Pull Factors and Migration Preferences: Evidence from the Middle East and North Africa

Abstract

Which national characteristics do voluntary migrants prioritize when considering destinations? Although this question is salient for policymakers, extant studies face challenges when seeking to identify how various pull factors shape destination preferences. Surveys of migrants are typically conducted after they arrive, introducing selection bias and post-hoc reasoning. Moreover, desirable national characteristics tend to co-vary, implying that observed relationships with migration flows may be confounded. In this article, we identify the destination preferences of prospective migrants by drawing on a sample of 8,500 respondents from five sending states across the Middle East and North Africa. Prospective migrants completed a series of conjoint survey tasks in which they chose between two destinations with randomly varying characteristics. The results reveal a clear hierarchy of preferences, with prospective migrants placing the greatest priority on liberal democratic governance and employment prospects. The availability of welfare benefits acted as a secondary consideration, while geographic distance and co-ethnic stock did not strongly predict initial destination preferences. While the rank order of these considerations remains consistent across national samples, our results suggest that respondents from different economic and political backgrounds vary in how they navigate potential tradeoffs between national characteristics. These findings address post-arrival bias in extant studies by revealing prospective migrants’ preferences before they interact with the opportunity structures that facilitate and restrict entry into desirable destinations.

Keywords

survey destination preferences

Introduction

Which national characteristics do voluntary migrants prioritize when considering potential destinations? This question has become increasingly salient worldwide, as the costs of migration have fallen across the developing world (King 2012; Van Der Velde and Van Naerssen 2015; Crawley and Hagen-Zanker 2019). Spurred by dislocation and reduced barriers to exit, the number of non-citizens living abroad has increased by 70 percent since 2000 (United Nations, Department of Economic and Social Affairs 2016). Within this mobile context, a nuanced understanding of the factors that motivate voluntary migration is essential for governments and policymakers seeking to anticipate, encourage, or manage migratory flows. Yet despite a comprehensive literature on the determinants of migration, disagreement persists regarding the set of factors that lead migrants to prioritize certain destinations over others (Pederson et al 2008; Mayda 2010; Geis, Uebelmesser, and Werding 2013; Sporlein 2015). While populist politicians in high-income countries have increasingly emphasized material motivations, ranging from migrants’ desire for higher wages to their interest in generous welfare benefits (e.g., Ivarsflaten 2008; Rooduijn, de Lange, and van der Brug 2016), other observers have de-emphasized material concerns and stressed pull factors such as liberal democratic rights and established expatriate communities (Crawley 2010; Fitzgerald, Leblang, and Teets 2014).

These factors need not be mutually exclusive. Yet empirically, their relative hierarchy remains opaque, largely because studies that ascertain destination preferences face two methodological challenges. First, studies that assess migrants’ motivations are generally conducted after migrants arrive within a destination state, introducing the possibility of post-hoc reasoning. As Schewel (2019) argues, focusing on the post-arrival context generates a “mobility bias” that influences perceptions of prospective migrants’ aspirations and intentions. Second, destination states with desirable attributes, such as liberal democratic institutions, tend to possess other pull factors that have been posited to spur migration (Van Der Velde and Van Naerssen 2015; James and Mayblin 2016; Gest and Boucher 2020). If pull factors co-vary, quantitative analyses of the relationship between national characteristics and migration flows may be confounded.

In this article, we seek to address these issues by leveraging conjoint survey analysis to assess the latent preferences of voluntary migrants before they emigrate. Conjoint analysis is a survey technique derived from marketing research that employs randomization and repeated choice tasks to identify respondents’ latent preferences (Hainmueller, Hopkins, and Yamamoto 2014). Drawing on samples across five sending contexts — Jordan, Lebanon, Libya, the Palestinian Territories, and Tunisia — we query respondents’ migration intentions and ask them to report their interest in hypothetical destination countries with randomly varying characteristics. This approach generates novel evidence concerning the role that national attributes play in shaping hierarchies of potential migrant destinations. Moreover, the conjoint design allows us to examine how these preferences interact, while minimizing social desirability bias associated with direct questioning.

Our results suggest that liberal democratic governance and employment prospects are the primary factors that shape the relative desirability of migrant destinations within our sample. Prospective migrants were on average 26.2 percent more likely to prefer a liberal democratic destination and 25.3 percent more to prefer a state with high job availability, holding all other national characteristics constant. We also find that welfare remains salient after controlling for other factors, with prospective migrants 17.4 percent more likely to prefer a country with accessible benefits. Finally, the data suggest that the presence of established communities of co-nationals and geographic distance do not strongly shape latent preferences. The rank order of these preferences is consistent across countries and survey fielding method (in-person vs. phone), suggesting broad consensus on desirable destination traits within our sample. However, our results also suggest that prospective migrants’ socio-economic background and political context nevertheless shape how they manage tradeoffs between destination characteristics. Prospective migrants who were female, older, or from active conflict regions, for example, were most likely to prioritize democratic rights over economic factors. Moreover, we find evidence that demand for welfare benefits is highest within destination states that lack liberal democratic rights and have weak economies, suggesting that the salience of this pull factor is driven by risk aversion rather than by income maximization.

Understanding migrants’ initial preferences provides several contributions to ongoing policy and scholarly debates. First, receiving states’ immigration policies are frequently informed by assumptions concerning migrants’ motivations (Boucher and Gest 2018, 66–93). Without a full understanding of these preferences, states may implement restrictive policies to reduce the perceived attractiveness of a pull factor, such as restrictions on welfare access or nationality quotas. If these pull factors do not in fact influence destination decisions, such restrictions could impose substantial costs on resident immigrants within destination states, without meaningfully altering future migration patterns. Second, in contrast to the analysis of historic flows, an assessment of pre-departure preferences is forward-looking. Past migration flows are a function of migrants’ preferences at the time of departure (Creighton 2013; Tjaden, Auer, and Laczko 2018), as well as of a set of contingent opportunity structures that constrain or encourage migration to particular destinations (Carling 2002, 2004; De Haas 2010; Boucher and Gest 2018). Given that these opportunity structures are likely to change over time, an analysis of pre-departure preferences enables scholars and policymakers to better anticipate future migration patterns and provides insight into how demand for destinations might vary if restrictions on international flows were relaxed.

In sum, then, a robust theoretical account of migration requires an understanding of the factors that shape decision-making across the migration process (Carling and Schewel 2018). This article’s design allows us to shed light on the early stages of this process by evaluating the relative draw of various pull factors. In exploring the hierarchy of destination-state characteristics, our findings also demonstrate how migrants navigate a series of realistic tradeoffs, such as the relative importance placed on democratic rights versus economic opportunity. This evidence holds the potential to contribute to an emerging research agenda exploring how these preferences subsequently interact with information and opportunity structures to shape final destinations.

To develop these ideas, we first review research on the various pull factors that have been hypothesized to shape migrants’ destination preferences. We then consider how these findings may be influenced by the post-arrival setting in which they are typically examined and we derive a set of hypotheses concerning migrants’ preferences prior to departure. After outlining our survey sampling strategy and research design, we present results from five sending contexts. We conclude by discussing the implications of our findings for public and scholarly understandings of migrants’ destination preferences.

Pull Factors and Hierarchies of Destination Preferences

Early scholarship on the migration process sought to better understand the abstract considerations that motivate or constrain individuals’ decision-making (see King 2012 for an overview). Over the last decade, the field has increasingly focused on how various national pull factors interact with these latent preferences and aspirations (e.g., Thielemann 2008; Mayda 2010; Castles, de Haas, and Miller 2013; Van Der Velde and Van Naerssen 2015; James and Mayblin 2016). Broadly, this new research agenda can be partitioned into theories that stress economic motivations and those that emphasize non-material factors.

The traditional understanding of migration preferences argues that economic considerations, specifically employment opportunities, dominate migrant decision-making. Seminal work by Sjaastad (1962) and Harris and Todaro (1970) argued that migration preferences were a direct function of wage differentials between origin and destination countries. Subsequent arguments have extended this perspective to explain migration’s observed inelasticity to changes in wage differentials by noting that employment gradients must also account for household-level considerations, labor market conditions within the origin country, and insurance and credit market failures (Stark and Bloom 1985; Taylor 1987; De Giorgi and Pellizzari 2009; Geis, Uebelmesser, and Werding 2013). The resulting synthesis, termed the “new economics of labor migration,” remains arguably the dominant approach for explaining variation in immigrant destinations within scholarly and public accounts (Abreu 2012).

Although the majority of economic accounts use wage differentials between sending and receiving states to predict migrant flows, a parallel literature has focused on the availability of welfare benefits within the destination state. In particular, Borjas (1999) assessed data on migratory flows within the United States to argue that immigrants seek destinations with generous welfare programs. In Europe, studies have similarly suggested that generous welfare provisions attract immigrants (Freeman 1986; Péridy 2006; Warin and Svaton 2008; De Giorgi and Pellizzari 2009). Yet the concept of “welfare migration” has also been highly contested. Around the same time that Borjas published his study of “welfare magnets” in the United States, several scholars leveraged more detailed samples and found no substantial evidence of welfare migration across state borders (Levine and Zimmerman 1999; Allard and Danziger 2000). In a European context, the majority of studies examining immigration flows have also documented a weak correlation between immigration and social expenditures (Brücker et al. 2002; Pedersen, Pytlikova, and Smith 2008). However, other accounts have argued that international migration remains highly responsive to differences in welfare benefits (Boeri 2010; Agersnap, Jensen, and Kleven 2019). As a result, the question of whether welfare benefits function as a meaningful pull factor in shaping international migration flows remains unsettled.

Another prominent strand of scholarship deemphasizes the role of economic considerations in shaping migrants’ destination choice in favor of non-material concerns, especially the role of social ties. This view was popularized by Massey’s seminal article on “cumulative causation” (1990). Focusing largely on Mexican immigration to the United States, Massey argued that migration strengthens and expands bilateral social networks and makes future migration more likely (see also Massey and España 1987). While acknowledging the role played by direct ties, more recent work has demonstrated how the presence of co-ethnic communities within destination countries may provide advantages in the absence of familial or social connections. For instance, established co-ethnic communities may provide new arrivals with information about how to find employment and services (Aguilera and Massey 2003; Munshi 2003; Drever and Hoffmeister 2008) and how to diminish the costs of migration and integration more broadly (Portes and Zhou 1993; Korinek, Entwisle, and Jampaklay 2005). These studies predict that migration flows will be self-sustaining, with migrants most likely to select destinations with established co-ethnic communities.

Most recently, and in the context of increased migration across the developing world, scholars have argued that international migrants prioritize moving to societies where they can be assured a baseline of liberal democratic rights and freedoms. For instance, Fitzgerald Leblang, and Teets (2014) identified the importance of democratic rights to political voice, labor-market access, social service provision, and legal protection in a large sample of bilateral migrant flows from 178 origin countries into 18 destinations between 1980 and 2006. This work builds upon earlier studies that documented migrants’ prioritization of political rights (Crawley 2010) and discrimination (Iosifides et al. 2007; Henry 2009). By incorporating political factors, these perspectives further nuance a debate previously defined by social or economic considerations.

Determining Relative Salience

While these perspectives each offer persuasive arguments in favor of the importance of a specific pull factor, empirical evidence on which factors predominate remains conflicted. This uncertainty can be traced to two distinct challenges researchers face when measuring destination preferences. First, scholars face a significant selection problem. Settled foreign residents within destination countries do not reflect the pool of future migrants to those states; rather, they correspond to the subset of individuals who previously had the capacity to complete a move (Carling 2002, 2004; De Haas 2010). Barriers to migration within receiving and sending states imply that this capacity is in turn a function of time-varying and contingent opportunity structures, including admissions regimes, migration costs, and informational flows (Guiraudon and Joppke 2003; Boucher and Gest 2018).¹ To reduce cognitive dissonance, migrants who have successfully navigated these constraints and completed the migration process may plausibly engage in post-hoc reasoning when queried about original destination preferences. If the migration decision-making process entails a “comparison of places,” scholars are thus selecting on the dependent variable by focusing on individuals who have already reached their destination (Carling and Schewel 2018).

Recently, several studies have sought to address this issue by surveying prospective migrants prior to departure. Although not all individuals surveyed within the sending state will successfully complete a move, measuring these initial preferences arguably provides a more valid indicator for the set of considerations that motivate the migration process. Following this logic, Gallup, Inc. currently solicits interest in emigration and respondents’ preferred destination across their international polls (Esipova, Ray, and Srinivasan 2011). A similar effort motivated the EUMAGINE project (Hemmerechts et al. 2014), which triangulated qualitative and quantitative evidence among prospective migrants in Morocco, Senegal, Turkey, and Ukraine to ascertain their relative appraisals of conditions in home states versus perceptions of life in the European Union.²

While these studies provide valuable insights, the survey methods used do not permit a clear determination of which country characteristics are assigned the highest priority. This uncertainty derives from a second methodological challenge confronting any assessment of migrants’ destination choices: the pull factors hypothesized to attract migrants are multidimensional and tend to co-vary (James and Mayblin 2016; Gest and Boucher 2020). A compounding issue is that not all possible configurations of pull factors are visible within observational data. As a result, it is difficult to separate correlated national characteristics within a standard regression framework. To address these problems, our design leverages conjoint analysis to isolate the independent draw of pull factors among prospective migrants.

Hypotheses and Design

Following Carling (2002), we begin from the perspective that the aspiration to migrate and the consideration of potential destinations occur simultaneously. The migration process begins when individuals first imagine their quality of life within alternate locations (Koikkalainen and Kyle 2016; Carling and Schewel 2018). These aspirations are informed by the political and social context in which they live, but conceptions of what constitutes a good life may be relatively fixed as migrants consider alternative destinations (Ibid.). Preferences regarding desirable destination characteristics (pull factors) are in turn linked to specific countries via informational flows that inform prospective migrants about how their preferences map onto objective conditions (e.g., Massey 1990; Aguilera and Massey 2003; Portes and Rumbaut 2006). Given that prospective migrants will be motivated to acquire information that maps onto their aspirations, the initial set of preferences is likely to affect subsequent destinations in a path-dependent manner.

While our design does not track prospective migrants through the various informational flows and contingencies that shape the final destination, we seek to contribute to an understanding of the initial stages of this process by evaluating the relative draw of prominent pull factors identified within the previous section.³ Specifically, we focus on the availability of welfare benefits upon arrival, the availability of employment, the presence of democratic rights, the size of the co-ethnic population, and the geographic distance from the origin country in shaping the relative hierarchy of desirable destination characteristics.⁴

Our expectation is that voluntary migrants seek to improve their relative status, broadly conceived. Accordingly, we expect that characteristics that proxy for opportunities and rights within the destination state will be more salient than factors that reduce the cost of migration. While characteristics such as spatial distance or co-ethnic stock are strongly predictive of cross-national flows (Mayda 2010; Fafchamps and Shilpi 2012; Fagiolo and Santoni 2016), we hypothesize that this correlation is largely driven by opportunity structures that facilitate entry into particular destinations rather than by migrants’ first-order preferences. As a result, we expect that respondents will assign limited priority to such pull factors within the hierarchy of desirable country characteristics.

Given that our survey design focuses on voluntary migrants, we hypothesize that economic conditions will have the strongest influence on destination preferences. Of the two factors that tap into economic status, we expect the availability of jobs to be more salient than the near-term availability of welfare benefits. Critiques of the welfare magnet hypothesis point to the fact that demand for labor tends to co-vary with accessible welfare regimes (e.g., Pedersen, Pytlikova, and Smith 2008, Thielemann 2008). We expect that when considered separately from labor markets, preferences for welfare access upon arrival will be minimal. Prospective migrants in our sample operate in contexts without generous transfers and may have limited expectations regarding welfare access within destination states. Within this context, welfare availability, we expect, is likely to be viewed through the lens of insurance rather than subsistence.

Finally, following Fitzgerald Leblang, and Teets (2014), we expect to observe consistent preferences for democratic states, independent of economic factors. However, we hypothesize that these preferences will vary as a function of the origin state’s political regime. While research suggests that autocratic states encourage the outmigration of democratic dissidents (Tsourapas 2015, Miller and Peters 2018), this process is gradual and unlikely to be visible within national surveys. Our hypothesis is that individuals living in contexts where they are afforded democratic rights will be reluctant to abandon such rights, leading to a higher preference for democratic destinations among prospective migrants who are citizens of fully or partially consolidated democracies.

Survey Design

To assess these hypotheses, we leverage a conjoint survey design (Hainmueller, Hopkins, and Yamamoto 2014). This approach has been widely used in applied research to model instances in which individuals decide between options that vary across a range of characteristics (Hainmueller, Hopkins, and Yamamoto 2014). In contrast to direct questioning, conjoint designs are less likely to generate social desirability bias (Horiuchi, Markovich, and Yamamoto 2018) and may map more closely to respondents’ actual behavior (Hainmueller, Hangartner, and Yamamoto 2015). Within the field of migration studies, conjoint designs have recently been leveraged to understand citizens’ multidimensional preferences regarding immigrant admissions (Hainmueller and Hopkins 2014; Clayton, Ferwerda, and Horiuchi 2018), asylum seekers and refugees (Bansak, Hainmueller, and Hangartner 2016; Adida, Lo, and Platas 2017), and immigration policies more generally (Wright et al. 2016). To our knowledge, however, the conjoint design has not yet been leveraged to understand migrants’ multidimensional destination preferences.

In our application, respondents are presented with two hypothetical country profiles, each of which is composed of a series of national characteristics. Although every respondent views the same set of characteristics, the attributes within each profile vary randomly across each conjoint comparison. After viewing the table, respondents were asked to select the country they preferred and to provide a rating on a seven-point scale reflecting their level of interest in moving to the hypothetical destination. Given that the components of each profile were randomly assigned, the degree to which each characteristic shaped respondents’ choices can be directly identified (Hainmueller, Hopkins, and Yamamoto 2014). Specifically, the results can be interpreted as the marginal effect of a characteristic on the probability of selecting a destination, averaged across all other combinations of national characteristics.⁵

The survey design, shown in Figure 1, varies five characteristics for each country, each of which randomly takes one of two values. We limit variation to binary options for two reasons. First, since the number of prospective migrants within each country was unknown a priori, a conservative design was necessary to maximize statistical power. Second, given the diversity across our cross-national sample, mapping responses to binary extremes (i.e., liberal democracy versus autocracy) increases the probability that the design will be evaluated consistently across national samples. More subtle gradations, such as the relative strength of democracy or the difference between a “high” and a “medium” unemployment rate, would likely be interpreted differently depending on respondents’ context-specific experiences.

Figure 1.

Destination Characteristics and Conjoint Design.

The national characteristics within the conjoint table parallel our hypotheses and indicate job availability, regime type, welfare accessibility,⁶ co-ethnic presence,⁷ and geographic distance from the origin country. While these factors often co-vary across destination states, it is important to note that they are not deterministic. For instance, Gulf states may provide immigrants with strong employment opportunities and, in some instances, welfare access in the absence of liberal democratic governance. Likewise, liberal democratic countries may offer access to labor markets while fencing off non-contributory public benefits from third country nationals. Similarly, variation in historic opportunity structures imply that co-ethnic communities are asymmetrically distributed. The choice set thus represents plausible real-world variation in abstract destination characteristics.

Although potential destinations vary along additional dimensions than those tested here, our design intentionally restricts the scope of such variation. Since the pool of real-world countries is limited, over-specificity would encourage respondents to link hypothetical countries with a single real-world example or, alternately, present an implausible combination of country characteristics. Either possibility would undermine the comparability of estimates across country pairs. In addition, constraining the number of characteristics limits the cognitive demands imposed on survey respondents, who may be unfamiliar with conjoint survey tasks.

Sample

Our approach seeks to measure the preferences of voluntary migrants before they depart, thus avoiding selection bias associated with surveying immigrants after arrival. Accordingly, in partnership with an international nongovernmental organization (NGO), in 2016, we collected data from samples of citizens in five sending regions: Jordan, Lebanon, Libya, the Palestinian Territories,⁸ and Tunisia.⁹ These regions represent a key set of migrant-sending states proximal to Europe. According to 2013 United Nations data (preceding the 2015 influx of immigrants and asylum-seekers from this region), there were six million expatriates in these countries’ worldwide diasporas, which was 17 percent of their current populations — significantly higher than the world average (3 percent). The selected sending regions also feature a variety of regime types (democracies in Lebanon and Tunisia), fragility profiles (ongoing conflict in Libya, Lebanon, and Palestine), and economic profiles (low per capita incomes in Jordan and Palestine), which enable the identification of heterogeneous effects across contexts.¹⁰

Working with national polling companies in each country, we implemented a standardized questionnaire translated into the local dialect. Surveys were translated by dual-language nationals accustomed to working within each country and pre-tested to ensure comprehension. These questions were appended to public opinion surveys, unrelated to migration, that were designed and implemented by the NGO. Sample size varied across countries, according to prior arrangements between the polling firm and the NGO. In total, we surveyed 1,000 people in Jordan (April 2016), 3,111 people in Libya (May 2016), 1,225 people in Tunisia (May 2016), 1,200 people in the Palestinian Territories (July 2016), and 2,055 people in Lebanon (September 2016) — a total of 8,591 respondents across the region. Further information on the survey samples is available in the Supplemental Appendix.¹¹

In Tunisia and the Palestinian Territories, surveys were conducted using tablets. In Jordan and Lebanon, surveys were fielded in paper format. To conduct the randomization in these latter cases, we developed a web script that randomly generated individual questionnaires for printing. In Libya, due to ongoing conflict, the survey was administered over the phone. While the phone questionnaire’s content was identical to in-person fieldings, to ensure comprehension, we instructed enumerators to summarize the list of similarities and differences between the two countries after discussing each country’s characteristics.

Enumerators were nationals from each respective country, employed by polling companies but trained by the research team, either in-person or via conference call, to implement the conjoint design. When administering hypothetical questions, enumerators were instructed to offer only the information provided and nothing further. Enumerators were instructed to clarify the differences between the country choices by presenting subjects with a table that made discrepancies visually apparent. If subjects felt equally displeased or equally pleased by both country options or did not want to respond, enumerators urged them to select the better of the two and to reserve their impressions for the following question, which rated their interest in the countries as described. They were also instructed to ensure comprehension by re-administering the question if respondents’ binary choice did not correspond to their country ratings. Response rates to the conjoint question among prospective migrants varied across contexts in a manner unrelated to fielding method (Supplemental Table A4). However, we observed lower rates of response among older and female respondents (Supplemental Table A5).

The survey asked three questions to ascertain respondents’ migration intentions. First, respondents were asked whether they were considering migrating to another state within the next 12 months. This measure was derived from Gallup, Inc. and has been previously validated on cross-national samples (e.g., Migali and Scipioni 2018). Second, respondents were asked the name of the state to which they were considering migrating (intended destination). Finally, as a behavioral measure, respondents were asked whether they had made concrete preparations for a move, such as applying for a residency permit or visa or purchasing a ticket (see Supplemental Figure A1 for question wording).

Within the pooled sample, 2,275 respondents indicated that they were considering migrating to another state within the next 12 months (26.48%).¹² As Supplemental Table A2 demonstrates, on average, self-identified prospective migrants tended to be younger, have higher educational attainment, and have similar incomes to respondents who did not express an interest in leaving. About 88.7 percent of prospective migrants were able to name a specific country to which they were considering migrating (see Supplemental Table A3), and 23.7 percent indicated that they had already taken concrete steps to prepare for a move abroad. For the main analysis, we focus on all respondents who indicated that they were considering moving in the next 12 months. In the robustness checks, we limit the sample to those who indicated that they had made concrete plans, with similar results. Respondents were asked to complete the conjoint task multiple times; in total, self-identified prospective migrants evaluated 7,986 hypothetical destination country profiles (see Supplemental Table A1).¹³

Cross-national Results

Focusing on prospective migrants within the cross-national sample, Figure 2 displays the marginal effect of each national characteristic on the probability of selection. As seen in Supplemental Figure A2, the forced response and the seven-point rating question provide similar results. However, given that the average rating assigned to destinations differs across origin countries (from 3.21 in Lebanon to 3.66 in Tunisia), in the main text, we present results using the forced response to ensure comparability across country subsamples.

Figure 2.

Marginal Effect of Destination Characteristics.

Our findings suggest that preferences were not dominated by a single pull factor. However, the results nevertheless suggest a clear hierarchy in migrants’ initial preferences. Respondents within the sample placed approximately equal priority on the regime type and the availability of work in the destination country. Holding other characteristics constant, prospective migrants were most likely to select liberal democratic countries (26.2%) and those with high job availability (25.3%). The third most important factor was the availability of welfare: respondents were 17.4 percent more likely to select a country with welfare benefits available upon arrival. Contrary to expectations, although countries within the sample differed in terms of regime type, economic development, and security environment, we observe a similar rank order of preferences across each national case.¹⁴ We also observe consistent preferences when focusing on individuals who had already taken concrete steps toward a move (Supplemental Figure A3). Together, these patterns suggest a broad consensus on desirable country traits.¹⁵

Although these results are largely consistent with our initial hypotheses, we note two deviations. First, although our expectation was that concerns for employment would predominate, we did not expect to observe a substantial independent effect for welfare considerations, given the low level of state provision within many sending states. In contrast, our results suggest that a significant subset of respondents within the sample did, in fact, optimize on welfare availability, as well as on labor markets, when considering potential destinations. Second, although we hypothesized that voluntary migrants would have an independent preference for democratic rights, the magnitude of this preference is notable and provides novel evidence in support of the nascent literature on rights-based migration (e.g., Fitzgerald Leblang and Teets 2014). Given that a subset of prospective migrants within our sample was primarily motivated by democratic rights rather than by economic factors, these findings also indicate that democratization within the developing world could hold the potential to substantially alter prevailing migration patterns.

Finally, the weak preferences with respect to geographic distance and co-ethnic stock within our sample are notable.¹⁶ Although this finding challenges the persistent correlation scholars have documented between these factors and bilateral migration flows (Fagiolo and Santoni 2016), it is consistent with a model of migration decision-making in which these characteristics primarily influence flows via opportunity structures rather than preferences. In other words, while geographic distance and the presence of co-ethnics may not independently shape migrants’ first-order prioritization of potential destinations, they nevertheless facilitate migration to specific destinations within the context of high barriers to mobility. Distinguishing between preferences and facilitating factors is theoretically important: if admission barriers were reduced, our results suggest that the prospective migrants in our sample would not necessarily prioritize states selected by co-ethnics in prior waves.

Tradeoffs between Democracy, Employment, and Welfare Considerations

In addition to providing information on each characteristic’s rank order, the conjoint design permits an evaluation of how characteristics interact. Following Hainmueller Hopkins, and Yamamoto (2014), we expand the specification to include two-way interactions between national characteristics. In the analysis that follows, we focus on a subset of these interactions and examine how prospective migrants navigated tradeoffs between different national characteristics as a function of their respective sociodemographic and political context.

Given the observed importance of democratic rights within the pooled sample, we first examined the degree to which respondents were willing to trade democratic rights for a robust economy, in the form of either job availability or welfare benefits. The left-hand panel of Figure 3 plots the predicted difference in selection probability between a democratic state with a poor economy and an autocratic state with a strong economy across relevant cross-national and demographic subsets.¹⁷ Positive values indicate that the average respondent was more likely to select the democratic country when presented with this tradeoff, while negative values indicate that the average respondent was more likely to select the autocracy.

Figure 3.

Tradeoffs across National Samples and Demographic Subgroups.

The results suggest that willingness to favor democratic rights over labor market opportunities varied across national samples and socio-demographic groups. Respondents from Libya and the Palestinian Territories, both of which were characterized by ongoing conflict during the survey period, were the least willing to trade democratic rights for employment, while respondents within an autocracy (Jordan) were the most willing to prioritize labor market factors over such rights. Respondents within weak democracies (Tunisia and Lebanon) were on average more likely to select employment opportunities over democratic rights, although this tendency was statistically insignificant at conventional levels. This variation runs counter to our initial hypothesis that preferences for democratic rights would be highest among respondents habituated to democratic governance. Rather, these findings suggest a process in which the relative demand for democratic rights versus labor market considerations varies as a function of perceived insecurity.¹⁸ Our results also suggest that the willingness to engage in this tradeoff varied with respect to demographic factors: we find that younger and male respondents were relatively more likely to prioritize labor considerations over democratic rights. In contrast, differentiation with respect to education or income was not statistically significant.

The middle panel of Figure 3 presents the results of a similar tradeoff between democratic governance and welfare benefits. This tradeoff is more decisive, with the majority of national and demographic subgroups preferring a liberal democracy with inaccessible benefits to an autocracy with accessible welfare benefits. The results also suggest that this tradeoff maps more closely onto socio-economic status. Consistent with economic self-interest, we find that individuals with low incomes, and to a lesser extent lower educational levels, were less likely to prefer democracy over welfare benefits relative to respondents with higher socio-economic status (SES), although on average low SES respondents remained more likely to choose the democratic option.

To shed further light on the tradeoff between employment and welfare, the right-hand panel of Figure 3 holds democratic governance constant and evaluates the welfare magnet hypothesis directly. Positive values indicate that respondents were more likely to select a state with a robust labor market but no immediate access to welfare benefits, while negative values indicate respondents were less likely to select a state with a weaker labor market and immediate benefit access. The results suggest that the majority of respondents within the sample preferred employment opportunities to welfare. However, within conflict regions such as Libya and the Palestinian Territories, this tendency is no longer statistically significant, suggesting a more contentious decision-making process. Similarly, we find that differences with respect to gender and education remain statistically significant, with males and individuals with higher education levels less likely to optimize on welfare benefits.

These results suggest that welfare-seeking preferences may be correlated with a tendency toward risk aversion among individuals living in unstable situations. We further explore this hypothesis by examining how the marginal effect of the welfare attribute varied conditional on the presence of democracy and a strong economy. If the demand for welfare was a first-order preference, we should expect to observe limited interactions with other national characteristics. However, if prospective migrants viewed welfare benefits primarily as a form of insurance, we should expect demand for welfare to be higher in instances in which perceived levels of insecurity were higher, such as within autocracies or states with weak economies. Supplemental Figure A5 demonstrates that this latter pattern holds: destination attributes that proxied for insecurity increased the relative demand for welfare. Although the statistical significance of this relationship varied across subgroups, the results suggest that relative demand for welfare was conditional on employment prospects for individuals living in potentially precarious situations, including youth, respondents with below-median incomes, and those with lower educations. Thus, while our findings suggest that prospective migrants do indeed consider welfare accessibility when evaluating destinations, our results are inconsistent with the logic advanced by Borjas (1999), who argues that welfare-seeking behavior emerges due to migrants’ tendency to rationally maximize their income. Rather, our results suggest that demand for welfare is largely a function of risk aversion and that this demand is relatively weaker among individuals considering migrating to high-income democracies.

Information and Targeting

Although the response to specific tradeoffs varied across subgroups, the results suggest that migrants within our sample had broadly consistent preferences regarding destination state characteristics. But to what extent are these preferences linked with behavior? While our design does not track individuals across subsequent migration decisions, in this section, we examine the degree to which latent preferences were linked to potential outcomes.

First, we leverage a survey question that asked respondents to indicate the specific country to which they were most likely to migrate. Using these responses, we examine how conjoint results differed across respondents who indicated destinations with varying characteristics. For the primary validation exercise, we focus on an objective measure — distance from the origin country — under the assumption that individuals would be most likely to have accurate perceptions regarding this item. Our expectation is that respondents who selected a destination close to their origin country will be more likely to select hypothetical national profiles that were geographically proximate, relative to other respondents. In Supplemental Table A6, we demonstrate that this relationship is as expected: respondents who indicated a preferred destination in proximity to their home state (<2,000 km) were more likely to select a hypothetical conjoint profile close to the origin state. This difference is statistically significant and substantively large relative to the baseline effect for distance.¹⁹

In Supplemental Table A6, we also demonstrate that a similar finding holds when assessing how the Polity IV Score regime-type indicator (Marshall, Gurr, and Jaggers 2017) influences selection rates.²⁰ Respondents who selected intended destinations characterized by liberal democratic governance (Polity score >=8) were more likely to select a democratic country within the conjoint. However, this relationship is conditional: it does not hold among respondents with a primary education or less, or among respondents residing in non-democratic regimes. This divergence suggests that while the conjoint responses are linked to specific destination intentions, local context and the availability of information nevertheless mediate the relationship between latent preferences and specific destinations.

Finally, to directly evaluate how levels of information vary across the sample, we fielded an additional survey question in the Lebanon sample to measure objective knowledge.²¹ Respondents in Lebanon were presented with a list of seven countries and were asked to select the two countries that provided the most generous and least generous welfare benefits for immigrants, respectively.²² Each of the seven countries presented was a developed state in Europe or North America. Of the seven, at the time of fielding, three imposed initial restrictions on welfare access for third-country nationals (Austria, United States, and the United Kingdom), one imposed few restrictions but had low per-capita spending on social assistance (Italy), and three had high per-capita social assistance expenditures and limited restrictions on immigrant welfare usage (Belgium, Norway, and Sweden).²³

Table 1.

Knowledge of Welfare System Generosity towards Immigrants.

	% Identifying As “Most Generous”		% Identifying as “Least Generous”
	Prospective Migrants	Non-Migrants	Prospective Migrants	Non-Migrants
Sweden	29.2	31.0	5.1	6.1
USA	13.5	11.4	11.5	13.0
Norway	13.4	12.7	9.2	10.5
Italy	9.5	10.2	14.4	15.3
Austria	9.4	10.3	15.3	17.3
Belgium	7.1	8.4	11.1	12.4
UK	6.0	6.3	11.8	12.0
Don’t Know	12.0	9.7	21.6	13.4
K-S p-value		0.68		0.99

Note. Percentages correspond to the proportion of respondents who selected the country as of the two most generous, or two least generous, respectively. p-values refer to a Komogorov-Smirnov test for the equality of distributions, with ‘Don’t Know’ responses omitted.

The results, shown in Table 1, demonstrate that while prospective migrants were sensitive to the generosity of welfare programs, few individuals within the sample possessed accurate knowledge of welfare accessibility across destination states.²⁴ While a sizable minority of respondents were able to identify Sweden as a country with a generous safety net (29.2%), recognition of other policy regimes was limited. In total, 49.7 percent of prospective migrants selected one of the three countries with the most generous benefit regimes — a rate similar to a coin flip. Indeed, the United States, which imposes a five-year ban on initial welfare access (Office of the Assistant Secretary for Planning and Evaluation 2009), was identified by prospective migrants as the second most generous state. This misidentification may suggest that economic strength and history of immigrant admissions are conflated with the availability of benefits.

To investigate how knowledge concerning welfare states’ relative generosity varied as a function of socio-demographic characteristics, we fit a linear probability model regressing an indicator for an answer correctly identifying a generous welfare state on a set of individual-level covariates. The results, reported in the first column of Supplemental Table A7, suggest that the primary factor shaping accurate knowledge was education. Compared to those with a primary education level, respondents with a secondary education were 15.4 percent more likely to answer correctly, while those with a tertiary education were 19.9 percent more likely to answer correctly. We also find that gender and ethnic identity predict accurate responses within the Lebanese sample. For instance, Druze respondents were 17.6 percent less likely to answer correctly than Christian respondents. Given that the Druze diaspora is small relative to other ethnic groups, this finding is consistent with the argument that while co-ethnic networks may not independently shape preferences, they may provide information that allows potential migrants to accurately target destinations (Radu 2008). Finally, the right-hand columns of Supplemental Table A7 suggest that while prospective migrants were no more likely than non-migrants to identify favorable welfare regimes in aggregate, there is an interaction between planned emigration and educational levels, with educated migrants possessing the most knowledge concerning destination country characteristics. However, we observe no such interaction with employment status.

These results emphasize the strengths and limitations of focusing on latent preferences to illuminate the migration process. Although the preferences identified in the conjoint analysis are correlated with intended destinations and reveal the ways in which characteristics of these destinations interact, accurate knowledge plays a critical role in mediating the translation of these preferences into final destinations (see also Shrestha 2017). Our results suggest that while educated migrants are likely to overcome these informational barriers, migrants with lower levels of human capital may not. This distinction does not imply that initial preferences are immaterial, given that individuals may still act on inaccurate information. However, it does suggest the opportunity for future research to identify how misperceptions concerning desirable destination country characteristics are distributed across sending states.

Discussion

This article provides new evidence on the initial destination preferences of prospective migrants in one of the world’s most prominent regions of emigration. While the preferences we isolate are not the only factors influencing migration flows between states, they are a critical and largely unobserved input into the migration process. Without a robust understanding of migrants’ initial preferences, it is difficult to distinguish between contingent and motivating factors. In turn, this uncertainty presents challenges for policymakers and scholars seeking to understand how migration patterns shift as conditions within receiving states change.

Rather than deriving these preferences from observed flows, we measure them by fielding surveys across five sending states and querying prospective migrants’ intentions and destination preferences prior to departure. This approach avoids the selection bias and post-hoc reasoning associated with soliciting immigrants’ original intentions after arrival. Although relatively few studies to date have examined preferences prior to departure, we also depart from extant work by implementing a conjoint survey to measure multi-dimensional preferences. In contrast to open-ended questions, this approach enables us to separate the effect of co-varying national characteristics, to rank order destination preferences, and to examine how these preferences interact across demographic subgroups.

Although our results confirm the multidimensionality of prospective migrants’ preferences, the hierarchy of preferred destination characteristics we document has implications for the literature. First, consistent with theoretical perspectives that stress economic motivations for migration (Sjaastad 1962; Harris and Todaro 1970; Stark and Bloom 1985), our findings suggest that employment availability is a primary factor shaping the relative desirability of hypothetical destinations within our sample. However, our results also indicate a salient concern for liberal democratic rights. This independent preference for democracy provides novel empirical support to the expanding literature on rights-based migration (e.g., Crawley 2010; Fitzgerald Leblang, and Teets 2014) and is inconsistent with neo-liberal accounts that assume that elevated migration flows to democracies reflect the relative economic strength of these regimes.²⁵ Thus, this article’s conclusions about the relative pull between democratic and economic factors is likely to become increasingly relevant in a global context where autocracies exhibit strong economic growth (Collier and Hoeffler 2009; Chandra and Rudra 2015) and developed democracies become less welcoming to immigrants.

Second, in underscoring the salience of democratic and economic factors in shaping prospective migrants’ decision-making, the results presented here suggest that these considerations significantly outstrip concern for the number of co-nationals currently residing in the destination country and the destination’s proximity to the origin country. While the latter finding is less surprising in an era of increased cross-national and cross-continental mobility, the former relationship challenges accounts that suggest that the presence of an established community of co-ethnics shapes first-order migration preferences. While the relationship between co-ethnic stock and bilateral flows remains undisputed at the aggregate level, evidence within our sample suggests that these patterns may not be predominantly driven by initial preferences for co-ethnic communities. Rather, this persistent relationship may be a function of opportunity structures, such as admissions or enforcement policies that facilitate the entry of people with specific linguistic skills, colonial affinities, or family ties (Guiraudon and Joppke 2003; Ireland 2006; Boucher and Gest 2018). This distinction has important implications for policymakers and scholars seeking to understand how migration flows might change under different admission regimes.

Third, beyond establishing the rank order of desirable destination characteristics within our sample, the conjoint design enables us to clarify how migrants navigate potential tradeoffs between democratic rights and economic factors. Specifically, while our results indicate a consistent preference for democratic rights over welfare benefits, they also suggest that the tradeoff between employment and democracy is highly contested. Respondents from unstable regimes (Libya and the Palestinian Territories) were likely to prioritize democratic freedoms over employment prospects, while respondents within autocracies (Jordan) and weak democracies (Tunisia and Lebanon) were more likely, on average, to favor autocratic destinations with robust employment prospects. In addition, older and female respondents were more likely to prioritize democratic governance over economic factors. These findings nuance our understandings of democratic proclivities in the Middle East and North Africa and provide attitudinal evidence that can be leveraged to explain observed variation in flows between democratic and autocratic states.

Finally, although our findings demonstrate that welfare availability is an important consideration for prospective migrants, the data also suggest that the fiscal impact of this tendency may be more limited than typically assumed by opponents of immigration. Across all five national samples, our results indicate that voluntary migrants tended to be drawn from the class of individuals who were among the least likely to consume welfare benefits: those who were disproportionately young and possessed higher educational credentials than their peers who did not wish to migrate (Supplemental Table A2). Moreover, in exploring how demand for welfare accessibility varied as a function of regime type and economic considerations, we find that observed patterns are consistent with a logic of risk aversion rather than income maximization. Perhaps most critically, evidence from a subgroup of respondents suggests that prospective migrants lack actionable information that would allow them to optimize destinations according to welfare generosity.

We expect that informational constraints play a similar role in mediating the impact of other preferences (van Dalen and Henkens 2012; Shreshtha 2017). In addition to implementing conjoint designs to evaluate whether the initial preferences we document in our samples from the Middle East and North Africa generalize across sending regions and other categories of migrants (Holland, Peters, and Sanchez 2018), future work could fruitfully explore how objective knowledge of desirable country characteristics varies across national contexts. Similarly, panel surveys of prospective migrants would shed additional light on how preferences and information evolve over time. While a great deal of work remains to understand the full migration process, this article offers a new lens on how the process begins.

Supplemental Material

Supplemental Material, supplementary_materials - Pull Factors and Migration Preferences: Evidence from the Middle East and North Africa

Supplemental Material, supplementary_materials for Pull Factors and Migration Preferences: Evidence from the Middle East and North Africa by Jeremy Ferwerda and Justin Gest in International Migration Review

Supplemental Material

Supplemental Material, supplementary_materials - Pull Factors and Migration Preferences: Evidence from the Middle East and North Africa

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Jeremy Ferwerda

Supplemental Material

Supplemental material for this article is available online.

Notes

References

Abreu

, 2012. “The New Economics of Labor Migration: Beware of Neoclassicals Bearing Gifts.” Forum for Social Economics 41(1): 46–67. Routledge.

Adida

C. L

Platas

. 2017. “Engendering Empathy, Begetting Backlash: American Attitudes towards Syrian Refugees.” Working paper.

Agersnap

Jensen

A. S.

Kleven

. 2019. “The Welfare Magnet Hypothesis: Evidence From an Immigrant Welfare Scheme in Denmark.” Report No. w26454, Cambrige, MA: National Bureau of Economic Research.

Aguilera

M. B.

Massey

D. S.

. 2003. “Social Capital and the Wages of Mexican Migrants: New Hypotheses and Tests.” Social Forces 82(2): 671–701.

Allard

S. W.

Danziger

. 2000. “Welfare Magnets: Myth or Reality?” Journal of Politics 62(2): 350–368.

Bansak

Hainmueller

Hangartner

. 2016. “How Economic, Humanitarian, and Religious Concerns Shape European Attitudes toward Asylum Seekers.” Science, 354(6309): 217–222.

Boeri

, 2010. “Immigration to the Land of Redistribution.” Economica 77(308): 651–687.

Borjas

G. J.

1999. “Immigration and Welfare Magnets.” Journal of Labor Economics 17(4): 607–637.

Boucher

A. K.

Gest

. 2018. Crossroads: Comparative Immigration Regimes in a World of Demographic Change. New York: Cambridge University Press.

10.

Brücker

Epstein

G. S.

McCormick

Saint-Paul

Venturini

Zimermann

. 2002. “Managing Migration in the European Welfare State.” In Immigration Policy and the Welfare System, edited by Boeri

Hanson

McCormick

. Oxford: Oxford University Press, 1–167.

11.

Carling

2002. “Migration in the Age of Involuntary Immobility: Theoretical Reflections and Cape Verdean Experiences.” Journal of Ethnic and Migration Studies 28(1): 5–42.

12.

Carling

2004. “Emigration, Return and Development in Cape Verde: The Impact of Closing Borders.” Population, Space and Place 10(2): 113–132.

13.

Carling

Schewel

. (2018). “Revisiting Aspiration and Ability in International Migration.” Journal of Ethnic and Migration Studies, 44(6).

14.

Castles

de Haas

Miller

. 2013. The Age of Migration: International Population Movements in the Modern World, 5th ed. New York: Guilford Press.

15.

Chandra

Rudra

. 2015. “Reassessing the Links between Regime Type and Economic Performance: Why Some Authoritarian Regimes Show Stable Growth and Others Do Not.” British Journal of Political Science, 45(2): 253–285.

16.

Clayton

Ferwerda

Horiuchi

. 2018. “Exposure to Immigration and Admission Preferences: Evidence from France.” Political Behavior, 1–26.

17.

Collier

Hoeffler

. 2009. “Testing the Neocon Agenda: Democracy in Resource-rich Societies.” European Economic Review 53(3): 293–308.

18.

Crawley

2010. Chance or Choice? Understanding Why Asylum Seekers Come to the UK. London: Refugee Council.

19.

Crawley

Hagen-Zanker

. (2019). “Deciding Where to go: Policies, People and Perceptions Shaping Destination Preferences.” International Migration, 57: 20–35.

20.

Creighton

M. J.

2013. “The Role of Aspirations in Domestic and International Migration.” The Social Science Journal 50(1): 79–88

21.

De Giorgi

Pellizzari

. 2009. “Welfare Migration in Europe.” Labour Economics 16(4): 353–363.

22.

De Haas

, 2010. “Migration and Development: A Theoretical Perspective.” International Migration Review 44(1): 227–264.

23.

Drever

A. I.

Hoffmeister

. 2008. “Immigrants and Social Networks in a Job-Scarce Environment: The Case of Germany.” International Migration Review 42(2): 425–448.

24.

Esipova

Ray

Srinivasan

. 2011. “The World’s Potential Migrants: Who They Are, Where They Want to Go, and Why It Matters.” Gallup, Inc. https://www.imi.ox.ac.uk/files/news/gallup_whitepaper_migration-1.pdf

25.

European Commission Report. 2014. “Migrant Access to Social Security and Healthcare: Policies and Practice.” Brussels: European Commission Report.

26.

Fafchamps

Shilpi

. 2012. “Determinants of the Choice of Migrant Destination.” Oxford Bulletin of Economics and Statistics 75(3): 388–409.

27.

Fagiolo

Santoni

. 2016. “Revisiting the Role of Migrant Social Networks as Determinants of International Migration Flows.” Applied Economics Letters 23(3): 188–193.

28.

Fitzgerald

Leblang

Teets

J. C.

. 2014. “Defying the Law of Gravity: The Political Economy of International Migration.” World Politics 66(3): 406–445.

29.

Freeman

G. P.

1986. “Migration and the Political Economy of the Welfare State.” Annals of the American Academy of Political and Social Science 495: 51–63.

30.

Geis

Uebelmesser

Werding

. 2013. “How Do Migrants Choose their Destination Country? An Analysis of Institutional Determinants.” Review of International Economics 21(5): 825–840.

31.

Gest

Boucher

. 2020, forthcoming. “A Segmented Theory of Immigration Regimes.” Polity.

32.

Goodman

S. W.

2014. Immigration and Membership Politics in Western Europe. New York: Cambridge University Press.

33.

Goodman

S. W.

Baldi

G. C.

. 2015. “Migrants into Members: Social Rights Civic Requirements, and Citizenship in Western Europe.” West European Politics 38(6): 1152–1173.

34.

Guiraudon

Joppke

. 2003. “Controlling a New Migration World”. In Controlling a New Migration World, 13–40. Routledge.

35.

Hainmueller

Hopkins

D. J.

Yamamoto

. 2014. “Causal Inference in Conjoint Analysis: Understanding Multidimensional Choices via Stated Preference Experiments.” Political Analysis 22: 1–30.

36.

Hainmueller

Hangartner

Yamamoto

. 2015. “Validating vignette and conjoint survey experiments against real-world behavior.” Proceedings of the National Academy of Sciences, 112(8), 2395–2400.

37.

Hainmueller

Hopkins

. 2014. “Public Attitudes Toward Immigration.” Annual Review of Political Science, 17 (1): 225–249.

38.

Harris

J. R.

Todaro

M. P.

. 1970. “Migration, Unemployment and Development: A Two-Sector Analysis.” American Economic Review 60(1): 126–142.

39.

Hemmerechts

De Clerck

H. M.

Willems

Timmerman

. 2014. “Project Paper 14: Eumagine final report with policy considerations”. EUMAGINE. http://www.eumagine.org/outputs/Project%20paper%2014%20-%20Final%20report%20with%20policy%20considerations.pdf.

40.

Henry

2009. “Does Racism Affect a Migrant’s Choice of Destination? A Case Study of African Americans, 1995-2000.” Paper presented at 10th European Summer Symposium in Labour Economics, IZA, Buch, Germany, October 8–12, 2008.

41.

Holland

Peters

Sanchez

. 2018. “Migrants’ Destination Choices.” Working Paper.

42.

Horiuchi

Markovich

Z. D.

Yamamoto

. 2018. “Can Conjoint Analysis Mitigate Social Desirability Bias?.” Working Paper.

43.

Iosifides

Lavrentiadou

Petracou

Kontis

. 2007. “Forms of Social Capital and the Incorporation of Albanian Immigrants in Greece.” Journal of Ethnic and Migration Studies 33(8): 1343–1361.

44.

Ireland

, 2006. “Institutions, Political Opportunity Structures, and the Shaping of Migration Policies”. In Dialogues on Migration Policy, edited by Giugni

Passy

. Oxford: Lexington Books 137.

45.

Ivarsflaten

2008. “What Unites Right-Wing Populists in Western Europe? : Re-Examining Grievance Mobilization Models in Seven Successful Cases.” Comparative Political Studies, 41(1): 3–23.

46.

James

Mayblin

. 2016. “Factors Influencing Asylum Destination Choice: A Review of the Evidence.” University of Sheffield Working Papers No. 04/16.1.

47.

King

. 2012. “Theories and Typologies of Migration: An Overview and a Primer.” in Willy Brandt Series of Working Papers in International Migration and Ethnic Relations 3/12, Malmo.

48.

Koikkalainen

Kyle

. 2016. “Imagining mobility: the prospective cognition question in migration research.” Journal of Ethnic and Migration Studies 42(5): 759–776.

49.

Koning

2011. “The Real and Perceived Economics of Immigration: Welfare Chauvinism and Immigrants’ Use of Government Transfers in Twelve Countries.” Working Paper.

50.

Korinek

Entwisle

Jampaklay

. 2005. “Through Thick and Thin: Layers of Social Ties and Urban Settlement among Thai Migrants.” American Sociological Review 70(5): 779–800.

51.

Levine

P. B.

Zimmerman

D. J.

. 1999. “An Empirical Analysis of the Welfare Magnet Debate Using the NLSY.” Journal of Population Economics 12(3): 391–409.

52.

Marshall

M. G.

Gurr

T. R.

Jaggers

. 2017. “Polity IV Project: Political Regime Characteristics and Transitions, 1800–2016.” Virgina: Center for Systemic Peace.

53.

Massey

D. S.

1990. “Social Structure, Household Strategies, and the Cumulative Causation of Migration.” Population Index 56(1): 3–26.

54.

Massey

D. S.

Durand

. 2017. “The Mexican Migration Project”. Princeton University. http://mmp.opr.princeton.edu/.

55.

Massey

D. S.

España

F. G.

. 1987. “The Social Process of International Migration.” Science 237(4816): 733–738.

56.

Mayda

. 2010. “International Migration: A Panel Data Analysis of the Determinants of Bilateral Flows,” Journal of Population Economics 23(4): 1249–1274

57.

Migali

Scipioni

. 2018. A Gobal Analysis of Intentions to Migrate. European Commission.

58.

Miller

M. K.

Peters

M. E.

. 2018. “Restraining the huddled masses: Migration policy and autocratic survival.” British Journal of Political Science 50(2): 1–31.

59.

Munshi

2003. “Networks in the Modern Economy: Mexican Migrants in the U.S. Labor Market.” Quarterly Journal of Economics 118(2): 549–599.

60.

Office of the Assistant Secretary for Planning and Evaluation (ASPE). 2009, February 25. “Summary Of Immigrant Eligibility Restrictions Under Current Law,” U.S. Department of Health and Human Services.

61.

Pedersen

P. J.

Pytlikova

Smith

. 2008. “Selection and Network Effects–Migration Flows into OECD Countries 1990-2000.” European Economic Review 52(7): 1160–1186.

62.

Péridy

2006. “Welfare Magnets, Border Effects or Policy Regulations: What Determinants Drive Migration Flows into the EU?” Global Economy Journal 6(4): 1–35.

63.

Portes

Zhou

. 1993. “The New Second Generation: Segmented Assimilation and Its Variants.” The Annals of the American Academy of Political and Social Science 530(1): 74–96.

64.

Radu

2008. “Social Interactions in Economic Models of Migration: A Review and Appraisal.” Journal of Ethnic and Migration Studies 34(4): 531–548.

65.

Rooduijn

de Lange

S. L.

van der Brug

. 2016. “A Populist Zeitgeist? Programmatic contagion by populist parties in Western Europe.” Party Politics 20(4): 563–575.

66.

Schewel

2019. “Understanding Immobility: Moving Beyond the Mobility Bias in Migration Studies.” International Migration Review 54(2), 328–355.

67.

Sainsbury

2012. Welfare States and Immigrant Rights: The Politics of Inclusion and Exclusion. Oxford: Oxford University Press.

68.

Shrestha

2017. “Get Rich or Die Tryin’: Perceived Earnings, Perceived Mortality Rate and the Value of a Statistical Life of Potential Work-migrants from Nepal.” World Bank Policy Research Working Paper No. 7945.

69.

Sjaastad

L. A.

1962. “The Costs and Returns on Human Migration.” Journal of Political Economy 70(5): 80–93.

70.

Spörlein

, 2015. “Destination Choices of Recent Pan–American Migrants: Opportunities, Costs, and Migrant Selectivity.” International Migration Review 49(2): 523–552.

71.

Stark

Bloom

. 1985. “The New Economics of Labor Migration.” American Economic Review 75(2): 173–178.

72.

Taylor

J. E.

1987. “Undocumented Mexico–U.S. migration and the returns to households in rural Mexico.” American Journal of Agricultural Economics 69(3): 626–638.

73.

Thielemann

2008. “The Effectiveness of Governments’ Attempts to Control Unwanted Migration.” In Immigration and the Transformation of Europe, edited by Parsons

C. A.

Smeeding

T. M.

. Cambridge: Cambridge University Press, 442–472.

74.

Tjaden

Auer

Laczko

. 2018. “Linking Migration Intentions with Flows: Evidence and Potential Use” International Migration. 57(1): 36–57.

75.

Tsourapas

. 2015. “Why Do States Develop Multi-tier Emigrant Policies? Evidence from Egypt.” Journal of Ethnic and Migration Studies. 41(13): 2192–2214.

76.

United Nations, Department of Economic and Social Affairs, population division. 2016. “International migration report 2015.” United Nations, Department of Economic and Social Affairs.

77.

Van Dalen

H. P.

Henkens

. 2012. “Explaining Low International Labour Mobility: The Role of Networks, Personality, and Perceived Labour Market Opportunities.” Population, Space and Place 18(1): 31–44.

78.

Van Der Velde

Van Naerssen

. 2015. Mobility and Migration Choices: Threshold to Crossing Borders. London: Routledge.

79.

Warin

Svaton

. 2008. “European Migration: Welfare Migration or Economic Migration?” Global Economy Journal 8(3): 1–30.

80.

Wright

Levy

Citrin

. 2016. “Public Attitudes Toward Immigration Policy Across the Legal/Illegal Divide: The Role of Categorical and Attribute-Based Decision-Making.” Polit Behav 38: 229–253.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

1.41 MB

0.03 MB