to predict the chance of leaving an inheritance for individuals aged 50 and over using data from waves 1 (2004/5) and 2 (2006/7) of the Survey of Health and Ageing and Retirement in Europe (SHARE). 1

2 Introduction It is widely acknowledged that intergenerational transfers are a major source of social bonds. In Europe the setting up of welfare systems and the widespread existence of pensions systems have been crucial in making older people as providers of financial support to their offspring. Gains in life expectancy combined with a decrease in the number of children per family have also contributed greatly to the number of people who inherit assets (Attias-Donfut and Wolff, 2000). Other factors such as the growth of homeownership have led to more bequests, thus affecting the distribution of wealth between different population subgroups depending on their housing wealth (Rowlingson and McKay, 2005). The existence of disparities in advantage and disadvantage that produce patterns of inequality within and across societies is likely to put various population subgroups in different positions in terms of the distribution of wealth and the way inheritance moves down generations. These patterns are seen as crucial as they affect the well-being of individuals and families as well as issues of social policy, social stratification and social integration (Szydlik, 2004; Avery and Rendall, 2002). Therefore, the receipt of inheritance by future generations, particularly one of any value, is expected to be heavily skewed by socioeconomic and ethnic characteristics which, in turn, is likely to lead to a further increase of the concentration of wealth among some groups more than others. While a significant number of studies have analysed the effect of inheritance on wealth inequality (see, for example, Forrest et al, 1990; Gokhale et al, 2001; Wolff, 2002), much less is known about the comparative evidence concerning differences on inheritance transfers among immigrants and native-born populations across countries. Research to date has generally focused either on particular country experiences (see, for example, Izuhara, 2002 for a comparison of the Japanese and English perspectives) or particular groups (see, for example, Nesbitt and Neary, 2001 for a comparison of the Pakistani, Bangladeshi and White groups in England). In the USA, some scholars such as Thomas Shapiro (2004: 5) have also argued that differences between minority groups such as the African American and the White group might be subject to the effect of the baby boomer generation which is in the midst of benefiting from the greatest inheritance of wealth in history. It is assumed that the majority of housing wealth is generally left to the next generation (Munro, 1988; Hamnett et al, 1991). However, attitudes to leaving an inheritance are currently shaped by two extreme views (Rowlingson and McKay, 2005; Hancock et al, 2002). One that fully supports the principles of intergenerational solidarity and inheritance so that deliberate saving and/or accummulation of housing assets becomes the norm in order to leave as much as possible to future generations (i.e. living poor to die rich); and at the other extreme, there is also the idea that some people will deliberately Spend thier Kids Inheritance (SKI) by not saving anything and by using housing equity so that nothing is left to their children. The latter group, the SKIers, has become a social policy concern given the high levels of borrowing and the growing low levels of pension provision. The balance of evidence from previous studies which incorporate an immigrant/ethnic dimension suggest that people from minority groups clearly worry about leaving an inheritance, even if this means that they will not enjoy their retirement, thus reflecting variations in attitudes to family ties and intergenerational contracts (Rowlingson and McKay, 2005). These views are, however, expected to be clearly influenced by welfare policy (Paxton, 2003) 2

3 as well as macro-level or contextual factors (Albertini et al, 2007; Kaufmann et al, 2002). Of course, attitudes to assets and inheritance itself are not the only important variables here. Demographic ageing and international migration have added a further dimension to the study of inheritance due to the significant changes in the age and ethnic composition of national populations in Europe. Due to the widespread migratory movements of peoples seen into, and across, Europe over the past fifty years, the population of immigrants aged 50 and over has increased from 12.2 million to 15.3 million between 2002 and 2011 (Eurostat, 2012). In 2011, older migrants aged 50 and over accounted for 31.4% of all migrants in the European Union (EU), a value slightly lower than their native counterparts (37.6%). The demographic situation of ageing migrants is clearly subject to the effects of the periodization of migration and the age structure of migrants (see Figure 1). Hence the bulk of young adults, for instance, who arrived during the large-scale immigration period (i.e. between the 1950s and the early 1970s) are already reaching older ages (e.g. migrants aged 20 in 1955 turned 65 in 2000, whereas migrants aged 20 in 1965 turned 65 in 2010). This tendency of population growth of older migrants is expected to continue due to the large younger generation beginning to age into the group and the significant effect of family-joining migrants. The issue of growing old in a second homeland is a topic that is too often overlooked, even among migration scholars. This is partly as a result of the complexity (and diversity) of the phenomenon as well as the lack of representative data sources on many aspects of ageing, including that of international migrants. In Europe, the growth of older foreign-born populations generally consists of immigrants who migrated either from south to north or from east to west within Europe as a result of economic or political reasons (Fassmann and Münz, 1992), including those who came from areas that experienced severe agriculture decline in Southern Europe (Italy, Spain, Portugal) or political instability in Eastern Europe (Czechoslovakia, Hungary, Poland). However, foreign-born populations are also the result of colonial legacies (for instance, Algeria and Morocco, in the case of France), limited opportunities in their country of origin and asylum-seekers responding to political and economic crisis. Europe is a multi-ethnic community and it is assumed that there may be cultural differences in practices and attitudes related to assets and inheritance. These will include groups who might have a very strong notion of the intergenerational contract as well as others who might be more prone to become SKIers, or somewhere in the middle. Within this context, the role of homeownership not only represents an expression of long-term economic progress, wealth accumulation and financial wellbeing, it is also seen as something of value that can either be used via liquidation during retirement or via bequest of housing assessts to inheritors. Yet, as noted by the literature on immigrant housing, a significant gap remains between native-born and foreign-born populations. Although homeownership rates vary significantly by country, immigrant populations have lower homeownership rates than thier native counterparts (Constant et al, 2009). This has implications for the development of immigrants economic and social incorporation in the host society as homeownership attainment is likely to be accompanied by access to neighbourhoods with desirable characteristics (Alba and Logan, 1992). In general, one can assume that although some groups might be more prone to homeownership as well as more supportive of 3

4 the idea of leaving an inheritance, their lack of assets, including housing wealth, makes intergenerational transfers to the next generation very difficult for them. Indeed, this is likely to put some foreign-born populations at a significant disadvantage in terms of the distribution of wealth, to the extent that some scholars (Szydlik, 2004) argue that support within society for intergenerational solidarity via inheritance is at odds with support for greater socioeconomic equality. When studying immigrant populations and socioeconomic integration, an important dimension to consider is the timing of migration, which refers to both age at migration and years since migration. Although the latter is usually seen as the main crucial factor for social integration, there exist two common key detereminant factors in different types of assimilation/integration or exclusion: the family s social capital and origins, and acquisition during childhood or adolescense of human capital (Glick and White, 2003; Portes and Rumbaut, 2001; Zhou, 1997). This paper builds on the previous work in the areas of inheritance and international migration by employing secondary data from the Survey of Health, Ageing and Retirement in Europe (SHARE). Despite there is very little quantitative data to look at the complex relationship between inheritance and migration, SHARE s multidisciplinary and cross-national database delivers a valid picture of the ageing process, including information on wealth transfer and data on nativity (i.e. place of birth) of the population. Since countries have historically pursued different types of integration policies, a cross-national comparative design provides a good framework for investigation, which is possible using SHARE data too. Inheritance patterns among older native-born and foreign-born populations in eight European countries are examined in this paper (Austria, Belgium, Denmark, France, Germany, the Netherlands, Sweden and Switzerland), which became the most important European receiving countries after the Second World War and are usually known as old immigration countries (Fassmann and Reeger, 2008). Using SHARE data for these countries, this paper examines, first, differences in the chances of leaving an inheritance between older native-born and foreign-born populations in Europe; and second, the role of tenure statuses and timing effects (age at migration and length of residence) in relation to leaving an inheritance. The paper addresses three specific questions: 1) Are there differences/similarities between the older native-born and foreignborn population in the chances of leaving an inheritance; how does this vary by origin and country of residence? 2) To what extent is homeownership an advantage for the chances of leaving an inheritance for both ageing native-born and foreign-born populations in Europe? 3) Have timing effects resulted in increased chances of leaving an inheritance for older immigrants? 4

5 Data and method As mentioned above, this paper uses data from the Survey of Health, Ageing and Retirement (SHARE). Based on probability samples in all participating countries, SHARE data is unique in so far as it provides a consistent set of questions on inheritance, thus allowing cross-country comparisons on wealth transmission in Europe for both native-born and foreign-born populations (Börsch-Supan and Jürges, 2005). The dataset contains samples of non-institutionalized people aged 50 and older in 12 (wave 1) and 15 (wave 2) European countries. Excluded from this analysis were Israel and three Mediterranean countries -Spain, Greece and Italy- (wave 1) and Czech Republic, Poland and Ireland (wave 2). Since the number of available observations limits the statistical power, the entire wave 1 of the survey (year 2004/5) and the refresher sample of wave 2 (year 2006/7) are used together so as to have a pooled sample while accounting for the inherent correlation structure within the data. Given that waves 1 and 2 are pooled, same respondents are only observed once, and standard errors are computed for clustering at the level of the individual. Therefore, the analysis is based on a pooled sample of respondents for which information on all required variables is reported. The final subsample consists of 20,597 individuals, of whom 13,306 were interviewed in both waves of SHARE, 6,046 were interviewed in wave 1 only and 1,245 were interviewed only in wave 2. Leaving an inheritance was identified as a positive numerical response to the question, Including property and other valuables, what are the chances that you or your husband/wife/partner will leave an inheritance totalling 50,000 euros (in local currency) or more? (0, 100). Individuals were then classified into two groups: those who had a zero probability of leaving an inheritance, and those who had a positive probability of leaving an inheritance. The foreign-born or immigrant population was identified using the question regarding country of birth. Survey respondents indicated whether they had been born in the country of residence where the interview was taking place. Their responses were used to divide the residents of each country into native-born and foreign-born. Individuals born outside the country in which they were interviewed were also asked in what year they had come to the country as well as the country of birth. Age at migration was derived from the information on year of birth and year they had come to live to the country of current residence. Due to the diversity of origins of the foreign-born populations, aggregation into consistent geographical regions from the United Nations Statistics Division (2011) was used as follows: Western, Northern, Eastern and Southern Europe, Asia, Africa and Americas. The role of immigration was analysed in several ways: An interaction variable which considered citizenship status of foreign-born in the country of residence was included. Also, in order to estimate simultaneously both the effects of origin and destination countries, interaction variables between immigrants countries of origin and residence were created. Similarly, other interaction variables were computed such as nativity and tenure status, and including in some cases country of residence, age at migration and length of residence. The main explicative variable is the immigration status of the person, and a set of demographic and socioeconomic variables which allowed testing the robustness of estimations. 5

6 Methodologically, this paper used logistic regression models for predicting a binary outcome (0 or 1): the chance of leaving an inheritance. The dependent variable was treated as a categorical (zero probability versus positive probability), thus allowing the adoption of the maximum likelihood estimation method. Binary responses were modelled so as to explore the effect of nativity and other immigrant-related variables, as opposed to the native-born population. The analysis using binary logit regressions includes various stages, which were added sequentially to capture the effects of demographics (Model 1, with age, sex, marital status and having children), education (Model 2, with highest educational attainment), health status (Model 3, with selfperceived health using good or very good and less than good), tenure status (Model 4, with tenant, outright owner, owner with mortgage and rent free), and the chance of receiving an inheritance (Model 5). In order to capture the complexity of the foreign-born populations, various interaction variables were derived, thus allowing the computation of odds ratios for leaving an inheritance by (1) nativity and citizenship, (2) origin and country of residence, (3) nativity and tenure status, (4) nativity, tenure status and country of residence, (5) nativity, tenure status and age at migration, and, (6) nativity, tenure status and length of residence, while controlling rival variables (i.e. holding other variables constant). Country-fixed effects in the analysis of leaving an inheritance were used with all eight European countries, thus mean variations in the explanatory variables are relative to one country reference (France). The country-fixed effects approach was not implemented when country of residence-specific interaction variables were used. Results Descriptive analysis Limiting eligible observations from the pooled sample of waves 1 and 2 from SHARE as outlined above in the Data section led to an analysis comprising 20,597 individuals aged 50 and over (18,408 native-born and 2,189 foreign-born). Table 1 provides the first set of descriptive statistics for native-born and foreign-born populations in the eight European countries under consideration. The results indicate that the proportion of respondents that stated leaving an inheritance is higher among the native-born (72.4%) compared to the foreign-born population (59.1%). In terms of age, the proportions between these two groups are very similar, with a concentration of more than half of these two groups within the youngest old group, aged less than 65 years old (55.7% and 57.5% for natives and foreign-born respectively). Similarly, the proportion of males and females among the native-born and foreign-born does not differ very much, with females slightly overrepresented (53.7% and 54.2% for natives and foreign-born respectively). The variable on marital status illustrates that the majority of respondents are married or in a partnership (73.3% and 71.3% for natives and foreign-born respectively). The proportion of respondents with children is clearly dominant (87.1% and 87.4% for natives and foreign-born respectively). The variable on human capital reflects that a majority have attained at least a secondary level of education (76% and 72.8% for natives and foreign-born respectively). Information on self-reported health denotes important differences with regard to experiencing poor health/less than good between natives (65.3%) and foreign-born populations (73.2%). The variable on tenure status depicts important differences, with more natives living as outright owners (38.2%) and more foreign-born living as tenants (45.6%). The 6

7 proportion of respondents who expected receiving inheritance is smaller than the group of non-receivers, with significant differences among the latter group for natives and the foreign-born (68.3% and 74.9% respectively). The origins of the foreign-born population with the largest percentages are Western Europeans (29.9%), Southern Europeans (17.4%) and Eastern Europeans (16.7%). The timing variables indicate that the foreign-born population predominantly migrated when they were aged between 16 and 40 years old (57.2%), and the great majority have resided in the host country more than 30 years (73.4%). Table 2 offers detailed information of the characteristics of immigrants by country of residence. The proportion of immigrants is greatest in Germany (17.7%), Switzerland (17.3%), and France (16.8%), although the countries with the largest proportions of immigrants with citizenship are Germany (86.6%), The Netherlands (82.2%) and Austria (74.7%). The information on mean age at arrival reflects the expected young age profile which ranges from 21 years of immigrants in Belgium to 25.7 years of immigrants in Austria and Sweden. Finally, the mean year of immigration illustrates the significance of international migration during the economic expansion of the 1960s in the eight European countries under consideration. Table 3 offers information on the share of respondents leaving inheritance and vice versa by country of residence and nativity. The results indicate that the greatest proportions of native-born leaving inheritance are found in Belgium (82.2%), Denmark (77.4%) and Switzerland (77.1%). The largest proportions of foreign-born populations leaving inheritance are also found in the same countries: Belgium (73%), Switzerland (66.7%) and Denmark (65.7%). Multivariate analyses Figure 2 provides odds ratio (OR) coefficients for leaving inheritance by nativity and citizenship in the eight European countries. Since the OR are all below 1, this indicates that the foreign-born population (with and without citizenship) is less likely to leave inheritance than their native counterparts. The outcome is significant after sequential controls for demographics (Model I), education (Model II), health status (Model III), tenure status (Model IV), and the chance of receiving an inheritance (Model V). The results also indicate that capturing each one of these effects is important and significant as shown by the positive gradient between Model I and Model V. Table 4 illustrates the estimated OR (Model I and V) for leaving an inheritance by nativity and other covariates in eight European countries. The probability of leaving an inheritance was lower and significant compared to that of natives using the two model specifications. The inclusion of socioeconomic and health effects did not change the results significantly, although it reduced the difference between immigrants and natives. The full model clearly indicates that noteworthy gains are made after including tenure status, which is by far the most powerful predictor in the model. The results also show that the chance of receiving an inheritance is a robust predictor for leaving an inheritance, thus signalling the importance of intergenerational solidarity. Finally, it can be seen how the probability of leaving an inheritance is higher in some countries (Sweden, Switzerland and the Netherlands). 7

8 Coincidently, these countries have all abolished inheritance tax compared to the country of reference France, which still maintains it (droits de succession). Table 5 shows the logistic regression results of the effect of being foreign-born on leaving inheritance, examined through the interaction variable country of origin x country of residence. Results are provided using the basic model with demographics (Model 1) and one which includes all the explanatory variables (Model 5). The results illustrate two distinctive patterns. First, immigrant populations with a non-european origin generally appear to have the lowest odds compared to natives (see, for instance, Asians in Austria, Asians and Africans in Belgium, Africans in France, Asians in Sweden, and Asians and Americans in The Netherlands). Second, immigrant populations from Western, Northern and, in some cases, also from Eastern Europe, have higher odds of leaving an inheritance than natives (see, for example, Western and Eastern Europeans in Austria, Northern Europeans in Denmark, or Western Europeans in France). This contrasts with the overall experience of Southern European migrants in these old immigration countries, thus suggesting the existence of a North-South European gradient. Table 6 show the OR for leaving inheritance by nativity and tenure status in the eight European countries. The table is divided into two parts. The first part captures differences using native-born tenants as the reference category, and the second part displays differences between native-born and foreign-born for each tenures status. The results from the first part of the table clearly highlight the importance of homeownership for leaving inheritance, which is the most robust explanatory variable of all independent variables considered. With the exception of foreign-born tenants who have lower odds compared to their native counterparts, all other respondents (natives and foreign-born) have higher odds of leaving inheritance, particularly outright owners and owners with mortgage (between to times more likely). The results from the lower part of the table indicate that the foreign-born population is systematically less likely to leave inheritance than their native counterparts regardless of their tenure status. Figure 3 displays OR for leaving inheritance by age at migration in the eight European countries. Although the OR are all below 1, thus indicating that the foreignborn population is less likely to leave inheritance than their native counterparts, there is a visible negative gradient between those who migrated during their childhood or adolescense and those who migrated at older ages. This suggests that the timing of migration can be used as a robust predictor for leaving an inheritane among immigrant groups. The results from Table 7 also confirm this tendency, with the addition of the interaction variable nativity x tenure status x age at migration. Hence, while foreign-born tenants who arrived during their childhood or adolescence still show lower odds than their native counterparts, foreign-born outright owners constitute a group which appears to be significantly more likely to leave an inheritance compared to natives in the same tenure position, using the final model with all the explanatory variables (Model 5). The OR go in the same direction for the foreign-born owners with mortgage although the coefficients are not statistically significant, most likely to due to power limitations. Finally, Figure 4 shows the importance of another timing effect, namely the length of residence among immigrant groups. Although the OR are all below 1, a positive 8

9 gradient can be detected between those immigrants whose residence in the host country has lasted less than 15 years and those whose length of residence is greater than 30 years. The results from Table 8 reflect these differences by incorporating the interaction variable nativity x tenure status x length of residence. Despite there being clear empirical evidence of higher OR for the foreign-born with the longer length of residence, immigrant groups in any of the tenure types still appear to be less likely to leave an inheritance compared to the equivalent tenure group of native-born using the basic model with demographics (Model 1) and the final one with all the explanatory variables, including receiving inheritance (Model 5). Some conclusions This paper contributes to the literature on inheritance and international migration in Europe by using SHARE data (to the best of my knowledge, for the first time). The findings clearly suggest that foreign-born populations are less likely to leave an inheritance compared to their native counterparts. In each country studied, the majority of immigrant groups reported a lower probability of leaving an inheritance. However, being born in a particular country or region and living in a specific country or region also determined an increased chance of leaving an inheritance. In particular, respondents born in Western and Northern Europe, predominantly, showed the highest OR for leaving an inheritance with respect to natives in the great majority of the countries examined in this study. Immigration-related variables such as age at migration and length of residence clearly contributed in explaining disparities on wealth transfer between immigrants and natives. Unsurprisingly, there is a strong relationship between respondents who own their property (outright or with mortgage) and the chance of leaving an inheritance. For instance, those who owned their own property were up to 12 times more likely to leave an inheritance compared to tenants, thus demonstrating the strong link between tenure of the respondent and inheritance. Comparisons using specific interaction variables with tenure-status and age at migration, or length of residence, have also demonstrated that the foreign-born population is systematically worse off and report lower chances of leaving an inheritance compared to natives. The results depicted in this study are seen as relevant in the wider scheme of things, particularly regarding intergenerational transfers and the role of inheritance in structuring wealth inequality. While the issue of inheritance is normally linked to the longstanding debate on intergeneration transmission of inequality (Menchik, 1979) and wealth mobiliy (Szydlik, 2004), the relationship between inheritance and immigration is too often overlooked. The findings from this paper around the issue of leaving inheritance between natives and foreign-born populations can be seen as worrisome, as they clearly illustrate that there is a dividing line by nativity/ethnicity which is likely to add to the already persistence of ethnic inequalities. Paradoxically, studies on the attitudes to leaving bequests among ethnic minorities and immigrants tend to show strong support for the concept of family solidarity (see, for instance, Nesbitt and Neary, 2001), thus suggesting that there is an important gap between the aspiration and the reality of inheritance among ethnic and immigrant groups. 9

10 References ALBA, R.D. and LOGAN, J.R. (1992) Assimilation and stratification in the homeownership patterns of racial and ethnic groups, International Migration Review, 26(4): ALBERTINI, M.; KOHLI, M. and VOGEL, C. (2007) Intergenerational transfers of time and money in European families: common patterns different regions?, Journal of European Social Policy, 17(4): ATTIAS-DONFUT, C. and WOLFF, F.C. (2000) The redistributive effects of generational transfers, in ARBER, S. and ATTIAS-DONFUT C. (Eds), The myth of generational conflict. The family and state in ageing societies, London: Routledge, pp AVERY, R. and RENDALL, M. (2002) Lifetime inheritances of three generations of Whites and Blacks, American Journal of Sociology, 107(5): BÖRSCH-SUPAN, A. and JÜRGES, H. (2005) The survey of health, ageing and retirement in Europe Methodology. Mannheim: MEA. CONSTANT, A., ROBERTS, R. and ZIMMERMANN, K. (2009) Ethnic Identity and Immigrant Homeownership, Urban Studies, 46(9): EUROSTAT (2012) Population Database, Luxembourg: Directorate-General of the European Commission. FASSMANN, H. (2009) Synthesis report: Old Immigration Countries, Conference Paper presented at the IDEA Final Conference, Krakow, 3-5 June FASSMANN, H. and MÜNZ, R. (1992) Patterns and trends of international migration in Western Europe, Population and Development Review, 18(3): FORREST, R.; MURIE, A. and WILLIAMS, P. (1990) Home ownership, London: Unwin Hyman. GLICK, J. and WHITE, M. (2003) The Academic Trajectories of Immigrant Youths: Analysis Within and Across Cohorts, Demography, 40(4): GOKHALE, J.; KOTLIKOFF, L.; SEFTON, J. and WEALE, M. (2001) Simulating the transmission of wealth inequality via bequests, Journal of Public Economics, 79: HAMNETT, C., HARMER, M. and WILLIAMS, P. (1991) Safe as houses: Housing inheritance in Britain, London: Paul Chapman. HANCOCK, R., KATBAMNA, S., MARTIN, G., CLARKE, H. and STUCHBURY, R. (2002) Attitudes to inheritance: An exploratory study, York: Joseph Rowntree Foundation. IZUHARA, M. (2002) Care and inheritance: Japanese and English perspectives on the generational contract, Ageing and society, 22: KAUFMANN, F. X.; KUIJSTEN, A.; SCHULZE, H.J. and STROHMEIER, K.P. (2002) Family life and family policies in Europe: Problems and issues in comparative perspective. Oxford: Oxford University Press. MENCHIK, P. (1979) Intergeneration transmission of inequality: an empirical study of wealth mobility, Economica: 46, MUNRO, M. (1988) Housing wealth and inheritance Journal of Social Policy, 17(4): NESBITT, S. and NEARY, D. (2001) Ethnic minorities and their pensions decisions: A qualitative study of Pakistani, Bangladeshi and White men in Oldham, York: Joseph Rowntree Foundation. 10

11 PAXTON, W. (Ed) (2003) Equal shares? Building a progressive and coherent assetbased welfare policy, London: Institute of Public Policy Research. PORTES, A. and RUMBAUT, R.G. (2001) Legacies. The Story of the Immigrant Second Generation. Berkeley, CA: University of California Press. ROWLINGSON, K. and McKAY, S. (2005) Attitudes to inheritance in Britain, Bristol: The Policy Press. SHAPIRO, T. (2004) The hidden cost of being African American: How wealth perpetuates inequality. New York: Oxford University Press. SZYDLIK, M. (2004) Inheritance and inequality: Theoretical reasoning and empirical evidence, European Sociological Review, 20: UNITED NATIONS STATISTICS DIVISION (2011) Composition of macro geographical (continental) regions, geographical sub-regions, and selected economic and other groupings, New York: United Nations. [ WOLFF, E. (2002) Inheritances and wealth inequality, , American Economics Association Papers and Proceedings, 92(2): ZHOU, M. (1997) Segmented Assimilation: Issues, Controversies, and Recent Research on the New Second Generation, International Migration Review, 31(4):

12 Figure 1. Age structures of native-born and foreign-born populations in the EU27, 2011 Native-born (shaded) and Foreign-born ,0% 8,0% 4,0% 0,0% 4,0% 8,0% 12,0% Source: own elaboration with EUROSTAT data. 12

13 Table 1. Descriptive statistics for native-born and foreign-born in eight European countries Variables Native-born Foreign-born Percentages Chance of leaving inheritance No 27,6 40,8 Yes 72,4 59,2 Age group Youngest old <65 55,7 57,5 Young old ,3 25,6 Middle old ,9 14,3 Oldest old >85 3,1 2,6 Gender Male 46,3 45,8 Female 53,7 54,2 Marital status Married / Partnership 73,3 71,4 Never married 5,3 5,4 Divorced 7,5 8,7 Widowed 13,8 14,5 Children No 12,9 12,6 Yes 87,1 87,4 Education Primary 15,2 19,0 Secondary 76,0 72,9 Post-secondary + 8,8 8,2 Health Good or very good 34,6 26,7 Less than good 65,4 73,3 Tenure status Tenant 27,5 45,6 Outright owner 38,2 28,1 Owner with mortgage 30,3 21,5 Rent free 4,0 4,9 Chance of receiving inheritance No 68,3 74,9 Yes 31,7 25,1 Origin Western Europe 29,9 Northern Europe 9,6 Eastern Europe 16,7 Southern Europe 17,5 As ia 9,2 Africa 12,2 Americas 4,4 Other,5 Age at migration <16 years 28, years 57,3 >40 years 14,0 Length of residence <15 years 11, years 15,5 >30 years 73,5 Source: SHARE, waves 1 (2004/5) and 2 (2006/7), 20,597 individuals aged 50 and over (18,408 native-born and 2,189 foreign-born). 13

14 Table 2. Characteristics of immigrants by country of residence Country Immigrants (%) Immigrants with citizenship (%) Mean age at arrival Mean year of immigration Aus tria Belgium Denmark France Germany Sweden Switzerland The Netherlands Source: SHARE, waves 1 (2004/5) and 2 (2006/7), 20,597 individuals aged 50 and over (18,408 native-born and 2,189 foreign-born). 14

15 Table 3. Share of respondents leaving an inheritance and vice versa by country of residence and nativity Country Total sample Leaving inheritance Not leaving inheritance Total sample Nativeborn Foreignborn Nativeborn Foreignborn Percentages Aus tria Belgium Denmark France Germany Sweden Switzerland The Netherlands Source: SHARE, waves 1 (2004/5) and 2 (2006/7), 20,597 individuals aged 50 and over (18,408 native-born and 2,189 foreign-born). 15

16 Table 4. Binary logistic regression models: odds ratio for leaving an inheritance by nativity and other covariates in eight European countries Variables Model I Leaving inheritance Model V Nativity Native-born (ref) Foreign-born 0.55 ( 0.02 ) *** 0.86 ( 0.03 ) *** Citizenship No (ref) Yes 1.21 ( 0.03 ) *** 0.88 ( 0.07 ) ** Country of residence France (ref) Aus tria 0.59 ( 0.02 ) *** 0.67 ( 0.04 ) *** Belgium 0.66 ( 0.02 ) *** 0.97 ( 0.04 ) Denmark 1.04 ( 0.02 ) * 0.78 ( 0.04 ) *** Germany 0.43 ( 0.02 ) *** 0.49 ( 0.04 ) *** Sweden 1.08 ( 0.02 ) *** 1.10 ( 0.04 ) ** Switzerland 1.14 ( 0.03 ) *** 2.34 ( 0.05 ) *** The Netherlands 1.67 ( 0.02 ) *** 1.58 ( 0.04 ) *** Gender Male (ref) Female 0.94 ( 0.01 ) *** 0.93 ( 0.02 ) *** Marital status Married / Partnership (ref) Never married 0.42 ( 0.02 ) *** 0.82 ( 0.04 ) *** Divorced 0.33 ( 0.02 ) *** 0.65 ( 0.03 ) *** Widowed 0.49 ( 0.02 ) *** 0.77 ( 0.03 ) *** Children No (ref) Yes 0.97 ( 0.02 ) 1.02 ( 0.03 ) Education Primary (ref) Secondary 1.45 ( 0.03 ) *** Post-secondary ( 0.04 ) *** Health Good or very good (ref) Less than good 0.66 ( 0.02 ) *** Tenure status Tenant (ref) Outright owner ( 0.02 ) *** Owner with mortgage ( 0.03 ) *** Rent free 1.98 ( 0.04 ) *** Chance of receiving inheritance No (ref) Yes 2.46 ( 0.02 ) *** Source: SHARE, waves 1 (2004/5) and 2 (2006/7), 20,597 individuals aged 50 and over (18,408 native-born and 2,189 foreign-born). Notes: Standard errors are reported in brackets. ref: reference category. Significance levels: * p<0.1, ** p<0.05, *** p<

17 Figure 2. Binary logistic regression models: odds ratio for leaving an inheritance by nativity and citizenship in eight European countries 1,00 Model I II III IV V 0,80 0,60 0,40 0,20 0,00 All foreign-born Foreign-born citizens Source: SHARE, waves 1 (2004/5) and 2 (2006/7), 20,597 individuals aged 50 and over (18,408 native-born and 2,189 foreign-born). Notes: Native-born (reference category). Significance levels defined as P 0.05 unless otherwise specified (dotted). 17

18 Table 5. Binary logistic regression models: odds ratio for leaving an inheritance by origin and country of residence Variables Model I Leaving inheritance Model V Country of origin x residence: Native-born x Austria (ref) Western Europe x Austria 1.39 ( 0.11 ) *** 1.39 ( 0.16 ) ** Eastern Europe x Austria 0.95 ( 0.11 ) 1.60 ( 0.16 ) *** Southern Europe x Austria 0.48 ( 0.11 ) *** 0.70 ( 0.20 ) * Asia x Austria 0.06 ( 0.34 ) *** 0.47 ( 0.38 ) ** Native-born x Belgium (ref) Western Europe x Belgium 0.42 ( 0.09 ) *** 0.97 ( 0.15 ) Northern Europe x Belgium 0.79 ( 0.26 ) 1.38 ( 0.31 ) Eastern Europe x Belgium 0.38 ( 0.28 ) *** 0.51 ( 0.47 ) Southern Europe x Belgium 0.26 ( 0.14 ) *** 0.65 ( 0.27 ) Asia x Belgium 0.28 ( 0.23 ) *** 0.24 ( 0.49 ) *** Africa x Belgium 0.96 ( 0.22 ) 0.13 ( 0.36 ) *** Native-born x Denmark (ref) Northern Europe x Denmark ** 5.13 ( 0.32 ) *** Eastern Europe x Denmark *** 0.76 ( 0.34 ) *** Southern Europe x Denmark *** 0.57 ( 0.42 ) Asia x Denmark *** 0.92 ( 0.35 ) Native-born x France (ref) Western Europe x France 1.07 ( 0.09 ) 3.48 ( 0.20 ) *** Southern Europe x France 0.83 ( 0.08 ) * 0.96 ( 0.15 ) *** Asia x France 0.80 ( 0.13 ) * 2.18 ( 0.24 ) Africa x France 0.39 ( 0.06 ) *** 0.87 ( 0.10 ) * Native-born x Germany (ref) Western Europe x Germany 0.81 ( 0.05 ) *** 1.12 ( 0.09 ) Northern Europe x Germany 3.81 ( 0.30 ) *** 2.55 ( 0.46 ) ** Eastern Europe x Germany 0.39 ( 0.05 ) *** 0.49 ( 0.09 ) *** Southern Europe x Germany 0.58 ( 0.13 ) *** 0.44 ( 0.22 ) *** Asia x Germany 0.42 ( 0.14 ) *** 0.63 ( 0.24 ) ** Native-born x Sweden (ref) Northern Europe x Sweden 0.74 ( 0.09 ) *** 0.62 ( 0.14 ) *** Eastern Europe x Sweden 0.31 ( 0.11 ) *** 0.57 ( 0.19 ) *** Southern Europe x Sweden 0.29 ( 0.18 ) *** 0.69 ( 0.30 ) Asia x Sweden 0.14 ( 0.20 ) *** 0.44 ( 0.40 ) ** Americas x Sweden 0.28 ( 0.18 ) *** 0.69 ( 0.30 ) Native-born x Switzerland (ref) Western Europe x Switzerland 0.53 ( 0.09 ) *** 0.26 ( 0.19 ) *** Eastern Europe x Switzerland 0.69 ( 0.16 ) ** 0.17 ( 0.44 ) *** Southern Europe x Switzerland 0.24 ( 0.12 ) *** 0.12 ( 0.25 ) *** Asia x Switzerland 0.32 ( 0.20 ) *** 0.73 ( 0.38 ) Africa x Switzerland 0.39 ( 0.26 ) *** 0.89 ( 0.42 ) Native-born x The Netherlands (ref) Western Europe x The Netherlands 0.81 ( 0.14 ) 1.14 ( 0.26 ) Asia x The Netherlands 0.42 ( 0.10 ) *** 0.73 ( 0.19 ) ** Americas x The Netherlands 0.41 ( 0.12 ) *** 0.41 ( 0.21 ) *** Source: SHARE, waves 1 (2004/5) and 2 (2006/7), 20,597 individuals aged 50 and over (18,408 native-born and 2,189 foreign-born). Notes: Standard errors are reported in brackets. ref: reference category. Significance levels: * p<0.1, ** p<0.05, *** p<

19 Table 6. Binary logistic regression models: odds ratio for leaving an inheritance by nativity and tenure status in eight European countries Variables Model I Leaving inheritance Model V Nativity x tenure status NB Tenant (ref) FB Tenant 0.62 ( 0.03 ) *** 0.86 ( 0.05 ) *** NB Outright owner ( 0.02 ) *** ( 0.03 ) *** FB Outright owner ( 0.04 ) *** ( 0.06 ) *** NB Owner with mortgage ( 0.02 ) *** ( 0.03 ) *** FB Owner with mortgage 7.30 ( 0.05 ) *** 9.22 ( 0.07 ) *** NB Rent free 1.98 ( 0.03 ) *** 2.06 ( 0.04 ) *** FB Rent free 1.17 ( 0.08 ) ** 1.26 ( 0.11 ) ** Nativity x tenure status NB Tenant (ref) FB Tenant 0.68 ( 0.04 ) *** 0.83 ( 0.05 ) *** NB Outright owner (ref) FB Outright owner 0.76 ( 0.05 ) *** 0.95 ( 0.07 ) NB Owner with mortgage (ref) FB Owner with mortgage 0.66 ( 0.05 ) *** 0.81 ( 0.08 ) *** NB Rent free (ref) FB Rent free 0.68 ( 0.09 ) *** 0.60 ( 0.15 ) *** Source: SHARE, waves 1 (2004/5) and 2 (2006/7), 20,597 individuals aged 50 and over (18,408 native-born and 2,189 foreign-born). Notes: Standard errors are reported in brackets. ref: reference category. Significance levels: * p<0.1, ** p<0.05, *** p<

20 Figure 3. Binary logistic regression models: odds ratio for leaving an inheritance by age at migration in eight European countries 1,20 1,00 0,80 0,60 0,40 0,20 0,00 Model I II III IV V <16 years years >40 years Source: SHARE, waves 1 (2004/5) and 2 (2006/7), 20,597 individuals aged 50 and over (18,408 native-born and 2,189 foreign-born). Notes: Native-born (reference category). Significance levels defined as P 0.05 unless otherwise specified (dotted). 20

21 Table 7. Binary logistic regression models: odds ratio for leaving an inheritance by nativity, tenure status and age at migration in eight European countries Variables Model I Leaving inheritance Model V Nativity x tenure status x age at migration (All) NB Tenant (ref) FB Tenant 0.68 ( 0.04 ) *** 0.83 ( 0.05 ) *** NB Outright owner (ref) FB Outright owner 0.76 ( 0.05 ) *** 0.95 ( 0.07 ) NB Owner with mortgage (ref) FB Owner with mortgage 0.66 ( 0.05 ) *** 0.81 ( 0.08 ) *** NB Rent free (ref) FB Rent free 0.68 ( 0.09 ) *** 0.60 ( 0.15 ) *** Nativity x tenure status x age at migration (<16 years) NB Tenant (ref) FB Tenant 0.89 ( 0.06 ) ** 1.09 ( 0.08 ) NB Outright owner (ref) FB Outright owner 1.08 ( 0.07 ) 1.47 ( 0.11 ) *** NB Owner with mortgage (ref) FB Owner with mortgage 1.15 ( 0.10 ) 1.23 ( 0.14 ) NB Rent free (ref) FB Rent free 1.08 ( 0.13 ) 0.95 ( 0.18 ) Nativity x tenure status x age at migration (16-40 years) NB Tenant (ref) FB Tenant 0.64 ( 0.05 ) *** 0.82 ( 0.07 ) *** NB Outright owner (ref) FB Outright owner 0.57 ( 0.06 ) *** 0.67 ( 0.09 ) *** NB Owner with mortgage (ref) FB Owner with mortgage 0.53 ( 0.07 ) *** 0.69 ( 0.10 ) *** NB Rent free (ref) FB Rent free 0.40 ( 0.16 ) *** 0.29 ( 0.26 ) *** Nativity x tenure status x age at migration (>40 years) NB Tenant (ref) FB Tenant 0.31 ( 0.10 ) *** 0.33 ( 0.17 ) *** NB Outright owner (ref) FB Outright owner 0.58 ( 0.19 ) *** 0.66 ( 0.26 ) NB Owner with mortgage (ref) FB Owner with mortgage 0.30 ( 0.20 ) *** 0.76 ( 0.29 ) NB Rent free (ref) FB Rent free 0.49 ( 0.27 ) *** 0.22 ( 0.91 ) * Source: SHARE, waves 1 (2004/5) and 2 (2006/7), 20,597 individuals aged 50 and over (18,408 native-born and 2,189 foreign-born). Notes: Standard errors are reported in brackets. ref: reference category. Significance levels: * p<0.1, ** p<0.05, *** p<

22 Figure 4. Binary logistic regression models: odds ratio for leaving an inheritance by length of residence in eight European countries 1,20 1,00 0,80 0,60 0,40 0,20 0,00 Model I II III IV V <15 years years >30 years Source: SHARE, waves 1 (2004/5) and 2 (2006/7), 20,597 individuals aged 50 and over (18,408 native-born and 2,189 foreign-born). Notes: Native-born (reference category). Significance levels defined as P 0.05 unless otherwise specified (shaded). 22

23 Table 8. Binary logistic regression models: odds ratio for leaving an inheritance by nativity, tenure status and length of residence in eight European countries Variables Model I Leaving inheritance Model V Nativity x tenure status x age at migration (All) NB Tenant (ref) FB Tenant 0.68 ( 0.04 ) *** 0.83 ( 0.05 ) *** NB Outright owner (ref) FB Outright owner 0.76 ( 0.05 ) *** 0.95 ( 0.07 ) NB Owner with mortgage (ref) FB Owner with mortgage 0.66 ( 0.05 ) *** 0.81 ( 0.08 ) *** NB Rent free (ref) FB Rent free 0.68 ( 0.09 ) *** 0.60 ( 0.15 ) *** Nativity x tenure status x length of residence (<15 years) NB Tenant (ref) FB Tenant 0.44 ( 0.10 ) *** 0.37 ( 0.19 ) *** NB Outright owner (ref) FB Outright owner 0.16 ( 0.30 ) *** 0.15 ( 0.35 ) *** NB Owner with mortgage (ref) FB Owner with mortgage 0.27 ( 0.16 ) *** 0.56 ( 0.23 ) ** NB Rent free (ref) FB Rent free 0.34 ( 0.38 ) *** 0.32 ( 1.07 ) Nativity x tenure status x length of residence (15-30 years) NB Tenant (ref) FB Tenant 0.36 ( 0.08 ) *** 0.43 ( 0.12 ) *** NB Outright owner (ref) FB Outright owner 0.33 ( 0.17 ) *** 0.38 ( 0.28 ) *** NB Owner with mortgage (ref) FB Owner with mortgage 0.38 ( 0.12 ) *** 0.55 ( 0.18 ) *** NB Rent free (ref) FB Rent free 0.93 ( 0.45 ) 0.66 ( 0.26 ) Nativity x tenure status x length of residence (>30 years) NB Tenant (ref) FB Tenant 0.82 ( 0.04 ) *** 1.05 ( 0.06 ) NB Outright owner (ref) FB Outright owner 0.82 ( 0.05 ) *** 1.04 ( 0.07 ) NB Owner with mortgage (ref) FB Owner with mortgage 0.85 ( 0.07 ) ** 0.98 ( 0.09 ) NB Rent free (ref) FB Rent free 0.71 ( 0.10 ) *** 0.64 ( 0.15 ) *** Source: SHARE, waves 1 (2004/5) and 2 (2006/7), 20,597 individuals aged 50 and over (18,408 native-born and 2,189 foreign-born). Notes: Standard errors are reported in brackets. ref: reference category. Significance levels: * p<0.1, ** p<0.05, *** p<