Life Satisfaction of Immigrants and Length of Stay in the New Country

Abstract

A large amount of research is dedicated to the measurement of immigration trends and integration processes, but comparative studies of the components of immigrants’ well-being are rare. This paper investigates the link between the length of stay and the subjective well-being (SWB) of immigrants. A step-by-step regression method is used to understand interactions between different individual- and macro-level factors in the life satisfaction of immigrants. The results of the European Social Survey (ESS) show that the effect of length of stay on SWB is mitigated by numerous individual- and country-level variables. After all background variables are considered, newly arrived immigrants and those who have been in the new country for more than 20 years had a similar life satisfaction to that of the local-born population. Immigrants with 10 to 20 years’ experience in the country seem to have lower life satisfaction than the local-born population. The SWB of different immigrant groups is shaped by cultural background, economic coping, number of social contacts, perception of discrimination, and democracy. These factors influence the life satisfaction of migrants with different lengths of stay differently. The most vulnerable immigrant groups in Europe are those from Africa region.

1. Introduction

Since the 1990s, Europe has received growing numbers of immigrants, and the number of persons with an immigrant background in the European population has grown. The well-being of the migrant population will have a strong influence on the well-being of all European societies in the future. A better knowledge of the well-being of the immigration population is crucial for both policy making and scientific knowledge. Several surveys show that well-being and life satisfaction do not only influence individual coping, but determine future individual success in several life domains (Parr 2010; Sprecher 2002).

Mobility might have a twofold influence on people’s well-being. Migration is a tool to achieve some objectives in life, and it might provide therefore a better life for those who move (Kofman and Meetoo 2008; Alpaslan et al. 2014; Helliwell et al. 2018); however, moving is not a simple process, and it presents different challenges. The empirical literature generally confirms that immigration might lead to progress in multiple well-being domains in life (Chiswick et al. 2005; Farley and Alba 2002; Depalo et al. 2006; Manning and Roy 2010), although the SWB level of immigrants may remain below the average well-being level in the host country (Helliwell et al. 2018). It means that despite making some progress in life, many immigrants fail to gain the same SWB level as the native population (Hendriks 2015; Safi 2010; Obućina 2013; Stillman et al. 2015; Calvo and Cheung 2018; Hendriks et al. 2018; Nesterko et al. 2013).

Length of stay in the country seems to be an important determinant of immigrants’ SWB. For example, Hendriks et al. (2018) state that migration is considered a long-term investment in one’s future and that life is expected to gradually improve over time after the initial hurdles are overcome. The authors measured the positive and negative effects on immigrants over time and found that most migrants experience more positive effects of migration over time and fewer negative ones. Previous studies have shown that the main problems faced by immigrants causing lower SWB are separation from loved ones, perceived discrimination in the new location, a feeling of relative deprivation, less prosperous living conditions, strong feelings of social isolation, and loss of cultural heritage (Helliwell et al. 2018; De Vroome and Hooghe 2014; Safi 2010; Senik 2014), but moving from one place to another can also help people to develop their careers, find new friends, and experience new opportunities. Among other factors, state-level policies (e.g., immigration policies, housing policies, and regulations regarding healthcare access) and the attitudes of the receiving society shape the well-being of newly arrived persons (Vignoli et al. 2014), all of which are among the indicators covered in this paper.

This paper aims to analyse the SWB of immigrants in Europe according to their length of stay and examine the role of individual- and macro-level factors in the SWB of immigrants. The paper uses European Social Survey (ESS) data and analyses immigrants in three categories, according to the length of time they have lived in their new country.

2. Subjective Well-Being and Migration

Voluntary migration can be seen as a rational act to improve personal well-being. SWB reflects how people experience their lives (Ryan and Deci 2001; Ryff and Keyes 1995; Huppert 2014). The overall evaluation of life is often seen as a combination of evaluations of life in different life domains (Sirgy 2021). Domains can be non-compensatory, compensatory, conjunctive, or lexicographic (see Ainsaar 2004). Compensatory and non-compensatory rules mean that some components of well-being can and cannot be replaced by others, respectively. According to the conjunctive rule, each component of life satisfaction must be achieved at least to a minimally acceptable level. According to the lexicographic rule, people rank the components of SWB according to the importance they ascribe to each one; therefore, each domain might be given a different weight. In summary, we can expect that some factors are more important for immigrants than others and that their importance can be dependent on the individual characteristics of an immigrant.

Although the life satisfaction of immigrants can be analysed within the same framework as general factors of SWB, such as age, health, income, partnership, employment status, social trust, trust of institutions, and perception of democracy (Graham 2009; Sirgy 2021), migration experience adds additional culture- and environment-related adaptation challenges. One general factor, but with some migration-specific features, is the role gap between expectations and reality (Czaika and Vothknecht 2014; Böhme 2015). According to adaptation theories (Luhmann et al. 2012), integration and changing aspirations change the way people experience and evaluate their lives. Perceptions are mostly based on the gap between what one wants (aspirations) and what one has. Aspirations, in turn, depend on comparison with the situation in the past. For example, Fozdar and Torezani (2008) describe how previous experiences shape the life evaluation of refugees in Australia.

Information processing plays an important role in this process. The evaluation of a new environment depends on the availability of information and the information-processing capacity of immigrants. Better-informed immigrants might start preparations for integration into the new country before they arrive in the country (Demireva 2019) and might also have better options to achieve higher well-being.

Social networks have been an important source of information processing for centuries, and, as information technology advances, immigrants’ pre-knowledge about their future destinations is steadily improving. Therefore, large individual social networks usually improve the information-processing ability of persons. For example, Coletto and Fullin (2019) show how medium- and low-skilled migrants use both social networks and social media during their preparations to migrate to Europe. Positive social contacts might be more important for vulnerable groups, such as young, elderly, and poor people (Hartshorne 1992). However, regardless of improved information opportunities, people make errors, for example, tending to overestimate opportunities in their destination (Knight and Gunatilaka 2010; Bartram 2011), which might lead to disappointment. Some studies (De Jong et al. 2002) have estimated the correctness of information by different life domains. They found, that immigrants were less satisfied with the living environment in the destination country than they had anticipated, and more satisfied with their employment. The result might depend on the purpose of migration. For example, temporary immigrants (e.g., students) had lower post-move employment satisfaction than single permanent labour force migrants.

Finally, the SWB of immigrants is influenced by the efforts of the receiving society and the acceptance of immigrants by the local population. Higher acceptance leads to higher well-being (Helliwell et al. 2018). The well-being of all people is dependent on state support systems (Alba et al. 2011; Lewin-Epstein et al. 2003; Whiteley et al. 2010). Social protection and integration systems vary between countries and shape different environments for immigrants as well as for local people. More generous social protection and integration systems should increase the SWB of immigrants. The impact of the overall immigration situation on SWB is unclear. Some surveys show that increasing immigration erodes social trust (Ziller 2015), but the SWB of migrant-receiving destinations is higher than that of countries with few immigrants (Betz and Simpson 2013; Alpaslan et al. 2014).

The perception and selection of a reference group or standard of comparison plays an important role in the process of evaluating well-being (Festinger 1954; Michalos 1985; Diener et al. 2006; Luhmann et al. 2012). Studies (Gelatt 2013; Reese 2001; Menjívar and Bejarano 2004) show that most immigrants refer to the situations in both the home and the host country when evaluating their situation, but the group of comparison tends to change during the stay in the country.

Discrimination and difficulties in adaptation and assimilation are also common explanations for the low SWB of immigrants. Safi (2010) tested two hypotheses explaining the lower level of immigrants’ SWB: assimilation and the effect of discrimination. She found that immigrants’ relative dissatisfaction did not diminish with time and across generations, which refuted the assimilation paradigm. For some ethnic groups, perceived discrimination provided a consistent explanation for their lower level of life satisfaction compared to local people.

To conclude, we can formulate two hypotheses about the processes shaping the well-being of immigrants after they arrive in a new country. The first is the assumption that gradual integration will improve the objective situation of immigrants in the new destination and their SWB should therefore also gradually improve. The second hypothesis is related to the perception of equality: if immigrants do not achieve their expected goals, they might experience growing disappointment and the gradual erosion of SWB. Both processes are influenced by all the other factors of SWB.

3. Materials and Methods

Data from the cross-national European Social Survey (ESS) are used for the analyses in this paper. The ESS is a highly standardised international survey with country-representative samples of people aged 15 and over who are residents in private households. Among other topics, the ESS covers respondents’ socio-demographic background, ethnic identity, migration background, self-reported SWB, and a large range of variables related to SWB. A detailed description of the used variables is given below.

This paper first presents the SWB distribution by immigrant and non-immigrant populations in Europe according to time and migrants’ length of stay in the country.

Secondly, ESS pooled data are used from rounds seven and eight (ESS Round 7 2014; ESS Round 8 2016) to analyse the SWB of immigrants by date of arrival: before 1994, 1994–2003, and 2004–2016. The periods represent different immigration stages in Europe. The period before 1994 is characterised by a rather low level of immigration into Europe. The liberation of East European countries in the 1990s saw the rise of immigration pressure in Europe, although general migration activity was still lower than in the later periods. The period 1994–2003 marks the gradual growth of international immigration flows in Europe. Since 2004, immigration has become more intensive for various international and internal reasons, and the migration area has gradually expanded.

Initial analyses of the original survey data indicated that some respondents could not recall the exact year in which they immigrated to Europe. The original survey data show a clear overrepresentation of immigrants reporting that they arrived in years ending with 0 or 5. This finding indicates the well-known bias of memory recall in population research when persons attribute an event to years ending with 0 or 5 if they do not remember the exact year. We decided to avoid these years as cut-off points of periods as they were not a reliable means of measurement.

The main criteria for the selection of the countries were first, the availability of data from the ESS databank and second, the presence of at least 100 immigrants in all periods in all country samples. After those criteria were applied, six countries remained in the analyses, namely, Austria, Belgium, Switzerland, Germany, Great Britain, and Ireland, which are also the countries with the highest share of the immigrant population in Europe (Eurostat 2019a).

Immigrant respondents are first-generation immigrants. Second- and third-generation immigrants were consciously left out of the analyses to avoid having to consider overly complex processes in explanations of SWB formation. It is known (Maxwell 2010) that first-generation migrants are different from second- and third-generation ones, principally because of their lower expectations and more positive evaluations of the host society. At the same time, they experience more objective problems in a new country. The language of the receiving society presents one of the major difficulties and barriers to communication and accessing institutions for the first generation, and they are more likely to accept lower-ranking jobs (Krieger 2008; Demireva 2019).

Knowing immigrants’ region of origin can provide additional understanding of the roots of SWB. For example, cultural distance and language familiarity influence ease of integration. Different rules of access to European countries apply to immigrants from within and outside the European Union (EU). For example, immigrants from EU countries are granted equal access to the labour market, social protection, etc., yet they are sometimes not entitled to the integration services provided to third-country migrants, such as free language courses. In this paper, a question about ancestry was used to measure the main cultural-ethnic background of immigrants, namely, ‘How would you describe your ancestry?’ The original ancestry codes were grouped into seven large groups for analyses: European; North Africa, Central Asia, and the Middle East; Sub-Saharan Africa; South Asia; East Asia; South America and the Caribbean; and North America and Australia.

Table 1. Main variables (means) and comparison of groups of immigrants by their length of stay in the host country with local-born population.

	Country Born	Period of Stay in the Country
		Before 1994 >20 Years		1994–2003 10–20 Years		2004–2016 <10 Years
How satisfied with life (0 extremely dissatisfied, 10 extremely satisfied)	7.5	7.4	**	7.2	**	7.5
How often socially meet with friends, relatives or colleagues (1 never, 7 every day)	4.9	4.7	**	4.8	**	4.9
Without partner (%)	18.0	20.4	**	10.4	**	13.4	**
Trust of institutions (0–50)	24.8	26.5	**	27.3	**	29.3	**
Trust of people (0–30)	16.9	16.1	**	16.3	**	19.9
How satisfied with the way democracy works in country (0 extremely dissatisfied, 10 extremely satisfied)	5.6	6.1	**	6.5	**	6.9	**
Incomes (1 comfortable, 3 difficult)	1.7	1.9	**	1.9	**	1.9	**
Subjective general health (1 good, 3 bad)	1.3	1.4	**	1.2	**	1.1	**
In paid work (%)	54.0	52.5		65.8	**	62.2	**
In education (%)	9.5	1.8	**	10.2		16.4	**
Unemployed, actively looking for job (%)	3.4	4.5	**	7.6	**	8.1	**
Age of respondent, calculated	48.8	55.8	**	39.9	**	34.2	**
Gender male (%)	48.9	44.8	**	50.1	**	51.0	**
Member of a group discriminated (%)	5.7	14.4	**	17.4	**	15.6	**
With tertiary education	31.3	32.9		39.2		46.7	**
Ancestry
Europe	98.4	80.8		66.7		67.1
North Africa, Central Asia, Middle East	0.8	9.5		14.6		10.3
Sub-Sahara Africa	0.1	2.0		4.9		5.5
South Asia	0.4	4.6		7.1		8.9
East Asia	0.0	0.3		1.7		2.2
South America, Carib	0.2	1.6		3.9		4.9
North America, Australia	0.0	1.2		1.0		1.2
N	21,809	1521		895		1412

** differences with local born population, p < 0.05.

shows the distribution of people with different backgrounds in three immigrant and one local-born group. People with European identity form 98% of the local-born group, and European identity also prevails in all the immigrant groups, although the share of non-Europeans is rising among the recent immigrants. The second largest group is people with North African, Central Asian, and Middle Eastern backgrounds, and the third group is South Asian people.

For all people in the analyses, we included a large range of individual and social environment characteristics to understand the formation of SWB. The variable selection process entailed a careful combination of preventing collinearity while including all the essential variables.

The main dependent variable is subjective life satisfaction, measured with the question ‘All things considered, how satisfied are you with your life as a whole nowadays?’ Answers ran from 0 (‘extremely dissatisfied’) to 10 (‘extremely satisfied’).

Additionally, we used traditional individual characteristics shaping well-being, such as age, health, partnership status, and a subjective evaluation of household economic coping, generalised trust and trust in power institutions, number of social contacts, perceived discrimination, perception of democracy, and information about the self-reported ancestry background in the models. The exact questions and response options are as follows:

Subjective evaluation of health was measured with the question ‘How is your health in general?’ (Very good or good—1; fair—2; bad and very bad—3).

Income level was measured with the question ‘Which of the descriptions on this card comes closest to how you feel about your household’s income nowadays?’ (Living comfortably on present income—1; coping on present income—2; finding it difficult or very difficult with present income—3).

The index of generalised trust was measured with three questions: (1) Would you say that most people can be trusted or that you can’t be too careful in dealing with people?; (2) Do you think that most people would try to take advantage of you if they got the chance, or would they try to be fair?; and (3) Would you say that most of the time people try to be helpful or that they are mostly looking out for themselves? The scores of individual answers ranged from 0 to 10, and the final index value from 0 to 30.

Trust in state institutions was measured through five questions about trust in the country’s parliament, legal system, politicians, political parties, and police, and scores were added to form a final index. The scores of the final index ranged from 0 to 50.

Social contacts were measured with the question ‘How often do you meet socially with friends, relatives, or work colleagues?’ (never—1; every day—7).

Perception of discrimination, an intermediate variable between individual perception and country characteristics, was measured with the question ‘Would you describe yourself as being a member of a group that is discriminated against in this country?’ (yes—1; no—2).

Conditions of integration, including the context of reception and support systems, differ by country and influence the coping of immigrants and their SWB. We used four macro-level variables to describe country-specific policy situations: the Migrant Integration Policy Index (MIPEX), the average crude immigration rate for the period 2014–2016 (Eurostat 2019b), the immigrant’s acceptance index, and the average life satisfaction level in the receiving country.

MIPEX is a tool that measures the level of policies dealing with migrant integration. It comprises 167 policy indicators that give a multi-dimensional picture of migrants’ opportunities to participate in society. The index is a useful tool to compare government actions in the field of integration (Huddleston et al. 2015). We used only the final aggregate index for analyses. The higher the index, the more comprehensive the integration policy.

The question about the democracy in the country, which belongs to the group of questions about perceptions of a state, asked ‘On the whole, how satisfied are you with the way democracy works in country?’ (extremely dissatisfied—0; extremely satisfied—10).

Migrants often experience a great shock due to the perceived lack of empathy towards them as they have not yet acquired insider knowledge and feel disorientated (Demireva 2019). The attitudes of local people towards immigrants influence immigrants’ SWB (Helliwell et al. 2018; Hendriks and Bartram 2016), adaptation process, and self-perception. To capture these phenomena, we included a country-level aggregate index about acceptance on the basis of attitudes toward immigrants. This is used as a country average index, and it is derived from three questions about attitudes toward immigrants: (1) To what extent do you think the country should allow people of the same race or ethnic group as most of the country’s population to come and live here?; (2) How about people of a different race or ethnic group?; and (3) How about people from poorer countries outside Europe? (allow many to come—1; none—4).

Gender roles and the absence of social networks might determine the integration experiences of professional female spouses (Lutz 2010; Phan et al. 2015). We also tested all the models with gender, but the main results about immigrants did not change, and we left gender out of the models. Two further variables left out of the analyses for reasons of collinearity were education and country-level social protection expenditure. Helliwell and Putnam (2004) assume that education has a mainly indirect impact on measures of well-being, and its effect is therefore explained with other variables. Social protection expenditure per inhabitant in the countries was a powerful influencer of SWB, but it was too closely correlated with country-level life satisfaction level (k = 0.976) and produced collinearity. Ultimately, we had to omit this indicator from the analyses.

4. Results

Figure 1 gives an overview of a level of SWB according to the time of arrival. The SWB of immigrants who arrived most recently (after 2004) is equal to the non-immigrant group, while migrants who had stayed in the country longest period—arrived before 1994—are somehow less satisfied with their lives compared to non-immigrants, and immigrants who have been in the country for 10 to 20 years are the most dissatisfied (Figure 1). This result does not coincide with our expectation that the SWB of immigrants would improve linearly the longer they spent in the country. The following analyses provide some explanations of the reasons for these differences in SWB. At the same time, it should be kept in mind that the differences between the groups are not large and vary within the range of only one point on the 11-point scale.

For a better understanding of the groups, we begin analyses with a simple description of the migrant groups, which also help interpret the regression models later. All these characteristics affect people’s well-being in general. The local-born population is the most numerous group and a reference group for comparison of all the immigrant groups in Table 1.

The local-born population is distinguished from the immigration groups by higher incomes and lower trust in state institutions (Table 1). It is difficult to explain why local people have lower trust, but it might be due to higher expectations of institutions. We see the same pattern of lower satisfaction with democracy among local people: all groups of immigrants have significantly better opinions of democracy than the country-born population. This positive result for immigrants may be a result of a comparison of a new destination country with the countries from which the immigrants had come.

Immigrants who arrived before 1994 differ significantly from the other groups (Table 1). They are the oldest group, being even older than the local-born group, with an average age of 55.8 years. At the time of arrival, they were young, as immigrants usually are, but they have stayed more than 20 years in the country. Therefore, it is not surprising that they also have worse health than the other groups.

People who arrived before 1994 have lower trust in other people, and they have fewer social contacts than the locals and other immigrant groups (Table 1). Eighty-one percent of the immigrants in the group are of European origin, and they have more trust in state institutions and are more satisfied with democracy in the country than the local people. About 14% of these immigrants feel that the group to which they belong is discriminated against in the country.

Immigrants who arrived between 1994 and 2003 are, on average, 40 years old and in better health than the older immigrants (Table 1). They have the lowest SWB. Fifty percent of them are women, and 90% have a partner, which is a higher number than in the other groups. Like the oldest immigrants, this group has a lower general trust in other people and fewer social contacts than locals and later immigrants. Their satisfaction with their income is low, but they trust state institutions more than the local people. A higher percent, compared to other groups, believe that their group is discriminated against in the country.

Immigrants who arrived in 2004 and later are the youngest group, with an average age of 34, and they have the best health, on average (Table 1). Eighty-seven percent have a partner with whom they live. Recent immigrants are also the best-educated, and there is an equal share of men and women in the group. They trust other people to the same extent as local people, and their trust in state institutions is the highest of all the groups. The recent group of immigrants also has the highest share of students. The main problem faced by the most recent arrivals is economic difficulties, and 16% of people in this group state that their group is discriminated against in the country.

Regression models give more detailed analyses of the sources of differences in the well-being of immigrants who arrived at different times (

Table 2. Generalised linear regression models for life satisfaction.

	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6
Arrival 2004–2016	−0.206 **	−0.116	0.03	0.064	−0.023	−0.075
Arrival 1994–2003	−0.36 **	−0.257 **	−0.09	−0.056	−0.094	−0.135 **
Arrival before 1993	−0.136 **	−0.074	0.057	0.066	0.071	0.048
Local-born (reference)	1	1	1	1	1	1
Age	0.009 **	0.009 **	0.005 **	0.008 **	0.006 **	0.005 **
Health bad	−2.062 **	−2.061 **	−1.74 **	−1.661 **	−1.501 **	−1.478 **
Health average	−0.82 **	−0.818 **	−0.647 **	−0.617 **	−0.552 **	−0.543 **
Health good (reference)	1	1	1	1	1	1
Average LS in country	0.916 **	0.912 **	0.694 **	0.630 **	0.333 **	0.191 **
Alone	−0.466 **	−0.464 **	−0.341 **	−0.400 **	−0.365 **	−0.369 **
Living with a partner (reference)		1	1	1	1	1
North America, Australia		−0.382	−0.594 **	−0.591 **	−0.645 **	−0.648 **
South America, Caribean		−0.014	0.068	0.099	0.212	0.213
East Asia		0.387	0.064	0.047	0.031	0.024
South Asia		−0.151	−0.178	−0.161	−0.147	−0.171
Sub-Sahara Africa		−0.864 **	−0.5 **	−0.438 **	−0.353 **	−0.416 **
North Africa, Central Asia, Middle East		−0.423 **	−0.151	−0.189**	−0.107	−0.171 **
Europe (reference)			1	1	1	1
Incomes difficult			−1.664 **	−1.598 **	−1.322 **	−1.282 **
Incomes average			−0.64 **	−0.602 **	−0.489 **	−0.469 **
Incomes easy to cope (reference)			1	1	1	1
Unemployed, actively looking for job			−0.811 **	−0.784 **	−0.738 **	−0.735 **
Something else (reference)			1	1	1	1
Student			0.038	−0.015	−0.120**	−0.154 **
Something else (reference)			1	1	1	1
In paid work			−0.150 **	−0.109 **	−0.130 **	−0.128 **
Something else (reference)			1	1	1	1
How often socially meet with friends, relatives or colleagues			0.160 **		0.149 **	0.145 **
Member of a group discriminated against in this country				0.306 **		0.234 **
Not discriminated (reference)				1		1
Trusting state institutions				0.020 **		0.008 **
Trusting people				0.049 **		0.047 **
Integration index				0.000		−0.001
Immigration attitude				−0.035 **		−0.014
Immigration rate				−0.164 **		−0.196 **
Satisfaction with democracy in the country						0.105 **

** p < 0.05.

). The first model shows differences in the SWB of immigrants according to time of arrival, considering differences in age, partnership status, health, and the average life satisfaction of the country as a country-level variable. All variables in this model are statistically significant. As expected, life satisfaction generally depends on health, partnership status, age, and the average level of life satisfaction in the country. Considering these variables in the model, all immigrant groups have a lower level of SWB than local-born people. Immigrants who arrived between 1994 and 2003, and have lived in the country for between 10 and 20 years, have the lowest life satisfaction of all immigrants after considering demographic background and the country’s average SWB.

The second model adds ancestry information. The results show that immigrants with African, Central Asian, and Middle Eastern backgrounds are essentially less satisfied with their lives in Europe. The ancestry composition of arrivals also negatively influences the life satisfaction of immigrants who arrived before 1993 and the latest immigrant group.

Model 3 includes economic coping variables and shows that income level and work activity status explain the lower life satisfaction of the immigrants who arrived in 1994–2003. After income and social activity indicators are added, the SWB of all immigrants becomes like that of the local people. This finding indicates that income and economic activity background are significant determinants of the SWB of immigrant groups. Interestingly, after considering all variables in model three, employment status makes people more dissatisfied with their lives. Income and activity status is also an important determinant of SWB of North Africa, Central Asia, and Middle East immigrant. Their SWB difference with European origin people disappears. Immigrants with North American, Australian background seem to have equal SWB with European people, mainly due to economically better situations. After considering their economic status, they have an even lower SWB. However, this is a rather limited group in analyses and the final interpretation needs to be careful.

After adding an indicator about social relations into Model 4, the SWB of North Africa, Central Asia, and Middle East immigrant groups changes once again. Social contacts are an explanation for the higher SWB in this immigrant group. Model 4 shows that after considering the positive impact of social contacts, the SWB of this group decreases once again (all other factors are already considered).

Model 5 tests whether national policies and the evaluation of the state have an additional impact on the SWB of immigrants. Non-discrimination, trust in people and state institutions, and lower immigration rates have a generally positive influence on the SWB of all groups, but they also help to explain the lower SWB of people with certain backgrounds, namely persons with North Africa, Central Asia, and Middle East origin. These factors do not change the coefficients of people with a Sub-Sahara African and North American, Australian background people, who retain the lowest SWB after all variables in Model 5 are considered. Also, country-level policies do not change the SWB result for arrival periods.

As a last step, satisfaction with democracy in the country was added to Model 6, and this introduced some changes. A simple comparison of groups (Table 1) showed, that especially recent immigrants evaluated democracy in the country more highly than non-immigrants and satisfaction with democracy is an essential source of SWB for them. After considering satisfaction with democracy, the life satisfaction of immigrants who arrived between 1994 and 2004 became lower compared to local-born people. The results remained the same even after other potential SWB indicators, such as physical safety and education (not presented in the paper), were added to the model. Of all the immigration period variables, only the 1994–2003 arrival group has a worse SWB than local-born people in the final model. This finding indicates that the life satisfaction of immigrants who arrived 10 to 20 years ago is lower than that of the local people and the other immigrant groups, which is not explainable by the variables in the model. At the same time, all the main effects in the model remain the same as described in the previous models, except for the impact of country-level measured immigration attitude. Integration activities and positive attitudes to immigration do not influence SWB in the final model, and higher immigration rates are associated with lower SWB.

5. Discussion

Immigration flows will continue to arrive in Europe in the future (Eurostat 2019c), and studies of the formation of immigrants’ life satisfaction therefore become more urgent.

This paper provides complex analyses of the differences in immigrants’ SWB according to the length of stay in the country and the impact of individual- and country-level factors on SWB. Knowledge of the mechanisms underlying the formation of migrants’ SWB is useful for policymakers and other stakeholders working with immigrant issues. The effect of the period on subjective well-being consists of the interaction of integration factors and many other psychological factors of well-being. These factors partially can overlap.

Previous analyses had to choose between two approaches: including many countries with limited explanative variables in the analyses or concentrating on analyses of one country with rich background variables. Due to the collection of rich information from the ESS, this paper ran a complex analysis of multiple mechanisms in many countries. The paper is also innovative by not treating immigrants as a single group but according to the length of stay in the country. The analyses showed that this approach might be a useful tool to understand the lives of immigrants. Immigrants were analysed in three groups according to their time of arrival in the country.

The immigrant integration hypothesis argues that a longer stay in the country will lead to better integration and improvement of SWB, but this hypothesis was not supported by the empirical results. Empirical results showed, contrary, lower SWB after 10–20 years of stay, which cannot be explained by the usual demographics and country-specific indicators of well-being. This SWB difference with local-born people persisted after testing for many explanatory variables. Limited social contacts, but higher satisfaction with democracy in combination with other studied factors explained some difference in SWB compared to that of local born, but some of their lower SWB remained unexplained. Because of the remaining difference between SWB with that of local people, we can conclude that 10 to 20 years after arrival might be the most critical period for immigrants.

One possible explanation for the lack of improvement in SWB over time is the change of the reference group during the stay in the country. The initial optimism of immigrants about the future will be replaced by pessimism over time if initially higher expectations are not met or the situation does not improve as quickly as expected. However, these hypotheses need to be tested in future research projects, our models were not able to capture the influence of a reference group, nor optimism–pessimism variables about the future in this paper. It is also important to keep in mind that the overall difference in well-being of this group compared to non-immigrants is not large and current immigrant groups are heterogenous. For example, low-income, low-trust groups or un-partnered immigrants will need separate analyses in the future.

Immigrants who arrived in the host country more than 20 years ago had, somehow, lower SWB compared to the local-born population in a simple comparison of groups; however, this difference was entirely explained by the demographic and cultural background of the immigrants. Thus, there is no reason to believe that the SWB of immigrants who have been in the country for more than 20 years is actually different from that of natives.

SWB of immigrants who arrived less than 10 years ago did not differ from the local-born population SWB. The most important driver of the SWB of immigrants who arrived in the country during the past 10 years was low income in combination with the employment/unemployment situation.

Furthermore, groups of immigrants with different ethnic backgrounds had different SWB coefficients. The most vulnerable immigrant group was that of African origin, whose SWB needs further investigation.

One peculiar but consistent outcome, was the negative impact of the immigration rate on people’s SWB. This result conflicts with the results of some studies reporting a small positive effect (Betz and Simpson 2013; Alpaslan et al. 2014). In this study, we used official migration statistics to measure immigration intensity in the years when survey data were collected, and this is a rather different approach to those taken in previous papers that found positive effects. The result is also influenced by factors already included in the models. For example, many economic and state policy variables were already included in the models in our study, and the immigration effect was additional to those. Some previous studies have also indicated the possibility of a negative influence. For example, Longhi (2014) demonstrated negative effects in specific contexts, and Hendriks (2018) stated that small positive effects may become negative when there are sharply rising numbers of immigrants. Many studies report the impact of immigration on the labour market and that wages drop because of immigration in the region (see, for an overview, Betz and Simpson 2013). Lower incomes, in turn, might have a negative influence on SWB.

The current approach also has several limitations. Although the ESS data used in the study are high in quality and rich in content, they are still cross-sectional by nature and cannot replace longitudinal data. On the other hand, identical international longitudinal surveys, especially for immigration issues, are scarce. Until such data are available, studies such as the ESS will be useful in understanding integration and SWB formation.

Due to the need to guarantee the high quality of the data, the selection of countries was limited with countries with at least 100 respondents in all time-period categories. Future research should include a large range of countries.