Nowhere Better Than Here? The Subjective Well-Being of German Emigrants and Remigrants Marcel Erlinghagen Abstract: The paper investigates in the question if and how the subjective well- being (SWB) of German emigrants, German non-migrants, and German remigrants differ. Based on regression analyses of data from the European Social Survey (ESS) the analyses focus on life satisfaction and happiness as main indicators of SWB. It turns out that German emigrants show increased SWB compared to German non- migrants or remigrants. However, these fi ndings cannot be explained by differences in the socio-economic or socio-demographic group structure. In fact, the increased SWB of emigrants is much more an effect of psychosocial differences and differ- ences in the individual evaluation of household income. Keywords: Life-satisfaction · Well-being · Emigration · Remigration 1 Introduction Considering the not only quantitatively but also culturally signifi cant immigration to Germany in the second half of the 20th century in particular, until the recent past the phenomenon of emigration from Germany appeared to be neither an issue in public debate nor in migration research. If at all, the term “emigrant” was mainly associ- ated with the cliché of the “dropout” seeking new or apparently old freedoms, pref- erably in climatically pleasant regions of the world, beyond the confi nes of the mod- ern working society. Not until after the most recent economic crises and mainly in the face of yet high unemployment rates has the issue of emigration again become more the focus of the public, politics, and science since the mid-1990s. Additionally, interest in emigration from Germany has lately increased due to the growing under- standing of migration as an uncompleted and at least partly transnational process of mobility (Pries 1996; 2007). Unlike the analysis of immigration, as yet there are only a few studies dealing with emigration from Germany. Most of these studies examine the re-migration of Comparative Population Studies – Zeitschrift für Bevölkerungswissenschaft Vol. 36, 4 (2011): 899-926 (Date of release: 12.07.2012) © Federal Institute for Population Research 2012 URL: www.comparativepopulationstudies.de DOI: 10.4232/10.CPoS-2011-15en URN: urn:nbn:de:bib-cpos-2011-15en3 • Marcel Erlinghagen900 immigrants to their former homelands (cf. e.g. Jankowitsch et al. 2000; Constant/ Massey 2003). With regard to the emigration of Germans, there is hardly any scien- tifi cally valid data apart from anecdotal evidence and individual case descriptions in the media (cf. e.g. Preuß 2009; Heinrich 2010). Furthermore there is hardly any information on the living conditions of emigrants from Germany after their arrival in their new home – aside from television soaps such as “Good Bye Germany” or “Auf und davon” [up and away]. The number of research works dealing with the phe- nomenon of German emigration is growing only gradually. These works fi rst were limited to mainly aggregated migration data in the offi cial statistics, which were then supplemented by non-representative quantitative and qualitative studies of specifi c emigrant groups, in particular highly skilled workers (cf. Enders/Bornmann 2002; Mohr 2002; Diehl/Dixon 2005). Representative analyses based on large indi- vidual datasets remain rare and can be counted on the fi ngers of one hand. Promi- nent studies are those of Schupp et al. (2005) and Erlinghagen et al. (2009), who analysed the social structure of German emigrants on the basis of the data of the German Socio-Economic Panel (GSOEP). There is also an extensive study by Ette/ Sauer (2010), who examined the socio-economic characteristics of Germans living abroad in particular on the basis of the European Labour Force Survey.1 While Ette/Sauer (2010) focus mainly on the earning situation of German emi- grants, so far – with the exception of a small GSOEP pilot study (Schupp et al. 2008; Erlinghagen/Stegmann 2009) based on a few individual cases – no studies have examined the subjective well-being of emigrants in their new domicile. If we under- stand individuals as (bounded) rational actors who try to maximise their subjective expected utility, it would be particularly interesting to learn the extent to which the search for a “better life” assumedly associated with emigration was successful. This article takes this up by aiming to compare the subjective well-being of German emi- grants, German non-migrants, and German remigrants. The chief aspects of subjec- tive well-being in this context examine subjective general life satisfaction and hap- piness. The explanations begin in Section 2 with the theoretical background of the following analysis, an outline of the state of research and the wording of the research questions to be examined in the following. Section 3 presents the study’s data basis and explains the analysis strategy before the results of the analysis are presented in Section 4. The article closes with a summarising conclusion (Section 5). 1 There are also a few studies on the emigration willingness of migrants (cf. e.g. Haug 2001) or Germans (cf. e.g. Diehl et al. 2008). However, we must keep in mind that ensuing actual migra- tion movements can only be conditionally predicted on the basis of expressed migration inten- tions for a number of reasons. Nowhere Better Than Here? • 901 2 Theoretical background and state of research 2.1 Reasons for emigration Emigration is not a monocausal process and the reasons that people leave their homeland can be highly different and diverse. In addition to personal reasons, for example, unemployment in the native country or a better job market abroad, the economic, social and political situation in the home and host societies also play a chief role. Moreover, the literature draws attention to specifi c personality traits such as adventurousness, readiness to assume risks, or even an optimistic and confi - dent basic attitude, which can favour emigration (cf. e.g. van Dalen/Henkens 2007). Furthermore it is important to stress that emigration decisions should be observed in a family context when possible, in other words emigration can often also be a consequence of the emigration of family members and that therefore in such cases migration is only indirectly a consequence of the living conditions of the “followers” (Haug 2000 and Kalter 2000 offer an overview of the different migration theory ap- proaches). In the “classical migration model” (Braun/Recchi 2008: 162), migration is the more probable the greater the difference is in the living conditions in the home and host societies. In this respect it is understandable that migration research in highly developed industrial countries such as Germany focus in particular on the causes and consequences of immigration from far poorer regions of Europe and the world that is again increasing in the course of globalisation. Even if the different migration theory approaches are particularly explanatory in these cases, if e.g. the prosperity gap is great or the differences with regard to political liberties are particularly strik- ing, basic assumptions should also apply when explaining emigration from coun- tries with an overall relatively high standard of living (such as Germany) (cf. van Dalen/Henkens 2007). Santacreu et al. (2009) refer in this context to various person- al reasons for emigration, whereby those surveyed repeatedly indicate in addition to vocational and family-related reasons the (aspired for) improvement of quality of life. If we base our assumptions on the premise of the action theory of limited rational actors who base their decisions on expectations (not on knowledge!), who are at the same time embedded in (in the broadest sense) social contexts, and under these preconditions attempt (!) to maximize their benefi t (Lindenberg 1989; 1990, Esser 1999), then we can anticipate that German emigrants also (should) promise themselves an increased benefi t from the decision to emigrate, regardless of their personal reasons for migration. First, we must clarify what we understand as “benefi t” or how “benefi t” can be empirically measured and operationalised. Braun/Recchi (2008), for example, exam- ine the extent to which emigration is linked to social mobility. Moreover, the relevant primary economic research discusses other options for measuring benefi ts (Diener/ Suh 1997). While for a long time benefi t was frequently equated with income, in economics not least there has increasingly been a departure from such objective indicators and now indicators that illustrate people’s subjective assessment of their quality of life (Frey/Stutzer 2002) are instead favoured to measure benefi ts. Here, • Marcel Erlinghagen902 various facets of subjective well-being can be considered, whereby in practice in particular the question of subjective general life satisfaction and subjective hap- piness plays a chief role in research (cf. Haller/Hadler 2006) and therefore will be drawn upon in this study to measure subjective well-being. Regardless of the actual operationalisation of individual benefi ts, it is clear that the increased benefi t supposedly expected from emigration is dependent upon a number of socio-economic and socio-demographic attributes and correspondingly differs among different groups of the population. In other words, emigration is more worthwhile for some people than for others. Theoretically, it is at fi rst unclear what personal, family or social factors systematically enhance the emigration probability for Germans. Empirically, however, it is shown that German emigrants are recruit- ed primarily from younger, well-educated, and unattached persons. Furthermore, there are more women than men among the emigrants. Also, there is increased emigration probability both among executives and among the unemployed (Erling- hagen et al. 2009). Moreover, the great majority of German emigrants move to other highly developed industrial nations. Frequently, these are directly neighbouring states (in particular Switzerland and Austria), but also the United Kingdom and the classical emigration country of the USA are the goal of German emigrants relatively frequently (Erlinghagen/Stegmann 2009). In their extensive study, Ette/Sauer (2010) furthermore point out that for Germans in most cases emigration is merely a limited episode in their life course and that therefore there is a considerable amount of re- migration. The group of remigrants is also chiefl y made up of younger and highly qualifi ed individuals. It is not the primary aim of this essay to explain socio-structural differences in the composition of the emigration or re-migration populations (cf. in particular Chapter 2 in Ette/Sauer 2010; Erlinghagen et al. 2009; Braun/Recchi 2008). Here it is only important to ascertain that German emigrants and remigrants are a selective group. Hence, if we would wish to compare the subjective well-being of German emigrants and remigrants with that of German non-migrants, this selectivity would have to be correspondingly taken into consideration in the empirical analysis. This is im- portant in particular because the relevant research has shown that the subjective assessment of one’s own quality of life is systematically infl uenced by fundamental personal, family, and context-dependent factors. 2.2 Determinants of subjective well-being A large number of studies fi nd a correlation between age and subjective well-being – however with inconsistent fi ndings (cf. Brockmann 2010). If we at fi rst do not con- sider familial, health, and economic circumstances, there is an inverted U-shaped correlation and in particular people of middle age appear to have the highest qual- ity of life (cf. e.g. Easterlin 2006). However, if we consider infl uencing factors such as family and health there is instead a U-shaped correlation, i.e. young and older people show greater well-being than middle-aged people (cf. e.g. Blachfl ower/Os- wald 2008). Yang (2008) and Brockmann (2010) additionally refer to cohort effects. Consequently, we can assume distinct interactions between age, life course, and Nowhere Better Than Here? • 903 subjective well-being. Familial events in particular (marriage, birth of children, sepa- ration) and state of health have a distinct infl uence on subjective quality of life (Plag- nol 2010 offers a detailed overview of the literature). Gender differences with regard to subjective well-being are seen in particular in the differing signifi cance of different factors infl uencing quality of life. For example, employment status: “In the overall population, men’s happiness is signifi cantly de- pendent on employment status. Compared to those who are not employed, nearly any other employment status makes men happier, higher ranked positions do in particular. […] In contrast, West German women report no signifi cant emotional benefi t from any engagement in the labour market” (Brockmann 2010: 34). Ac- cordingly, unemployment has a distinctly negative infl uence on the well-being of men (Winkelmann/Winkelmann 1998), while this is comparatively less important for women (Lucas et al. 2004). Independent of gender, Lucas et al. (2004) also reveal that unemployment lowers the quality of life for the long term even if the individuals have found a new job. In addition, the endowment with resources (fi nancial capital, human capital, so- cial capital) is signifi cant. Not surprisingly, wealthier individuals are in fact more satisfi ed with their lives (cf. e.g. Blanchfl ower/Oswald 2004; Shields/Wheatley Price 2005). However, there is no linear correlation between income growth and subjec- tive well-being, but from a specifi c income level no further growth in quality of life is observed (cf. e.g. Frey/Stutzer 2002: 409-410). Rather there are indications that it is not the absolute income level or absolute income growth that have a positive effect on well-being, but that positive effects then result when an advantageous or better income status is achieved compared with other individuals (cf. e.g. Clark et al. 2008). The correlation between well-being and educational level appears to be unclear. Some authors fi nd a positive correlation between skills and well-being, oth- ers in turn prove no or even a signifi cantly negative correlation (for a brief overview cf. Dolan et al. 2008: 99-100). Furthermore, it has been shown that social capital in the form of involvement in social networks is also accompanied by increased well- being (cf. e.g. Bjornskov 2003). Finally, there are distinct indications of context-related infl uences through cul- tural factors and institutional circumstances. For example, there are clear national differences observed with regard to the social extent of subjective well-being (cf. e.g. Borooah 2006). Causes for these differences may lie, for one, in the actual eco- nomic, political, and social circumstances of the people. Haller/Hadler (2006) refer in particular to the fact that subjective well-being rises in societies with relatively equal income distribution, a functional democracy, and a well-structured social wel- fare state (cf. also Veenhoven 2009). In addition to such institutional infl uences, cul- tural factors must also be taken into consideration. It seems very dubious whether there are actually nations that are fundamentally happier or more satisfi ed than others (cf. Haller/Hadler 2006: 174-175). However, it seems plausible that national differences with regard to the extent of subjective well-being may well be related to a culturally different way in which emotions are dealt with or expressed: “A sub- stantive explanation for differences between nations in SWB [subjective well-being] is that cultures may differ in the norms regulating the momentary experience of • Marcel Erlinghagen904 emotions” (Diener et al. 1995). In assessing international comparisons of subjective well-being this basic problem in social research of the international comparability of survey responses must be taken into consideration accordingly (cf. for example Johnson et al. 2005). 2.3 The subjective well-being of emigrants and remigrants Up to now there have been comparatively few research works dealing with the sub- jective well-being of migrants. The great majority of these analyses examine the extent of and the infl uencing factors on the well-being of migrants either within a group of immigrants itself (cf. Neto 1995; Amit/Litwin 2010) or in a comparison with the domestic population in the host country (Verkuyten 2008; Safi 2010), whereby the latter reveal that immigrants exhibit generally lesser well-being. According to our state of knowledge, there is only one study that examined the correlation between subjective general satisfaction and emigration. Erlinghagen et al. (2009) were able to show that emigrants from Germany exhibited no greater gen- eral dissatisfaction and also no greater dissatisfaction with their household income than German non-emigrants. And, with regard to changes in general life satisfaction after emigration, there is – as far as we know – also only one fi nding, which, how- ever, is based on the non-representative data of the GSOEP pilot study mentioned above. In it, the levels of satisfaction indicated by the GSOEP participants before their emigration were compared with the results of the follow-up survey following their emigration. It showed that the general satisfaction of 20 of the 32 emigrants surveyed (hence about 62 %) increased, while it remained the same among seven persons (22 %). Only fi ve emigrants (16 %) indicated a lower rate of general life sat- isfaction (cf. Erlinghagen/Stegmann 2009: 21). Therefore, what is lacking to date is a comparison of the subjective well-being of non-migrants with emigrating or re-mi- grating fellow nationals. This paper intends to make an initial contribution to this. 3 Data and methods 3.1 Fundamental methodological problems in surveying emigrants Quantitative empirical analyses of emigrants are faced with special methodologi- cal challenges. It is less problematic that migration, as other social phenomena, is a process. In other words, not solely the singular event of crossing a border is of signifi cance for researching migration, but the interest actually refers, fi rstly, to the developments leading to the migration event and, secondly, to the events follow- ing this singular migration event in the life course of the migrants. Mapping such life course processes has always been the source of methodological problems in empirical social research. However, panel studies now conducted for many years offer the opportunity to map and study the processes of a large number of interest- ing social phenomena, surely fi rst and foremost the German Socio-Economic Panel (GSOEP) (Wagner et al. 2007). Nonetheless, the range of such panel data is restrict- Nowhere Better Than Here? • 905 ed not only chronologically due to the respective survey periods, but also spatially by the boundaries of national states. This national restrictedness can be explained in the context of the generation of such data. The objective of population surveys was and is primarily to provide representative information for one society, which can still be spatially described quite well (in spite of increasing internationalisation and globalisation) by means of national borders. For most questions in the social sciences, this national restrictedness is, for the rest, relatively unproblematic, since certain research issues are meant to be explored only in the institutional and social context of a single, specifi c society anyway. An international comparison is not fundamentally prevented by nationally restricted survey data, since harmonisation is possible, thus allowing for comparisons. Yet, the data situation is particularly dissatisfactory for migration research, since the the- oretically well-substantiable processes of the migration occurrence can commonly not be mapped continuously. Using panel data, emigrants from a home society can only be “followed” until the emigration event, since they then leave the geographi- cal observation region of the national panel. Conversely, immigrants do not enter the methodological sights of national surveys until they cross the border. Due to this assessment problem, migration research consequently is faced with the problem that it cannot map the complete life courses of migrants. Although it is basically possible for a national panel to continue to survey emigrating respond- ents in their new domicile, a pilot study based on the GSOEP data has also shown that follow-up surveys of emigrated panel members is highly problematic and as yet hold little promise of success (cf. Schupp et al. 2008). Therefore, retrospective surveys are one suitable means to nonetheless study the effects of the “emigration” event on the individual life course. This pathway was taken by the PIONEUR project, a project funded within the EU and considered innovative in a variety of respects (cf. Rother 2005). In order to analyse the living conditions of emigrants, separate samples were drawn in Germany, the United Kingdom, France, Italy, and Spain with immigrants from these very countries and compared with the populations of both the home and the host society (cf. e.g. Braun/Recchi 2008). Also, in the recent past initial studies have begun to make use of internationally harmonised cross-sectional surveys in order to identify emigrants in their host society and compare this group either with emigrants in other countries (Geis et al. 2011) or with respondents left behind in the home society (Ette/Sauer 2010). While Ette/Sauer (2010) make use of data from the European Labour Force Survey, Geis et al. (2011) work with national datasets, which they harmonise themselves for their own purposes. Given the lack of uncensored panel data, this procedure is quite adequate. Nonetheless, we must assume two fundamental methodological problems when using this strategy in particular, which lead to coverage problems of immigrants in national surveys and therefore also had to be considered in this study: Undercoverage of emigrants – There is a danger that emigrant populations are not suffi ciently covered even during sampling. The extent to which this problem of undercoverage occurs (cf. in general Biemer/Lyberg 2003: 63-64; Groves et al. 2004: 54-55; Lohr 2008) depends decisively on the procedure taken for sampling. Some sampling for surveys is based, for example, on voter registrations in which, • Marcel Erlinghagen906 of course, only those persons are listed who possess voting rights. Depending on voting law, this is probably only limited in the case of emigrants. Another sampling strategy is based on addresses found in telephone directories. Here, as well, emi- grants would probably only appear after a while and thus be undercovered (cf. van Goor/Rispens 2004). Non-response of emigrants – Non-response problems (cf. in general Schnell 1997; Biemer/Lyberg 2003: 63-64; Lynn 2008; Haunberger 2011) would occur pri- marily because internationally harmonised surveys are conducted in the national languages of the respective survey countries. In this respect, participation in the survey requires considerable knowledge of the national language (cf. Feskens et al. 2006; Deding et al. 2008). We can therefore assume that in internationally harmo- nised surveys mainly those emigrants are underrepresented who recently entered the country. In the scope of the PIONEUR project mentioned above, not only were bilingual questionnaires used for surveying emigrants, but also interviewers who speak the languages of the home and the host society (Santacreu Fernandez et al. 2006: 87). Due to undercoverage and non-response problems, foreigners or immigrants are frequently underrepresented in national population surveys (cf. Blohm/Diehl 2001; Rendall et al. 2003). We can therefore also assume that only a selective number of German emigrants are covered in foreign surveys.2 In the description of the analysis strategy applied further on (Section 3.3), we will explain how this problem can be dealt with in the analyses to be carried out. 3.2 Data basis The European Social Survey (ESS) served as the data basis for the following analy- ses. The ESS is a survey funded together by the European Commission, the Europe- an Science Foundation (ESF) and national research funding institutions. The survey has been conducted regularly every two years since 2002. At time of the analyses, four waves were available for evaluation (2002, 2004, 2006, and 2008). It is impor- tant to note that the ESS is not a longitudinal, but a repeated cross-sectional survey and that new representative samples are drawn for the individual countries in each wave. The core of the participating countries is represented in all waves, but there are also countries that have only taken part in single waves. The number of par- ticipating countries thus deviates from wave to wave. With the exception of some country-specifi c questions, the interview sheet primarily consists of a number of 2 Another fundamental problem is how long a person must have lived abroad to actually describe them as an “emigrant.” Since a chronological defi nition is problematic, emigration is commonly considered the case when the person has shifted their focus of life abroad. In the following, we assume that this is given when Germans abroad occur in the respective population samples and therefore also as respondents in the ESS and are therefore understood in the further course as “emigrants.” Moreover, it was discovered that far more than 80 % of the German ESS partici- pants surveyed abroad lived longer than fi ve years and almost 60 % even 20 years or longer abroad (cf. Table 2 below). Against the background of the sampling and the actual duration of stay, we can continue quite confi dently to speak of “emigrants.” Nowhere Better Than Here? • 907 modules that are uniform in all countries. Two of these modules form the solid ‘core’ of the survey, that is, these modules were used in all survey waves (core modules). The other modules are variable and have had different content in the survey waves conducted so far (rotating modules) (cf. Stoop et al. 2010; detailed information can also be found at www.europeansocialsurvey.com). A pooled dataset from all four available ESS waves was used to increase the number of cases of available German emigrants in the dataset, whereby understand- ably only information from the core modules surveyed in all of the waves could be used. In the following, we defi ne German emigrants as survey respondents who did not live in Germany at the time of the survey and who possess German citizenship or were born in Germany. A more restricted defi nition of German emigrants as only those who live abroad and were both born in Germany and possess German citizen- ship was not implementable since no citizenship information is available in the ESS for a large number of persons living abroad and born in Germany. We know only that they do not possess citizenship of the emigration country. This means that this also covers a certain number of persons as emigrants who were born in Germany as the second immigrant generation. However, this does not appear unduly problem- atical for our study purpose since it deals with the correlation between leaving the home country of Germany and for such a classifi cation there is no basic difference between people who were born here with and without foreign roots. Using this method, we were able to identify 1,010 German emigrants in 24 ESS countries. The largest group by far reside in Switzerland (311 emigrants), followed by Austria (126), Luxembourg (74), Greece (65), the Netherlands (57), and the United Kingdom (56). Cyprus and the Czech Republic are at the bottom end of the list, where only one German emigrant each is contained in the ESS data (Table 1). Two factors must be considered here. For one, using the data of the ESS means that only those emigrants are covered in countries that take part in the ESS and consequently important host countries such as the USA, Canada, and Australia can- not be included. The second is that there are deviations with regard to the quantita- tive signifi cance of emigration compared with the information on host countries re- corded by the Statistische Bundesamt (cf. relevant fi gures in Erlinghagen/Stegmann 2009: 12) or also by international bodies such as EUROSTAT (cf. for more informa- tion Herm 2008). It appears that emigrants to Greece and Luxembourg are overrep- resented in the ESS dataset used here, while emigrants in Turkey, France, and Spain are undercovered. This can be a result of the deviating coverage problems in each country (cf. explanation on “undercoverage” and “non-response” above) for the na- tional ESS surveys in those countries. Moreover, the data from the offi cial German statistics are not unproblematic since they can only record such emigrations ac- cording to the host countries if the migrants properly give notice of departure with their registration offi ce according to German registration law and thereby the host country is also effectively recorded. Similar haziness possibly also occurs with re- gard to the registration of the number of German immigrants by overseas statistical offi ces. In addition, the ESS data used here refers only to adult individuals, while the number of German emigrants recorded by the offi cial statistics commonly also includes children. In this respect there are a variety of methodological reasons for • Marcel Erlinghagen908 these deviations sometimes observed between the ESS data used here and the data from the offi cial statistics. Hence, when we speak of “emigrants” in the following, we are always referring to people living abroad at the time of the survey who originally came from Germany. Consequently all relevant assertions are not based in the present occurrence of emi- gration, but always on persons a great number of whom have lived for many years or even decades abroad. Under this assumption, we have broken down emigrants in the following into three groups according to their time spent abroad so far (Table 2). We differentiate between emigrants who have lived abroad for a maximum of fi ve years (150 cases), emigrants who have lived abroad for between six and twenty years (264 cases), and emigrants who left Germany over 20 years ago (596 cases). The following analyses compare this emigrant population with Germans who have remained in Germany, who were both born in Germany and possess German citi- zenship. Under these conditions, we have data for 10,122 Germans in Germany. The third group of persons, fi nally, are German remigrants who are compared both with the emigrants and with the non-migrants. We understand remigrants to be persons born in Germany with German citizenship who are living in Germany at the time of the survey, but indicate in the ESS that they have worked for at least six months Tab. 1: Number of the German emigrants identifi ed in the ESS in the respective survey countries and percentage among all identifi ed German emigrants N Percentage N Percentage Switzerland 311 30.8 Norway 21 2.1 Austria 126 12.5 Ireland 19 1.9 Luxembourg 74 7.3 Spain 15 1.5 Greece 65 6.4 Portugal 10 1.0 Netherlands 57 5.6 Russia 8 0.8 United Kingdom 56 5.5 Turkey 7 0.7 Sweden 42 4.2 Ukraine 7 0.7 Poland 35 3.5 Italy 6 0.6 Denmark 30 3.0 Estonia 5 0.5 Belgium 29 2.9 Finland 5 0.5 France 28 2.8 Hungary 3 0.3 Israel 27 2.7 Cyprus 1 0.1 Slovenia 22 2.2 Czech Republic 1 0.1 Total 1,010 100 Source: ESS 2002-2008 (own calculations) Nowhere Better Than Here? • 909 abroad in the previous 10 years. Under this condition, we could identify 202 persons as remigrants.3 3.3 Analysis strategy The analyses fi rst begin by estimating various binary logistical regression models (cf. Hosmer/Lemeshow 2000) in order to analyse the primary socio-structural dif- ferences between emigrants, remigrants, and non-migrants. This is then followed by the actual analyses of the subjective well-being of German emigrants and remi- grants. Subjective general life satisfaction and subjective happiness are primarily used as indicators in the international research of subjective well-being (cf. Haller/ Hadler 2006). Both are also measured in the four available ESS waves with the fol- lowing questions: All things considered, how satisfi ed are you with your life as a whole nowa- days? Taking all things together, how happy would you say you are? The respondents assess their general satisfaction and happiness on an eleven- fi gure scale, whereby zero means entirely dissatisfi ed or unhappy and 10 means entirely satisfi ed or happy. Multivariate regression analyses are a good strategy for examining the correlation between non-migration, emigration, or re-migration on these two indicators using further control variables. In the relevant research lit- erature, both ordered logit or probit regressions and Ordinary Least Square (OLS) regressions are used for this purpose. Corresponding method comparisons consist- ently fi nd, however “that assuming cardinality or ordinality of the answers to gen- 3 In the ESS, re-migrants can only be identifi ed to a limited extent, since the relevant question about longer stays abroad (longer than six months) during the past 10 years is only asked of those participants who were gainfully employed at least once in their lives. Tab. 2: Number and percentage of non-migrants, emigrants, and re-migrants in the dataset N Percentage in dataset Non-migrants 10,122 89.3 % Re-migrants 202 1.8 % Emigrants 1,010 8.9 % of which: Percentage of emigrants 0-5 years abroad 150 14.9 % 6-20 years abroad 264 26.1 % More than 20 years abroad 596 59.0 % Source: ESS 2002-2008 (own calculations) • Marcel Erlinghagen910 eral satisfaction questions is relatively unimportant to results” (Ferrer-i-Carbonell/ Frijters 2004: 655; cf. also Studer/Winkelmann 2011). In the following, we examined general life satisfaction and happiness using multivariate OLS regressions. The regression analyses are used to examine the extent to which emigrants and remigrants systematically differ in comparison to non-migrant Germans with regard to subjective happiness and general life satisfaction. A number of relevant factors (cf. Section 2) are used as control variables in the following regression models to take both the selectivity of emigration and re-migration and important general de- terminants of subjective well-being into account. In addition to gender, also age and squared age are used as socio-demographic variables. Socioeconomic status is recorded using the educational level (years of full-time education) and employment status. Others are household status (single vs. with partner; children vs. no children in the household), frequency of contacts with friends, and subjective health. Differ- ences in attitudes and convictions are recorded using the extent of general trustful- ness (0 = no trust, 10 = great trust in other people), satisfaction with the democracy in the (host) country (0 = completely dissatisfi ed, 10 = completely satisfi ed), and religiousness, which is measured by the frequency of prayer (regularly or frequently vs. rarely or never). Moreover, the assessment of income is used as an explanatory variable, whereby three groups were differentiated: comfortable livelihood with the income, income suffi cient for subsistence, and (very) problematical income. To ac- count for economic or political infl uences, moreover the survey time is used as a control variable. The basic composition of the analysis dataset is contained in Table 3. In order to counteract the above-described undercoverage and non-response problems, in addition to the analyses of the total dataset, also regressions are esti- mated that account only for the German emigrants living in Austria and Switzerland and compare them with the German non-migrants and remigrants. Since the ESS questionnaire is provided in the German language in both Austria and Switzerland, the non-responses due to language problems should be of little signifi cance for our study purpose. Nonetheless, undercoverage problems cannot be completely excluded even using this method due to the sampling strategy, in particular in Swit- zerland. While in Austria a mixed sampling strategy of telephone directory and the random route method should reduce undercoverage problems of immigrant house- holds (in particular in cases of relatively short stay durations), the Swiss ESS sample is drawn only from the national telephone directory (including unlisted numbers and mobile phone connections) (information on the sampling strategy can be found in ESS 2011). However, all in all a comparison of the estimation results of the total emigrant population and the emigrant population living in Austria and Switzerland is probably the best-possible strategy under these circumstances to assess possi- ble distortions of the fi ndings based on these methodological problems. Nowhere Better Than Here? • 911 4 Results 4.1 Structural differences between non-migrants, emigrants, and remigrants Table 4 documents the coeffi cients of the different binary logistical regression es- timations, which is initially intended to provide further information about the dif- Tab. 3: Descriptive statistics Non-migrants Emigrants Re-migrants Average/ Percentage n Average/ Percentage n Average/ Percentage n Subjective well-being Happiness (0-10) 7.1* 10,086 7.7* 1,008 7.2* 202 General satisfaction (0-10) 6.8* 10,103 7.5* 1,008 6.8* 202 Demographics Men 49.3 4,992 41.0 414 69.3 140 Women 50.7 5,130 59.0 596 30.7 62 Age (in years) 48.3* 10,122 49.6* 1,010 42.9* 202 Employment status & education Education duration (in years) 13.2 10,122 13.4 1,010 15.6 202 Employed 40.9 4,144 42.0 424 55.5 112 Self-employed 5.8 585 6.8 69 13.9 28 Unemployed 7.0 704 3.6 36 7.4 15 Retired. etc. 26.1 2,641 28.1 284 9.9 20 Non-employed 19.9 2,012 18.9 191 12.9 26 Meet friends Never/rarely 18.5 1,875 17.4 176 18.3 37 Regularly 44.9 4,543 39.5 399 40.6 82 Frequently 36.5 3,695 42.7 431 41.1 83 Family No partner 38.6 3,905 42.2 426 37.1 75 Has partner 61.4 6,217 57.8 584 62.9 127 No children 67.5 6,828 67.3 680 64.9 131 Has children 32.5 3,294 32.7 330 35.2 71 Subjective health Poor/satisfactory 40.2 4,068 25.6 259 22.3 45 (Very) good 59.8 6,054 74.4 751 77.7 157 Attitudes General trustfulness (0-10) 4.7* 10,122 5.3* 1,010 5.1* 202 Satisfaction with democracy (0-10) 5.0* 10,122 6.1* 1,010 5.1* 202 Pray: never/rarely 70.7 7,158 56.0 566 75.7 153 Pray: regularly/frequently 29.3 2,964 44.0 444 24.3 49 Assessment of income Comfortable 27.2 2,802 43.9 443 36.6 74 Adequate 55.8 5,651 38.3 387 47.0 95 (Very) problematical 15.6 1,583 15.0 151 14.4 29 ESS wave 2002 26.3 2,663 28.9 292 - - 2004 24.7 2,497 26.8 271 28.2 57 2006 25.2 2,552 21.5 217 38.1 77 2008 23.8 2,410 22.8 230 33.7 68 * Arithmetic mean (metric variable). Source: ESS 2002-2008 (own calculations) • Marcel Erlinghagen912 ferent structure of the non-migrants, emigrants, and remigrants. We emphasise once more that this study is not (!) an analysis of the present migration occurrence; rather it covers individuals who are presently living abroad and at some point in their lives emigrated from Germany. This study therefore refers to emigrants who (still) live abroad at the time of the survey and whose emigration in many cases took place years or even decades ago. Under these conditions, model 1 compares non-migrants (0) with emigrants (1), whereby model 1a considers all emigrants and model 1b only those Germans living in Austria and Switzerland. Model 2 compares non-migrants (0) with remigrants (1), and model 3 fi nally compares emigrants (0) with remigrants (1). This reveals that even when controlling for the other factors, a larger number of women are among the emigrants; however the group of remigrants is dominated by men. This possibly indicates different gender-specifi c migration motives: per- haps women remain abroad permanently because their emigration is more strongly family-related and they have perhaps followed their (foreign) husbands. Possibly there are so many men among the remigrants because their emigration is more employment-related and therefore to a greater extent was planned from the outset as a limited episode. For both emigrants and remigrants an inverted U-shaped correlation is recognis- able with regard to the ages, i.e. both emigration and re-migration is a phenomenon of people of middle age. Model 3 furthermore shows that there is also an inverted U-shaped age correlation when comparing the remigrants with the emigrants. This is not surprising and reveals that emigration and ensuing re-migration occur in a chronological sequence in the life course and in this respect remigrants tend to be older than emigrants. It is moreover interesting that there is no difference with regard to the skills of non-migrants and emigrants, however at the same time the remigrants on average are better educated than the emigrants and the Germans who remain at home. This fi nding is clear proof that emigration from Germany does not lead to a loss of hu- man capital in society as a whole (brain drain), but rather that we can anticipate an increase in human capital (brain gain) by highly qualifi ed returning persons enriched by their experience of living abroad (cf. also the corresponding results in Ette/Sauer 2010). Compared with the total German population, German emigrants are with signifi - cantly lesser probability not unemployed, which indicates the strong employment orientation of emigrants. At the same time, there is no notable difference with regard to employment status if we compare remigrants with the total population. However, this also means that remigrants exhibit an increased risk of unemployment in com- parison to Germans who remain abroad (model 3). In my opinion, these fi ndings are lesser an indication of problems of remigrants, but underscore in general how successful German emigrants are and that the majority of those who remain abroad have (temporarily) better job opportunities than in their homeland. With regard to their social involvement, the group of German emigrants mani- fests a confl icting picture. Compared with non-migrants their probability is higher both to meet friends more frequently and more rarely at the same time. There are Nowhere Better Than Here? • 913 Tab. 4: Coeffi cients of the binary logistic regressions for the comparison of non-migrants, emigrants, and re-migrants Model 1a Model 1b Model 2 Model 3 Demographics Women (ref.: men) 0.32 (0.08) *** 0.51 (0.11) *** -0.55 (0.17) *** -1.02 (0.20) *** Age 0.03 (0.01) ** 0.10 (0.02) *** 0.13 (0.04) *** 0.11 (0.04) *** Age2 -0.00 (0.00) * -0.00 (0.00) *** -0.00 (0.00) *** -0.00 (0.00) *** Education & employment status Education in years -0.00 (0.01) -0.00 (0.02) 0.14 (0.02) *** 0.09 (0.02) *** Employed Ref. Ref. Ref. Ref. Self-employed 0.13 (0.17) -0.20 (0.25) 0.60 (0.33) * 0.43 (0.40) Unemployed -0.46 (0.21) ** -0.53 (0.34) 0.44 (0.39) 1.12 (0.48) ** Retired, etc. 0.17 (0.15) 0.17 (0.21) 0.45 (0.41) 0.05 (0.45) Non-employed -0.26 (0.13) ** -0.49 (0.20) ** 0.08 (0.34) 0.41 (0.38) Meet friends Never/rarely 0.23 (0.10) ** 0.06 (0.16) 0.34 (0.21) -0.14 (0.26) Regularly Ref. Ref. Ref. Ref. Frequently 0.22 (0.08) *** 0.22 (0.12) * 0.20 (0.17) 0.07 (0.20) Family Has partner (ref.: no partner) -0.26 (0.08) *** -0.50 (0.12) *** 0.02 (0.18) 0.45 (0.22) ** Has children (ref.: no children) 0.10 (0.08) -0.25 (0.13) * -0.32 (0.18) * -0.50 (0.22) ** Subjective Health (Very) good (ref.: poor/satisfactory) 0.69 (0.08) *** 1.05 (0.14) *** 0.50 (0.18) *** -0.31 (0.23) Attitudes General trustfulness 0.05 (0.02) *** 0.07 (0.02) *** 0.01 (0.03) -0.03 (0.04) Satisfaction with democracy 0.16 (0.02) *** 0.29 (0.03) *** -0.03 (0.03) -0.13 (0.04) *** Prays regularly/frequently (ref.: never/rarely) 0.51 (0.07) *** 0.61 (0.11) *** -0.01 (0.17) -0.56 (0.20) *** Assessment of income Comfortable 0.64 (0.08) *** 0.91 (0.11) *** 0.26 (0.17) -0.66 (0.21) *** Adequate Ref. Ref. Ref. Ref. (Very) problematical 0.63 (0.11) *** 0.45 (0.19) ** 0.17 (0.24) -0.72 (0.28) ** Constants -5.23*** -8.83*** -8.99*** -3.89*** N 11,121 10,506 10,175 1,171 Pseudo-R2 0.084 0.154 0.111 0.207 Model 1: non-migrants (0), emigrants (1); (1a: all emigrants; 1b: only emigrants in Austria and Switzerland). Model 2: non-migrants (0), re-migrants (1). Model 3: emigrants (0), re- migrants (1). *** p < 0.01 ** p < .0.05 * p <= 0.1; standard errors in brackets. All models controlled for ESS wave (results not shown). Source: ESS 2002-2008 (own calculations) • Marcel Erlinghagen914 two possible explanations for this. It is possible that the German emigrant popula- tion is polarised, so that some of this group is well integrated and actively maintains social contacts while the other part of this group is more strongly affected by so- cial isolation. However, the more plausible assumption is that this fi nding uncovers dynamic processes, which cannot be further examined in this analysis. By nature, establishing new social networks takes a certain amount of time, meaning that a time effect is hidden behind the simultaneous isolation and integration in social networks. Correspondingly, the results are then not evidence of polarisation, but should correlate highly with the duration of residence in the new domicile. Compared with the total German population, emigrants are more often single (cf. also the fi ndings in Erlinghagen et al. 2009), while remigrants are more often child- less. These fi ndings are clear evidence that the family or household context has a major infl uence on emigration and re-migration decisions. It would be easier for singles to decide to emigrate, since they need not reach consensus with a partner, whereas having children possibly makes the decision to re-migrate more diffi cult, because remigration by their parents would mean removing the children from their familiar school surroundings. Moreover both emigrants and remigrants report sig- nifi cantly better health than the total German population, which is not surprising since moving (particularly across national borders) is a physical and psychological burden that can only be undertaken in good health. An inverted causal direction is also imaginable (although rather unlikely), i.e. experience living abroad increases health and well-being (for example, through greater self-confi dence). In addition to the socio-demographic and socio-economic structural differences outlined above, there are furthermore distinct differences observable in individual attitudes. Compared with non-migrants, emigrants have signifi cantly greater gen- eral trustfulness in other people, are more satisfi ed with the democracy in their respective country of residence, and are more religious. All of these correlations are statistically highly signifi cant and indicate the special signifi cance of psychoso- cial factors in the emigration decision. If we assume that trustfulness and religious- ness are long-term behaviours, which formed prior to the decision to emigrate, then these results are evidence that primarily confi dent and optimistic character traits enable emigration decisions. The fact that additionally the democracy in the new host society is better assessed than the situation in Germany by the total German population could possibly also be evidence of greater optimism in the emigrant population. It may possibly also be a sign that understandably the political events in a foreign land are less critically assessed by immigrants. However, it may be that the status of the democracy in at least some of the countries is simply essentially (“objectively”) better than in Germany and in this respect has nothing to do with emigration as such. It is furthermore striking that compared with the emigrants who remain abroad remigrants do not manifest lesser general trustfulness but signifi - cantly lesser religiousness. There appears to be further need for research here in future in order to better understand the correlation between religiousness, emigra- tion, and re-migration. Moreover there are striking differences between emigrants and the total German population with regard to assessment of personal income. Compared with non- Nowhere Better Than Here? • 915 migrants, emigrants not only assess their household income to a greater degree as comfortable, but also report problematic incomes more often. This indicates that emigration is not always successful per se, but can very well also lead (temporarily) to poorer living conditions than may be the case in comparison to Germany. 4.2 The subjective well-being of emigrants and remigrants The results of the regressions used to determine the determinants of subjective happiness and general life satisfaction are found in Table 5 and Table 6 respectively. Six models were estimated for each, whereby models 1 to 5 use reduced variable sets and model 6a/b (full dataset and emigrants in Austria and Switzerland) include all control variables.4 The aim of the reduced models is mainly to learn more about the reciprocity between subjective well-being, emigration status, and psychosocial factors by always only adopting single, isolated variables in the model to measure personal attitudes or the assessment of income. This article is particularly interested in the correlation between emigration/re- migration status and subjective well-being, which is why the other explanatory vari- ables take up mainly a control function in the models. Nonetheless, before interpret- ing the main results, we will take a look at the signifi cance of the control variables in a brief summary. We will look at the fi ndings concerning happiness and general life satisfaction together, since there are hardly any differences in the comparison of the two partial analyses. In summary, we can say that this calculation on the basis of the ESS data largely confi rms the chief determinants of subjective well-being (cf. Sec- tion 2.2) known from relevant research. Women appear to have greater well-being than men, there is a U-shaped correlation between age and well-being, a higher ed- ucational level makes people happier, while unemployment has a negative effect on well-being. Having many friendship contacts, good health, and living in a partner- ship make people more satisfi ed and happier, while having children does not lead to any original effect. The assessment of income as comfortable correlates with greater well-being while a problematic fi nancial situation has a negative effect on happiness and life satisfaction. Moreover, religious people tend to feel better, and both trustfulness in other people and satisfaction with the political system increase well-being. In the correlation with psychosocial factors it is furthermore interesting that women in particular show a correlation between well-being and religiousness (cf. in particular model 3). It is also striking that in each of the models 1 to 5 parents have signifi cantly decreased satisfaction or decreased happiness – with the excep- tion of model 4, in which the assessment of income is included. This indicates that the subjective well-being of parents in particular is more highly infl uenced by their fi nancial situation. The two latter fi ndings in particular of the correlation between 4 The results of the dataset reduced to emigrants in Austria and Switzerland are very similar to the results achieved on the basis of the whole analysis dataset. This indicates for one that methodological problems such as non-response or undercoverage are in this case negligible. It is also an indicati on that the ‘broad’ operati onalizati on chosen for this study of emigrants and the parti al coverage of emigrants with migrati on backgrounds born in Germany do not notably infl uence the results. • Marcel Erlinghagen916 Tab. 5: Coeffi cients of OLS regressions on subjective happiness Model 1 Model 2 Model 3 Model 4 Model 5 Model 6a Model 6b1 Emigration & re-migration Non-migrants Ref. Ref. Ref. Ref. Ref. Ref. Ref. Emigrants (<= 5 years) 0.36 (0.15) ** 0.34 (0.15) ** 0.29 (0.15) * 0.28 (0.15) * 0.19 (0.15) 0.12 (0.15) 0.16 (0.19) Emigrants (6-20 years) 0.34 (0.11) *** 0.32 (0.11) *** 0.30 (0.11) *** 0.24 (0.11) ** 0.18 (0.11) * 0.12 (0.11) 0.11 (0.15) Emigrants (> 20 years) 0.44 (0.07) *** 0.41 (0.07) *** 0.39 (0.07) *** 0.43 (0.07) *** 0.33 (0.07) *** 0.31 (0.07) *** 0.42 (0.11) *** Re-migrants -0.04 (0.13) -0.04 (0.13) -0.03 (0.12) -0.05 (0.12) -0.03 (0.12) -0.04 (0.12) -0.04 (0.12) Demographics Women (ref.: men) 0.08 (0.04) ** 0.05 (0.04) 0.08 (0.04) ** 0.09 (0.03) ** 0.11 (0.03) *** 0.10 (0.03) *** 0.10 (0.04) *** Age -0.05 (0.01) *** -0.05 (0.01) *** -0.05 (0.01) *** -0.04 (0.01) *** -0.05 (0.01) *** -0.04 (0.01) *** -0.04 (0.01) *** Age2 0.00 (0.00) *** 0.00 (0.00) *** 0.00 (0.00) *** 0.00 (0.00) *** 0.00 (0.00) *** 0.00 (0.00) *** 0.00 (0.00) *** Education & employment status Education in years 0.05 (0.01) *** 0.05 (0.01) *** 0.04 (0.01) *** 0.03 (0.01) *** 0.04 (0.01) *** 0.02 (0.01) *** 0.02 (0.01) *** Employed Ref. Ref. Ref. Ref. Ref. Ref. Ref. Self-employed -0.13 (0.08) -0.12 (0.09) -0.10 (0.09) -0.16 (0.08) ** -0.11 (0.08) -0.13 (0.08) -0.12 (0.08) Unemployed -1.17 (0.08) *** -1.14 (0.08) *** -1.09 (0.08) *** -0.65 (0.08) *** -0.99 (0.08) ** -0.53 (0.08) *** -0.56 (0.08) *** Retired. etc. -0.12 (0.08) -0.10 (0.08) -0.10 (0.07) -0.04 (0.07) -0.09 (0.07) -0.01 (0.07) 0.00 (0.07) Non-employed 0.07 (0.07) 0.06 (0.07) 0.05 (0.07) 0.09 (0.06) 0.05 (0.06) 0.05 (0.06) 0.06 (0.07) Meets friends Never/rarely -0.57 (0.05) *** -0.57 (0.05) *** -0.52 (0.05) *** -0.48*** -0.52*** -0.42 (0.04) *** -0.42 (0.05) *** Regularly Ref. Ref. Ref. Ref. Ref. Ref. Ref. Frequently 0.23 (0.04) *** 0.22 (0.04) *** 0.20 (0.04) *** 0.21 (0.04) *** 0.21 (0.04) *** 0.18 (0.04) *** 0.18 (0.04) *** Family Has partner (ref.: no partner) 0.97 (0.04) *** 0.97 (0.04) *** 0.97 (0.04) *** 0.81 (0.04) *** 0.96 (0.04) *** 0.82 (0.04) *** 0.84 (0.04) *** Has children (ref.: no children) -0.12 (0.04) *** -0.13 (0.04) *** -0.12 (0.04) *** -0.04 (0.04) -0.12 (0.04) *** -0.04 (0.04) -0.05 (0.04) Subjective Health (Very) good (ref.: poor/satisfactory) 0.89 (0.04) *** 0.89 (0.04) *** 0.83 (0.04) *** 0.76 (0.04) *** 0.81 (0.04) *** 0.67 (0.04) *** 0.69 (0.04) *** Attitudes General trustfulness - - 0.11 (0.01) *** - - 0.07 (0.01) *** 0.07 (0.01) *** Satisfaction with democracy - - - - 0.14 (0.01) *** 0.10 (0.01) *** 0.10 (0.01) *** Prays: frequently (ref.: never/rarely) - 0.20 (0.04) *** - - - 0.10 (0.04) *** 0.11 (0.04) *** Assessment of income Comfortable - - - 0.45 (0.04) *** - 0.35 (0.04) *** 0.33 (0.04) *** Adequate Ref. Ref. Ref. Ref. Ref. Ref. Ref. (Very) problematical - - - -1.05 (0.05) *** - -0.95 (0.05) *** -0.95 (0.05) *** Constants 6.50*** 6.42*** 6.21*** 6.69*** 5.86*** 6.02 *** 6.07*** N 11,296 11,296 11,296 11,296 11,296 11,296 10,725 R2 0.176 0.178 0.193 0.227 0.208 0.254 0.254 Adjusted R2 0.174 0.176 0.191 0.225 0.206 0.252 0.252 1 Model 6b: only emigrants in Austria and Switzerland. *** p < 0,01 ** p < 0,05 * p <= 0,1; standard errors in brackets. All models controlled for ESS wave (results not shown). Source: ESS 2002-2008 (own calculations) Nowhere Better Than Here? • 917 religiousness, gender, and, happiness/satisfaction and between parenthood, as- sessment of fi nancial situation and happiness/satisfaction should be more closely examined in future studies. If we now turn to the underlying question of this study, we fi rst see that without taking psychosocial factors into consideration such as attitudes or the assessment of income (model 1), regardless of their duration of residence, emigrants are hap- pier and more satisfi ed than non-migrants. If we then insert individual attitude fac- tors in the estimations (models 2 to 5), it becomes clear that there are defi nitely reciprocal effects between religiousness (model 2), general trustfulness (model 3), and above all the assessment of income (model 4) and the well-being of emigrants. However, the correlation between happiness/satisfaction and emigration always re- mains highly signifi cant. Not until we control for satisfaction with the democracy in the respective country of residence (model 5), can no more signifi cantly increased levels of happiness and life satisfaction in emigrants be found in comparison to Germans who remain in Germany. The differentiated observation of the infl uences in particular of psychosocial factors and of attitude parameters shows that emi- grants have better well-being than non-migrants, but this better well-being can be explained mainly by a better-assessed personal income and by greater satisfaction with the respective political system. All in all, we therefore see that with regard to their subjective well-being non- migrants and remigrants basically do not differ (model 6). This is true, however, only with one restriction: if emigrants have already been abroad for a very long time and left their homeland more than 20 years ago, even when controlled for all of the variables integrated in the model, they manifest a greater feeling of happiness and greater general life satisfaction in comparison with Germans who remained in Ger- many. No signifi cant correlation can be ascertained for rather short to medium-term stays abroad of up to 20 years. The explanation of this fi nding may be selective re- migrations, which extend over a more or less long period of time and where those who are especially well integrated in the host country or are bound to a long-term life abroad remain in the new domicile permanently. In such cases, these are very deliberate decisions, which actually do lead to improved quality of life. If we return again to the signifi cance of attitudes and psychosocial factors, it was revealed that satisfaction with the democracy in the respective country of residence is an important element for understanding the greater well-being of emigrants. Yet, it is the case that the emigrants covered in the ESS live in very different countries (cf. Table 2) and in this respect it is interesting to question in what host countries the German emigrants are more satisfi ed with the function of democracy. This is quite conceivable at least for some countries. For example, the German emigrants living in these countries may quite positively assess Switzerland with its direct elements of democracy or Sweden with its well-developed welfare state. But it is hard to con- ceive the same for Russia, Hungary, or the Ukraine. In order to gain a bit more clarity about this, Figure 1 shows the assessment of satisfaction with the democracy on a scale of eleven (0 to 10). The fi gures refer to German emigrants in the countries un- der each column (Switzerland, Luxembourg, Sweden, Austria, the Netherlands, the • Marcel Erlinghagen918 Tab. 6: Coeffi cients of OLS regressions on general satisfaction Model 1 Model 2 Model 3 Model 4 Model 5 Model 6a Model 6b1 Emigration & re-migration Non-migrants Ref. Ref. Ref. Ref. Ref. Ref. Ref. Emigrants (<= 5 years) 0.49 (0.18) *** 0.46 (0.18) ** 0.39 (0.18) ** 0.36 (0.17) ** 0.20 (0.17) 0.09 (0.16) 0.02 (0.21) Emigrants (6-20 years) 0.37 (0.13) *** 0.34 (0.13) *** 0.30 (0.13) ** 0.20 (0.12) * 0.12 (0.12) -0.00 (0.12) 0.03 (0.17) Emigrants (> 20 years) 0.45 (0.09) *** 0.40 (0.09) *** 0.38 (0.09) *** 0.42 (0.08) *** 0.27 (0.08) *** 0.23 (0.08) *** 0.44 (0.13) *** Re-migrants -0.12 (0.15) -0.12 (0.15) -0.12 (0.14) -0.14 (0.14) -0.11 (0.14) -0.13 (0.13) -0.13 (0.13) Demographics Women (ref.: men) 0.10 (0.04) ** 0.06 (0.04) 0.11 (0.04) *** 0.12 (0.04) *** 0.15 (0.04) *** 0.14 (0.04) *** 0.13 (0.04) *** Age -0.06 (0.01) *** -0.06 (0.01) *** -0.06 (0.01) *** -0.04 (0.01) *** -0.05 (0.01) *** -0.04 (0.01) *** -0.04 (0.01) *** Age2 0.00 (0.00) *** 0.00 (0.00) *** 0.00 (0.00) *** 0.00 (0.00) *** 0.00 (0.00) *** 0.00 (0.00) *** 0.00 (0.00) *** Education & employment status Education in years 0.05 (0.01) *** 0.05 (0.01) *** 0.04 (0.01) *** 0.02 (0.01) *** 0.04 (0.01) *** 0.01 (0.01) 0.01 (0.01) Employed Ref. Ref. Ref. Ref. Ref. Ref. Ref. Self-employed -0.08 (0.09) -0.07 (0.09) -0.04 (0.09) -0.13 (0.09) -0.05 (0.09 -0.09 (0.09) -0.08 (0.09) Unemployed -2.05 (0.09) *** -2.00 (0.09) *** -1.94 (0.09) *** -1.23 (0.09) *** -1.79 (0.09) *** -1.04 (0.09) *** -1.08 (0.09) *** Retired. etc. -0.06 (0.08) -0.04 (0.08) -0.04 (0.08) 0.06 (0.08) -0.01 (0.08) 0.11 (0.08) 0.11 (0.08) Non-employed 0.14 (0.07) * 0.13 (0.07) * 0.10 (0.07) 0.17 (0.07) ** 0.10 (0.07) 0.11 (0.07) 0.10 (0.07) Meets friends Never/rarely -0.62 (0.04) *** -0.61 (0.04) *** -0.54 (0.04) *** -0.48 (0.04) *** -0.53 (0.04) *** -0.38 (0.04) *** -0.38 (0.05) *** Regularly Ref. Ref. Ref. Ref. Ref. Ref. Ref. Frequently 0.25 (0.04) *** 0.24 (0.04) *** 0.21 (0.04) *** 0.22 (0.04) *** 0.23 (0.04) *** 0.18 (0.04) *** 0.19 (0.04) *** Family Has partner (ref.: no partner) 0.73 (0.05) *** 0.73 (0.05) *** 0.72 (0.05) *** 0.48 (0.04) *** 0.72 (0.04) *** 0.49 (0.04) ** 0.51 (0.04) *** Has children (ref.: no children) -0.15 (0.05) *** -0.16 (0.05) *** -0.14 (0.05) *** -0.01 (0.04) -0.15 (0.04) *** -0.03 (0.04) -0.02 (0.04) Subjective Health (Very) good (ref.: poor/satisfactory) 1.10 (0.04) *** 1.10 (0.04) *** 1.00 (0.04) *** 0.89 (0.04) *** 0.97 (0.04) *** 0.76 (0.04) *** 0.76 (0.04) *** Attitudes General trustfulness - - 0.16 (0.01) *** - - 0.10 (0.01) *** 0.10 (0.01) *** Satisfaction with democracy - - - - 0.22 (0.01) *** 0.15 (0.01) *** 0.15 (0.01) *** Prays: frequently (ref.: never/rarely) - 0.31 (0.04) *** - - - 0.16 (0.04) *** 0.16 (0.04) *** Assessment of income Comfortable - - - 0.77 (0.04) *** - 0.62 (0.04) *** 0.60 (0.04) *** Adequate Ref. Ref. Ref. Ref. Ref. Ref. Ref. (Very) problematical - - - -1.62 (0.06) *** - -1.47 (0.05) *** -1.48 (0.06) *** Constants 6.03*** 6.05*** 5.66*** 6.37*** 5.07*** 5.33 *** 5.39*** N 11,313 11,313 11,313 11,313 11,313 11,313 10,742 R2 0.184 0.188 0.210 0.278 0.238 0.322 0.322 Adjusted R2 0.182 0.186 0.208 0.276 0.236 0.319 0.320 1 Model 6b: only emigrants in Austria and Switzerland. *** p < 0,01 ** p < 0,05 * p <= 0,1; standard errors in brackets. All models controlled for ESS wave (results not shown). Source: ESS 2002-2008 (own calculations). Nowhere Better Than Here? • 919 United Kingdom, Greece, and Poland) and to German non-migrants, who continue to live in Germany.5 The black crossbars depict the confi dence intervals (95 %). Figure 1 shows that German emigrants in Switzerland assess the political system there most highly at 7.2 on average, followed by Luxembourg (7.0), Sweden (6.6), and Austria (6.2). As the displayed confi dence intervals show, these differences are also signifi cant. In the other countries as well (with the exception of Poland), the emigrants assess the function of the democracy on average more positively than Germans living in Germany – however these differences are not signifi cant. These are indications that future studies of the living conditions of emigrants and therefore in that context of the possible infl uencing factors on re-migration decisions should take the institutional context including the corresponding personal assessment by the emigrants into consideration. Similar questions can be raised with regard to the actual wage situation of Germans abroad, the assessment of income, and subjective 5 This restricted selection of countries is due to the corresponding numbers of cases, since only those countries are taken into account in which more than 30 emigrants could be identifi ed in the ESS dataset. Fig. 1: Average satisfaction of German non-migrants / emigrants with the democracy in the respective country 4.85.0 5.35.35.5 6.2 6.6 7.0 7.2 CH LUX SWE AUT NL GB GR D POL Source: ESS 1-4, own calculations • Marcel Erlinghagen920 well-being. However, this is beyond the scope of the questions chosen for this study and can therefore not be taken into consideration here.6 5 Summary and concluding remarks Existing studies have shown that on average, German emigrants are happier and more satisfi ed than Germans who remain in Germany or German remigrants. This greater sense of well-being is however not due to differences in the socio-demo- graphic or socio-economic structure of the various groups under study. We were able to show that the enhanced subjective well-being of emigrants is related prima- rily with psychosocial differences or better assessment of their income abroad. The better assessment of the political situation in particular, at least in some of the main emigration countries (e.g. Switzerland, Sweden) has a key role in understanding the positive emotional state of emigrants. If such factors are taken into consideration, there are no more ascertainable differences between non-migrants and emigrants with regard to happiness and life satisfaction. The exception here, however, are those emigrants who left Germany over 20 years ago and, even when considering attitude variables, manifest signifi cantly improved life satisfaction and increased happiness. This may be due to the effects of positive selection, i.e. these are per- sons who remain abroad for a very long time, feel especially comfortable there, and are well integrated in the host society. Pertaining to remigrants, there is absolutely no evidence that they differ sig- nifi cantly with regard to happiness or life satisfaction than Germans who remain in their homeland. Since no causal analyses are possible on the basis of the cross- sectional data used here, we can only speculate when interpreting this fi nding. One the one hand, the remigrants could be failed emigrants who are now happy to be back in their homeland, but this is not expressed in increased levels of satisfac- tion or happiness, because this relief only compensates the previous reduced well- being while abroad to a “standard measure.” On the other hand, these results may indicate that temporary emigration and ensuing re-migration are phases in the life courses of people that appear appropriate under the respective life circumstances, but in comparison to non-migrants are not accompanied by any general increased benefi t but also not by any general loss in benefi t. Since no causal analyses are possible based on the ESS data used here, it must ultimately remain uncertain how emigration and subjective well-being are precisely related to one another. Nonetheless, it appears plausible that enhanced well-being actually only arises once one resides abroad and is not due to the fact that generally more satisfi ed or happier people tend to emigrate. Erlinghagen et al. (2009) were 6 On principle it may well be possible to methodologically take into consideration differences with regard to institutional contextual circumstances, e.g. by including the host countries as control variables in the regression model. However, due to the in part very low number of cases in some host countries, this approach did not appear adequate and must therefore be left to future research works. Nowhere Better Than Here? • 921 able to show, for one, that there is no correlation between general satisfaction prior to emigration and the emigration decision. Moreover, the results of Erlinghagen/ Stegmann (2009) also indicate that the general satisfaction of emigrants tends to increase following their arrival in the new domicile. The particular signifi cance of psychosocial factors in the assessment of life circumstances in the new domicile, fi nally, indicates that emigrants possibly react especially positively to new impres- sions and experiences. This, in turn, would mean that emigration per se does not make them happy and satisfi ed, but that emigrants are the type of people who, due to their personality make-up, are easier to make happy or for whom the challenge of new life circumstances in particular increases their satisfaction. References Amit, Karin; Litwin, Howard 2010: The Subjective Well-Being of Immigrants Aged 50 and Older in Israel. In: Social Indicators Research 98: 89-104 [doi: 10.1007/ s11205-009-9519-5]. Biemer, Paul P.; Lyberg, Lars E. 2003: Introduction to Survey Quality. London: Wiley [doi: 10.1002/0471458740]. Bjornskov, Christian 2003: The Happy Few: Cross-Country Evidence on Social Capital and Life Satisfaction. In: Kyklos 56: 3-16 [doi: 10.1111/1467-6435.00207]. Blanchfl ower, David G.; Oswald, Andrew J. 2004: Well-Being over Time in Brit- ain and the USA. In: Journal of Public Economics 88: 1359-1386 [doi: 10.1016/ S0047-2727(02)00168-8]. Blanchfl ower, David G.; Oswald, Andrew J. 2008: Is Well-Being U-Shaped Over the Life Cycle? In: Social Science and Medicine 66: 1733-1749 [doi: 10.1016/j.socsci- med.2008.01.030]. Blohm, Michael; Diehl, Claudia 2001: Wenn Migranten Migranten befragen. Zum Teil- nahmeverhalten von Einwanderern bei Bevölkerungsbefragungen. In: Zeitschrift für Soziologie 30: 223-242. Borooah, Vani K. 2006: How Much Happiness is there in the World? A Cross-Country Study. In: Applied Economics Letters 13: 483-488 [doi: 10.1080/13504850500400652]. Braun, Michael; Recchi, Ettore 2008: Keine Grenzen, mehr Opportunitäten? Migra- tion und soziale Mobilität innerhalb der EU. In: Berger, Peter A.; Weiß, Anja (Eds.): Transnationalisierung sozialer Ungleichheit. Wiesbaden: VS Verlag: 161-183 [doi: 10.1007/978-3-531-91160-1_8]. Brockmann, Hilke 2010: Why are Middle-Aged People so Depressed? Evidence from West Germany. In: Social Indicators Research 97: 23-42 [doi: 10.1007/s11205-009-9560-4]. Clark, Andrew; Frijters, Paul; Shields, Michael 2008: Relative Income, Happiness, and Utility: An Explanation for the Easterlin Paradox and Other Puzzles. In: Journal of Eco- nomic Literature 46: 95-144 [doi: 10.1257/jel.46.1.95]. Constant, Amelie; Massey, Douglas S. 2003: Self-selection, Earnings, and Out-Migra- tion: A Longitudinal Study of Immigrants to Germany. In: Journal of Population Eco- nomics 16: 631-653 [doi: 10.1007/s00148-003-0168-8]. Deding, Mette; Fridberg, Torben; Jakobsen, Vibeke 2008: Non-response in a Survey among Immigrants in Denmark. In: Survey Research Methods 2: 107-121. • Marcel Erlinghagen922 Diehl, Claudia; Dixon, David 2005: Zieht es die Besten fort? Ausmaß und Formen der Abwanderung deutscher Hochqualifi zierter in die USA. In: Kölner Zeitschrift für Sozio- logie und Sozialpsychologie 57: 714-734 [doi: 10.1007/s11577-005-0222-5]. Diehl, Claudia; Mau, Steffen; Schupp, Jürgen 2008: Auswanderung von Deutschen: Kein dauerhafter Verlust von Hochschulabsolventen. In: DIW-Wochenbericht 75: 49-55. Diener, Ed; Suh, Eunkook; Smith, Heidi; Shao, Liang 1995: National Differences in Re- ported Subjective Well-Being: Why do they Occur? In: Social Indicators Research 34: 7-32 [doi: 10.1007/BF01078966]. Diener, Ed; Suh, Eunkook 1997: Measuring Quality of Life: Economic, Social, and Subjective Indicators. In: Social Indicators Research 40: 189-216 [doi: 10.1023/A:1006859511756]. Dolan, Paul; Peasgood, Tessa; White, Mathew 2008: Do We Really Know what Makes Us Happy? A Review of the Economic Literature on the Factors Associated with Sub- jective Well-Being. In: Journal of Economic Psychology 29: 94-122 [doi: 10.1016/j. joep.2007.09.001]. Easterlin, Richard A. 2006: Life Cycle Happiness and its Sources: Intersections of Psychology, Economics, and Demography. In: Journal of Economic Psychology 27: 463-482 [doi: 10.1016/j.joep.2006.05.002]. Enders, Jürgen; Bornmann, Lutz 2002: Internationale Mobilität von bundesdeutschen Promovierten. In: Bellmann, Lutz; Velling, Johannes (Eds.): Arbeitsmärkte für Hoch- qualifi zierte. Nürnberg: IAB: 357-374. Erlinghagen, Marcel; Stegmann, Tim 2009: Goodbye Germany – und dann? Erste Ergeb- nisse einer Pilotstudie zur Befragung von Auswanderern aus Deutschland. In: SOEP- papers on Multidisciplinary Panel Data Research No. 193, Berlin: DIW [doi: 10.2139/ ssrn.1413828]. Erlinghagen, Marcel; Stegmann, Tim; Wagner, Gert G. 2009: Deutschland ein Auswan- derungsland? In: DIW Wochenbericht 76: 663-669. ESS (European Social Survey) 2011: ESS3 – 2006 Documentation Report. Edition 3.3. URL: http://ess.nsd.uib.no/ess/round3/surveydoc.html, 09.05.2011. Esser, Hartmut 1999: Soziologie. Spezielle Grundlagen. Vol. 1: Situationslogik und Han- deln. Frankfurt a.M./New York: Campus. Ette, Andreas; Sauer, Lenore 2010: Auswanderung aus Deutschland. Daten und Analy- sen zur internationalen Migration deutscher Staatsbürger. Wiesbaden: VS Verlag. Ferrer-i-Carbonell, Ada; Frijters, Paul 2004: How Important is Methodology for the Es- timates of the Determinants of Happiness? In: Economic Journal 114: 641-659 [doi: 10.1111/j.1468-0297.2004.00235.x]. Feskens, Remco; Hox, Joop; Lensvelt-Mulders, Gerty; Schmeets, Hans 2006: Collecting Data among Ethnic Minorities in an International Perspective. In: Field Methods 18: 284-304 [doi: 10.1177/1525822X06288756]. Frey, Bruno S.; Stutzer, Alois 2002: What Can Economists Learn from Happiness Re- search? In: Journal of Economic Literature 40: 402-435. Geis, Wido; Übelmesser, Silke; Werding, Martin 2011: Why go to France or Germany, if You Could as Well Go to the UK or the US? Selective Features of Immigration to the EU ‘Big Three’ and the United States. In: Journal of Common Market Studies 49, 767-796 [doi: 10.1111/j.1468-5965.2010.02154.x]. Groves, Robert M. et al. 2004: Survey Methodology. London: Wiley. Nowhere Better Than Here? • 923 Haller, Max; Hadler, Markus 2006: How Social relations and Structures can Produce Happiness and Unhappiness: An International Comparative Analysis. In: Social Indi- cators Research 75: 169-216 [doi: 10.1007/s11205-004-6297-y]. Haug, Sonja 2000: Klassische und neuere Theorien der Migration. In: MZES Arbeitspa- pier Nr. 30/2000. Mannheim: MZES. Haug, Sonja 2001: Bleiben oder Zurückkehren? Zur Messung, Erklärung und Prognose der Rückkehr von Immigranten in Deutschland. In: Zeitschrift für Bevölkerungswissen- schaft 26: 231–270. Haunberger, Sigrid 2011: Teilnahmeverweigerung in Panelstudien. Wiesbaden: VS Ver- lag [doi: 10.1007/978-3-531-92685-8_2]. Heinrich, Christian 2010: Diagnose: Mangelnde Wertschätzung. Nordeuropa zieht deut- sche Ärzte an. Doch bald dürfte sich das Auswandern nicht mehr lohnen. In: Die Zeit 17, 22.04.2010. Herm, Anne 2008: Recent Migration Trends: Citizens of EU-27 Member States become ever more Mobile while EU Remains Attractive to Non-EU Citizens. Statistics in Focus 98/2008. Brüssel: Eurostat. Hosmer, David W.; Lemeshow, Stanley 2000: Applied Logistic Regression. 2nd ed. New York: Wiley. Jankowitsch, Beate; Klein, Thomas; Weick, Stefan 2000: Die Rückkehr ausländischer Arbeitsmigranten seit Mitte der achtziger Jahre. In: Alba, Richard; Schmidt, Peter; Wasmer, Martina (Eds.): Deutsche und Ausländer: Freunde, Fremde oder Feinde? Em- pirische Befunde und theoretische Erklärungen. Blickpunkt Gesellschaft 5. Wiesba- den: Westdeutscher Verlag: 93-109. Johnson, Timothy; Kulesa, Patrick; Cho, Young Ik; Shavitt, Sharon 2005: The Relation between Culture and Response Styles. Evidence from 19 Countries. In: Journal of Cross-Cultural Psychology 36: 264-277 [doi: 10.1177/0022022104272905]. Kalter, Frank 2000: Theorien der Migration. In: Müller, Ulrich; Nauck, Bernhard; Diek- mann, Andreas (Eds.): Handbuch der Demographie. Vol. 1: Modelle und Methoden. Berlin: Springer: 438-475. Lindenberg, Siegwart 1989: Choice and Culture: The Behavioral Basis of Cultural Impact on Transactions. In: Haferkamp, Hans (Eds.): Social Structure and Culture. Berlin/New York: De Gruyter: 175-200. Lindenberg, Siegwart 1990: Homo Socio-oeconomicus: The Emergence of a General Model of Man in Social Sciences. In: Journal of Institutional and Theoretical Econom- ics 146: 727-748. Lohr, Sharon L. 2008: Coverage and Sampling. In: Leeuw, Edith D.; Hox, Joop J.; Dill- man, Don A. (Eds.): International Handbook of Survey Methodology. New York: Psy- chological Press: 97-112. Lucas, Richard E.; Clark, Andrew E.; Georgellis, Yannis; Diener, Ed 2004: Unemploy- ment Alters the Set Point for Life Satisfaction. In: Psychological Science 15: 8-13 [doi: 10.1111/j.0963-7214.2004.01501002.x]. Lynn, Peter 2008: The Problem of Nonresponse. In: Leeuw, Edith D.; Hox, Joop J.; Dill- man, Don A. (Eds.): International Handbook of Survey Methodology. New York: Psy- chological Press: 35-55. Mohr, Henrike 2002: Räumliche Mobilität von Hochschulabsolventen. In: Bellmann, Lutz; Velling, Johannes (Eds.): Arbeitsmärkte für Hochqualifi zierte. Nürnberg: IAB: 249-277. • Marcel Erlinghagen924 Neto, Felix 1995: Predictors of Satisfaction with Life among Second Generation Mi- grants. In: Social Indicators Research 35: 93-116 [doi: 10.1007/BF01079240]. Plagnol, Anke 2010: Subjective Well-Being over the Life Course: Conceptualizations and Evaluations. In: Social Research 77: 749-768. Preuß, Roland 2009: Abwanderung von Fachkräften: Ein Arzt im Wert von einer Million Euro. In: Süddeutsche Zeitung, 27.05.2009. Pries, Ludger 1996: Transnationale Soziale Räume. Theoretisch-empirische Skizze am Beispiel der Arbeitswanderungen Mexiko-USA. In: Zeitschrift für Soziologie 25: 437-453. Pries, Ludger 2007: Die Transnationalisierung der sozialen Welt. Sozialräume jenseits von Nationalgesellschaften. Frankfurt/Main: Suhrkamp. Rendall, Michael S.; Tomassini, Cecilia; Elliot, David J. 2003: Estimation of Annual In- ternational Migration from the Labour Force Surveys of the United Kingdom and the Continental European Union. In: Statistical Journal of the United Nations ECE 20,3-4: 219-234. Rother, Nina 2005: Wer zieht innerhalb der EU wohin und warum? Das PIONEUR-Projekt. In: ZUMA-Nachrichten 29: 94-97. Safi , Mirna 2010: Immigrants Life Satisfaction in Europe: Between Assimilation and Dis- crimination. In: European Sociological Review 26: 159-176 [doi: 10.1093/esr/jcp013]. Santacreu, Oscar; Baldoni, Emiliana; Albert, Maria C. 2009: Deciding to Move. Migrating Projects in an Integrating Europe. In: Recchi, Ettore; Favell, Adrian (Eds.): Pioneers of European Integration – Citizenship and Mobility in the EU. Cheltenham: Edward Elgar: 52-71. Santacreu, Oscar; Rother, Nina; Braun, Michael 2006: Stichprobenziehung für Migran- tenpopulationen in fünf Ländern. Eine Darstellung des methodischen Vorgehens im PIONEUR-Projekt. In: ZUMA-Nachrichten 30: 72-88. Schnell, Rainer 1997: Nonresponse in Bevölkerungsumfragen: Ausmaß, Entwicklung und Ursachen. Opladen: Leske + Budrich. Schupp, Jürgen; Siegel, Nico A.; Erlinghagen, Marcel; Stegmann, Tim; Wagner, Gert G. 2008: Leben außerhalb Deutschlands. Eine Machbarkeitsstudie zur Realisierung von Auslandsbefragungen auf Basis des Sozio-oekonomischen Panels (SOEP). In: SOEP- papers on Multidisciplinary Panel Data Reseach 120. Berlin: DIW. Schupp, Jürgen; Söhn, Janina; Schmiade, Nicole 2005: Internationale Mobilität von deutschen Staatsbürgern. Chance für Arbeitslose oder Abwanderung von Leistungs- trägern? In: Zeitschrift für Bevölkerungswissenschaft 30: 279-292. Shields, Michael A.; Wheatley Price, Stephen 2005: Exploring the Economic and Social Determinants of Psychological Well-Being and Perceived Social Support in England. In: Journal of the Royal Statistical Society: Series A (Statistics in Society) 168: 513-537 [doi: 10.1111/j.1467-985X.2005.00361.x]. Stoop, Ineken; Billiet, Jaak; Koch, Achim; Fitzgerald, Rory 2010: Improving Survey Re- sponse. Lessons Learned from the European Social Survey. Chichester: Wiley. Studer, Raphael; Winkelmann, Rainer 2011: Specifi cation and Estimation of Rating Scale Models – with an Application to the Determinants of Life Satisfaction. In: De- partment of Economics Working Paper No. 3. Zürich: Universität Zürich [doi: 10.2139/ ssrn.1846465]. van Dalen, Hendrik P.; Henkens, Kène 2007: Longing for the Good Life: Understanding Emigration from High-Income Country. In: Population and Development Review 33: 37-65 [doi: 10.1111/j.1728-4457.2007.00158.x]. Nowhere Better Than Here? • 925 van Goor, Henk; Rispens, Sonja 2004: A Middle Class Image of Society. A Study of Un- dercoverage and Nonresponse Bias in Telephone Surveys. In: Quality & Quantity 38: 35-49 [doi: 10.1023/B:QUQU.0000013237.59150.ff]. Veenhoven, Ruut 2009: Well Being in Nations and Well-Being of Nations. Is there a Confl ict Between Individual and Society? In: Social Indicators Research 91: 5-21 [doi: 10.1007/s11205-008-9323-7]. Verkuyten, Maykel 2008: Life Satisfaction Among Ethnic Minorities: The Role of Dis- crimination and Group Identifi cation. In: Social Indicators Research 89: 391-404 [doi: 10.1007/s11205-008-9239-2]. Wagner, Gert G.; Frick, Joachim R.; Schupp, Jürgen 2007: The German Socio-Economic Panel Study (SOEP) – Scope, Evolution and Enhancements. In: Schmollers Jahrbuch 127: 139-169 [doi: 10.2139/ssrn.1028709]. Winkelmann, Liliana; Winkelmann, Rainer 1998: Why Are the Unemployed So Unhappy? Evidence from Panel Data. In: Economica 65: 1-15 [doi: 10.1111/1468-0335.00111]. Yang, Yang 2008: Social Inequalities in Happiness in the United States, 1972 to 2004: An Age-Period-Cohort Analysis. In: American Sociological Review 73: 204-226 [doi: 10.1177/000312240807300202]. Translated from the original text by the Federal Institute for Population Research, for information only. The reviewed and author’s authorised original article in German is available under the title “Kein schöner Land? Glück und Zufriedenheit deutscher Aus- und Rückwanderer”, DOI 10.4232/10. CPoS-2011-15de or URN urn:nbn:de:bib-cpos-2011-15de9, at http://www.comparativepopulation- studies.de. Date of submission: 07.11.2011 Date of Acceptance: 08.03.2012 Prof. Dr. Marcel Erlinghagen ( ). Institut für Soziologie, Fakultät für Gesellschafts- wissenschaften, Universität Duisburg-Essen, Duisburg, Germany. E-Mail: marcel.erlinghagen@uni-due.de, URL: http://www.uni-due.de/soziologie © Federal Institute for Population Research 2012 – All rights reserved Published by / Herausgegeben von Prof. Dr. Norbert F. Schneider Federal Institute for Population Research D-65180 Wiesbaden / Germany Managing Editor / Verantwortlicher Redakteur Frank Swiaczny Editorial Assistant / Redaktionsassistenz Katrin Schiefer Language & Copy Editor (English) / Lektorat & Übersetzungen (englisch) Amelie Franke Copy Editor (German) / Lektorat (deutsch) Dr. Evelyn Grünheid Layout / Satz Beatriz Feiler-Fuchs E-mail: cpos@destatis.de Scientifi c Advisory Board / Wissenschaftlicher Beirat Jürgen Dorbritz (Wiesbaden) Paul Gans (Mannheim) Johannes Huinink (Bremen) Marc Luy (Wien) Clara H. Mulder (Groningen) Notburga Ott (Bochum) Peter Preisendörfer (Mainz) Board of Reviewers / Gutachterbeirat Martin Abraham (Erlangen) Laura Bernardi (Lausanne) Hansjörg Bucher (Bonn) Claudia Diehl (Göttingen) Andreas Diekmann (Zürich) Gabriele Doblhammer-Reiter (Rostock) Henriette Engelhardt-Wölfl er (Bamberg) E.-Jürgen Flöthmann (Bielefeld) Alexia Fürnkranz-Prskawetz (Wien) Beat Fux (Zürich) Joshua Goldstein (Rostock) Karsten Hank (Köln) Sonja Haug (Regensburg) Franz-Josef Kemper (Berlin) Michaela Kreyenfeld (Rostock) Aart C. Liefbroer (Den Haag) Kurt Lüscher (Konstanz) Dimiter Philipov (Wien) Tomáš Sobotka (Wien) Heike Trappe (Rostock) Comparative Population Studies – Zeitschrift für Bevölkerungswissenschaft www.comparativepopulationstudies.de ISSN: 1869-8980 (Print) – 1869-8999 (Internet)