Translating migration theory into empirical propositions

Jør gen Carl ing, Mat hias Czaika and Mart a Bivand Erdal Transl at ing migrat ion t heory int o empirical proposit ions Del iverabl e 1. 2 QuantMig has received funding from the European Union’ s Horizon 2020 research and innovation programme under grant agreement No. 870299. i History of changes Version Date Changes 1.0 27 October 2020 Issued for Consortium Review 1.1 30 October 2020 First version submitted as official deliverable to the EC Suggested citation Carling J, M Czaika, and M.B. Erdal (2020) Translating migration theory into empirical propositions. QuantMig Project Deliverable 1.2. Dissemination level PU Public Acknowledgments This document reflects the authors’ view and the Research Executive Agency of the European Commission are not responsible for any use that may be made of the information it contains. We are very grateful for useful comments and advice from Jakub Bijak, Michaela Potancokova, and Jackie Wahba. Cover photo: iStockphoto.com/Guenter Guni Translating Migration Theory into Empirical Propositions 1 Table of Contents Acknowledgments.................................................................................................................................. i Table of Contents ................................................................................................................................... 1 1. Introduction ......................................................................................................................................... 2 2. What is migration theory? ................................................................................................................. 4 3. Propositions ......................................................................................................................................... 7 3.1 Migration flows often reflect pre-existing connections between countries ....................... 7 3.2 Migration flows beyond a certain threshold become self-sustaining ................................ 8 3.3 Migrants are outnumbered by involuntary non-migrants .................................................. 9 3.4 Migrants with fewer resources make more fragmented, convoluted and unpredictable journeys ................................................................................................................................................... 9 3.5 Environmental stress separates populations into trapped and mobilised groups......... 10 3.6 Violent conflict causes short-distance migration, often followed by geographically dispersed secondary migration ......................................................................................................... 12 3.7 Emigration rises with economic development until a certain level and thereafter falls 13 3.8 Development aid can deter emigration from low- and middle-income countries......... 14 3.9 Migration policies are largely ineffective in producing desired outcomes ..................... 16 3.10 Migration restrictions affect migration flows in unintended ways .................................. 17 4. Conclusion ......................................................................................................................................... 18 5. References .......................................................................................................................................... 19 Translating Migration Theory into Empirical Propositions 2 1. Introduction With the steady rise in absolute numbers of international migrants globally, especially since the turn of the Millennium, scientific and policy interest in migration dynamics, both to understand (and seek to manage) present developments, and to predict future trends, has only increased. Meanwhile, the challenge of untangling the causes of migration, offering a clear-cut and context-independent explanation of who migrates, why, when, where, and how, remains to be overcome (Arango 2000). As recent contributions show, the lack of a grand theory of migration might be understood in the light of the well-evidenced reality of composite sets of migration drivers (Czaika and Reinprecht 2020). Attempts at explaining migration processes, also for the purpose of enabling the development of plausible scenarios of future migration, must build on approaches and models that recognise both this complexity and the uncertainty which profoundly shapes migration (Bijak and Czaika 2020). It seems self-evident that the connection between theorisation and empirical insight is crucial. Meanwhile, it is pertinent to ask: What can we know? If the grand theory of migration does not exist, is this because such a migration theory is (tentatively) unconceivable, due to epistemic uncertainty of the very complex nature of migration? Or, is it because we lack appropriate tools, both conceptually and methodologically perhaps, to allow for the collection of data that would yield necessary empirical insight, and which could enable empirical validation of different hypotheses and associated theorisation? Conversely, could reasons be linked to not posing the right questions, in the first place, and not at the relevant scale? This paper sets out to translate migration theory into empirically testable propositions. Drawing actively on elements from different corners of the fragmented landscape of migration theory, we formulate ten propositions, selected based on their relevance to current societal and academic debates on international migration, its dynamics and patterns. We deliberately refer to ‘migration theory’ rather than ‘theories’ because the relevant insights and assertions are packaged so diversely. In fact, Everett Lee’s (1966) seminal article remains exceptional in unabashedly putting forth ‘a theory’ of migration. Some of the more recent contributions have been composite bundles of relationships and mechanisms that, in variable measure, concern the causes, effects and dynamics of migration. Examples include the new economics of labour migration (Stark and Bloom 1985; Taylor 1999), migration systems (Mabogunje 1970; Bakewell 2014) and the aspirations and capabilities framework (de Haas 2014). Neither fits the label of ‘a theory’ but they encompass several more focused proffers. For instance, the ‘migration hump’—an empirical relationship between development and migration that we later discuss in detail (Martin and Taylor 1996)— fits within a broader notion of migration systems. Similarly, the aspiration/ability model of migration (Carling 2002) is incorporated in the comprehensive aspirations and capabilities framework of migration and development. In such a multi-layered landscape, ‘theories’ are hardly distinct and countable units. Moreover, the disciplinary and methodological diversity of migration studies is reflected not only in diverse elements of theory, but also in diverse ways of using and relating to theory in the research process. The notion of testing a theory with empirical data is central to some approaches and alien to others. But in formulating empirically testable propositions, it would be a misstep to exclude theoretical insights from the latter approaches. For instance, the transnational turn in migration studies grew out of anthropological and other ethnographic research before being adopted by researchers that searched for empirical regularities with quantitative methods. Before setting out our approach to the present the landscape of migration theory, and its relevance and applicability for the development of empirical propositions and testable hypotheses, a view to the past is justified. Looking back then, Ravenstein’s ‘laws of migration’ published in the 1880-1890s, QuantMig Translating Migration Theory into Empirical Propositions 3 and more appropriately referred to as hypotheses emanating from empirical observation (Grigg 1977; Samers 2010), were developed in the context of understanding and explaining internal migration within the UK, but arguably grappled with many of the same issues as contemporary efforts (Rees and Lomax 2019). Ravenstein’s ideas about the roles of a ‘push-pull’ comparative evaluation between contexts, has had a profound impact, and found support across empirical contexts, at a generalised level. His basic, initial hypotheses about migration, first, underscore the value of a context of reference for theorization efforts which is well-known and where empirical patterns are readily available and intelligible. Second, Ravenstein’s acknowledgment of the value in exploring women’s and men’s different, and not always predictably different migration experiences, illustrates the necessity for differentiation (Samers 2010: 56-57). And, third, the above raises the question of why today, nearly a century and a half later, the impasse on agreement even about basic aspects of a migration theory at a grand scale appear to prevail, within the interdisciplinary field of migration studies. In the absence of a grand theory, migration scholars have mobilised a wide range of concepts in efforts to demonstrate and to explain the mechanisms of migration. These mechanisms are conceptualised in migration theorization which have been developed across social science disciplines, as well as some humanities disciplines (Brettell and Hollifield 2014; Garip 2012). Consequently, a fragmented body of migration theorizations has developed, in a sense perpetuating a self-fulfilling prophecy that the complex, multi-faceted and multi-layered processes of migration cannot be cast into a comprehensive “grand theory” of migration (Arango 2000). The fragmentation apparent across this body of existing migration theories is demonstrated and summarised by some comprehensive reviews (Brettell and Hollifield 2014; Ghatak et al. 1996; Hagen-Zanker 2008; King 2012; Massey et al. 1993; Morawska 2007). Drawing on these reviews and our own reading of existing work, we approach this body of work with a problem-oriented focus, asking how these existing theoretical approaches may explain some forms of contemporary, and potentially, future forms of migration. Evidently, our ability as migration researchers to ask the right questions is key to the opportunities and limitations, which lie in present theorisation efforts. The types of questions which we ask, and the spatial, temporal, and aggregation scales at which these are anchored reflect our different disciplines. In the context of an effort to translate migration theory into empirical propositions it is therefore key to acknowledge that we must understand each theoretical contribution appropriate to the inner logic of the scientific context it grows out of (Massey et al. 1993: 433). Inspired by previous efforts to ‘talk across disciplines’ (Brettell and Hollifield 2014) this is also the approach taken in this paper. Simultaneously, we recognise the opportunities and limitations in our ability to ask the right questions as impacted not just by our disciplinary or methodological lenses, but also by where in the world we are based (Portes and DeWind 2008). In the next section, we further explore the state of migration theory and its conceptual limitations in grasping the broader picture and propose ‘theoretical triangulation’ as a pragmatic approach for understanding migration in its multifaceted spatial and temporal manifestations. We then elaborate on a set of theoretically informed propositions for empirical tests, for use in the empirical evaluation of theories, factors and drivers, with focus on new temporalities and geographies. This paper extracts and formulates ten empirically testable propositions. These propositions are based on empirical evidence and establish stylized facts, useful for further analysis and empirical validation. A stylized fact must be true in general, but not necessarily in every case or context. For example, it is a stylized fact that it is not the poorest migrating first, or the most. This is approximately true for most countries, but it is not exactly true for any country or context. Empirical analyses typically require operationalization and specifications of various elements in each proposition, and we engage QuantMig Translating Migration Theory into Empirical Propositions 4 with those needs where they are substantively important. However, given the conceptual slant of this paper, we stop short of practical issues of data availability and methodological considerations such as research design and model specifications. 2. What is migration theory? Migration experts or policy makers refer to migration theory only reluctantly when making claims or statements about contemporary and future migration. There are several possible reasons for this. First, there is not one theory of migration but multiple pieces of theory that explain diverse forms or aspects of migration. In this sense, migration theory can be perceived as fragmented and messy. Alternative approaches are sometimes conflicting in their assumptions and claims, and they often explain only specific forms or features of the observed phenomenon. Consequently, migration theory is a toolbox where those interested in understanding migration as a complex social phenomenon are referred to in order to choose the best fitting concept for explaining certain forms or manifestations of migration. The untidiness of migration theory is not a flaw, but rather reflects the nature of migration processes. It would be futile seek universal ‘laws’ of migration that can be used for prediction in a nomological fashion (Bijak and Czaika 2020). The meaning of ‘migration theory’ is not self-evident. In fact, both the ‘migration’ and the ‘theory’ parts of the term are open to different perspectives. In terms of thematic scope, we can differentiate between migration theory in the narrow sense of explaining migration outcomes and in a broader sense of also explaining the consequences of migration for societies of origin and destination, and for the people who migrate (cf. White and Woods 1980, King 2020). The two sometimes overlap, as in the case of neoclassical theory that sees migration as a response to wage differentials and therefore predicts a convergence of wage levels as a result of migration (Lewis 1954). Our emphasis here, however, is on migration theory in the narrow sense of explaining migration outcomes. Such theory should offer answers to the following questions (paraphrased from White and Woods 1980, King 2020): - Why does migration occur? - Who migrates and who stays? - What are the spatial and temporal patterns of migration? - How does migration self-perpetuate? The second word in ‘migration theory’ is far more ambiguous. One seminal definition of theory in the social sciences is Merton’s (1967:39) reference to ‘logically interconnected sets of propositions from which empirical uniformities can be derived’. Some migration theory meets this expectation. Ravenstein’s (1885:199) laws of migration were claims to empirical uniformity, such as ‘each main current of migration produces a compensating counter-current’. Similarly, the more recent notion of a ‘migration hump’ postulates a specific empirical pattern of migration in response to economic development (proposition 7 below). Migration theory of this kind are testable; that is, the general validity and or context-specific applicability can be empirically verified. New data and empirical evidence should be compatible with the theory. If propositions of a theory are repeatedly refuted, the theory must be refined or rejected altogether (Popper 1935, 2005). But such theoretical claims are the exception rather than the rule in migration studies. Arango (2000) argues that ‘rather than fulfilling the function of guiding empirical research and providing testable hypotheses that can be contrasted with facts, existing migration theories are mainly useful for providing explanations ex-post'. Indeed, migration theory has pursued diverse objectives since its QuantMig Translating Migration Theory into Empirical Propositions 5 infancy. In contrast to Ravenstein’s (1885) laws of migration, Lee’s (1966) theory of migration was as much a framework for interpreting empirical uniformities. In this sense, his theory resembles what Rueschemeyer (2009:1) calls theory frames, which ‘guide hypothesis formation but do not themselves contain or logically entail a body of testable hypotheses’. More recent theoretical advances in migration studies have also enlightened without implying particular empirical uniformities. Often, such advances are focused on a concept that opens a new analytical perspective on a complex reality. Cases in point are ‘transnational social fields’ (Basch et al. 1996), ‘social remittances’ (Levitt 1998), and ‘migration infrastructure’ (Xiang and Lindquist 2014). Such elements of theory cannot be empirically ‘tested’ and refuted. However, their usefulness is tried in the encounter with empirical data. At times, the result is a call for revision or rejection. Migration theories claim to explain certain aspects of the observable phenomenon of people moving in time and space, but migration theory - as any theory- shall also allow predictions of the future in migration trends and outcomes if the theoretical account is true. However, the fact that not any of the established migration theories is capable in explaining all forms of migration – they do not even claim to do so - does not suggest that existing ‘partial’ migration theories are wrong and must be rejected due to their limited explanatory scope. Rather, migration theories are to be contextualised, and where possible, combined to bridge the gap between theoretical assumptions, conceptual claims, and empirical observation. Theoretical accounts are inherently limited in addressing the complex interplay of economic, social, political, environmental, and technological factors configuring context-specific migration driveroutcome relationships. However, the existence of multiple causal configurations of migration drivers and mechanisms does not imply that different theories are mutually exclusive. Quite the opposite, neoclassical income-maximizing migration (Harris and Todaro 1970) may simultaneously be a form of migration that seeks to diversify risks (cf. new economics of labour migration, Stark 1991) and relies on social network and cumulative causation theory along the way (Garip 2012, Massey 1990). In this sense, different elements of migration theory can be combined to provide composite explanations. Several accounts of migration theory have also stressed the distinction in how a theory relates to the theoretical landscape within which it is set, either in actively seeking integration with other theories, or operating as a standalone entity (Samers 2010). The fragmented body of theories, which seek to provide explanations and generalisable insight about the causes, dynamics and impact of migration, are inherently interdisciplinary in nature. Thus, the weight given from the outset to economic, social, cultural, religious, or political factors that might affect migration aspirations, decision-making and actual migration journeys, is also different. Migration studies is an interdisciplinary field, and migration theory reflects this diversity. However, there is a risk of exaggerating the importance of disciplines as such. It is common for migration scholars to draw upon literature from multiple disciplines and many works in interdisciplinary teams or institutions. Moreover, much migration research is published in interdisciplinary journals, including the more than 40 journals that specialize in migration studies (Pisarevskaya et al. 2019). The leading textbook on migration theory (Brettell and Hollifield 2014) is nevertheless organized by discipline. Its ambition to foster dialogue across disciplinary boundaries is unquestionably fruitful, but the presentation of each discipline’s dominant approach can at times seem caricatured. An image of researchers from various disciplines approaching migration with ‘their own theories’ about migration is only partially true. But there is great disciplinary variation in how researchers engage with theory in their work. The insistence on testable hypotheses, for instance, is extraneous to the research of anthropologists and many other qualitatively oriented researchers in other social sciences (Boyer et al. 2015). QuantMig Translating Migration Theory into Empirical Propositions 6 Across disciplines, the appreciation both for the complexity of migration processes, and for the need of multiple angles in order to make sense of them, is evident. How to marry the insights from across the interdisciplinary field of migration studies concretely, and in empirically researchable ways, remains less clear-cut. However, one might, by contrast, start from categories of migrants, leaning on labelling actively used by nation-states in the governance of migration, and by humanitarian actors concerned with the protection of human life, thereby e.g. considering terms such as refugees, IDPs, persons subject to modern slavery, those trafficked or exposed to human smuggling, and probably also students and (temporary) labour migrants, whether high-or-low skilled. A further approach, proposed by Samers and Collyer (2016: 8), appears more fruitful, in both recognizing that migrant lives rarely fit squarely into the boxes which categories offer, but simultaneously, that categorisation is nonetheless necessary, for the study of migration: Migrants seem to fit both into and across different types or categories of migration that imply a certain citizenship or residency status (e.g. internal or international, temporary and permanent, legal and undocumented), and different modes of entry (e.g. as asylumseekers, refugees, low-income and highly skilled workers, students, and so on). Concerning their mode of entry, they may also be classified by academics, policymakers, or statisticians as ‘forced’ or ‘voluntary’. Given the apparent fluidity of migrant lives across these different categories and modes of entry, it has now become common in migration studies to reject the significance of categorizations in some cases (e.g. Faist 2008; Richmond 2002). The contextualization of migration theory suggests that the applicability of theoretical claims and propositions is conditional to the time-, location-, and group-specific circumstances and characteristics. Massey and Taylor (2004) have identified as a major step in advancing migration research “the testing of various theoretical explanations comparatively to determine which ones prevail under what circumstances and why” (p.384) as a major step in advancing migration research. This suggests that any attempt to explain or predict migration must identify the most relevant and best fitting theory or theories for accounting variations in the number, rate, forms and characteristics of migration patterns over space and time. In this context, it becomes especially salient to pay attention to the nature of migration flows at the individual level. For instance, we know that ‘the feminization of labour migration’ is a core trait of contemporary migration dynamics (Castles et al. 2013), underscoring a need for attention to gender in empirical, and therefore also, theoretical terms. Similarly, ethnic, religious, sexual or other minority status, as well as political dissident roles, are other key traits that may at the individual (or group) level trump other key variables in understanding migration dynamics and must therefore also be accounted for. The ‘transnational turn’ in the 1990s, where transnational social spaces, constituted by the ties which migrants forged and maintained between places of residence and of origin, played a central role in migration studies (Pries 1996, Portes et al. 1999). The 2010s have similarly seen what is referred to as a ‘temporal turn’ in migration studies, foregrounding the different ways in which time – as chronological time, but also as experienced and narrated – profoundly impact migration processes (Griffiths et al. 2013). Meanwhile, it may be argued that neither the spatial nor the temporal dimensions of migration, which are being explored, have to date influenced migration theories in ways that reflect their overall relevance and potentially also contribution to explanations of migration. Given the proliferation of work, both referencing transnationalism and temporalities, this might be a surprising claim. Nevertheless, plausibly because both these bodies of work encompass the complexity of migration – across space and over time – meaningful incorporation into e.g. economically framed theories foregrounding wage differentials, is not straightforward. QuantMig Translating Migration Theory into Empirical Propositions 7 3. Propositions The following ten propositions state in a declarative form tentative and conjectural relationships between established driver constructs and specific forms and features of migration – with migration always as the phenomenon to be explained. The selection of themes for the ten propositions has been informed by their relative prominence in policy debates as well as their salience in more recent migration scholarship. We hereby consider these propositions and the underlying hypotheses plausible but also contestable. This means that all propositions refer to a substantive but nonconclusive body of empirical work with an explicit need and potential for future research that may bring a stronger evidence base on the subject. But these propositions have also been selected for their analytical relevance in assessing future migration scenarios which at the same time ensures their relevance for forward-looking policymaking. 3.1 Migration flows reflect pre-existing connections between countries Whereas international migration is often approached both as an exception to otherwise sedentary norms, and as separate from pre-existing connections between countries, in fact, migration flows to a large extent reflect long-standing connections. Such pre-existing connections refer to ties of different nature, such as trade, or ties connected to a different period in time, such as a colonial past, or these pre-existing connections might take the form of linguistic or ethno-cultural commonalities, but also quite simply of geographic proximity regionally between countries. In part reflecting an enduring presentism in migration research (Schmidt 2017), much attention is given to relatively recent transnational social fields, which are the result of international migration flows and ensuing transnational connections (Lubbers et al. 2020). However, considering global migration patterns, and despite diversification of migration destinations, which include increasing South-South migration, it remains true that ‘the former colonial past continues to play a role’ (Hooghe et al. 2008: 502). Former colonial powers, such as Spain and Portugal, France and the UK, continue to see migration patterns, which also reflect this history of former colonial connections. While it might be tempting to equate such former connections between countries simply with past human mobility, other kinds of pre-existing connections, whether or not part of colonial legacies, also matter (Morawska 2007). More often than not, such pre-existing connections might include migration, which is embedded within a larger tapestry of ties, where economic, as well as political ties play mutually constitutive roles. Taking the example of the history of Chinese intra-Asian migrations, Zhou and Benton (2017) place the history of Chinese diasporas not only in the context of, but as inextricably tied to, the formations of economic and political connections between and, significantly, on behalf of states over time. Trade has historically been a crucial driver of connections across geographic distances, where the migration of people has often been both instrumental to – and entirely framed by it. Meanwhile, also political and geopolitical considerations have shaped ties between countries, which may be seen in the roles they play as pre-existing connections reflected in current migration flows. Indeed, when considering ‘human geopolitics’ (Gamlen 2019), it is hard to imagine the leverage which current migration might have, without its firm anchoring in different forms and degrees of pre-existing connections between countries, set within the context of global and regional power hierarchies. Considering pre-exiting connections between countries, and how they shape current migration flows, it is necessary to recognise that the movement of nation-state borders themselves, both happens and has an impact on patterns of migration seen today (Gorodzeisky and Leykin 2020). Whether considering post-colonial Africa, or the array of border movements in Europe in the past QuantMig Translating Migration Theory into Empirical Propositions 8 century, it is evident that the ways in which pre-existing ties shape present migration, are tempered by the location of countries’ borders over time. Unsurprisingly therefore, linguistic and ethnocultural ties, across borders, which spur ‘ethnic return’ for instance, are a significant component of how pre-existing connections produce present-day migration flows, even before longer historical lines are considered (Mylonas and Žilović 2019). The impact of pre-existing connections between countries on migration flows and patterns, necessitates a focus methodologically beyond the present, but also beyond a focus explicitly on migrants in migration studies (Dahinden 2016). 3.2 Migration flows beyond a certain threshold become self-sustaining While much migration theory focuses on individual-level migration decision-making, the salience of the meso-level and collectives to which migrants belong, are well evidenced for both destinationchoices and migration decision-making (Banerjee 1983; Boyd 1989; Massey et al. 1993). The phenomenon of ‘chain migration’ has been empirically observed over time across different contexts, where kinship-based and location-specific social networks contribute to self-sustaining migration flows, often in a specific migration corridor (Choldin 1973; Eurenius 2020; MacDonald and MacDonald 1964; Shah and Menon 1999). The role of migration becoming self-perpetuating, has been theorized in terms of cumulative causation (Massey et al. 1993; Fussell 2010). The roles of social networks in migration processes usually emphasises the roles of family members, extended family, kinship group or co-ethnics in the locations of both destination and of origin. The roles of social networks in decision-making about migration have been hypothesized to manifest through affinity, information, facilitation, conflict and encouragement (Haug 2008: 589). The strength and nature of ties in places of destination, as well as in places of origin, temper the effects of social networks for migration decision-making – for both initial and subsequent return or onward movement (Massey et al. 1993; Ryan 2011). The initial stage of migration, which might become chain migration, fuelled by social networks, is often referred to as pioneer migration (Bakewell et al. 2012; Wessendorff 2019), and is sometimes dominated by migrants from a more elite background, with more education or social capital, often in the form of connections (Poros 2001; Friberg 2012). The different stages of migration that may become a self-sustaining chain-migration flow, typically involve a sequence from pioneers, who by virtue of remaining part of social networks with people in places of origin, and elsewhere, subsequently play a role in the sustained migration from the same areas of origin to the same areas of destination (Paasche 2020). Given that the conditions for chain-migration in legal and structural ways are in place, such as e.g. was the case for a long time from Mexico to the US, from Turkey to Germany, and from a number of newly independent colonies to the UK and France, once such chain migration gains force – in volume and speed of growth – its momentum is arguable self-sustaining, for a period of time. Depending on the regulation of mobility, where free intra-EU mobility offers a very different structural background, to that of international migration between most other states globally, chain migration may be more or less encouraged, or constrained by nation-states (Massey and Espinosa 1997; Wolf 2016). While the emphasis in the context of migration flows becoming self-sustaining is often on specific migration corridors, such as from a region of country of origin x, to a region in country of destination y, such migration corridors are often embedded within multi-sited transnational networks (Caarls et al. 2020; Lubbers et al. 2020). The roles of migrants’ social networks in creating and sustaining migration to particular destination locations, whether cities, regions or even countries, holds some QuantMig Translating Migration Theory into Empirical Propositions 9 key similarities with mechanisms which enable migration, referred to as the ‘migration industry’, facilitating, enabling or brokering migration (Fernandez 2013; Lindquist et al. 2012; Spaan and van Naerssen 2018). Thus, the roles of migrant networks – as well as institutions (formal or informal) – across contexts of origin, destination and beyond, play a part in contributing to make certain migration flows self-sustaining (Massey et al. 1993; Fussell 2010). 3.3 Migrants are outnumbered by involuntary non-migrants Fewer than 4 per cent of the global population are international migrants, living outside their country of birth. However, a higher proportion of people would have migrated if they had the opportunity. These individuals, who remain despite their preference to leave, are involuntary nonmigrants (Carling 2002). Data on migration aspirations, intentions and plans have proliferated over the past decade or two. It is now possible to say with considerable certainty that the number of involuntary non-migrants is at least twice the number of migrants worldwide (Esipova et al. 2015, Laczko et al. 2017, Clemens and Mendola 2020). The extent of involuntary immobility matter for the explanation, modelling and prediction of migration. If actual migration is just the tip of the iceberg of migration desires, then the limiting factor is the ability to migrate (Carling 2002). Consequently, the size of migration flows will be determined primarily by changes in the obstacles to migration. Obstacles first and foremost take the form of legal restrictions on migration; their effect depend on the regulations in their own right as well as their implementation and the opportunities for circumventing them (cf Czaika and de Haas 2013). In some cases, these obstacles represent a credit constraint, as when large amounts of money are needed to pay smugglers or other intermediaries. Even if migrants are outnumbered by involuntary non-migrants worldwide, this does not hold true for every country of origin and every population segment. For instance, the dynamics of migration aspirations and actual migration are stratified by education levels (Docquier et al. 2014). Identifying involuntary non-migrants requires survey data. The amount of data has swelled and now include several multi-round multi-country surveys as well as specialized surveys from various parts of the world. 1 However, conceptual and methodological challenges remain. The value of existing data is often hampered by haphazard formulations in survey instruments (Carling and Schewel 2018, Carling 2019). Future empirical work would benefit from a stronger conceptual and methodological foundation. 2 3.4 Migrants with fewer resources make more fragmented, intricate and unpredictable journeys Key migration theories seek to explain why people migrate, foregrounding specific drivers or composite sets of drivers (Czaika and Reinprecht 2020). Meanwhile, the question of how migration can and does take place, has also received some attention, exploring question both of how migration is financed and in practice realised, but also of where people migrate. In particular, research sought to understand how migration destinations are chosen (Brunarska 2019; Neumayer 2004). Here, the implications of permanent vs. temporary migration durations has been found to make a difference See QuantMig deliverable 2.1 (in preparation, due 2021). Developing such a foundation is central to the ongoing ERC-funded project Future Migration as Present Fact (www.prio.org/fumi) 1 2 QuantMig Translating Migration Theory into Empirical Propositions 10 (Liu and Xu 2017). Interest in migration management and migration control, has also contributed to an emphasis on how migration is realised, in addition to the question of why, using concepts such as ‘migration infrastructures’ as vehicles for multifaceted engagements with how migration comes about (Lindquist et al. 2012; Xiang and Lindquist 2014). The relationship between resources available to migrants, and the distance and duration of migration journeys, is well-captured by the title of a much-cited contribution to debates on how class, capital and migration are interconnected: ‘I went as far as my money would take me’ (Van Hear 2006). These links are rather intuitive yet have only to a limited extent been central to empirical investigations of migration. This is perhaps in part due to a pre-occupation with drivers at a massgeneralised scale, and perhaps in part due to challenges with access to baseline data which might easily offer opportunities of analysing prospective and actual migrants’ capital endowments of various kinds systematically. Nevertheless, the basic insight that there is a relationship between resources, migration journeys and destinations holds. This is reflected in labour migration patterns from South Asia to the Gulf states, where the direction of migration to particular locations is based on specific resources, which in this case grant the possibility of participating in highly regulated temporary labour migration schemes. Conversely, it is probably even more strongly reflected in patterns of internal migration, such as where female migrants travel to cities to find employment as domestic workers. Meanwhile, qualitative research offers much insight into the fragmentation of migrant journeys, indicative of how migration actually happens, often allowing for time and opportunity to accumulate further needed resources along the way (Schapendonk et al. 2020). Migration journeys come with a cost, and migrants with different types of human capital may therefore plan for stepwise migration, toward their destination of preference (Paul 2011). Other research also demonstrates the same dynamics at work, in the context of onward migration of third country nationals upon naturalisation in one EU country, to another (Ahrens et al. 2016; Hoon et al. 2020; Moret 2016). An important insight from these studies is that the distinction between migration which may be associated with particular drivers including conflict, in relation to the degree of fragmentation of migration journeys, becomes extremely elusive (Erdal and Oeppen 2018). International student migration, which has tripled in volume since the year 2000 (King and Raghuram 2013), further illustrates the connections between resources and complex migration trajectories. Student migration both has the prerequisite of own or family resources to invest in higher education, and the inherent possibility of becoming a long-term investment in mobility capital if the student upon completion of the degree remains in the country of study, and is able to naturalise and acquire a ‘stronger passport’ in mobility and residence-rights terms, e.g. such as a European or US citizenship. The example of student migration combines migration where a driver is the pursuit of education, with a reality where migration simultaneously may also be conceived of as a stepwise strategy, of going as far as resources allow, in order to accrue further resources, in this case in the form of human capital. As such, student migration illustrates how access to resources affects the type and degree of fragmentation of migration journeys over time, but also how fragmented migration journeys also reflect relationships between amount of resources and accessible migration pathways. 3.5 Environmental stress separates populations into trapped and mobilised groups Climate change and environmental degradation has multiple implications on people’s livelihoods. Environmental change in its multidimensional manifestations causes stressful situations for a QuantMig Translating Migration Theory into Empirical Propositions 11 growing population around the globe affecting people’s necessity, willingness and ability to migrate. Environmental change forms a predisposing factor that makes population movements increasingly likely in the longer term, while more sudden events have the potential to trigger immediate and large-scale internal displacements but may also force people to permanently move abroad (Black et al. 2011; Shen and Binns 2012; Martin et al. 2014; Veronis and McLeman 2014; Islam 2018). Spatial disparities in the availability of natural resources such as fertile soil, woodland or freshwater may establish a migration-conducive context. Detrimental effects of climate change as well as sudden environmental shocks, such as earthquakes, droughts, or floods, destroy gradually or rapidly livelihoods and may lead to food insecurity and scarcity. Environmental shocks may also detrimentally affect public health systems due to flooding and power shortages (Kishore et al. 2018). Climate change as a predisposing driver of both internal and international migration is extensively studied (Migali et al. 2018). Studies find that slow-onset changes in temperatures and precipitation are associated with emigration, particularly from more agricultural countries and rural areas (Backhaus et al. 2015; Bohra-Mishra et al. 2014; Cai et al. 2016; Nawrotzki et al. 2015). However, if climatic factors are evaluated alongside economic factors, the latter’s effects are often stronger (Joseph and Wodon 2013), and several studies suggest that climate change does not directly influence migration (Abu et al. 2014; Beinebe and Parsons 2015; Codjoe et al. 2017; Mortreux and Barnett 2009). Climate change affects migration primarily by its impact on economic factors including agricultural incomes, livelihood opportunities, or food security (Martin et al. 2014; Khavarian-Garmsir et al. 2019), health-related risks (Marchiori et al. 2012), or resource scarcity and conflict (Abel et al. 2019). At the same time, environmentally stressful situations often affect the poorest and financially constrained most adversely, which often increases their inability to move (Veronis and McLeman 2014). Migration as an adaptation strategy is often not available for a poorly resourced population, with the effect of a rapidly growing global population trapped in deprived and vulnerable situations (Black et al. 2011; Cattaneo et al. 2019). Besides gradually worsening environmental conditions, sudden environmental shocks such as floods, storms, droughts, or earthquakes may also trigger large-scale displacements. Natural disasters predominantly lead to internal, and to a lesser extent, to international migration (Beine and Parsons 2015; Islam 2018; World Food Program 2017). Disaster-induced migration is often temporary (Mallick and Vogt 2014), and sometimes indirectly driven by resource conflicts (Naudé 2010). Sudden onset disasters are often the underlying cause for the deterioration of more proximate economic drivers, such as deprivation, loss of assets, and joblessness (Warner et al. 2010; Wodon et al. 2014). In surveys, respondents often refer to economic factors, such as wages and market inaccessibility, as reasons for migration; even though the underlying cause, however, may be a gradual or sudden environmental deterioration (Afifi 2011). Overall, the relationship between environmentally induced deprivation and migration is complex but bifurcated. While some people have the capacity and are willing to move outside risk zones, most others who are vulnerable and impoverished must stay put. Adversely affected people are often involuntary immobile due to a lack of resources, but also aspirations, information, or other factors that limit their ability to move (Carling 2002; Schewel 2019). Not all immobility in the context of environmental stress is involuntary though but can be self-determined. Reasons mentioned are a variety of factors, such as place attachment (Adams and Kay 2019; Nawrotzki and DeWaard 2018), family responsibilities (Schewel and Fransen 2020), or care for assets. QuantMig Translating Migration Theory into Empirical Propositions 12 3.6 Violent conflict causes short-distance migration, often followed by geographically dispersed secondary migration Civil, ethnic, or religious conflicts and violations of human rights are known as key drivers of involuntary and forced migration including displaced persons such as asylum seekers, refugees, irregular migrants, unaccompanied migrant minors, and internally displaced persons (IDPs). Safety and security concerns usually do not immediately trigger out-migration as it may be unsafe to prepare for exit or individuals anticipate and hope for an improving security situation. But once insecurity and life-threatening situations surpass a certain tolerance level, people consider emigration or flight as the last available option for survival (Bohra-Mishra and Massey 2011). Individuals often migrate (or flee) due to direct or indirect exposure to violence but also more generic feelings of insecurity and danger for life (Lundquist and Massey 2005; van Wijk 2010). A plethora of qualitative and quantitative studies implemented at the micro and more aggregated levels have investigated the link between different measures of insecurity in sending countries and out-migration (Castles et al. 2013; Correa-Velez et al. 2017; Davenport et al. 2003; Hagen-Zanker and Mallett 2016; Hatton 2004; McAuliffe 2017; Migali et al. 2018; Moore and Shellman 2007, 2004; Ruyssen and Rayp 2014). However, the complex nexus between conflict and migration is not fully explored yet. Specifically, not all forms and contexts of violent conflicts trigger large-scale emigration, the best contemporary example is the conflict and scale of refugee movements from Syria compared to conflict situations in Yemen or Eritrea. While about a quarter of the Syrian population (about 5.6 million people) has fled from Syria since 2011, only about 10 per cent in the case of Eritrea and even a smaller fraction of Yemen’s population has been able to leave for seeking refuge abroad (UNHCR 2020). A majority of the forcefully displaced persons and refugees relocate - often only temporarily - within their home country and only a minority and self-selected group of the primarily younger, male, and better endowed and educated people (can) move abroad. Distance and costs of moving play an important role in most forms of migration but, in particular, for refugees with limited choice regarding domestic or international destinations. For most refugees, it is neighbouring countries that are predominantly providing a first safe place which most (even though not all) people in need of protection can or aspire to reach. It is therefore these geographically proximate first countries of asylum that usually absorb between 80 and 90 percent of those who flee their home country often with the intention to return as soon as the situation allows (Moore and Shellman 2007). Contemporarily, most “new” forcibly displaced persons seeking protection have moved across the border from their home country in conflict to a neighbouring country, such as Syrians to Turkey, South Sudanese to Sudan, or Congolese to Uganda (UNHCR 2020). Only a small, albeit increasing fraction of those who seek refuge outside their home country travel longer distances for claiming asylum in Western destinations. A key characteristic of the global refugee distribution is its concentration by origin and destination, respectively, that is the very limited number of countries (or sometimes even regions within countries) where refugees are coming from, but also the relatively small number of destinations where they are seeking protection. War and conflict drive involuntary displacements also indirectly through their effects on the functioning of the public infrastructure, availability of economic opportunities, and sustainability of livelihoods (Khavarian-Garmsir et al. 2019). While conflict and violence may directly predispose and trigger migration, conflict as such is often mediated or reinforced by its detrimental impact on economic, environmental or political factors (Moore and Shellman 2004; Naudé 2010). Refugees and other conflict-displaced persons often share similar motivations with other types of migrants. Like economic migrants, refugees are concerned about economic opportunities and employment, sustainable livelihoods, and prospects for their children. Migration decisions are decisions taken QuantMig Translating Migration Theory into Empirical Propositions 13 under risk and uncertainty, but in conflict and life-threatening situations, insecurity creates additional emotional stress that diminishes the role and relative importance of other socioeconomic factors. In such situations, the decision whether and where to migrate to depends on people’s risk and insecurity perceptions, emotional resilience, and attitudes towards uncertain outcomes (Czaika 2015). 3.7 Emigration rises with economic development until a certain level and thereafter falls It would seem logical that when incomes rise in poor countries, more people prefer to stay rather than emigrate. However, there is compelling evidence of the opposite. Emigration appears to rise with national income in low- and lower-middle-income countries, creating an inverted U-shaped relationship between economic development and emigration. Based on census and register data on migrant stocks 1960–2010, Clemens (2014) estimated that the turning point is a per capita purchasing power that corresponds to around 7000–8000 USD. The proposition of an inverted U-curve is an aspect of migration transitions and holds a central place in migration theory (de Haas 2010; Hatton and Williamson 2005; Martin and Taylor 1996; Zelinsky 1971). The theoretical relationship has been buttressed by the image of middle-income countries such as Mexico, Morocco and the Philippines as emigration countries par excellence. In the early 1990s, Philip Martin popularized the idea of a ‘migration hump’ in his predictions about the impacts of the North American Free Trade Agreement (NAFTA). He concluded that “NAFTA is likely to produce a temporary migration hump, slightly raising already high migration levels in the 1990s, but reducing the volume of Mexico-to-U.S. migration that would otherwise occur over subsequent decades” (Martin 1993, p. 329). Still, the conviction that economic development will stem migration, especially in poor societies, has been entrenched in policy circles (Collyer 2019). There are several reasons for the counterintuitive upward-sloping curve at lower income levels (Clemens 2014; de Haas 2010; Lucas 2005). Most are rooted in all the other societal changes that tend to accompany economic growth. First, emigration may be spawned by the demographic transition, which, in early stages, produces large cohorts of young adults. Second, even as average incomes grow, traditional labour-intensive sectors, especially agriculture, may provide shrinking livelihoods and free up large numbers of potential migrants. Third, economic development tends to produce greater inequality, especially in early stages of growth. The result can be that, for much of the population, the reference level of a decent income rises faster than their actual income. This experience of relative deprivation could be a powerful driver of emigration (Czaika and de Haas 2012). In addition to rising local inequality, increasing exposure to media and social media can spur awareness of differences and relative deprivation. In other words, the reference group against which people assess their own lives could shift away from the local community. Analyses of migration drivers call for a breakdown of effects on the aspiration and the ability to migrate (Carling 2002; Carling and Schewel 2018). Experiences of relative deprivation increase only the appeal of migration; they do not make migration more feasible. However, other aspects of economic development do. Economists have stressed the easing of credit constraints, meaning that with higher average incomes, more people can afford the initial investment in migrating. Moreover, at higher levels of economic development, higher levels of education could increase opportunities for employment abroad. The proposition that emigration from low- and middle-income countries rises with economic development could potentially be challenged in two ways. First, one could imagine crude empirical results to the contrary. That is, analyses of income and emigration levels could show a pattern that QuantMig Translating Migration Theory into Empirical Propositions 14 undermines the idea of a hump. In a recent paper, using long time series of global data, however, Clemens and Mendola (2020) finds that the expected relationship is robust to a variation in measures, samples, time periods and model specifications. A second way of challenging the migration hump would be to undercut the appropriateness of ‘economic development’ as the independent variable. The problem with this measure, of course, is that it is simultaneously a clear-cut individual variable and a proxy for complex societal changes. With relevant additional data, it is possible to decompose the relationship and examine diverse corollaries of rising income (Clemens and Mendola 2020). However, econometrics alone might not resolve the epistemological and pedagogical challenges of explaining how migration evolves with the societal transformations subsumed under the term ‘development’. 3.8 Development aid can deter emigration from low- and middle-income countries ‘It has often been suggested that […] aid, trade and investment could be utilized to influence migration pressures in major sending countries’ wrote Morrison (1982, p. 4) almost four decades ago. The strategic deployment of aid to curb migration remains an appealing prospect to many policy makers—with added emphasis and funding in the wake of the 2015 migration and refugee crisis. The merits of aid as a migration policy tool has been continuously contested, with disagreement revolving around two core issues: is such a strategy ethically and politically justifiable? Moreover, can it ever work? Even if the strategy affects migration, it could be hard to justify as a humanitarian one if it, for instance, means diverting aid from the countries where it is needed the most or would have the greatest impact on development (Bakewell 2008; Carling and Talleraas 2016; Collyer 2019). Here we address only the issue of effectiveness, captured in the proposition that development aid can deter emigration from low- and middle-income countries. However, assessing this claim with sufficient nuance is also a prerequisite for considering the ethical and political implications. The reasoning behind this proposition is that international aid can stimulate development, which in turn reduces the incentives to emigrate. Since the 1980s, it has been rejected primarily on two grounds. First, international aid is claimed to have limited effect on development (Marchal et al. 2020; Morrison 1982). That is, even if individual programs are deemed successful, they are unlikely to sway the development trajectory of a country and thereby affect the drivers of migration. Second, it is claimed that development in poor countries tends to raise rather than lower emigration (Berthelemy et al. 2009; Clemens and Postel 2018a; Faini and Venturini 1993). Such an effect of development on migration is exactly what proposition 7 above asserts. Yet, the two propositions are independent of each other. Even if the migration hump hypothesis (proposition 7) holds true, certain forms of aid might affect migration in other ways than via impacts on overall economic development. It is necessary at the outset to disaggregate ‘aid’. Development assistance is extremely diverse in objectives and implementation. For instance, Lanati and Thiele (2020) point out that 20–25% of official development assistance is spent within the donor’s borders as ‘non-transferred aid’. By excluding this category, they find a modest but significant negative effect of aid transfers on emigration rates. The diversity of in-country targets and objectives suggest that different programmes or funding mechanisms will have diverging consequences for migration. It could also matter how the aid is organized in term of national origins. Inflows of development aid are partly bilateral and partly multilateral, with potentially different impacts (Azam and Berlinschi 2009; Marchal et al. 2020). In the case of bilateral aid, it is an open question whether aid from, say, Germany, affects migration to Germany any differently from migration to other destinations QuantMig Translating Migration Theory into Empirical Propositions 15 (Berthelemy et al. 2009; Lanati and Thiele 2018c). The economics literature—which dominates research on the aid-migration relationship—has identified several possible channels, or mechanisms, through which aid could have an effect on migration (Lanati and Thiele 2018b; Marchal et al. 2020). However, they have often muddled or overlooked relevant insights from migration theory. First, there needs to be a fundamental distinction between effects on the aspiration to migrate and effects on the ability to migrate. Even if the ultimate goal is to explain migration flows, the impact of a determinant such as aid can only be understood by disaggregating the two types of effect (Carling 2002; Carling and Schewel 2018). In fact, when aid has no observable effect on migration flows, the reasons might be that effects on aspiration and ability are cancelling each other out. With this in mind, and drawing upon migration theory more broadly, we can identify several mechanisms through which inflows of development aid could affect migration either positively or negatively: The destination appeal mechanism: Inflows of aid could boost the recognition and prestige of donor countries—or high-income countries more generally—as migration destinations. Not only are overseas contributions often acknowledged in the media, inauguration events and the like, but many national development agencies have branding requirements that are reflected in signage such as “UK aid - from the British people” on infrastructure, assets, and handouts. Such multi-channel promotional messaging could plausibly have a positive effect on migration aspirations. The migration resource mechanism: Development aid could potentially raise prospective migrants' ability to migrate by adding to the resources that are required to overcome obstacles to migration. A case in point is relaxing credit constraints, as development economists have pointed out (Clemens and Postel 2018b; Faini and Venturini 1993; Marchal et al. 2020). But, depending on the context, the bottlenecks are typically policy-driven rather than financial. Aid could provide significant resources via human capital formation and employment that increases the prospect of obtaining a visa (Gaibazzi 2014; Infantino and Rea 2012). The bridgehead mechanism: When bilateral aid is accompanied by deepening social, economic, and political ties, migration can become more feasible (Clemens and Postel 2018a; Morrison 1982). Personal connections, short-term travelling, exchange programs, and the like can reduce the obstacles to migration. The direct link from development cooperation to migration can be clear in individual cases. However, if there is a broader deepening of bilateral ties, migration and aid could both be implicated in ways that make it hard to disentangle causal effects. The opportunity cost mechanism: The migration decision is always a comparative one, in the sense that the expected outcome of migrating must be related to the expected outcome of staying. This is, at least, the essence of proposition 3 above. If aid enhances the scenario of staying, the opportunity cost of migration increases, and migration aspirations decline. Development might have such an effect in specific ways. For instance, Angelucci (2004) shows that a programme of secondary-school subsidies reduces the migration propensity of participants in the short term. Similarly, Gamso and Yuldashev (2018) find that emigration can be reduced by investments in agricultural capacity building. In terms of broader societal changes, improvement in public services could improve the appeal of staying and lower migration outflows (Lanati and Thiele 2018a). Similar effects are plausible if aid has a tangible impact on corruption levels. Policy leverage mechanism: Aid transfers give donor countries bargaining power that can produce policy concessions that make migration control more effective. A case in point is making aid conditional upon readmission agreements, which facilitate the removal of unauthorized migrants and possibly dissuade future migration attempts (Cassarino 2009; Lavenex 2006) This is perhaps the mechanism that most persuasively links aid to a reduction in migration. Ironically, it is completely QuantMig Translating Migration Theory into Empirical Propositions 16 independent of development dynamics. Beyond these five specific mechanisms, there are residual and indirect ways in which aid could affect migration through broader societal changes, such as increasing levels of inequality and an accompanying rise in life aspirations. These are perhaps less pertinent because development aid will at most be a contributing factor to such fundamental transformation. The proposition that development aid can deter emigration from low- and middle-income countries remains rightfully contested. Research to date makes it impossible to confirm or reject outright, and only gives pointers as to when it might hold true. Future research should examine the diversity of mechanisms by breaking down the overall proposition into specific hypotheses. 3.9 Migration policies on their own are ineffective in producing desired outcomes The effectiveness of migration policies is continuously contested even though most migrants travel through legal channels and abide to immigration regulations. A growing number of quantitative studies indicate that immigration restrictions have a statistically significant deterrence effect on migrant inflows (Hatton 2004; Mayda 2010; Ortega and Peri 2013, Czaika and de Haas 2017). Scholars do not disagree about the stylized fact that restrictive migration policies adversely affect immigration flows, but to what extent, and under what circumstances, pre-and post-entry restrictions shape the volume, dynamic, direction, and the degree of legality of cross-border movements. Empirical evaluation of this proposition requires that policy effects can be measured and adequately assessed, and for this purpose, multiple methodological challenges including the definition, specification, and operationalisation of migration policies must be resolved. Migration policy analyses can have a range of reference points for evaluating migration outcomes. For instance, whether we evaluate migration outcomes against non-binding rhetorical claims or goals of some policymakers, or policy objectives as stated in legislative provisions and executive orders, or policies as implemented on the ground, has fundamental implications on the assessment of their effectiveness. Also, whether the object of investigation is a specific policy instrument targeting for instance a particular form of migration, or some ad hoc (one-off) interventions, or broader regulatory frameworks and reform packages can have major implications on actual magnitude (Czaika and de Haas 2013). Also, the effect size of a particular migration policy instrument is conditional upon mediating migration drivers including other policies of the broader policy mix. A further challenge is the identification of the intertemporal dynamics of causal relationship between migration policy outcomes and effects on migration outcomes, in particular, the responsiveness and sustainability of certain policy effects. Policy effects may unfold anytime between instantly (or even before implementation if policy changes are expected by potential migrants) and long after the policy change took place. Therefore, before-after assessments of policy effects should not only focus on the immediate effects on migration outcomes but consider also the long-term implications including both some “knock-on” effects on other than the targeted migration forms. Moreover, endogeneity of migration policies, that is the fact that (expected or actual) migration outcomes influence the formation and adaptation of migration policy goals, designs, and their implementation, are to be considered when assessing their effectiveness. Besides migration policies that aim to deter unwanted immigration, policies for attracting and selecting certain types of migrants such as skilled workers or students are increasingly implemented in most Westerns and many non-Western economies (Czaika and Parsons 2017). Despite the continuous rise in the number of skilled and highly skilled migrants worldwide and the proliferation QuantMig Translating Migration Theory into Empirical Propositions 17 of skill-selective migration policies, the degree to which such policies are truly effective is still contested (Bhagwati and Hanson 2009). Doomernik et al. (2009) argue that the potential of attracting high-skilled migrants in large numbers rather depend upon broader economic and social factors than on immigration policies per se. Migrants in general, but highly skilled in particular, value a myriad of factors such as the standard of living, the quality of schools, health services, and infrastructure, and the presence of a well-established professional network (Papademetriou et al. 2008). Skill-selective migration and other public policies are only forming part of a broader ‘immigration package’ (Papademetriou et al. 2008), which skilled migrants may evaluate in the migration decision-making process. Future research should further explore the evolution, impact, and implications of the broader mix of migration and migration-related policies that are instrumental in attracting, selecting, deterring, and retaining specific migrant groups in the context of other economic, social, and political factors that directly and indirectly affect migration decisions of diverse migrant populations. Policy impact analyses must address the interference of heterogenous region-, occupation-, and gender-specific drivers, dynamics, and policies in the assessment of specific or generic policy interventions. For this purpose, advanced methodological tools for policy impact evaluation are available (cf. Treasury 2020). 3.10 Migration restrictions affect migration flows in unintended ways Most migration control policies have the potential to develop both politically and ethically unwanted outcomes. Obviously, whether policy effects are wanted (i.e. intended), or not, depends on who is evaluating policy outcomes, and how, i.e. by what standards. Certain policy-induced migration outcomes can, for instance, be wanted by employers and unwanted by unions, or wanted by nativist groups and unwanted by human rights groups. The most objective yardstick for policy evaluation is the policy aims as stated by the legislative provision or the authority that is responsible for implementing a migration policy intervention (see Provision 9). Many if not most migration policy interventions result in some fundamental externalities or, side effects, which often trigger unintended consequences for migrants and host countries alike, and therefore, limiting or obscuring the realisation of migration policy goals (Castles 2004; Czaika and de Haas 2013). For instance, policy restrictions may be spatially diverting migrants to alternative locations and destinations, may be categorically deflecting migrants towards other legal entry channels or even to illegal routes and status. Migrants may also be intertemporally affected in their decision-making by pre- or postponing the timing of migration, or by procrastinating the realisation of return and onward migration plans as a reaction tightening migration policies. All these externalities often come in conjunction with an intended deterrence effect. The overall effect of a restrictive policy intervention on the volume and composition of short- and long-term immigration is a combination of a range of effects on the timing, direction and duration of individual migration events. Disentangling these multiple and multidirectional effects establishes a methodological challenge but would be a major step towards more realistic, unbiased, and comprehensive conclusions about the combination of intended and unintended effects of migration policy restrictions. Moreover, migration policies tend to produce asymmetric policy effects with different effect sizes, that is policy changes towards more liberalisation do not exactly mirror policy changes towards more restrictiveness. For instance, Czaika and de Haas (2017) find that the introduction of a travel visa restriction reduces migrant inflows to a much smaller extent compared to the effect size of a migration-inducing visa policy liberalisation (visa waiver). Migration flows respond usually more rapidly and much stronger to liberal policy changes than policy changes towards more restrictive regimes (e.g. entry conditions). We may hypothesize that such asymmetric policy effects are QuantMig Translating Migration Theory into Empirical Propositions 18 produced by complex interactions with, for instance, the facilitating role of migrant networks. But also behavioural scientific explanations such as prospect theory (Kahneman and Tversky 1979) can explain such asymmetric migration behaviour (Czaika 2015). In general, the effectiveness and the existence of significant side effects of migration policy interventions depends on the way and extent migration policy is mediated or reinforced by the interaction with other policy and contextual factors conjointly forming complex migration driver configurations. Addressing conceptual and methodological challenges in assessing the ways migration policy interventions are affected by complex configurations of ‘third factors’ in producing intended and unintended outcomes remains to be resolved by future research. 4. Conclusion Migration studies is still a fragmented and largely under-theorised research field. Arango (2000: 283) states that “migration is too diverse and multifaceted to be explained by a single theory”, yet through triangulation of various theoretical accounts and concepts, we may state propositions that address even diverse and multifaceted migration patterns and regularities. This paper outlines a limited set of ten propositions that can be empirically approached from very different angles. It is beyond the scope of this paper to propose explicitly how these propositions may be empirically approached. This was a deliberate decision since a comprehensive evaluation of the validity of each proposition does require an ‘empirical triangulation’ including a mix of hypotheses, data, methods, and contexts. Proposition 7, for instance, which claims that emigration is non-linearly associated with economic development, can be assessed through multiple analyses across analytical levels, geographic locations, scales of analyses, longitudinal coverage, methodological tools, definitions and operationalisations of the underlying concepts “migration” and “development”, and of course, the exact specification of testable hypotheses. For instance, proposition 7 can be turned into hypotheses stating that “middle-income countries have higher emigration rates than low-income countries”, or “semi-rural regions are more migratory than metropolitan areas”, or hypotheses claiming that “better educated segments of society are more mobile than the less educated”. And, as always, the specification of the research question in terms of the hypotheses informs the choice of data. Again, based on proposition 7, data and analyses can range from global assessments of emigration rates across many developing and developed countries e.g. using the World Bank bilateral stocks database (Özden et al. 2011), to large-scale cross-national surveys of migration intentions and events e.g. based on Gallup World Poll data (Migali and Scipioni 2019), to assessments of internal migration between locations within countries, e.g. based on nationally representative surveys such as the Indian National Sample Survey (NSS) or the Indonesian Family and Life Survey (IFLS), to the analysis of individual life courses of smaller, via not necessarily nationally representative surveys (e.g. MAFE project). However, even if a proposition is approached by a wide range of empirical angles and strategies, providing a solid amount of supportive evidence, the validity of the proposition remains inconclusive. We know that a proposition and its underlying theory is only tentatively true – basically until a new and better concept has been formulated which may be more convincing and powerful in explaining observable aspects of migration as a multifaceted social process and phenomenon. However, despite this tentative validity, is migration theory nevertheless useful for prediction of future migration? QuantMig Translating Migration Theory into Empirical Propositions 19 A long-standing debate in the theory of science dwells upon the question whether explanation of a phenomenon and its prediction is one or distinct (Hempel and Oppenheim 1948, Shmueli 2010). As far as migration theory is concerned, the predictive power of most theories and concepts is limited due to their partial nature and context-specific validity. Propositions of most migration theories are conditional upon the influence of ‘third factors’ that may interfere with the underlying theoretical relationship. The omnipresence of intervening factors in broader migration driver complexes, however, should not hinder attempts to use theory to inform predictions or projections of future migration trends. These predictions are likely to be even more accurate if contextualised to specific groups, geographical locations, scales of analyses, forms of migration, and temporal frames. Established but still contested ‘stylized facts’ about migration regularities have informed the selection and specification of the propositions put forward in this article. This means that these propositions already have a solid evidence base, but are nevertheless contentious – either for theoretical, empirical, or ideological reasons. More theoretical and empirical work therefore must consolidate and further substantiate our understanding of the regularities synthesized in these propositions (and the many more that we should have discussed). Theoretically and empirically triangulated and consolidated propositions are not only useful for ex post evaluations of historical migration patterns and dynamics but are also instrumental for ex ante projections of uncertain migration futures. 5. References Abel, G. J., Brottrager, M., Crespo Cuaresma, J., & Muttarak, R. (2019). Climate, conflict and forced migration. Global Environmental Change, 54(December 2018), 239–249 Abu, M., Codjoe, S. N. A., & Sward, J. (2014). Climate change and internal migration intentions in the forest-savanna transition zone of Ghana. Population and Environment, 35(4), 341–364 Adams, H., & Kay, S. (2019). Migration as a human affair: Integrating individual stress thresholds into quantitative models of climate migration. Environmental Science & Policy, 93, 129-138. Afifi, T. (2011). Economic or Environmental Migration? The Push Factors in Niger. International Migration, 49(SUPPL.1). Ahrens, J., Kelly, M., & Van Liempt, I. (2016). Free movement? The onward migration of EU citizens born in Somalia, Iran, and Nigeria. Population, Space and Place, 22(1), 84-98 Angelucci, M. (2004). Aid and Migration : An Analysis of the Impact of Progresa on the Timing and Size of Labour Migration (1187). Retrieved from Bonn: Arango, J. (2000). Explaining migration: a critical view. International social science journal, 52(165), 283-296. Azam, J.-P., & Berlinschi, R. (2009). L'aide contre l'immigration. Revue d'économie du développement, 17(4), 81-108. Backhaus, A., Martinez-Zarzoso, I., & Muris, C. (2015). Do climate variations explain bilateral migration? A gravity model analysis. IZA Journal of Migration, 4(1), 3. Bakewell, O. (2008). 'Keeping them in their place': The ambivalent relationship between migration and development in Africa. Third World Quarterly, 29(7), 1341-1358. Bakewell, O. (2014). Relaunching migration systems. Migration Studies, 2(3), 300-318. QuantMig Translating Migration Theory into Empirical Propositions 20 Bakewell, O., De Haas, H., & Kubal, A. (2012). Migration systems, pioneer migrants and the role of agency. Journal of critical realism, 11(4), 413-437. Banerjee, B. (1983) ‘Social networks in the migration process: empirical evidence on chain migration in India’, The Journal of Developing Areas, 17(2): 185-196. Basch, L., Glick-Schiller, N. and Szanton Blanc, C. (1994) Nations Unbound: Transnational Projects, Postcolonial Predicaments, and Deterritorialized Nation-States. Amsterdam: Gordon & Breach Science Publishers. Beine, M., & Parsons, C. R. (2015). Climatic factors as determinants of international migration. Scandinavian Journal of Economics, 117(2), 723–767 Berthelemy, J.-C., Beuran, M., & Maurel, M. (2009). Aid and Migration: Substitutes or Complements? World Development, 37(10), 1589-1599. Bhagwati, J. N., & Hanson, G. H. (Eds.). (2009). Skilled immigration today: Prospects, problems, and policies (Vol. 17). Oxford: Oxford University Press. Bijak, J., & Czaika, M. (2020). Assessing Uncertain Migration Futures: A Typology of the Unknown. QuantMig background paper D1.1, University of Southampton, UK Black, R., Adger, W. N., Arnell, N. W., Dercon, S., Geddes, A., & Thomas, D. (2011). The effect of environmental change on human migration. Global Environmental Change, 21(SUPPL. 1), S3–S11 Bohra-Mishra, P., & Massey, D. S. (2011). Individual decisions to migrate during civil conflict. Demography, 48(2), 401-424. Bohra-Mishra, P., Oppenheimer, M., & Hsiang, S. M. (2014). Nonlinear permanent migration response to climatic variations but minimal response to disasters. Proceedings of the National Academy of Sciences, 111(27), 9780-9785. Boyd, M. (1989) ‘Family and personal networks in international migration: recent developments and new agendas’, International Migration Review, 23(3): 638-670. Boyer, D., Faubion, J.D. and Marcus, G.E., eds (2015) Theory Can Be More than It Used to Be: Learning Anthropology's Method in a Time of Transition. Ithaca, NY: Cornell University Press. Brettell, C. B., & Hollifield, J. F. (Eds.). (2014). Migration theory: Talking across disciplines. Routledge. Brunarska, Z. (2019). A “Good Enough” Choice: Bounded Rationality in Migration Destination Choice. Studia Migracyjne-Przegląd Polonijny, 45(2 (172)). Caarls, K., Bilgili, Ö., & Fransen, S. (2020). Evolution of migration trajectories and transnational social networks over time: a study among sub-Saharan African migrants in Europe. Journal of Ethnic and Migration Studies, 1-19. Cai, R., Feng, S., Oppenheimer, M., & Pytlikova, M. (2016). Climate variability and international migration: The importance of the agricultural linkage. Journal of Environmental Economics and Management, 79, 135–151. Carling, J. (2002). Migration in the age of involuntary immobility: theoretical reflections and Cape Verdean experiences. Journal of Ethnic and Migration Studies, 28(1), 5-42. Carling, J. and Talleraas, C. (2016) Root causes and drivers of migration. Implications for humanitarian efforts and development cooperation. PRIO Paper. Oslo: Peace Research Institute Oslo. Carling, J., & Schewel, K. (2018). Revisiting aspiration and ability in international migration. Journal of Ethnic and Migration Studies, 44(6), 945-963. QuantMig Translating Migration Theory into Empirical Propositions 21 Cassarino, J. P. (2009). The Co-operation on Readmission and Enforced Return in the AfricanEuropean Context. In M. Trémolières (Ed.), Regional Challenges of West African Migration: African and European Perspectives, (pp. 49-72). Paris: OECD. Castles, S. (2004). Why migration policies fail. Ethnic and racial studies, 27(2), 205-227. Castles, S., de Haas, H., & Miller, M. J. 2013. The Age of Migration: International Population Movements in the Modern World (5th edition), Basingstoke: Palgrave Macmillan Cattaneo, C., Beine, M., Fröhlich, C. J., Kniveton, D., Martinez-Zarzoso, I., Mastrorillo, M., Schraven, B. (2019). Human Migration in the Era of Climate Change. Review of Environmental Economics and Policy, 1–19. https://doi.org/10.1093/reep/rez002 Choldin, H. M. (1973). Kinship networks in the migration process. International Migration Review, 7(2), 163-176. Clemens, M. A. (2014). Does development reduce migration? In R. E. B. Lucas (Ed.), International Handbook on Migration and Economic Development (pp. 152-185). Cheltenham: Edward Elgar Publishing. Clemens, M. A., & Mendola, M. (2020). Migration from Developing Countries: Selection, Income Elasticity, and Simpson's Paradox. IZA Discussion Paper no. 13612 Clemens, M. A., & Postel, H. M. (2018a). Can Development Assistance Deter Emigration? Retrieved from Washington: Clemens, M. A., & Postel, H. M. (2018b). Deterring Emigration with Foreign Aid: An Overview of Evidence from Low-Income Countries. Population and Development Review, 44(4), 667-693. doi:10.1111/padr.12184 Codjoe, S. N. A., Nyamedor, F. H., Sward, J., & Dovie, D. B. (2017). Environmental hazard and migration intentions in a coastal area in Ghana: a case of sea flooding. Population and Environment, 39(2), 39–146. Collyer, M. (2019). From preventive to repressive: the changing use of development and humanitarianism to control migration. In K. Mitchell, R. Jones, & J. L. Fluri (Eds.), Handbook on Critical Geographies of Migration (pp. 170-181). Cheltenham: Edward Elgar Publishing. Correa-Velez, I., Nardone, M., & Knoetze, K. (2017). Leaving family behind: Understanding the irregular migration of unaccompanied asylum-seeking minors. In M. McAuliffe & K. Koser (Eds.), A long way to go: Irregular Migration Patterns, Processes, Drivers and Decision-Making (pp. 141–165). Acton: Australian National University Press Czaika, M. (2015). Migration and Economic Prospects. Journal of Ethnic and Migration Studies, 41(1), 58–82. Czaika, M., & de Haas, H. (2012). The Role of Internal and International Relative Deprivation in Global Migration. Oxford Development Studies, 40(4), 423-442 Czaika, M., & de Haas, H. (2013). The effectiveness of immigration policies. Population and Development Review, 39(3), 487-508. Czaika, M., & de Haas, H. (2017). The Effect of Visas on Migration Processes. International Migration Review, 51(4), 893–926 Czaika, M., & Parsons, C. R. (2017). The gravity of high-skilled migration policies. Demography, 54(2), 603-630. Czaika, M., & Reinprecht, C. (2020). Drivers of migration. A synthesis of knowledge (Vol. 163). IMI QuantMig Translating Migration Theory into Empirical Propositions 22 Working Paper. Dahinden, J. 2016. A Plea for the 'Demigranticization' of Research on Migration and Integration. Ethnic and Racial Studies, 39 (13), 207-225. Davenport, C. A., Moore, W. H., & Poe, S. C. (2003). Sometimes you just have to leave: Domestic threats and forced migration, 1964-1989. International Interactions, 29(1), 27–55 De Haas, H. (2010). Migration transitions: a theoretical and empirical inquiry into the developmental drivers of international migration, Working Paper 24, Oxford: International Migration Institute, University of Oxford. De Haas, H. (2014). Migration Theory: Quo Vadis? Working Paper, 100. Oxford: International Migration Institute, University of Oxford. Docquier, F., Peri, G. and Ruyssen, I. (2014) 'The Cross-country Determinants of Potential and Actual Migration.' International Migration Review, 48(S1):37-99. Doomernik, J., Koslowski, R., & Thränhardt, D. (2009). The battle for the brains. Why Policy is not enough to Attract the Highly Skilled, Washington DC: Transatlantic Academy. Erdal, M. B., & Oeppen, C. (2018). Forced to leave? The discursive and analytical significance of describing migration as forced and voluntary. Journal of Ethnic and Migration Studies, 44(6), 981-998. Esipova, N., Srinivasan, R. and Ray, J. (2015) 'Global Desires to Migrate.' in Besharov, D.J. and Lopez, M.H. (eds) Adjusting to a World in Motion: Trends in Global Migration and Migration Policy. Oxford: Oxford University Press, 21-57. Eurenius, A. M. (2020). A family affair: Evidence of chain migration during the mass emigration from the county of Halland in Sweden to the United States in the 1890s. Population Studies, 74(1), 103-118. Faini, R., & Venturini, A. (1993). Trade, Aid and Migrations: Some Basic Policy Issues. European Economic Review, 37(2-3), 435-442. doi:Doi 10.1016/0014-2921(93)90032-6 Fernandez, B. (2013). Traffickers, brokers, employment agents, and social networks: The regulation of intermediaries in the migration of Ethiopian domestic workers to the Middle East. International Migration Review, 47(4), 814-843. Friberg, J. H. (2012). The stages of migration. From going abroad to settling down: Post-accession Polish migrant workers in Norway. Journal of Ethnic and Migration Studies, 38(10), 1589-1605. Fussell, E. (2010). The Cumulative Causation of International Migration in Latin America. The Annals of the American Academy of Political and Social Science, 630(1), 162-177. Gaibazzi, P. (2014). Visa problem: Certification, kinship, and the production of 'ineligibility' in the Gambia. Journal of the Royal Anthropological Institute, 20(1), 38-55. Gamlen, A. (2019). Human geopolitics: States, emigrants, and the rise of diaspora institutions. Oxford University Press. Gamso, J., & Yuldashev, F. (2018). Does rural development aid reduce international migration? World Development, 110, 268-282. Garip, F. (2012). Discovering diverse mechanisms of migration: The Mexico–US Stream 1970–2000. Population and Development Review, 38(3), 393-433. Ghatak, S., Levine, P., & Price, S. W. (1996). Migration theories and evidence: an assessment. Journal of Economic Surveys, 10(2), 159-198. QuantMig Translating Migration Theory into Empirical Propositions 23 Gorodzeisky, A., & Leykin, I. (2020). When borders migrate: Reconstructing the category of ‘international migrant’. Sociology, 54(1), 142-158. Griffiths, M., Rogers, A., & Anderson, B. (2013). Migration, time and temporalities: Review and prospect. COMPAS Research Resources Paper, 199-217. Grigg, D. B. (1977). E. G. Ravenstein and the “laws of migration”. Journal of Historical Geography,3(1), 41-54 Hagen-Zanker, J. (2008). Why do people migrate? A review of the theoretical literature. A Review of the Theoretical Literature (January 2008). Maastricht Graduate School of Governance Working Paper No. 2 Hagen-Zanker, J., & Mallett, R. (2016). Journeys to Europe: The role of policy in migrant decisionmaking. ODI Insights. ODI Harris, J. R., & Todaro, M. P. (1970). Migration, unemployment and development: a two-sector analysis. The American economic review, 60(1), 126-142. Hatton, T. J. (2004). Seeking asylum in Europe. Economic Policy, 19(38), 6–62. Hatton, T. J., & Williamson, J. G. (2005a). What Fundamentals Drive World Migration? In G. J. Borjas & J. Crisp (Eds.), Poverty, International Migration and Asylum (pp. 15-38). London: Palgrave Macmillan UK Haug, S. (2008). Migration networks and migration decision-making. Journal of Ethnic and Migration Studies, 34(4), 585-605 Hempel, C. G., & Oppenheim, P. (1948). Studies in the Logic of Explanation. Philosophy of science, 15(2), 135-175. Hooghe, M., Trappers, A., Meuleman, B., & Reeskens, T. (2008). Migration to European countries: A structural explanation of patterns, 1980–2004. International Migration Review, 42(2), 476-504. Hoon, M. D., Vink, M., & Schmeets, H. (2020). A ticket to mobility? Naturalisation and subsequent migration of refugees after obtaining asylum in the Netherlands. Journal of Ethnic and Migration Studies, 46(7), 1185-1204. Infantino, F., & Rea, A. (2012). La mobilisation d'un savoir pratique local: attribution des visas Schengen au Consulat général de Belgique à Casablanca. Sociologies pratiques, 1(24), 67-78. Islam, M. R. (2018). Climate change, natural disasters and socioeconomic livelihood vulnerabilities: migration decision among the Char land people in Bangladesh. Social Indicators Research, 136(2), 575-593. Joseph, G., & Wodon, Q. (2013). Is internal migration in Yemen driven by climate or socioeconomic factors? Review of International Economics, 21(2), 295–310. Kahneman, D. & Tversky, A. (1979). Prospect theory: an analysis of decision under risk. Econometrica, 47(2), 263-291. Khavarian-Garmsir, A. R., Pourahmad, A., Hataminejad, H., & Farhoodi, R. (2019). Climate change and environmental degradation and the drivers of migration in the context of shrinking cities: A case study of Khuzestan province, Iran. Sustainable Cities and Society, 47, 101480 King, R. (2012). Theories and typologies of migration: An overview and a primer. Malmö University King, R. (2020). On migration, geography, and epistemic communities. Comparative Migration Studies, 8(1), 1-10. QuantMig Translating Migration Theory into Empirical Propositions 24 King, R., & Raghuram, P. (2013). International student migration: Mapping the field and new research agendas. Population, Space and Place, 19(2), 127-137. Kishore, N., Marqués, D., Mahmud, A., Kiang, M. V., Rodriguez, I., Fuller, A., ... & Maas, L. (2018). Mortality in puerto rico after hurricane maria. New England journal of medicine, 379(2), 162-170. Khavarian-Garmsir, A. R., Pourahmad, A., Hataminejad, H., Farhoodi, R. (2019). Climate change and environmental degradation and the drivers of migration in the context of shrinking cities: A case study of Khuzestan province, Iran. Sustainable Cities and Society, 47. Laczko, F., Tjaden, J. and Auer, D. (2017) Measuring Global Migration Potential, 2010–2015, Data Briefing Series, 9. Berlin: IOM Global Migration Data Analysis Centre. Lanati, M., & Thiele, R. (2018a). Foreign aid can dampen migration if it improves public services (2). Retrieved from Lanati, M., & Thiele, R. (2018b). Foreign assistance and migration choices: Disentangling the channels. Economics Letters, 172, 148-151 Lanati, M., & Thiele, R. (2018c). The impact of foreign aid on migration revisited. World Development, 111, 59-74. doi:10.1016/j.worlddev.2018.06.021 Lanati, M., & Thiele, R. (2020). Foreign assistance and emigration: Accounting for the role of nontransferred aid. The World Economy, 43(7), 1951-1976. doi:10.1111/twec.12914 Lavenex, S. (2006). Shifting up and out: The foreign policy of European immigration control. West European Politics, 29(2), 329-350. Lee, E. (1966). A Theory of Migration. Demography, 3(1), 47-57. Levitt, P. (1998) 'Social remittances: Migration driven local-level forms of cultural diffusion.' International Migration Review, 32(4):926-948. Lewis, W. A. (1954) ‘Economic Development with Unlimited Supplies of Labour’. Manchester School of Economic and Social Studies, 22: 139–91. Lindquist, J., Xiang, B., & Yeoh, B. S. (2012). Opening the black box of migration: Brokers, the organization of transnational mobility and the changing political economy in Asia. Pacific Affairs, 85(1), 7-19. Liu, Y., & Xu, W. (2017). Destination choices of permanent and temporary migrants in China, 1985– 2005. Population, Space and Place, 23(1), e1963. Lubbers, M. J., Verdery, A. M., & Molina, J. L. (2020). Social networks and transnational social fields: A review of quantitative and mixed methods approaches. International Migration Review, 54(1), 177-204. Lucas, R. E. B. (2005). International Migration and Economic Development: Lessons from Low-Income Countries. Cheltenham: Edward Elgar. Lundquist, J. H., & Massey, D. S. (2005). Politics or economics? International migration during the Nicaraguan Contra War. Journal of Latin American Studies, 37(1), 29. Mabogunje A L. (1970) ‘Systems Approach to a Theory of Rural−Urban Migration’, Geographical Analysis, 2(1), 1-18. MacDonald, J. S., & MacDonald, L. D. (1964). Chain migration ethnic neighbourhood formation and social networks. The Milbank Memorial Fund Quarterly, 42(1), 82-97. Mallick, B., & Vogt, J. (2014). Population displacement after cyclone and its consequences: QuantMig Translating Migration Theory into Empirical Propositions 25 Empirical evidence from coastal Bangladesh. Natural hazards, 73(2), 191-212. Marchal, L., Naiditch, C., & Simsek, B. (2020). How Aid Impacts Migration Flows Once Again (202001). Retrieved from Lille: Marchiori, L., Maystadt, J.-F., & Schumacher, I. (2012). The impact of weather anomalies on migration in sub-Saharan Africa. Journal of Environmental Economics and Management, 63(3), 355–374 Martin, M., Billah, M., Siddiqui, T., Abrar, C., Black, R., & Kniveton, D. (2014). Climate-related migration in rural Bangladesh: a behavioural model. Population and Environment, 36(1), 85-110. Martin, P. L. (1993). Trade and Migration: The Case of NAFTA. Asian and Pacific Migration Journal, 2(3), 329-367 Martin, P., & Taylor, J. E. (1996). The anatomy of a migration hump. In J. E. Taylor (Ed.), Development strategy, employment and migration: Insight from models, Development Centre (pp. 43-62). Paris: OECD Martin, P., & Taylor, J. E. (1996). The anatomy of a migration hump. In J. E. Taylor (Ed.), Development strategy, employment and migration: Insight from models, Development Centre (pp. 43-62). Paris: OECD. Massey, D. (1990). 1990: Social structure, household strategies, and the cumulative causation of migration. Population Index 56, 3-26 Massey, D. S., & Espinosa, K. E. (1997). What's driving Mexico-US migration? A theoretical, empirical, and policy analysis. American journal of sociology, 102(4), 939-999. Massey, D. S., & Taylor, J. E. (Eds.). (2004). International migration: Prospects and policies in a global market. OUP Oxford. Massey, D.S., Arango, J., Hugo, G., Kouaouci, A., Pellegrino, A. and Taylor, J.E. (1993) ‘Theories of international migration: a review and appraisal’, Population and Development Review, 19(3): 431-466. Mayda, A. M. (2010). International migration: A panel data analysis of the determinants of bilateral flows. Journal of Population Economics, 23(4), 1249–1274. McAuliffe, M. (2017). Seeking the views of irregular migrants: Decision-making, drivers and migration journeys. In M. McAuliffe & K. Koser (Eds.), A Long Way to Go: Irregular Migration Patterns, Processes, Drivers and Decision-making (pp. 103–139). Acton: Australian National University Press. Merton, R. K. (1967) On Theoretical Sociology, New York: Free Press. Migali, S., & Scipioni, M. (2019). Who’s About to Leave? A Global Survey of Aspirations and Intentions to Migrate. International Migration, 57(5), 181-200. Migali, S., Natale, F., Tintori, G., Kalantaryan, S., Grubanov-Boskovic, S., Scipioni, M., Farinosi, F., Cattaneo, C., Benandi, B., Follador, M., & Bidoglio, G. (2018). International Migration Drivers, JRC No. 112622, Publications Office of the European Union Moore, W. H., & Shellman, S. M. (2004). Fear of Persecution: Forced migration 1952-1995. Journal of Conflict Resolution, 48(5), 723–745. Moore, W. H., & Shellman, S. M. (2007). Whither Will They Go. A Global Study of Refugees’ Destinations, 1965–1995. International Studies Quarterly, 51, 811–834 Morawska, E. (2007). International migration: Its various mechanisms and different theories that try to explain it. Willy Brandt Series of Working Papers in International Migration and Ethnic QuantMig Translating Migration Theory into Empirical Propositions 26 Relations; 1/07. Moret, J. (2016). Cross-border mobility, transnationality and ethnicity as resources: european Somalis’ post-migration mobility practices. Journal of Ethnic and Migration Studies, 42(9), 1455-1472. Morrison, T. K. (1982). The Relationship of U.S. Aid, Trade and Investment to Migration Pressures in Major Sending Countries. International Migration Review, 16(1), 4-26. Mortreux, C., & Barnett, J. (2009). Climate change, migration and adaptation in Funafuti, Tuvalu. Global Environmental Change, 19(1), 105–112. Mylonas, H., & Žilović, M. (2019). Foreign policy priorities and ethnic return migration policies: group-level variation in Greece and Serbia. Journal of Ethnic and Migration Studies, 45(4), 613-635. Naudé, W. (2010). The determinants of migration from Sub-Saharan African countries. Journal of African Economies, 19(3), 330–356. Nawrotzki, R. J., & DeWaard, J. (2018). Putting trapped populations into place: Climate change and inter-district migration flows in Zambia. Regional environmental change, 18(2), 533-546. Nawrotzki, R. J., Hunter, L. M., Runfola, D. M., & Riosmena, F. (2015). Climate change as a migration driver from rural and urban Mexico. Environmental Research Letters, 10(11), 114023 Neumayer, E. (2004). Asylum destination choice: what makes some West European countries more attractive than others? European Union Politics, 5(2), 155-180. Ortega, F., & Peri, G. (2013). The effect of income and immigration policies on international migration. Migration Studies, 1(1), 47-74. Özden, Ç., Parsons, C. R., Schiff, M., & Walmsley, T. L. (2011). Where on earth is everybody? The evolution of global bilateral migration 1960–2000. The World Bank Economic Review, 25(1), 12-56. Paasche, E. (2020). Elites and emulators: The evolution of Iraqi Kurdish asylum migration to Europe. Migration Studies, 8(2), 189-208. Papademetriou, D. G., Somerville, W., & Tanaka, H. (2008). Talent in the 21st century economy. Migration Policy Institute. Paul, A. M. (2011). Stepwise international migration: A multistage migration pattern for the aspiring migrant. American Journal of Sociology, 116(6), 1842-86. Pisarevskaya, A., Levy, N., Scholten, P., & Jansen, J. (2019). Mapping migration studies: An empirical analysis of the coming of age of a research field. Migration Studies, 8(3), 455-481 Popper, K. (2005, 1935). The logic of scientific discovery. Routledge (reprint). Poros, M. V. (2001). The role of migrant networks in linking local labour markets: the case of Asian Indian migration to New York and London. Global networks, 1(3), 243-260. Portes, A., & DeWind, J. (2008). Rethinking Migration: New Theoretical and Empirical Perspectives. Berghahn Books. Portes, A., Guarnizo, L. E., & Landolt, P. (1999). The study of transnationalism: pitfalls and promise of an emergent research field. Ethnic and racial studies, 22(2), 217-237. Pries, L. (1996). Transnationale Soziale Räume: Theoretisch-empirische Skizze am Beispiel der Arbeitswanderungen Mexico–USA. Zeitschrift für Soziologie, 25(6), 456-472. Ravenstein, E. G. (1885). The laws of migration. London: Royal Statistical Society. Rees, P., & Lomax, N. (2019). Ravenstein Revisited: The Analysis of Migration, Then and Now. QuantMig Translating Migration Theory into Empirical Propositions 27 Comparative Population Studies, 44. Rueschemeyer, D. (2009) Usable theory. Analytic tools for social and political research. Princeton: Princeton University Press. Ruyssen, I., & Rayp, G. (2014). Determinants of Intraregional Migration in Sub-Saharan Africa 1980-2000. Journal of Development Studies, 50(3), 426–443 Ryan, L. (2011). Migrants' social networks and weak ties: accessing resources and constructing relationships post-migration. The Sociological Review, 59(4), 707-724. Samers, M. (2010). Migration. Routledge. Samers, M., & Collyer, M. (2016). Migration. Routledge. Schapendonk , J., Bolay, M. and Dahinden, J. (2020): The conceptual limits of the ‘migration journey’. De-exceptionalising mobility in the context of West African trajectories, Journal of Ethnic and Migration Studies, DOI: 10.1080/1369183X.2020.1804191 Schewel, K. (2019) ‘Understanding Immobility: Moving Beyond the Mobility Bias in Migration Studies’, International Migration Review, pp. 1–28. Schewel, K. and Fransen, S. (2020) ‘Who aspires to stay? Immobility aspirations among youth in Ethiopia, India, Peru, and Vietnam’. IMI Working Paper No 161. Schmidt, G. (2017). Going beyond methodological presentism: examples from a Copenhagen neighbourhood 1885–2010. Immigrants & Minorities, 35(1), 40-58. Shah, N. M., & Menon, I. (1999). Chain migration through the social network: experience of labour migrants in Kuwait. International migration, 37(2), 361-382. Shen, S., & Binns, T. (2012). Pathways, motivations and challenges: contemporary Tuvaluan migration to New Zealand. GeoJournal, 77(1), 63-82. Shmueli, G. (2010). To explain or to predict? Statistical science, 25(3), 289-310. Spaan, E., & van Naerssen, T. (2018). Migration decision-making and migration industry in the Indonesia–Malaysia corridor. Journal of Ethnic and Migration Studies, 44(4), 680-695. Stark, O. (1991). The migration of labour. Blackwell. Stark, O., & Bloom, D. E. (1985). The new economics of labor migration. American Economic Review, 75(2), 173-178. Taylor, J. E. (1999). The new economics of labour migration and the role of remittances in the migration process. International Migration, 37(1), 63-86. Treasury, H. M. (2020). Magenta Book: Central Government guidance on evaluation, UK Government, https://www.gov.uk/government/publications/the-magenta-book UNHCR (2020) Global Trends: Forced Displacement in 2019, UNHCR: Geneva Van Hear, N. (2006) ”I went as far as my money would take me”: conflict, forced migration and class. Forced migration and global processes: A view from forced migration studies, pp. 125-158. Van Wijk, J. (2010). Luanda – Holanda: Irregular Migration from Angola to the Netherlands. International Migration, 48(2), 1–30 Veronis, L., & McLeman, R. (2014). Environmental influences on African migration to Canada: Focus group findings from Ottawa-Gatineau. Population and environment, 36(2), 234-251. Warner, K., Hamza, M., Oliver-Smith, A., Renaud, F., & Julca, A. (2010). Climate change, QuantMig Translating Migration Theory into Empirical Propositions 28 environmental degradation and migration. Natural Hazards, 55(3), 689–715 Wessendorf, S. (2019). Pioneer migrants and their social relations in super-diverse London. Ethnic and racial studies, 42(1), 17-34. White, P. E., & Woods, R. I. (1980). The foundations of migration study. In P. White, & R. Woods (Eds.), The geographical impact of migration (pp. 1–20). London: Longman. Wodon, Q., Liverani, A., Joseph, G., & Bougnoux, N. (2014). Climate Change and Migration: Evidence from the Middle East and North Africa. (Q. Wodon, A. Liverani, G. Joseph, & N. Bougnoux, Eds.). Washington, DC: World Bank Wolf, K. (2016). Marriage migration versus family reunification: How does the marriage and migration history affect the timing of first and second childbirth among Turkish immigrants in Germany? European Journal of Population, 32(5), 731-759. World Food Program. (2017). AT THE ROOT OF EXODUS: Food security, conflict and international migration Xiang, B., & Lindquist, J. (2014). Migration infrastructure. International Migration Review, 48(1s), 122-148. Zelinsky, W. (1971). The Hypothesis of the Mobility Transition. Geographical Review, 61(2), 219-249 Zhou, M., & Benton, G. (2017). Intra-Asian Chinese migrations: A historical overview. In Contemporary Chinese Diasporas (pp. 1-25). Palgrave, Singapore.  QuantMig Translating Migration Theory into Empirical Propositions