Abstract The September 2016 UN New York Declaration for Refugees and Migrants was welcomed with much enthusiasm, as the 193 UN Member States agreed to meet yet again to negotiate a Global Compact on Safe, Orderly and Regular Migration. In the year that followed, the process of consultations and negotiations laid out in Annex II of the Declaration moved ahead at full steam (IOM 2017). What will be the substantive outcome of this process? To answer this question, we provide a theoretical framework to explain the structure of international cooperation on migration. The structure consists of five elements: the patterns of migration flows in the post-World War II period, which divide states into countries of origin and countries of destination; the status quo of customary international law that privileges countries of destination; exogenous shocks that trigger changes in the costs of the status quo; the institutionalization of the international system that permits countries of origin to project their preferences onto the international stage; and finally, the ability of countries of destination to ignore these preferences. Examples illustrate the application of the theoretical model. Will the Global Compact provide more and better interstate cooperation on migration? Or, will states largely ignore it, as they have ignored the three previous multilateral migration treaties? We argue that, unfortunately, the structure of international cooperation on migration, as just described, leads to the conclusion that this Declaration is likely to be ignored. Given the limited possibility of international cooperation, we recommend that mobilization for migrant rights should focus at the national and local level. Introduction The September 2016 UN New York Declaration for Refugees and Migrants was welcomed with much enthusiasm, as the 193 UN Member States agreed to meet yet again to negotiate a Global Compact on Safe, Orderly and Regular Migration. In the year that followed, the process of consultations and negotiations laid out in Annex II of the Declaration moved ahead at full steam. By September 2017, the UN had held five informal thematic sessions and an informal interactive stakeholder meeting; issued multiple summaries, informational notes, and issue briefs; and planned a “stocktaking meeting” for December 2017 in Puerto Vallarta, Mexico (IOM 2017). What will be the substantive outcome of this process? Will the Global Compact provide more and better interstate cooperation on migration, or will states largely ignore it, as they have ignored the three multilateral migration treaties that came before it? We argue that the structure of international cooperation on migration suggests the latter outcome is more likely. International cooperation arises from five elements: the patterns of migration flows in the post-World War II period, which divide states into countries of origin and countries of destination; the status quo of customary international law that privileges countries of destination; exogenous shocks that trigger changes in the costs of the status quo; the institutionalization of the international system that permits countries of origin to project their preferences onto the international stage; and, finally, the ability of countries of destination to ignore these preferences. Globalization in the latter half of the twentieth century has knit together economies and societies across national boundaries. This process of globalization has been underpinned by a dense institutionalization of state interactions, yet the degree to which migration flows fit this pattern is highly contested. Alexander Betts (2011) points to a “tapestry” of global governance, whereas James Hollifield (2000) claims the migration regime is missing. Empirically, we observe a number of bilateral agreements between states. Labor migration is sometimes incorporated into regional agreements; and multilateral treaties have been negotiated at the International Labor Organization (ILO) and at the UN. Informal forums and dialogues abound. In this article, we make sense of these disparate views and the variegated migration landscape. First, we disaggregate international population movements into three types—forced, voluntary, and travel—and focus specifically on voluntary migration flows (Koslowski 2011).1 Second, we describe the patterns of migration over the last sixty years as characterized by nonreciprocal flows from poorer, less stable countries of origin to wealthier, more stable countries of destination. This pattern tends to generate bilateral externalities that are usually resolved with bilateral agreements rather than regional or multilateral agreements. Third, we argue that the current status quo of customary international law privileges the wealthier and more stable countries of destination—receiving states—and therefore negotiations to overturn customary international law are less likely. There are three conditions under which agreements may be reached: when exogenous shocks raise the costs of the status quo for countries of destination; when countries of origin are able to leverage their power in preexisting international institutions; and when migration flows are reciprocal. We provide examples to illustrate our claims. The article proceeds by first delineating the domain of our theoretical framework, voluntary international migration, and the definitions and assumption from which we proceed. We then present a bargaining model of interstate negotiations that provides for observable implications. We illustrate the reach of the model by examining the European migrant crisis, the International Convention on the Protection of the Rights of all Migrant Workers and Members of Their Families (ICRMW), and the freedom of movement provisions in the Gulf Cooperation Council (GCC). We conclude by pointing to the opportunities for action at the state and local level. Scope, Definitions, and Assumptions Voluntary Migration Some of the confusion surrounding issues of global migration governance stems from a failure to disaggregate the different types of movement across international borders. According to Rey Koslowski (2011), the entire picture of global migration governance can be divided into three subregimes: the travel or mobility regime, the voluntary migration regime, and the forced migration or refugee regime. Our focus is on voluntary migration, which encompasses 90 percent of migrants (roughly 258 million people in 2017) (UN, Department of Economic and Social Affairs 2017). This focus presumes a clear distinction between “voluntary” and “forced” migration. We know, however, that migrants often have multiple motives for moving and that the legal definition of “refugee,” found in the UN Convention Relating to the Status of Refugees, is so narrow as to leave many migrants unprotected from violence and other threats to their existence.2 However, this legal fiction is actually important in practice, and migrants are classified according to the legal criteria every day. Most of the individuals caught up in “migration crises,” such as the 2015 European migrant crisis, are classified as migrants rather than as refugees, even when they are fleeing war-torn countries, meaning that receiving states can decide to accept or reject them, depending on state preferences.3 If migrant rights activists and the international community more broadly hope to reduce the human tragedies involved in many migration flows, understanding the prospects for cooperation on voluntary migration is central. International Cooperation There is no standard definition of “international cooperation” employed in the scholarly literature. However, as the research agenda has deepened over the past three decades, we find that some scholars in international relations (IR) focus on international agreements and international organizations as markers for international cooperation (Koremenos, Lipson, and Snidal 2001). International agreements come in many forms (bilateral, regional, or multilateral) and can be formally binding or simply a memorandum of understanding that does not bind states to the letter of the text. These international agreements can create institutional structures or be informal, self-implementing, and self-enforcing. They define a set of behaviors to which state parties agree. International agreements bind states to action that is not unilaterally determined. We take up this position and, for the purposes of this article, define international cooperation as a formal or informal agreement among two or more states that binds states to adopt a joint solution to an issue area that requires action on the part of the signatories. In contrast to much of the global governance literature, which today explicitly incorporates the activities of nonstate actors (NSAs), both domestic and international, our focus remains on agreements between states as the locus of our theoretical and empirical inquiry for two reasons. First, it allows us to simplify our framework while still incorporating the effects of NSAs; they are incorporated into the analysis by way of delineating the costs of the status quo, and changes to it, for states. Those activities are then reflected in the negotiating outcomes should states choose to initiate international negotiations. Second, the practices adopted by NSAs exist only when those actors have both the preferences and resources to implement those practices, with no scope for enforcement of any type. Although it is not impossible for NSAs, such as firms or international organizations, to adopt and implement such governance practices, such practices appear largely absent in the international migration arena. Migration Patterns We observe that migration patterns in the post-World War II era are characterized both by nonreciprocity and by unique receiving country patterns (Hatton 2007; UNDP 2009; Sykes 2012). These characteristics affect the degree of cooperation and the “shape” of cooperation—bilateral, regional, or multilateral. There are complex reasons why individuals choose to migrate (Castles, de Haas, and Miller 2014). However, barring state barriers to egress and entry, the general pattern in the contemporary era is from poorer and less stable states to wealthier and more stable states (UNDP 2009). Wealth and stability are relative so that some states in the “Global South” are receiving states and about half of all voluntary migrant flows are among countries of the Global South. We characterize this movement as nonreciprocal flows between “sending” or source states and “receiving” or host states. This characteristic is unique to migration in the depth and breadth of international economic flows. The United Nations Development Program (UNDP 2009) reports that 37 percent of migrant flows are from developing countries to developed countries while only 3 percent move in the other direction. And 75 percent of all migrants enter a country with higher human development than their country of origin. These figures confirm that the flow of migrants among states is dominated by unidirectional flows. With the exception of movement within regional organizations such as the EU by citizens of Member States, this general pattern is reflected in individual countries (United Nations Department of Economics and Social Affairs 2013). If migration is nonreciprocal, why would this pattern of migration generate a barrier to international cooperation? Axelrod (1985) years ago pointed out the importance of reciprocity to the evolution of cooperation: it is based on the ability to retaliate. If states receive reciprocal flows, then efforts to cooperate can evolve with a Tit-for-Tat strategy, producing higher payoffs for both parties.4 The Tit-for-Tat strategy adopts a position of cooperation on the first move and then copies the opponent’s strategy thereafter. This means that when a state refuses to cooperate, it is punished in the next move. However, should it cooperate in the future, then forgiveness is demonstrated and this allows cooperation to ensue. The World Trade Organization, for example, has created a decentralized enforcement mechanism that allows states whose economies have been hurt by unfair trade practices to retaliate against the trade of the opposing state. The absence of reciprocity represents a stumbling block to cooperation on migration, as cooperation then requires linkage to some third issue. We argue that the central feature that shapes the type of international cooperation that arises on immigration is the fact that migration patterns are not only directional but country specific as well. These patterns are well known to migration experts, and flows can be reasonably well-modeled based on geography, historical ties, and wage differentials (Hatton and Williamson 2003a, 2003 b). One example is provided in Table 1, which illustrates the top “third country” immigrant groups to the largest recipient states in the EU in 2003. There is virtually no overlap in the migration source countries despite the fact that these are wealthy members of the EU, which has adopted free movement among its members (there are eighteen unique observations among the twenty data points; Turkey and China are each listed twice). This pattern suggests that any market failures or externalities that are generated by migration are dyadic in nature. For example, if there is a large population of undocumented Algerians in France, this does not concern Germany, the UK, or Sweden. The solution to this problem revolves around cooperation between France and Algeria. So, we argue, cooperation does exist but it is structured by the nature of the flows, and hence the stocks of migrants, which privileges bilateral cooperation and disadvantages multilateral cooperation. Table 1 Top non-EU migrant admissions, 2003 (select EU countries) France Germany UK Sweden Algeria Turkey Pakistan Iraq Morocco Russia India Thailand Tunisia Ukraine South Africa Serbia Turkey United States Nigeria China Congo China Afghanistan (2006) Somalia France Germany UK Sweden Algeria Turkey Pakistan Iraq Morocco Russia India Thailand Tunisia Ukraine South Africa Serbia Turkey United States Nigeria China Congo China Afghanistan (2006) Somalia Source: Migration Policy Institute 2012. Table 1 Top non-EU migrant admissions, 2003 (select EU countries) France Germany UK Sweden Algeria Turkey Pakistan Iraq Morocco Russia India Thailand Tunisia Ukraine South Africa Serbia Turkey United States Nigeria China Congo China Afghanistan (2006) Somalia France Germany UK Sweden Algeria Turkey Pakistan Iraq Morocco Russia India Thailand Tunisia Ukraine South Africa Serbia Turkey United States Nigeria China Congo China Afghanistan (2006) Somalia Source: Migration Policy Institute 2012. In contrast, reciprocal flows are defined by the exchange of migrants between two countries. In the limited number of cases where migration flows are reciprocal, states have been able to generate multilateral cooperation on migration—such as in the Nordic Union or the EU. But this type of cooperation is the exception rather than the rule. State Preferences We take the perspective that states are the primary actors in the international system, although we acknowledge that NSAs can play an important role in shaping international collaboration.5 This position is not controversial in the literature on international bargaining (Odell 2000; McKibben 2015). After all, we are interested in agreements that states sign, and this is one method of abstracting from the empirical reality to model the processes taking place. When examining the interests of states in the international system relative to migration, we divide states into two categories: “sending” states and “receiving” states.6 Just as John Odell (2000) refers to “market conditions” as central to understanding state preferences on trade and how those preferences change, we refer to market conditions that separate states into those states that attract immigrants and those states that generate emigrants.7 Of course, this is a simplification of the migration profile of states in the international system. All states in the international system send migrants as well as receive migrants. Some states are also geographically positioned to act as “transit” states between sending and receiving states. Yet all states have a net migration flow that weighs the number of emigrants against the number of immigrants and, we argue, serves to help understand the types of interests they pursue relative to migration issues. States may transition between a sending and receiving status. When they do, we argue their interests shift as well. The second attribute that we want to point out is the power characteristics of sending and receiving states. States that attract voluntary migrants are generally wealthier and more stable than countries of origin. As a result, we attribute to receiving states greater levels of external power resources—military power, diplomacy, and economic power. These attributes serve to reinforce the preferences of receiving states in the international system. The Bargaining Framework The Status Quo Ante, State Preferences, and Power Our bargaining framework begins with the status quo and the preferences of sending and receiving states. To illustrate, we can think about the accepted behavior of states in customary international law related to exit and entry of individuals. Table 2 summarizes customary international law in the area. Table 2 Customary law in international migration Departure Admission Right to leave any country except when restrictions are provided by law, necessary to protect public order and consistent with other fundamental rights Right to return to one’s own country Nonrefoulement (not turning away) (of those facing persecution as defined in the UN Refugee Convention (1951, 1967) Family reunion of children Prohibition of arbitrary detention Access to consular protection Prohibition of collective expulsion Departure Admission Right to leave any country except when restrictions are provided by law, necessary to protect public order and consistent with other fundamental rights Right to return to one’s own country Nonrefoulement (not turning away) (of those facing persecution as defined in the UN Refugee Convention (1951, 1967) Family reunion of children Prohibition of arbitrary detention Access to consular protection Prohibition of collective expulsion Source: Chetail (2014, 71). Table 2 Customary law in international migration Departure Admission Right to leave any country except when restrictions are provided by law, necessary to protect public order and consistent with other fundamental rights Right to return to one’s own country Nonrefoulement (not turning away) (of those facing persecution as defined in the UN Refugee Convention (1951, 1967) Family reunion of children Prohibition of arbitrary detention Access to consular protection Prohibition of collective expulsion Departure Admission Right to leave any country except when restrictions are provided by law, necessary to protect public order and consistent with other fundamental rights Right to return to one’s own country Nonrefoulement (not turning away) (of those facing persecution as defined in the UN Refugee Convention (1951, 1967) Family reunion of children Prohibition of arbitrary detention Access to consular protection Prohibition of collective expulsion Source: Chetail (2014, 71). This can be depicted by a two-dimensional issue space as shown in Figure 1. The most basic rule governing admissions is state sovereignty: the state has the right to turn away individuals seeking entry. This right, however, has been abridged in several ways in customary international law. For example, states are required to allow their own citizens to enter, should they leave the territory of the state. A second abridgement of the right is associated with the 1951 refugee convention that requires signatories to not refoule (turn away) refugees. This requires states to evaluate asylum claims of individuals seeking entry to determine whether they meet the refugee definition. In the two-dimensional issue space, the status quo on this dimension would reflect a point close to but not equivalent to total state sovereignty. Conversely, states are required by customary international law to permit their citizens to leave their country, save for concerns over public order. Thus, on the second issue dimension of state sovereignty over exit, the status quo would be reflected by low levels of state sovereignty. Figure 1 View largeDownload slide Conditions that promote retention of the status quo. SQ: Status quo (customary international law). State A: Receiving State A’s ideal point. State B: Sending State B’s ideal point. Notes: Receiving State A’s ideal point is identical to the status quo, whereas Sending State B prefers a different point in the policy space. Generically, the win set of the status quo is empty because there is no point in the policy space that both State A and State B prefer to the status quo. Because Receiving State A is more powerful than Sending State B, it can resist efforts of the sending state to change the status quo. Hence, the status quo represents a privileged outcome; negotiations are unlikely to ensue. Figure 1 View largeDownload slide Conditions that promote retention of the status quo. SQ: Status quo (customary international law). State A: Receiving State A’s ideal point. State B: Sending State B’s ideal point. Notes: Receiving State A’s ideal point is identical to the status quo, whereas Sending State B prefers a different point in the policy space. Generically, the win set of the status quo is empty because there is no point in the policy space that both State A and State B prefer to the status quo. Because Receiving State A is more powerful than Sending State B, it can resist efforts of the sending state to change the status quo. Hence, the status quo represents a privileged outcome; negotiations are unlikely to ensue. This represents one specific two-dimensional issue space. Generically, each migration issue area has a two (or multidimensional) issue space and a specific status quo ante. We point out that the win-set of the status quo, the policies that both states would prefer to the status quo, is empty if one party prefers the status quo. As illustrated in Figure 1, State A prefers the status quo and State B prefers a position in the policy space that is different from the status quo. Because we have defined receiving states as states with relatively more power and more stability, if State A is a receiving state and prefers the status quo, there is a strong likelihood that negotiations to change the status quo will not come to fruition, even though State B is dissatisfied with the status quo. This relationship holds, we argue, both on a bilateral basis and a multilateral basis. That is, States A and B can represent two individual states in the international system or can represent a coalition of all receiving states and all sending states. We conclude, therefore, that when powerful states prefer the status quo ante, international cooperation is less likely to occur. In light of this theoretical framework, we argue there are three conditions under which states will enter the international arena to bargain and to reach an agreement: when the presence of exogenous shocks raise the costs of the status quo; when sending states locate an international institution whose decision rules privilege their number; and when reciprocal flows modify the costs of cooperation. We elaborate below. Exogenous Shocks Although powerful states may generally prefer the status quo, exogenous events may change the costs of the status quo and modify the preferences of powerful states. Exogenous events come in many forms. Domestic political actors may create political costs for the government. In the realm of international migration, there are a number of anti-immigrant actors—political parties and interest groups—that may be able to embarrass the government and create electoral costs that the government finds unacceptable. Migrant rights organizations and domestic media outlets may bring attention to migration issues in ways that modify the government’s political calculus. International state and nongovernmental actors may also be important in increasing the costs of the status quo to the government in power. Even market forces may create rising costs. Peaks and valleys in the state’s economic cycle may generate demand for migrants that is unattainable through the status quo system, or increase calls to reduce the migrant population through methods that are unacceptable in the status quo system. Migrants themselves are also actors. They can mobilize within a polity in a way that create challenges to the government in power. Or, they can increase the costs to the government of the status quo by moving across international borders. Regardless of the source of the exogenous pressures or shocks, as the costs of the current status quo rise, the preferences of the powerful states may change. In this case, the powerful state is likely to initiate negotiations. The European migrant crisis is an example of an exogenous shock that mobilized a variety of different groups and put pressure on powerful receiving states to react. The full human tragedy of the European migration crisis came into focus with the photo of three-year-old Aylan Kurdi, whose lifeless body washed up on a Turkish beach in September 2015 after his family’s boat capsized as they tried to reach Greece. Refugee and migrant rights activists used this attention to advocate for a greater number of refugee spots, and indeed, countries like the United States and Canada acted unilaterally to expand their refugee resettlement programs slightly. But, more significantly, the migrant crisis is an example of the powerful receiving states of the EU entering the international arena in search of an agreement to reduce the costs of the status quo. Negotiations with Turkey, the main transit country during that crisis, began in late summer 2015 and concluded in March 2016. Thus, we argue that when the costs of the status quo ante rise, receiving states are more likely to initiate international negotiations. Once a powerful state initiates international negotiations, the distribution of power may shift based on each state’s best alternative to a negotiated agreement. The concept of “best alternative to a negotiated agreement”, or BATNA, is central to some of the bargaining literature and reflects the idea that parties to the negotiation will examine their alternatives and choose the alternative with the highest payoff. If an alternative to a negotiated agreement provides a higher payoff, then the state will select that alternative. If the negotiated agreement provides the highest payoff, then it is likely the agreement will be chosen. When receiving states experience an exogenous shock, the rising costs of the current status quo make no agreement costly. Time is not on the receiving state’s side; the state’s leaders need a resolution that reduces their costs. They may have tried unilateral solutions to address the rising costs as alternatives to a negotiated agreement, but these solutions have been ineffective. They need an agreement. On the other side of the negotiating table, the sending or transit state may well be better satisfied with the status quo than the receiving state. That is, although the sending or transit state prefers a policy position different than the status quo, their position is closer to the status quo than to the position desired by the receiving state. The BATNA that reflects the preferences of the receiving state may well be the status quo. In this case, they will not choose a negotiated agreement unless they are compensated for their compliance. The availability of alternatives to the negotiated agreement enhances the bargaining power of the sending state. As the receiving state’s costs rise, the sending state can extract higher side payments and/or a change in the international rules governing migration. To continue the same example mentioned above, when the EU was confronted with a large, and potentially unending, stream of migrants flowing through Turkey in 2015, it negotiated an agreement with Turkey to stop migrant departures—a change to the status quo. In the initial agreement in November 2015, Turkey was able to obtain three billion euros but ultimately realized that it could extract even larger side payments. The second agreement, which took effect in March 2016, upped the ante to six billion euros—payments promised by the EU to Turkey to help offset the costs of housing Syrians fleeing civil war within Turkey itself (European Commission 2016). Institutional Power Weak states, however, are not totally without resources (Odell 2000; Drahos 2003; Sebenius 2009; Whitfield and Fraser 2010). One of the resources they have gained in the international system is the power of numbers achieved through institutional rules (Schneider 2011; Panke 2012). An example is the UN where sending states, individually less powerful than the receiving states, can vote on provisions in the General Assembly and adopt instruments by majority vote, instruments that the receiving states may not prefer. However, this resource is limited in the sense that states are unbound by treaties that they choose not to ratify. So, when the less powerful states can agree on a policy and locate a forum that provides them with institutional power, negotiations may proceed and even result in an international agreement. However, the powerful states are not compelled to ratify the treaty. On issues of international migration, we argue that if sending states prefer a policy different than the status quo and can organize a coalition that allows them to employ institutional rules to initiate negotiations, then sending states may be able to obtain an international agreement. However, sending states are unlikely to have the power resources to persuade receiving states to modify their opposition to the treaty proposals to change the status quo. An example of this type of event is the negotiation of the UN ICRMW between 1980 and 1990. This form of weak state power can be summarized as follows: When less powerful sending states act as a coalition and find a forum that provides institutional power, bargaining is more likely to occur. However, when an agreement is negotiated without the support of powerful receiving states, those states are unlikely to ratify any negotiated treaty and will not be bound by the treaty provisions. We expect a similar outcome for the new Global Compact on Safe, Orderly and Regular Migration. Reciprocal Flows We have described most flows as nonreciprocal, yet there are some exceptions to the rule: cases of reciprocal flows. Specific conditions among a set of states may encourage reciprocal migration flows and allow states to improve efficiency in labor markets as well as minimize the potential costs of migration through freedom of movement. Similar levels of wealth, wages, and social benefits ensure that flows will tend toward reciprocity.8 Full employment mitigates social welfare costs for the state. Linguistic, cultural, and historical similarities also facilitate movement by decreasing the costs of integration for citizens of both home and host societies. Where flows are reciprocal, freedom of movement enhances labor market efficiency (Borjas 2001; Pritchett 2006; Zimmerman 2009). When the demand for labor is high, immigrants can fulfill unmet labor market needs; when the demand for labor is low, unemployed workers can find work abroad. Efficient labor markets generate full employment that reduces the demands on the state for social welfare and enhance state revenue through taxes. Under these conditions, international cooperation may be possible. The translation of this set of preferences into policy is conditional on similar standards of living and on “full” employment. International freedom of movement is extraordinary in the sense that, with or without border checks, citizens of Member States have permission to reside and to take up employment without regard to quotas or labor market conditions of the national labor force. The states that are party to such agreements relinquish sovereignty over immigration for a select group of countries. Britain discovered itself in this position, much to its dismay, when immigration from the ten new EU Member States in 2004 exceeded expectations by a factor of ten to one (Sherwood 2014).9 This policy is exceptional. Only European states and Australia and New Zealand, along with the GCC and the Organization of Eastern Caribbean States (OECS), have adopted and implemented freedom of movement. Our argument is complicated by attention to the institutional structures within which these bargains take place. Institutions can constrain states’ future actions, requiring them to accept free movement even when conditions are not met (Krasner 1976). This, we argue, was the road to Brexit.10 Alternatively, institutions can be flexible and allow states to renege on free movement when conditions change, as was the case with the Trans-Tasman Travel Arrangement between Australia and New Zealand. We note that states may respond differently within the same institutional structures when faced with similar exogenous shocks. There are other domestic factors, beyond unique migration profiles, which contribute to a state’s decision to pursue migration cooperation, pursue unilateral action, or withdraw from existing cooperative agreements. These include economic growth rates, unemployment rates, and the degree of social welfare provision, among other factors. For example, unlike the UK, Germany has remained committed to free movement within in the EU, even when confronted by a massive influx of EU migrants after the 2008 financial crisis.11 German tolerance for this unanticipated immigration under the Schengen agreement is partly due to a greater commitment to the European project. Because there are few institutional mechanisms for restricting intra-EU migration, Germany could do little without threatening European integration more broadly, which it was unwilling to do. Britain, on the other hand, has never been as committed, refraining from joining the monetary union and harboring a relatively strong anti-EU faction within the political discourse. Thus, it took less of a shock for Britain to decide that the costs of the cooperation outweigh the benefits. But, there are additional domestic differences that also explain Germany and Britain’s divergent responses. First, German economic growth rates are higher, and unemployment rates are lower (particularly youth unemployment rates) than in Britain, meaning that it was less politically costly for Germany to absorb EU migrants. In fact, these migrants often filled labor market needs, and Germany preferred EU migrants over other potential source countries, like Turkey and Middle Eastern states, in part because they anticipated that EU migrants would return home once the European economies stabilized.12 To summarize, we anticipate few instances of inter-state cooperation on issues of voluntary migration because powerful receiving states are often privileged by the status quo. However, when exogenous shocks increase the cost of the status quo, receiving states may initiate negotiations to modify the status quo through bilateral agreements, providing side payments to entice the sending states to enter into agreement. Alternatively, sending states may locate institutional forums with membership and decision rules that provide them with the ability to negotiate multilateral agreements. In these cases, sending states may initiate negotiations and conclude a multilateral agreement. However, such an agreement will have little impact because receiving states will be unwilling to ratify the agreement. Finally, where similar conditions create the possibility of reciprocal flows, states may negotiate freedom of movement. However, low flows are anticipated because of limited incentives to move. We note that the way in which states respond to exogenous shocks, even within the same institutional structures, is also conditioned by domestic factors that change their cost-benefit analysis of cooperation. Below, we provide illustrations of these three scenarios.13 Exogenous Shocks and Bilateral Cooperation: The European Migrant Crisis The European migrant crisis provides an illustration of the application of the model. In the second half of 2015, and the first quarter of 2016, more than one million irregular immigrants made their way to EU countries, most through Turkey. Many migrants came from Syria, fleeing the growing violence, but migrants also came from other parts of the Middle East, Africa, and Asia, as well as from poorer and less stable parts of Europe. In the first instance, this flood of migrants has exacted a heavy toll on the EU and its Member States. The cost of processing more than a million asylum claims is staggering, as is the cost of food, clothing, housing, and healthcare in the interim and, for those successful claimants, the continued cost of integration. The political price has also been high. These countries have experienced a rise in new anti-immigrant right-wing parties and the growing success of already established anti-immigrant right-wing parties. The soon to occur withdrawal of the UK from the EU, cemented by the June 2016 referendum, was fueled in part by anti-immigrant sentiment stirred by the European migrant crisis.14 In Germany, Chancellor Angela Merkel’s approval rating sank to a five-year low in 2016, after enjoying strong support over more than a decade of governing (Sharkov 2016). The radical right-wing party AfD received 13 percent of the vote in the 2017 German election, making it the third largest party in the Bundestag and the first far-right party to enter the Bundestag in sixty years. These costs elicited action on the part of the EU to negotiate an agreement with Turkey. We describe the trajectory of the negotiations below. One method of reducing the costs associated with migration has been bilateral readmission agreements (BRAs). These agreements require that states accept the return of not only their citizens but migrants who have transited their country to reach the destination country. Relative to migration flows, there are few BRAs in existence: the vast majority are negotiated between individual EU Member States and countries with significant migrant flows to those states. In contrast, the EU itself was only authorized to negotiate a small number of readmission agreements that would engage all Member States. One authorized agreement was with Turkey. However, because not all states were implicated in flows from Turkey, an agreement was not forthcoming until migrant crises raised the costs of the status quo. The European Commission received its negotiating directive for Turkey based on an EU Council meeting of November 28, 2002. Negotiations commenced in May 2005 and continued for eighteen months but ground to a halt in December 2006. It was only in response to an earlier European migrant crisis—flows from Turkey through the Western Balkan route and the Eastern Mediterranean route beginning in 2008—that the negotiations restarted (“European Migrant Crisis” 2016). Unilateral action continued apace with Greece ultimately constructing a wall on the land border between Greece and Turkey that was not marked by the flow of the Meriç River. In 2009 a new draft text was prepared and forwarded to the Turkish negotiators. This new agreement was signed on December 13, 2013 and entered into force on October 1, 2014. The agreement provided for reciprocal readmission, including third country nationals and stateless persons. However, the readmission of third country nationals, central to the 2015 European migrant crisis, would only commence on October 1, 2017 (European Commission 2012; European Union 2013).15 In response to the crisis, which mounted in the second half of 2015, the EU first negotiated the EU–Turkey Joint Action Plan, dated November 29, 2015. According to the European Commission (2015), the objective of the action plan was “to supplement Turkey’s efforts in managing the situation of massive influx of persons in need of temporary protection.” The three dimensions of the plan included: addressing the root causes of the Syrian influx; supporting Syrians and their host communities in Turkey; and “strengthening cooperation to prevent irregular migration flows to the EU” (European Commission 2015). The action plan notes the mobilization of 4.2 billion euros to support Syrian refugees in Turkey, in light of the 7 billion euros already expended by Turkey. Turkey agreed to police its borders to prevent the flow of irregular migrants toward the EU. Nonetheless, the flows continued and a new agreement with new incentives was needed to stem the tide. That agreement came in the form of the EU–Turkey Statement on March 19, 2016. This agreement brought the readmission of third country nationals into immediate effect more than eighteen months in advance of the official EU–Turkey readmission agreement date. The statement indicated that, “All new irregular migrants crossing from Turkey to the Greek islands as of 20 March will be returned to Turkey.” More importantly, Turkey again agreed and then acted to prevent the smuggling of third country nationals from its territory. The price paid by the EU was considerable and only agreed to in light of the high costs, financially and politically, of the more than a million irregular migrants that entered EU territory during 2015 and 2016. The EU agreed to speed up the disbursement of the 3 billion euros promised in the Action Plan and to provide an additional 3 billion euros to support the Syrian population in Turkey after the initial funding was spent.16 More importantly, the EU agreed to accelerate the visa liberalization road map and to lift the visa requirements for Turkish citizens at the latest by the end of June 2016 (Kirişci 2014). The Turkish accession process to the EU would also be “re-energized” (European Commission 2016). Visa-free travel for Turkish citizens is indeed a significant price to pay. Turkey is a country with more than 70 million citizens—larger than any EU country save Germany. The road map, however, does require Turkey “to implement in a full and effective manner the readmission agreement; manage the borders and the visa policy in such a manner as to effectively prevent irregular migration; have secure travel documents; establish migrant and asylum systems in line with international standards; have functioning structures for combating organized crime with a focus on migrant smuggling and trafficking in human beings; have in place and implement adequate forms of police and judicial cooperation with the EU Member States and the international community; and respect the fundamental rights of the citizens and foreigners, with a specific attention to persons belonging to minorities and vulnerable categories” (European Commission 2016).17 The European migrant crisis fits well with our theoretical expectations. The EU was authorized to negotiate a union-wide agreement with Turkey in 2002 but those negotiations went nowhere until flows from Turkey mounted and multiple EU Member States were affected. Then, a readmission agreement was negotiated, but the side payments were insufficient to actually achieve the readmission of third country nationals. The rise in irregular migration to Europe began in earnest in the second half of 2015, with more than one million migrants arriving on EU shores during the following months. Europe stepped up its pressure on Turkey to halt the flows, promising 3–4 billion euros to support Syrians in Turkey in the October 2015 Joint Action Plan. That payment was insufficient to elicit the needed cooperation, and the EU upped the ante to an additional 3 billion euros and a promise of visa-free travel in the EU–Turkey Statement. Institutional Power and Sending State Preferences: The ICRMW Migrant rights are an issue area in which receiving states prefer the status quo of national sovereignty, whereas sending states pay growing attention to the rights of their nationals abroad. This is reflected in the paucity of bilateral, regional, and multilateral agreements on migrant rights. Two multilateral ILO treaties have been negotiated but ratification is sparse: ILO Convention No. 97 of 1949 has been ratified by only fifty-one states. ILO Convention No. 143 of 1975 has received even fewer ratifications, currently just twenty-three. This evidence fits well with our theoretical frame. Here we trace the third multilateral treaty negotiations that illustrate the efforts of sending states to project their preferences on the international stage. The ICRMW expands the rights of both documented and undocumented workers beyond the workplace. Although the rights of documented workers exceed those of undocumented workers, the convention incorporates virtually all the rights included in the nine other “core” human rights treaties into a single document dealing with foreign residents (Nafziger and Bartel 1991; Niessen and Taran 1991; Cholewinski 1997). The story of the shift in migrant protection from the ILO to the UN General Assembly is well known and comes directly out of unhappiness with Convention No. 143 (Böhning 1991).18 Both sending and receiving countries were unhappy about aspects of Convention No. 143, but the impetus for change came from the developing world. Receiving states, in contrast, preferred to retain the ILO as the primary venue on worker rights. Mexico and Morocco were the primary promoters for a new multilateral migrant rights convention. Both states had significant populations of undocumented migrants in the United States and France, respectively, and were unhappy about the ban on undocumented migration and illegal employment contained in Part I of Convention No. 143 (Böhning 1991). These two countries led the charge to change the venue for negotiations on migration from the ILO to the UN General Assembly. They viewed the General Assembly of the UN as a more favorable environment for several reasons. UN conventions allow for ratification with reservations, whereas ILO conventions do not, which promised a better ratification record than reflected in Convention No. 143. According to Böhning (1991, 704), developing countries also wanted to avoid the ILO because: “1. The ILO would not propose anything that contradicted No. 143, which promised to close off remittances from undocumented workers; 2. The UN General Assembly had an automatic developing country majority. The ILO, with its tripartite representation, did not; 3. The ILO gave prominence to independent trade unions, which many developing countries did not like.” According to Lonnroth (1991), these states also wanted “to achieve a moral condemnation of some of the states of employment.” Central to our argument, we focus on the automatic sending country majority in the General Assembly, which meant that sending states would be able to endorse any negotiated convention without the support of receiving states and therefore would not have to compromise with receiving states on the text of the convention. In terms of our analysis, this is not a developing country coalition per se. Some states in the “Global South” have become receiving states, including the Gulf oil states starting in the 1970s and some East Asian “newly industrializing countries” beginning in the 1980s. Nonetheless, since the early 1950s, the vast majority of sending states are located in the “Global South” and, once they gained independence, they quickly came to outnumber receiving states in the UN General Assembly (Money and Lockhart 2018). Taking up the issue in the UN General Assembly was not a done deal. The emphasis was on spelling out the basic rights of migrant workers who are undocumented or in an irregular situation, and sending states argued that these basic rights lie primarily in the fields of civil and political rights or economic, social, and cultural rights rather than strictly on labor rights as is the case of the ILO. Morocco and Mexico worked for several years before obtaining a majority in the General Assembly to support General Assembly Resolution 34/172, which was adopted in December 1979 and established a working group to elaborate a new UN convention on the rights of migrant workers. The General Assembly working group on migrant workers was formed in October 1980. The Mexican ambassador to the UN was initially elected to chair the working group, a position he retained throughout the decade-long negotiations. The initial working draft was submitted by Mexico and Morocco, reflecting their leadership role and the priority they gave the issue. They were supported by the G77, including states with large emigrant populations. However, receiving countries in the working group were unhappy with the initial draft, which they viewed as condoning illegal migration and employment. Given that the receiving countries could not stop the working group from moving forward, a coalition of countries from the northern rim of Mediterranean and Scandinavian governments with social democratic parties worked to provide an alternative draft.19 This coalition came to be known as the MESCA group, and their draft became the working draft. While attentive to the fundamental human rights of migrants, regardless of their status in the host state, these states had for a “key objective … to discourage employers from seeking and hiring workers who are undocumented or in an irregular situation” (Böhning 1991, 702). The General Assembly finally adopted the International Convention on the Protection of Rights of All Migrant Workers and Members of Their Family in 1990, after a full decade of negotiations. However, Böhning (1988, 135) points out that the extension of rights achieved by the ICRMW is nominal. As of 2017, no major recipient state has signed or ratified the Convention, whereas fifty-one countries of emigration have become party to the Convention and an additional sixteen states have signed the Convention. Adherence to the Convention has not produced much in terms of additional protection on the ground for migrant workers, documented or undocumented.20 In light of the poor ratification and nominal oversight of the states that are party to the Convention, it would be difficult to call the Convention a success. This overview of the ICRMW is consistent with our hypotheses. When the status quo is preferred by powerful states, the status quo is likely to remain unchanged. When less powerful states prefer a change in the status quo, their ability to negotiate a multilateral treaty relies on their institutional power to achieve a majority. However, their meager power is insufficient to bring receiving states “on board,” so that these agreements remain poorly ratified. Over the nearly hundred-year history of formal multilateral institutions, few migrant rights conventions have been negotiated, despite an institutional context that acknowledged the significance of migrant rights. And, once negotiated, very few receiving states have signed onto the agreements. Most wealthy states that signed onto Convention No. 97 were sending states at the time of their ratification, even though their status has changed since then. Reciprocal Flows: The GCC The most prominent example of freedom of movement is the EU and the precursors to that agreement in the Nordic Union and the Benelux Union. The Trans-Tasman Travel Arrangement between Australia and New Zealand also permits freedom of movement. Given the similarities of conditions among the Member States, these examples fit with our theoretical framework. However, there is another example from the Global South that has been widely overlooked that illustrates well our theoretical propositions: the GCC.21 The GCC states are better known for the recruitment of workers, both high - and low-skilled, from other Arab states and, more recently, from Asia, as well as the concerns over the treatment of those workers, particularly low-skilled workers. The freedom of movement of citizens of Member States has received little attention. The cooperation among the six GCC Member States (Bahrain, Kuwait, Oman, Qatar, Saudi Arabia, and the United Arab Emirates) on labor mobility fits well with our model. Similarities in wealth, language, and culture, combined with low unemployment and a strong demand for labor, led to the negotiation of a freedom of movement clause in the economic agreement that accompanied the creation of the GCC in 1981. The GCC Member States have periodically adapted their regulations to account for the changing structure of national labor markets. Similar to the EU (prior to the accession of Central and Eastern European states in 2004), freedom of movement has not generated high levels of labor mobility. Wage differentials are insufficient to attract large flows in any direction. The Arabian Peninsula was peopled by Semitic tribes over many centuries, but the rise of Islam as a religious and political movement united the Peninsula in the seventh century. The intervening centuries brought innumerable conflicts among the various tribes and religious branches of Islam as well as between the Arab population and the Ottoman overlords. The establishment of independent states during the twentieth century set the stage for regional cooperation. Talks among the smaller Gulf oil states began in the early 1970s but it was not until 1981 that the Charter for the GCC was signed by the states of Bahrain, Kuwait, Oman, Qatar, Saudi Arabia, and the United Arab Emirates. The Charter begins with an acknowledgement of their common heritage. The states proclaim themselves: “fully aware of the ties of special relations, common characteristics, and similar systems founded on the creed of Islam which bind them” (Charter of the Gulf Cooperation Council 1981). There has even been an effort to promote a regional identity, the Khaleeji identity (Babar 2011). Yet this common heritage did not prevent the exclusion from the GCC of the seventh Arabian Peninsula country, Yemen. Central to our argument, the wealth disparities and potential for unilateral flows from Yemen prevented the inclusion of this country in the GCC and in the free movement provisions. Moreover, freedom of movement has not been extended to citizens of other Arab states which, according to many sources, reflect “centuries of common history, religion, and language [that] have resulted in a high degree of cultural, political and social integration in the Arab region” (Nassar 2010, 11; see also Babar 2011). The shock that disrupted the status quo ante was the growing oil wealth of the Gulf countries, especially after the oil price shocks of the 1970s, which catapulted these countries from low to high human development status in just two decades. The oil shocks of 1973 and 1979 transferred enormous amounts of wealth to these states, which then undertook vast development projects that required significant amounts of labor to implement. However, the populations of these states were small with low labor participation rates. In 1970, the entire region comprised only 7.8 million people, and labor market participation rates were between 30 percent and 40 percent. Moreover, formal employment was concentrated in the public sector. Thus, there was an enormous need for labor to complete the development projects that would diversify these oil economies (Winckler 1997). Responding to this demand for labor, the national strategies were two-fold: the first was to increase natality and education—to grow their own labor force. The second strategy was to import labor, at least in the interim. These strategies attest to the low unemployment despite the lack of systematic records of unemployment rates during this period, a fact which is confirmed by Winkler (1997, 2010). The import of labor included Arabs, especially Palestinians. But the GCC states ultimately turned to Asians to fill low-skilled labor positions. The Asian population working in the GCC states rose from 342,589 in 1975 to 3,258,500 ten years later, in 1985 (Winckler 1997). In the midst of this economic explosion, the six Gulf oil states negotiated the charter that created the GCC. At the same time, the six countries also adopted an economic agreement providing an impetus for economic integration. The Free Trade Agreement took effect in 1983. Article 8 provides for “freedom of movement, work, and residence” as well as “the freedom of engagement in economic activity.” It is difficult to confirm with certainty that freedom of movement is fully implemented. Nonetheless, several sources appear to agree that freedom of movement is partially or fully implemented (Babar 2011; Haftel 2012).22 The implementation of the policy was slow. A protocol was signed in 1993 that ensured equality treatment of GCC nationals, initially in the private sector where nationals were generally employed in very low numbers, followed by a protocol to facilitate employment and free movement. The guarantees of equal access were extended to the public sector in 2000. In 2001, the six states negotiated the Unified Economic Agreement to deepen the economic integration. As the population increased and was educated, reflecting earlier policy choices of the state governments, young GCC nationals began to enter the job market; states created labor market “nationalization” policies to ensure that they found employment. The 2001 Unified Economic Agreement ensured that GCC nationals were treated equally in terms of “nationalization targets,” goals for employment of nationals. This was followed in 2004 by the extension of social insurance to GCC nationals. Thus, the GCC Supreme Council appears to have taken very seriously efforts to ensure freedom of movement. Nonetheless, as is characteristic of other regions with freedom of movement, where living standards and wages are similar, the impetus to move is slight. As of 2013, only 35,000 nationals of GCC countries were living in a Member State different than their country of origin (Gulf Cooperation Council 2014). Our story of the construction of the GCC economic agreement that incorporates freedom of movement points to rising costs associated with labor market shortages that potentially could be filled by citizens of Member States. The similarities in wealth when regional cooperation was first established are substantial, despite the small population base, which distorts differences in income. The status of Arabian Peninsula countries is provided in Table 3. Table 3 GCC countries Country GDP/pc/PPP 1985–1988 Population 1981 GNI/pc/PPP 2013 HDI2013 Bahrain $9,490 373,000 $32,072 0.813 Kuwait $9,310 1,448,000 $85,826 0.813 Oman $9,290 247,000 $42,191 0.781 Qatar $11,800 1,217,000 $119,029 0.850 Saudi Arabia $9,350 10,521,000 $52,109 0.839 United Arab Emirates $19,440 1,092,000 $56,068 0.825 Total GCC population 14,898,000 Average across GCC $11,447 $64,549 Yemen (non-GCC) $2,410 (1992) 8, 344, 000 $3,945 0.499 Country GDP/pc/PPP 1985–1988 Population 1981 GNI/pc/PPP 2013 HDI2013 Bahrain $9,490 373,000 $32,072 0.813 Kuwait $9,310 1,448,000 $85,826 0.813 Oman $9,290 247,000 $42,191 0.781 Qatar $11,800 1,217,000 $119,029 0.850 Saudi Arabia $9,350 10,521,000 $52,109 0.839 United Arab Emirates $19,440 1,092,000 $56,068 0.825 Total GCC population 14,898,000 Average across GCC $11,447 $64,549 Yemen (non-GCC) $2,410 (1992) 8, 344, 000 $3,945 0.499 Source: UNDP 1990, 2001, 2014. Table 3 GCC countries Country GDP/pc/PPP 1985–1988 Population 1981 GNI/pc/PPP 2013 HDI2013 Bahrain $9,490 373,000 $32,072 0.813 Kuwait $9,310 1,448,000 $85,826 0.813 Oman $9,290 247,000 $42,191 0.781 Qatar $11,800 1,217,000 $119,029 0.850 Saudi Arabia $9,350 10,521,000 $52,109 0.839 United Arab Emirates $19,440 1,092,000 $56,068 0.825 Total GCC population 14,898,000 Average across GCC $11,447 $64,549 Yemen (non-GCC) $2,410 (1992) 8, 344, 000 $3,945 0.499 Country GDP/pc/PPP 1985–1988 Population 1981 GNI/pc/PPP 2013 HDI2013 Bahrain $9,490 373,000 $32,072 0.813 Kuwait $9,310 1,448,000 $85,826 0.813 Oman $9,290 247,000 $42,191 0.781 Qatar $11,800 1,217,000 $119,029 0.850 Saudi Arabia $9,350 10,521,000 $52,109 0.839 United Arab Emirates $19,440 1,092,000 $56,068 0.825 Total GCC population 14,898,000 Average across GCC $11,447 $64,549 Yemen (non-GCC) $2,410 (1992) 8, 344, 000 $3,945 0.499 Source: UNDP 1990, 2001, 2014. In 1985–1988, the poorest GCC country (Oman) had a GDP per capita that was 48 percent of that of the wealthiest (the United Arab Emirates). Taking the regional wealth average, which discounts the outlier UAE, the ratio rises to 81 percent—comparable to the ratios found in regional freedom of movement countries in Europe. What is additionally interesting is the place of the seventh Arabian Peninsula country that is not blessed with oil wealth. Yemen, the odd country out, had a GDP per capita that was only 12 percent of the GDP per capita of the wealthiest country. Moreover, even with their populations swollen by an enormous influx of international migrants, Yemen’s population was more than half of the GCC’s combined population in 1981. Admitting Yemen, even with the common history, language, and culture, would have generated substantial one-way migratory flows over which individual states desired unilateral control. As a result, Yemen was excluded. In later periods, wealth disparities among GCC countries broadened, as oil revenues are subject to economic shocks and because populations are small. Yet a second measure of similarities of living conditions, the Human Development Index, suggests highly similar standards of living (UNDP 2009). Yemen remains the outlier on the Arabian Peninsula and an outcast of the GCC and its freedom of movement provisions. It is interesting “to stumble across” the little-known policy of freedom of movement in the Global South and to see that it, too, fits well with the theoretical frame generated by the analysis of those dimensions of regionalization in the Global North that privilege freedom of movement. Conclusions Unlike other dimensions of globalization, voluntary international migration is not underpinned by a dense network of international treaties, institutions, or even informal norms. National sovereignty remains the most prominent feature. Caution is required, though, as our model applies only to voluntary migration. Other aspects of international population movement—the travel regime and the refugee regime—do not necessarily reflect these same dynamics. Voluntary international migration in the past 70 years has been underpinned by the pattern of flows from poorer, less stable countries to wealthier, more stable countries, dividing the globe into sending and receiving states and generating externalities that are predominantly bilateral in nature. We have argued that the status quo ante of customary international law frequently privileges the more powerful receiving states, thereby generating little desire for international cooperation. We describe three conditions likely to generate international negotiations and subsequent cooperation. First, when costs of the status quo ante increase, receiving states broker deals with sending states that generate the costs. In these circumstances, sending (or transit) states can often extract quid pro quo payments. Note, however, that these agreements tend to restrict rather than facilitate the movement of migrants, unlike similar agreements on trade and foreign direct investment that tend to enhance flows. Second, when sending states locate an international forum where decision rules favor larger numbers, they negotiate agreements that project their preferences onto the international stage. Receiving states actively negotiate to minimize their commitments but, almost without exception, these receiving states refuse to ratify the agreements that result. This undercuts the impact of these international agreements. Finally, where the standards of living are comparable, thereby generating reciprocal flows, labor market shortages, or other exogenous shocks may generate freedom of movement provisions to enhance labor market efficiency. We have provided illustrations of each of these phenomena and refer interested readers to a more detailed analysis in Migration Crises and the Structure of International Cooperation (Money and Lockhart 2018). So, what then, should we expect to see emerge from the consultations and negotiations over the Global Compact on Safe, Orderly and Regular Migration? Our analysis suggests a negative outlook. It is likely that an agreement can be negotiated, but it is also likely that any agreement will be ignored by the states whose implementation would be required to bring the agreement to life. There is the possibility that domestic advocacy groups might be able to use the Global Compact on Migration as a focal point for their own organizing, and they may find it to be a useful tool in creating domestic pressure on states to enact national laws that protect migrant populations. This can be done without presenting the challenge to national sovereignty that international obligations might. But, we remain unconvinced that the Global Compact will provide advocates with more leverage than more widely ratified agreements (like the Universal Declaration of Human Rights, or the many existing ILO instruments). Regardless, we do not foresee an increasingly institutionalized international migration regime emerging from the process. We are not uniformly pessimistic, however. International cooperation is only one possible route to protecting migrants, not necessarily a goal in and of itself. Migrant protections can be improved everywhere, but there remains substantial variation among the fates of migrants globally. This variation can be attributed in large part to local conditions. Thus, we suggest that, given the limited possibility of international cooperation on voluntary migration, mobilization should focus at the national and local level. And, given the dire situations of so many migrants around the world, those of us that care about migrants should direct our energies at mobilizing for migrant protections where it is going to be most effective. Footnotes 1 Voluntary international migration accounts for 90 percent of individuals living outside their country of birth (UNDP 2009). 2 Alexander Betts (2014) utilizes the term “survival migration” to encompass a broader notion of “forced” migration where human lives are at risk. 3 In 2015 and 2016, at the height of the European Migrant Crisis, 2.5 million people applied for asylum in the EU. In 2016 (most 2015 asylum-seekers were also processed in 2016), EU states processed 1.1 million asylum decisions, and 39 percent of claims were denied. Over the course of those two years, nearly 1.03 million people were told to return home (most did not leave). These figures do not include migrants who were denied entry (388,000 in 2016); successfully entered, avoided detection, and never claimed asylum; or entered legally for other reasons (European Union 2017). Irregular migration is notoriously difficult to track, but it is clear that most of those entering Europe, even during these years of peak refugee flows, would not be classified as refugees if they claimed this status. Prolonged conflict, exclusion, and economic deprivation did, however, factor into the complex motivations of many of these migrants (Crawley and Skleparis 2018). 4 There is a large literature on how international regimes and institutions emerge to facilitate Tit-for-Tat reciprocity, particularly in providing information and transparency, magnifying reputational effects, and establishing or reinforcing issue linkages (Keohane 1984, 1986; Oye 1986; Haggard and Simmons 1987; Milner 1992; Martin 1994; Young 1996). Additionally, the literature on various cooperative strategies in response to particular types of other cooperation challenges (coordination difficulties, distributional issues, bargaining problems, etc.) is large (Krasner 1991; Snidal 1991; Morrow 1994; Fearon 1998, Martin 1999). The basic concept of Tit-for-Tat reciprocity remains very powerful. 5 NSAs include, but are not limited to, domestic and international NGOs as well as “policy entrepreneurs.” Intergovernmental organizations (IGOs), by definition, are creations of governments. 6 The vocabulary for these two types of states includes the dichotomies of sending/receiving, source/host, and origin/destination. We employ these dichotomies interchangeably to provide some variation in vocabulary. 7 The international trade literature looks to the abundance and scarcity of factors of production relative to other states in the international system and labels some countries as labor abundant and capital scarce (with a comparative advantage in labor-intensive production) and other countries as capital abundant and labor scarce (with a comparative advantage in capital intensive production). These classifications are uncontroversial (Odell 2000). 8 Although the term freedom of movement implies that citizens of member states party to such an agreement can move and live at will, freedom of movement in the international sphere is generally constrained in some way. 9 EU Member States were permitted to restrict migration from new accession states for a seven-year period, but Britain was one of the three EU countries that did not, vastly underestimating the number that would come from Eastern Europe. 10 “Brexit” is the moniker for British exit from the EU, as the result of a referendum called in June 2016 by the UK government. 11 Total gross immigration to Germany held steady at about 600,000 between 2006 and 2009, before increasing by roughly 40 percent by 2011, to about 840,000. Most of this increase came from intra-EU immigration, which increased from 360,000 in 2009 to 550,000 in 2011 (Bertoli et al. 2013). 12 By the time the financial crisis hit, the wave of immigration from new ascension states in Eastern Europe had slowed and, in some cases, reversed. The financial crisis produced a new wave of migrants from Southern Europe (Spain, Italy, and Greece) who were, on average, more highly educated and easily absorbed into the economy than many other immigrant groups (Hanewinkel 2013). 13 We refer interested readers to our forthcoming book for a systematic evaluation of the theoretical arguments (Money and Lockhart 2018). 14 Immigration was not the only factor that contributed to the success of the referendum, although it may have been the decisive one. Dissatisfaction with the Labor Party, Boris Johnson’s endorsement, low youth voter turnout, and anti-elite sentiment also played a part. 15 For third country nationals with whom Turkey had a readmission agreement, the provisions of readmission took effect immediately. However, Turkey had few readmission agreements other than with the European countries. Moreover, individual readmission agreements negotiated between individual EU Member States and Turkey remained in force. 16 There is a discrepancy between the EU–Turkey Joint Action Plan, which mentions 4.2 billion euros, and the EU–Turkey Statement that indicates a prior commitment of 3 billion euros to which the agreement offers an additional 3 billion euros. These are both official documents of the European Union (European Commission 2015, 2016). 17 In light of the July 2016 coup attempt in Turkey and subsequent actions on the part of the Turkish government, the EU postponed the implementation of the road map. 18 There is a large secondary literature on various aspects of this Convention. Our discussion draws on Böhning (1988, 1991, n.d.); Cholewinski (1997); Cholewinski, de Guchteneire, and Pecoud (2009); Edelenbos (2009); Hasenau (1988, 1991); Hune (1985); Lonnroth (1991); Mattila (2000); Nafziger and Bartel (1991); Niessen and Taran (1991); and Taran (2000). 19 These countries included Greece, Italy, Portugal, and Spain and Finland, Norway, and Sweden. 20 See Edelenbos (2009) for an overview of the operational components of the Convention. 21 The OECS is the only other example in the Global South of an implemented freedom of movement agreement. A number of regional organizations in Africa and in Latin America have introduced freedom of movement in principle, but none is fully implemented (Money and Lockhart 2018). 22 Ibrahim (2010, 124) provides a note of caution, pointing to “the weak translation of these legislations and policies into effective execution.” Works Cited Axelrod Robert. 1985 . The evolution of cooperation . New York : Basic Books . Babar Zahra R. 2011 . Free mobility within the Gulf Cooperation Council . Doha, Qatar : Center for International and Regional Studies, Georgetown University School of Foreign Service in Qatar . Bertoli Simone , Brücker Herbert , Fernández-Huertas Moraga Jesús . 2013 . The European Crisis and Migration to Germany: Expectations and the Diversion of Migration Flows. IZA DP No. 7170. Bonn, Germany: Institute for the Study of Labor. Betts Alexander , ed. 2011 . Global migration governance . Oxford : Oxford University Press . Google Scholar CrossRef Search ADS Betts Alexander. 2014 . Survival migration. Failed governance and the crisis of displacement . Ithaca, NY : Cornell University Press . Böhning Roger . 1991 . The ILO and the New UN Convention on migrant workers: the past and future . International Migration Review 25 ( 4 ): 698 – 709 . Google Scholar CrossRef Search ADS PubMed Böhning Roger . n.d. A brief account of the ILO and policies on international migration . Geneva : ILO . Böhning W. R. 1988 . The protection of migrant workers and International Labor Standards . International Migration 26 ( 2 ): 133 – 46 . Google Scholar CrossRef Search ADS PubMed Borjas George J. 2001 . Does immigration grease the wheels of the labor market? Brookings Papers on Economic Activity 2001 ( 1 ): 69 – 119 . Google Scholar CrossRef Search ADS Castles Stephen , Haas Hein de , Miller Mark J . 2014 . The age of migration: International population movements in the modern world, 5th ed . New York : Guilford Press . Charter of the Gulf Cooperation Council . 1981 . https://www.files.ethz.ch/isn/125347/1426_GCC.pdf ( accessed April 6, 2018). Chetail Vincent. 2014 . The transnational movement of persons under general international law: Mapping the customary law foundations of international migration law. In Research Handbook on International Law and Migration , eds. Chetail V. , Bauloz C. , 1 – 72 . Cheltenham : Edward Elgar Publishing . Google Scholar CrossRef Search ADS Cholewinski Ryszard. 1997 . Migrant workers in international human rights law: Their protection in countries of employment . Oxford : Clarendon Press . Cholewinski Ryszard , Guchteneire Paul de , Pecoud Antoine , eds. 2009 . Migration and human rights: The United Nations Convention on migrant workers’ rights . Cambridge and Paris : Cambridge University Press and UNESCO . Google Scholar CrossRef Search ADS Crawley Heaven , Skleparis Dimitris . 2018 . Refugees, Migrants, Neither, Both: Categorical Fetishism and the Politics of bounding in Europe’s ‘migration crisis.’ . Journal of Ethnic and Migration Studies 44 ( 1 ): 48 – 64 . Google Scholar CrossRef Search ADS Drahos Peter. 2003 . When the Weak Bargain with the Strong: Negotiations in the World Trade Organization . International Negotiation 8 ( 1 ): 79 – 109 . Google Scholar CrossRef Search ADS Edelenbos Carla. 2009 . Committee on Migrant Workers and Implementation of the ICRMW. In Migration and Human Rights: The United Nations Convention on Migrant Workers’ Rights , eds. Cholewinski Ryszard , Guchteneire Paul de , Pécoud Anoine . Paris and Cambridge : UNESCO and Cambridge University Press . European Commission . 2012 . Proposal for a Council Decision Concerning the Conclusion of the Agreement Between the European Union and the Republic of Turkey on the Readmission of Persons Residing Without Authorization. Brussels: COM (2012) 239 final (22 June 2012). http://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX:52012PC0240 (April 6, 2018). European Commission . 2015 . “EU–Turkey Joint Action Plan,” Press Release Database (October 15, 2015). http://europa.eu/rapid/press-release_MEMO-15-5860_en.htm. European Commission . 2016 . “Factsheet on the EU–Turkey Statement,” Press Release Database (March 19, 2016). http://europa.eu/rapid/press-release_MEMO-16-963_en.htm ( accessed April 6, 2018). European Migrant Crisis. 2016 . Wikipedia. https://en.wikipedia.org/wiki/European_migrant_crisis ( accessed April 6, 2018). European Parliament . 2017 . “EU Migrant Crisis, Facts and Figures,” News, June 30, 2017. http://www.europarl.europa.eu/news/en/headlines/society/20170629STO78630/eu-migrant-crisis-facts-and-figures ( accessed April 6, 2018). European Union . 2013 . Agreement between the European Union and the Republic of Turkey on the readmission of persons residing without authorization. Official Journal of the European Union 57(L134): 3. Fearon James D. 1998 . Bargaining, enforcement, and international cooperation . International Organization 52 ( 2 ): 269 – 305 . Google Scholar CrossRef Search ADS Gulf Cooperation Council. Secretariat General . 2014 . The process and achievements . Riyadh, Saudi Arabia : GCC Division of Information Affairs . Haftel Yoram. 2012 . Regional economic institutions and conflict mitigation: Design, implementation, and the promise of peace . Ann Arbor : University of Michigan Press . Google Scholar CrossRef Search ADS Hanewinkel Vera. 2013 . EU internal migration before and during the economic and financial crisis – an overview . Bonn, Germany : Bundeszentrale für politische Bildung (Federal Agency for Civic Education ). Hasenau M. 1988 . Setting norms in the United Nations System: The draft convention on the protection of the rights of all migrant workers and their families in relation to ILO in standards on migrant workers. International Migration 26 ( 2 ): 133 – 57 . Google Scholar CrossRef Search ADS PubMed Hasenau Michael. 1991 . ILO Standards On Migrant Workers: The fundamentals of the UN Convention and their genesis . International Migration Review 25 ( 4 ): 687 – 97 . Google Scholar CrossRef Search ADS Haggard Stephen , Simmons Beth A . 1987 . Theories of international regimes . International Organization 41 ( 3 ): 491 – 517 . Google Scholar CrossRef Search ADS Hatton Timothy J. 2007 . Should we have a WTO for international migration? Economic Policy 22 ( 50 ): 339 – 83 . Google Scholar CrossRef Search ADS Hatton Timothy J. , Williamson Jeffrey G . 2003a . Demographic and economic pressure on emigration out of Africa . Scandinavian Journal of Economics 105 ( 3 ): 465 – 86 . Google Scholar CrossRef Search ADS Hatton Timothy J. , Williamson Jeffrey G . 2003b . What fundamentals drive world migration? Discussion Paper No. 2003/23, World Institute for Development Economics Research. Hollifield James F. 2000 . Migration and the ‘New’ international order: The missing regime. In Managing migration: Time for a new international regime? ed. Ghosh Bimal . Oxford : Oxford University Press . Hune Shirley. 1985 . Drafting an International Convention on the protection of the rights of all migrant workers and their families . International Migration Review 19 ( 3 ): 570 – 615 . Google Scholar CrossRef Search ADS PubMed Ibrahim Badr El Din A. 2010. Intra-national labour mobility among Arab Gulf Cooperation Council States. In Intra-Regional Labour Mobility in the Arab World . Geneva : International Office for Migration . IOM (International Organization for Migration). 2017. Global compact for migration. https://www.iom.int/global-compact-migration ( accessed April 6, 2018). Keohane Robert . 1984 . After Hegemony: Cooperation and discord in the world political economy . Princeton, NJ : Princeton University Press . Keohane Robert . 1986 . Reciprocity in international relations . International Organization 40 ( 1 ): 1 – 27 . Google Scholar CrossRef Search ADS Kirişci Kemal. 2014 . Will the readmission agreement bring the EU and Turkey together or pull them apart? Centre for European Policy Studies. Koremenos Barbara , Lipson Charles , Snidal Duncan . 2001 . The Rational design of international institutions . International Organization 55 ( 4 ): 761 – 99 . Google Scholar CrossRef Search ADS Koslowski Rey , ed. 2011 . Global mobility regimes . New York : Palgrave McMillan . Google Scholar CrossRef Search ADS Krasner Stephen D. 1976 . State power and the structure of international trade . World Politics 28 ( 3 ): 317 – 47 . Google Scholar CrossRef Search ADS Krasner Stephen D. 1991 . Global communications and national power: Life on the Pareto Frontier . World Politics 43 ( 3 ): 336 – 56 . Google Scholar CrossRef Search ADS Lake David A. 2009 . Open economy politics: A critical review . Review of International Organizations 4 ( 3 ): 219 – 44 . Google Scholar CrossRef Search ADS Lonnroth Juhani. 1991 . The international convention on the rights of all migrant workers and members of their families in the context of international migration policies: An analysis of ten years of negotiation . International Migration Review 25 ( 4 ): 710 – 36 . Google Scholar CrossRef Search ADS Martin Lisa L. 1994 . Heterogeneity, linkage and commons problems . Journal of Theoretical Politics 6 ( 4 ): 473 – 93 . Google Scholar CrossRef Search ADS Martin Lisa L. 1999 . The political economy of international cooperation. In Global public goods: International cooperation in the 21st century , eds. Kaul Inge , Grunberg Isabelle , Stern Marc A. , 51 – 64 . New York : United Nations Development Programme (UNDP ). Mattila Heikki S. 2000 . Protection of migrants’ human rights: principles and practice . International Migration 38 ( 6 ): 53 – 71 . Google Scholar CrossRef Search ADS Migration Policy Institute . 2012 . Migration information source. http://www.migrationpolicy.org/programs/migration-information-source ( accessed April 6, 2018). Milner Helen. 1992 . International theories of cooperation among nations: Strengths and weaknesses . World Politics 44 ( 3 ): 466 – 96 . Google Scholar CrossRef Search ADS Money Jeannette , Lockhart Sarah . 2018 . Migration crises and the structure of international cooperation . Athens, GA : Georgia University Press . Morrow James D. 1994 . Modeling the forms of international cooperation: Distribution versus information . International Organization 48 ( 3): 387 – 423 . Google Scholar CrossRef Search ADS Nafziger James A. R. , Bartel Barry C . 1991 . The migrant workers convention: Its place in Human Rights Law . International Migration Review 25 ( 4 ): 771 – 99 . Google Scholar CrossRef Search ADS Nassar Heba. 2010 . Intra-regional labour mobility in the Arab World: An overview. In Intra-Regional Labour Mobility in the Arab World . Geneva : International Office for Migration . Niessen Jan , Taran Patrick A . 1991 . Using the new migrant workers’ rights convention . International Migration Review 25 ( 4 ): 859 – 65 . Google Scholar CrossRef Search ADS © The Author(s) 2018. Published by Oxford University Press on behalf of Munk School and Rotman School of Management. All rights reserved. For permissions, please e-mail: email@example.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)
Global Summitry – Oxford University Press
Published: Apr 20, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera