Labor Migrants as Political Leverage: Migration Interdependence and Coercion in the Mediterranean

How do states attempt to use their position as destinations for labor migration to inﬂuence sending states, and under what conditions do they succeed? I argue that economically driven cross-border mobility generates reciprocal political economy effects on sending and host states. That is, it produces migration interdependence . Host states may leverage their position against a sending state by either deploying strategies of restriction— curbing remittances, strengthening immigration controls, or both— or displacement— forcefully expelling citizens of the sending state. These strategies’ success depends on whether the sending state is vulnerable to the political economy costs incurred by host states’ strategy, namely if it is unable to absorb them domestically and cannot procure the support of alternative host states. I also contend that displacement strategies involve higher costs than restriction efforts and are therefore more likely to succeed. I demonstrate my claims through a least-likely, two-case study design of Libyan and Jordanian coercive migration diplomacy against Egypt in the aftermath of the Arab Spring. I ex- amine how two weaker Arab states leveraged their position against Egypt, a stronger state but one vulnerable to migration interdependence, through the restriction and displacement of Egyptian migrants.


Introduction
"The number of people makes the wealth of states." Frederick the Great "It's an easy decision for Mexico," wrote Donald Trump in his March 2016 memo to the Washington Post. In it, Trump detailed how he intended to force Mexico to finance a border wall between the United States and its southern neighbor. He would threaten to cut off its migrants' remittances. "Make a one-time payment of $5-$10 billion to ensure that $24 billion [in remittances] continues to flow into their country year after year … we have all the leverage … Mexico is totally dependent on the United States as a release valve for its own poverty" (Washington Post 2016).
Cross-border population mobility often generates both tensions and opportunities for leverage in global politics. The European Union (EU)'s muddled response to migrant and refugee flows across the Mediterranean Sea has profoundly shaped politics in many of its member states. In June 2017, a number of Arab states heightened immigration controls and engaged in collective deportations of Qatari citizens in order to exert diplomatic pressure on the Gulf state. Meanwhile, European citizens in Great Britain accuse the British government of using them as bargaining chips in the ongoing negotiations with Brussels over "Brexit." This article aims to expand our understanding of the interplay between migration and power politics by examining how labor migration features in coercive interstate relations. It addresses two main questions: How do states attempt to leverage their position as destinations for labor migration to influence sending states? Under which conditions do these attempts succeed?
I introduce the term migration interdependence to describe the reciprocal political economy effects produced by labor migration between sending and host states. I argue that there are two ways through which host states are able to manipulate migration interdependence in order to coerce a sending state into specific policy shifts. Firstly, a host state may pursue a strategy of restriction by limiting the outflow of migrant remittances into a sending state, adopting tighter immigration controls for citizens of a sending state, or doing both. Secondly, a host state may pursue a strategy of displacement by forcibly expelling citizens of a sending state. The effectiveness of restriction or displacement depends on whether a sending state is vulnerable to migration interdependence. Vulnerable sending states are more likely to comply with host-state demands, particularly if a host state exercises a strategy of displacement rather than one of mere restriction.
The article proceeds as follows: I review the relevant literature and present my theoretical model. I then introduce two least-likely cases, selected for theorytesting purposes through covariation and within-case analysis. In the first case, Jordan sought to coerce the Egyptian state into increasing the supply of natural gas into the country in 2012. In the second case, Libya aimed, in 2013, to coerce the Egyptian state into complying with the extradition of ancien régime elites, who had relocated their political and economic activities in Egypt. The two host states attempted to leverage their positions against the sending state through strategies of restriction and displacement of their respective Egyptian migrant labor communities.
In neither case did restriction strategies produce the desired policy shift. But displacement led Egypt to comply with both Jordanian and Libyan demands. Egypt arguably yielded to Jordanian and Libyan pressures because it was, at the time, vulnerable to migration interdependence. I find little support for alternative explanations, and I also apply the study's framework on two attempts at coercion via displacement against Egypt by Libya and Iraq in the pre-2011 period. I demonstrate that, at the time, the two host states failed to produce compliance because Egypt was not vulnerable to migration interdependence. Finally, I briefly examine how my argument might shed light on other instances where host states attempt to coerce sending states and discuss avenues for future research on the politics of migration interdependence.

Investigating Labor Migration as Interstate Leverage
International-relations theorists have only recently treated the effects of labor migration on global politics as a separate field of inquiry. The realist and neorealist tradition, in particular, traditionally approached labor migration as a "low" rather than a "high" politics issue (Hollifield and Wong 2014, 228), save for a few notable exceptions of scholars working primarily on immigration or security politics (Zolberg 1989;Weiner and Russell 2001; see for comparison Teitelbaum 2001). The international-relations scholarship tends to also examine host states at the expense of sending states-that is, "to focus on the consequences of immigration in wealthy, migrant-receiving societies, and to ignore the causes and consequences of migration in origin countries" (Castles, De Haas, and Miller 2014, 26). This hinders our ability to theorize the international politics of migration, as it obscures the role of economically driven cross-border population mobility in interstate relations.
A first line of work that ties migration to coercive interstate relations sees cross-border mobility in terms of existing power imbalances between sending and host states. Early scholars of dependency theory, as well as recent critics of globalization, focus on migration's contribution to the "development of underdevelopment" (Frank 1966). Rooted in Marxist political economy, such research examines migration flows as part of broader patterns of uneven trade between "developed (migration receiving) and less-developed (migration sending) countries" (Hollifield 2012, 366; see for comparison Sassen 1988). Wallerstein and other worldsystems theorists identify that labor migration sustains "relations of domination" between these two groups of nations, confirming the expectations of mainstream internationalrelations theory regarding stronger powers exerting pressure on weaker ones (Castles et al. 2014, 33). For instance, Mexican migration into the United States features as a main dimension in the power imbalances between the two states (FitzGerald 2009, 55, 167-69). The presence of 15-20 million Bangladeshi immigrants across India is indicative of the unequal distribution of economic power between the two states (Sadiq 2008). Although such work affirms the importance of labor migration in interstate relations, it approaches economically driven cross-border population mobility not as an instrument of foreign policy or coercion, but as part and parcel of global power imbalances. How might a host state leverage its position against a sending state for specific policy goals?
Working broadly within the field of global power politics (Berenskoetter 2007;Goddard and Nexon 2016), three separate research agendas allow us to identify potential leverage mechanisms; firstly, work on economic remittances long established the importance of such capital inflows for sending states (Orrenius and Zavodny 2012;Escribà-Folch, Meseguer, and Wright 2015;Mosley and Singer 2015, 293-97;Hollifield, Martin, and Orrenius 2014, 7). Economic remittances constitute a valuable source of foreign exchange and produce multiplier effects on the aggregate economy (Kapur 2010). They enhance access to public services (Adida and Girod 2011) and decrease demand for social spending (Doyle 2015). The importance of migrant remittances is frequently identified in the extensive literature on state-diaspora relations (Délano and Gamlen 2014;Adamson 2017;Koinova 2014;Koinova and Tsourapas, forthcoming). An extensive political economy literature also highlights how remittances may also constitute a form of rent distribution, particularly in the Middle East, and suggests that it may contribute to the stability of sending states' political regimes (Korany 1986;Richter and Steiner 2008;Soliman 2011). Warnings regarding the risks of overreliance on remittances, of Dutch disease effects, or of associated inflationary risks notwithstanding (Chami, Barajas, Cosimano, Fullenkamp, Gapen, and Montiel 2008), sending states habitually place significant value upon the ability to attract economic remittances (Castles et al. 2014, 75-83).
Consequently, host states attempting to exercise leverage against sending states may target the latter's reliance on remittances, mainly via disrupting migrants' ability to send earnings home. The rationale behind such a strategy of restricting remittances is similar to states' adoption of negative economic sanctions (Lindsay 1986;Hufbauer, Schott, and Elliott 1990), as both mechanisms aim to force target states' compliance on a range of issues via disrupting capital inflows. "Because economic sanctions can impose costs … without carrying the degree of risk attached to military actions," Martin (1993, 3) argues, "governments use them to signal resolve and exert pressure for policy changes. In a similar fashion, a host state may curtail migrants' financial transfers as leverage. There is a wealth of empirical examples of host states targeting the outflow of economic remittances for political purposes. The United States' embargo against Cuba prohibited migrants from dispatching remittances, while Cubans had to use third countries to transfer funds back home (Vanderbush and Haney 2005). When Japanese-North Korean relations deteriorated in March 1999 as Pyongyang prepared to test-fire a new longrange ballistic missile, Japan threatened to ban all migrant remittances to North Korea, estimated at $600 million to $1 billion per year (Miyashita 2003, 163).
Beyond migrant remittances, cross-border mobility serves a second political economy purpose for sending states. Political scientists identify that economically driven migration allows for an outflow of domestic labor, constituting what Turner described as a "safety valve" (1920), or the reduction of labor surpluses and the easing of accompanying sociopolitical and economic pressures. Drawing on Hirschman's thesis on "voice" versus "exit," research demonstrates that sending states find it preferable to encourage the "exit" of disenfranchised social actors rather than allow them to "voice" their grievances (Hirschman 1970, 15-43;for comparison Dowding, John, Mergoupis, and Vugt 2000, 478-80;Burgess 2012). Sending states often attempt to curb overpopulation or unemployment, in particular, by aiming to "export" citizens abroad. Such policies are evident in India, Bangladesh, Pakistan, Sri Lanka, as well as poorer Middle East states, which traditionally promote labor migration to the oil-producing Arab states (Weiner 1985, 445). The 1942The -1964 Bracero program allowed for the outflow of Mexican labor into the United States (FitzGerald 2009, 48-50). In the second half of the twentieth century, Tunisia and Morocco began fostering emigration to Europe for similar political economy reasons (Brand 2006, 17), while Western European countries, notably Germany, engaged in recruitment of labor from Turkey and other Mediterranean states (Martin 1991). 1 By extension, I expect host states aiming to leverage their position against a sending state to also attempt to restrict the outflow of its citizens abroad. An extensive literature exists on various methods of immigration control; a host state may decide to curb immigration flows from a specific sending state through the tightening of border controls, an increase in security and policing, the introduction of new visa requirements or the strengthening of existing ones, the criminalization of irregular migration, and so on (Andreas and Snyder 2000). Following Turkey's shooting down of a Russian Su-24M military aircraft near the Turkish-Syrian border on November 24, 2015, for instance, President Putin responded by suspending visa-free travel for Turkish citizens (Schenk 2016, 481-86). The strengthening of immigration controls, when employed within a coercive intestate context, is similar to a potential disruption of remittances outflows; by restricting sending states' access to its labor market, a host state aims to elicit their compliance.
Finally, beyond a strategy of restriction-disrupting the outflow of migrant remittances, strengthening immigration controls, or both-a host state may adopt a second strategy, namely the expulsion of a sending state's migrant population, either in part or in its entirety. This occurs either through the deportation of a sending state's migrant community or through the implementation of policies that render a migrant group susceptible to deportation. Such policies may involve novel provisions on required documentation or on immigrants' family history (on the politics of deportation, see Ellermann 2009;De Genova and Peutz 2010;Oyen 2015, 99-127). Security studies scholars approach such policies of "demographic engineering" as extreme measures (Weiner and Teitelbaum 2001, 55-56, 65-74), for they are able to target both a state's reliance on migrant remittances and its ability to allow the outflow of domestic labor at the same time, effectively "capacity-swamping" it (Greenhill 2010, 39). Russia's 2008 law On the Legal Status of Foreign Citizens includes the option to declare certain foreigners "undesirable" (nezhelatel'nyi). Only in 2015, it was estimated that it deported 117,493 foreigners, particularly citizens of Central Asian and Western states (MVD Rossii-Grazhdanskoe Sodeistvie). Similarly, foreign policy analysts highlight how Saudi Arabia, Libya, and other Arab states do not hesitate to deport certain states' migrants, when bilateral relations deteriorate (Kapiszewski 2001;Thiollet 2011;Tsourapas 2015a;. 2 Under what conditions would sending states comply with host states' demands? Two research agendas are relevant. Firstly, work on linkage and leverage suggests that strong states are particularly successful when exercising coercion against weaker states. Levitsky and Way's seminal analysis of linkage and leverage mechanisms (albeit without examin-1 Similar to research on the effects of economic remittances on sending states, a debate continues on the benefits of such outward mobility, with a number of scholars arguing that it constitutes a form of "brain drain" for sending states. The article is merely interested in identifying the perceived importance of emigration for sending states. 2 At the same time, labor migration within the Middle East also arguably constitutes an element of cooperation-rather than coercion-between resource-rich/labor-poor and resource-poor/labor-rich autocracies (for comparison, Erdmann, Bank, Hoffmann, and Richter 2013; von Soest 2015; Tansey 2016). However, this is beyond the scope of the study's research questions.
ing cross-border mobility) focuses on how the West, namely the EU (Levitsky and Way 2011, 87-129) or the United States (130-80), exercised pressure upon weaker post-Cold War states. Some scholars examine "suasion" games that focus on the dilemmas facing a hegemonic power (indicatively, Martin 1992). When would weaker host states be able to leverage their position against stronger sending states? A small literature has theorized that weak states may coerce stronger states within conditions of military conflict and counterinsurgency (Mack 1975;Arreguin-Toft 2005;for comparison, Paul 1994), but issue-linkage in the field of labor migration has been examined only with regard to stronger states' "migration diplomacy" strategies (Tsourapas 2017). Similarly, Greenhill (2002Greenhill ( , 2010 does not incorporate weaker host states in her analysis and focuses primarily on forced migration. Secondly, work on economic coercion suggests that interstate economic linkages may produce political compliance, but may also produce resistance or nationalism. Not unlike the debated effectiveness of economic sanctions as a policy of coercion, placing restrictions on migrant remittances or strengthening immigration controls may have little effect on sending states' compliance, as the cases of Cuba or North Korea demonstrate. As a result, relevant work, particularly within security studies, is unclear in its conclusions. Brand, for instance, mentions how "it may make sense to think of … Jordanian expatriate labor in Kuwait" as a form of "economic statecraft," but she does not carry this argument any further (Brand 2013, 148). Oyen (2015) examines the interaction between US foreign policy and the deportation of Chinese citizens during the Cold War, but does not specify whether expulsion constitutes a host-state strategy or an outcome. Greenhill establishes that deportation produces compliance when employed by sending or transit states against liberal democratic host states, but her findings on its effect against sending states are inconclusive (see coding in 2010, 310-11, 312-13, 321). The following section provides a novel theoretical framework that explains the conditions under which such strategies will likely succeed in producing sending-state compliance.

Theorizing Migration Interdependence and Coercion: Sending-State Vulnerability to Restriction and Displacement Strategies
This section brings existing work on migration in conversation with the international-relations literature on economic interdependence. It puts forth a novel approach to understanding host states' use of coercion against sending states and the conditions under which it is successful. I identify how economically driven cross-border mobility generates reciprocal political economy effects on sending and host states, or migration interdependence. I argue that a host state may leverage its position against the sending state by either deploying strategies of restriction-curbing remittances, strengthening immigration controls, or bothor displacement-forcefully expelling citizens of the sending state. The success of these strategies depends on whether the sending state is vulnerable to the political economy cost incurred by the host state's strategy, namely if it is unable to absorb the incurred cost domestically and cannot procure the support of alternative host states. However, I contend that displacement is more likely than restriction to change the behavior of the sending state, because of the higher costs involved (see Table 1). The previous examination of cross-border mobility and interstate relations points to the need to conceptualize labor migration as form of trade between sovereign nations that generates interdependence between sending and host states (Hollifield 2012, 366; for a history of the concept, see Cooper 1972;Baldwin 1980). Interdependence is approached here as "reciprocal effects among countries or among actors in different countries" (Keohane and Nye 2012, 8). Similar to financial transactions or foreign trade, I expect economically driven cross-border population mobility to engender political economy effects among states, or migration interdependence. Drawing on the literature on states' labor emigration and immigration policies, I define migration interdependence in a minimalist fashion, as the reciprocal political economy effects arising from crossborder flows of people between a sending and host state. 3 According to Hirschman's (1945, xvi) landmark study, the international trade system is "vulnerable to the will of any government so minded to use it in the pursuit of power." Rather than constitute a Millian force for international cooperation, "commerce can become an alternative to war … by providing a method of coercion of its own in the relations between sovereign nations" (14-15). Hirschman also lays out the two mechanisms through which a state may employ trade as an instrument of power against a target state in conditions of interdependence, consisting "mainly of the threat of severance and ultimately of actual interruption of external economic relations with that country" (16).
While Hirschman does not discuss cross-border population mobility, I introduce his framework into migration interdependence. Building on existing work outlined in the previous section, I argue that a host state may leverage its position for coercive purposes against a sending state in two ways: either by reducing a sending state's migration interdependence through restriction, or by severing it completely through displacement. Restriction refers to host-state policies of curbing migrant remittances, strengthening immigration flows, or both. It corresponds to Hirschman's "threat of severance," (16) given that migration interdependence is not severed; even at cases of extreme restriction, as in the case of the American embargo against Cuba, for instance, Cuban citizens were able to remain in the United States and had devised ways to send remittances home via third countries. Displacement refers to host-state strategies of deportation and corresponds to Hirschman's expectation of "actual interruption of external economic relations" with the sending state (16).
When are these two mechanisms of coercive migration diplomacy successful in producing sending-state compliance? I introduce two key terms from the literature on interdependence: sensitivity and vulnerability. For Keohane and Nye, sensitivity "involves degrees of responsiveness within a policy framework"; as a result, "sensitivity interdependence is created by interactions within a framework of policies" (Keohane and Nye 2012, 12). The sensitivity of Western states' economies to increased oil prices in the 1970s was "a function of the greater costs of foreign oil and the proportion of petroleum they imported" (12). Similarly, in this article's framework, the existence of migration interdependence constitutes the threshold for establishing a sending state's sensitivity to migration interdependence. The sensitivity of the Mexican state to Washington's 1964 termination of the Bracero Program was a function of the political economy costs incurred once the United States restricted further immigration of Mexican contract laborers. Ghana's sensitivity to Nigeria's 1983 displacement of African labor immigrants was a function of the political economy costs incurred once Nigeria displaced more than one million Ghanaian citizens.
While sensitivity assumes that the policy framework remains unchanged, "the vulnerability dimension of interdependence rests on the relative availability and costliness of the alternatives that various actors face," and "can be measured only by the costliness of making effective adjustments to a changed environment over a period of time" (13). "The fact that the United States imports approximately 85 percent of its bauxite supply does not indicate American vulnerability to actions by bauxite exporters," Keohane and Nye argue, "until we know what it would cost (in time as well as money) to obtain substitutes" (13). In the context of migration interdependence, therefore, one would expect that sending states' vulnerability is established by their inability to make effective adjustments with regard to the political economy costs incurred by the change in the policy framework through a host-state strategy of restriction or displacement. In particular, two conditions must be satisfied in order for a sending state to become vulnerable to a host-state strategy: the sending state needs to be unable to both absorb the incurred cost domestically and to procure the support of alternative host states. 4 Empirical evidence attests as to why the two conditions' coexistence is necessary for establishing sending states' vulnerability interdependence. A state may not be able to find alternative states willing to bear the cost of a host-state strategy of restriction or displacement, but that would not render it vulnerable if it is able to absorb the cost of such a strategy domestically. For instance, European states' increasingly restrictive labor immigration policies in recent decades reduced Turkish immigration, but did not necessarily affect the vulnerability of the Turkish state; despite being "the main source of non-EU migration in the EEC between 1960 and 1990, Turkey has entered a migration transition" (Castles et al. 2014, 179) recent research suggests. "As a consequence of relatively high political stability and rapid economic growth, Turkey has transformed 4 One could argue that a sending state's ability to contain any political costs incurred by a host-state strategy (strikes, sit-ins, deportees' protests) is also important. However, I view such costs as spillover effects from a sending state's inability to absorb the political economy cost domestically or shift it to alternative host states. from a net emigration into a net immigration country" (Castles et al. 2014, 179). The strength of the Turkish economy suggests that it was able to bear the costs of European states' restrictive policies domestically. Similarly, a state's weak domestic economy would not matter if it is able to find alternative host states willing to bear the cost of restriction or displacement. President Trump's intentions to employ restriction and displacement of Mexican labor does not necessarily mean that the Mexican state will be vulnerable to such a strategy, even if it is unable to absorb the political economy cost of such strategies domestically. Canada's 2016 decision to lift visa requirements for Mexican citizens entering the country is expected to allow for a "surge of Mexican immigrants" unable to reach, or not allowed work in, the United States (Guardian 2016). The existence of an alternative host state suggests that Mexico will not necessarily be vulnerable to potential American strategies of restriction or displacement.
Most frequently, sending states' responses to host-state strategies involve both the identification of alternative host states and attempts to absorb (part of) the cost domestically; for instance, in 1985, Colonel Gaddafi expelled an estimated twenty thousand Tunisian workers from Libya (Vandewalle 2012, 144). Tunisia immediately launched emergency job programs for displaced migrants. Ismail Khelil, the country's planning minister, asserted that "it will take time, but we will integrate them into the economy" (quoted in New York Times 1985). At the same time, Tunisia diversified its emigration policy and encouraged formerly expatriate Tunisians to seek employment in Italy, which did not require a visa for entry at the time (Natter 2015, 7). Tunisia was not vulnerable to Libya's strategy of displacement. Sending states become vulnerable only once both conditions are met; when Yemen failed to denounce the Iraqi invasion of Kuwait at the United Nations Security Council (where it served as a nonpermanent member) in September 1990, Saudi Arabia expelled approximately eight hundred thousand Yemenis over the following two months. Yemen, diplomatically isolated and burdened with a struggling economy, barely provided for these returnees, as authorities sought the help of the Red Crescent Society and set up camps to house them (Okruhlik and Conge 1997, 560;Al-Rasheed 2010, 146). It was unable to relocate deported migrants elsewhere, as other Arab states followed Saudi's example and deported more Yemeni workers. Politically, the "monstrous" domestic upheaval that ensued has been identified as a major cause of the Yemeni Civil War (Dresch 2000, 186). This is not surprising as, arguably, Yemen was vulnerable to Saudi Arabia's displacement strategy.
As per Keohane and Nye, while sensitivity is not politically unimportant, "vulnerability interdependence includes the strategic dimension that sensitivity interdependence omits" (16). For the purposes of this study, vulnerability interdependence allows the theorization of the conditions under which a sending state is likely to comply with host-state coercion. When a nonvulnerable sending state is faced with hoststate coercion, via restriction and/or via displacement, its compliance is unlikely, as it is able to compensate for related political economy costs. When a vulnerable sending state is faced with host-state coercion via restriction, its compliance is only somewhat likely, as the strategy is merely a "threat" to sever migration interdependence, to borrow Hirschman's term. When a vulnerable sending state is faced with hoststate coercion via displacement, or an actual "severance" of interdependence, it is very likely to comply given the high political economy costs.

Methodology and Case Selection
I employ case-study methodology for the purposes of theorytesting through deduction, and I rely on covariation and within-case analysis (Bennett and Checkel 2015). Egypt constitutes the largest supplier of migrant labor in the Arab world (Tsourapas 2015a;. It is also one of the major sending states worldwide; it ranks as the eighth largest remittance-receiving state and received $18.32 billion in 2015 alone (International Monetary Fund 2016). Care has been taken so as to select two host states with diverse characteristics. Jordan is an established monarchy whose size is only 89,342 km 2 (or 0.7 percent of the total Arab world territory). Jordan also does not share a physical border with Egypt and is not an oil-exporting state. On the other hand, Libya in 2013 had just emerged from the first phase of a brutal civil war. It is one of the largest Arab states (at 1,759,540 km 2 , constituting 11.4 percent of the total Arab world territory), and shares a porous, 1,115-km border with Egypt. Since 1962, Libya is a member of the Organization of the Petroleum Exporting Countries.
The study focuses on Egyptian labor migration to Jordan and Libya under Mohamed Morsi, who became President in June 2012 under the Muslim Brotherhood-backed Freedom and Justice Party. In the first case, Jordan aimed to coerce the Egyptian state into increasing the supply of natural gas into the country while, in the second case, Libya aimed to coerce the Egyptian state into complying with the extradition of ancien régime elites, who had relocated their political and economic activities to Egypt. The two host states attempted to leverage their positions against the sending state through strategies of restriction and displacement of their respective Egyptian migrant labor communities. While the strategy of restriction did not produce a sending-state policy shift in either case, the strategy of displacement led to Egyptian compliance with both Jordanian and Libyan demands.
The study treats the two cases as least-likely ones, which "can strengthen support for theories that fit even cases where they should be weak" (George and Bennett 2005, 121; for contrast Rohlfing 2012, 62;Eckstein 1975). According to existing research, neither Jordan nor Libya should have been successful in their attempts at coercion. For one, both host states are considerably weaker than Egypt (Vandewalle 2012, 86;Brand 2013, 2, 5). Egyptian gross domestic product (GDP) exceeded $286 billion in 2013; in contrast, Libyan GDP amounted to $65.5 billion in the same year, a bit less than twice the Jordanian GDP ($33.59 billion). There is also no evidence that Egyptian compliance was forced through the intervention of any stronger, third state (be it regional or international). At the same time, in 2012, Jordan and Libya were not major sources of economic remittances, nor were they hosting a large number of Egyptians at the time (see Figures 1 and 2 below; World Bank 2015a, 2015b). One could argue that Egypt complied with Libyan demands because of future oil rent rewards, but such an expectation would be years away, given the extent of the Libyan state's disintegration. At the same time, it would not explain Egypt's compliance with the demands of Jordan, which is not an oil-exporting state. An argument that Egypt complied with Jordanian demands because the issue of natural gas provision was not crucial would also be inaccurate: Egypt was plagued with energy problems and power outages throughout Morsi's rule. In May 2013, amid scattered protests and social media calls to avoid paying energy bills, Morsi admitted to the Associated Press that only 80 percent of Egypt's electricity needs were being met, saying "we have a real energy problem in Egypt" (Associated Press 2013). A long discussion exists on the potential pitfalls of the case-study method (Geddes 1990;Collier and Mahoney 1996), particularly if cases are selected on the dependent variable. Yet a significant body of political science work highlights how "in the early stages of a research program, selection on the dependent variable can serve the heuristic purpose of identifying the potential causal paths and variables leading to the dependent variable of interest" (George and Bennett 2005, 23). Interestingly, both cases exhibit incase variance (cf. Blatter and Blume 2008), demonstrating noncompliance to restriction and compliance to displacement. Covariation and within-case analysis are employed to substantiate the study's theoretical claims (Gerring 2016). Within-case analysis is well-suited to the "systematic examination of diagnostic evidence selected and analyzed" (Collier 2011, 823), particularly in enabling qualitative tools to assess the causal claims and mechanisms outlined in the previous section (for comparison, Beach and Pedersen 2013).
A final note on data collection: fieldwork in authoritarian contexts presents unique challenges (Kapiszewski, MacLean, and Read 2015, 218), particularly in light of the fact that regional migration is traditionally considered a security issue for Arab elites (Kapiszewski 2001;Tsourapas 2014). At the same time, research is plagued by a lack of detailed, publicly available statistical data on intra-Arab flows (Fargues 2014), as well as by the fact that migration management is handled at the highest levels of the executive (Feiler 2003). As Brand (2013, 8) wrote on seeking statistical data on the Jordanian political economy, "one works under the assumption that such documents will probably never be released, or may never have existed in the first place." Key elites that would be able to substantiate the study's findings are either unlikely to share any information, such as Jordan's King Abdullah II, or unavailable, such as deposed Egyptian President Mohamed Morsi, who is currently facing a death sentence conviction. To overcome these issues, I rely upon a meticulous collection of the coverage of regional labor migration issues in the 2012-2013 period across the three main Egyptian newspapers (al-Ahram, al-Akhbar, al-Gomhuriya). In an attempt to fill in gaps and substantiate Egyptian reporting, I also draw on non-Egyptian Arabic media sources, particularly Jordanian and Libyan coverage of labor migration. Finally, for the purposes of triangulation, I employ semistructured expert and elite interviews conducted in Cairo (n = 31) between July 2013 and June 2014 (for comparison Tansey 2007). This includes former Prime Minister Abdel Aziz Hegazy, current and former ministers, and high-ranking Egyptian government officials.

Jordanian-Egyptian Relations and the Provision of Natural Gas 5
According to a 2004 agreement between Cairo and Amman, Egypt is contractually obliged to supply natural gas to Jordan via the Arab Gas Pipeline to the amount of 240 million cubic feet of gas per day until 2019 (al-Ahram, June 3, 2013). Egypt found it impossible to continue the supply of natural gas at these levels in the aftermath of the 2011 Arab Spring due to domestic energy shortages and frequent attacks against the pipeline itself, which is also supplying gas to Israel at reduced prices. Once in power in June 2012, Morsi unilaterally decided to curb the export of natural gas to Jordan, which saw its reliance on Egyptian natural gas fall from 89 percent in 2009 to 17 percent by the end of 2012. "As a consequence, the state-owned electricity company, the National Electric Power Company (NEPCO), made huge losses, in 2011 amounting to 5 percent of gross domestic product (GDP)" and was faced with a $3.5 billion deficit (Henderson 2015, 3). It was estimated that "every time the supply is halted as a result of the frequent interruptions caused by the bombing[s], it costs the Jordanian treasury $1 million per day" (al-Monitor, April 2014). Indicatively, the supply of gas to Jordan was interrupted more than fifteen times between February 2011 and November 2012 period alone (al-Ahram, November 21).
In the bilateral conflict over this issue, Jordan employed its status as host state of Egyptian labor in its coercive migration diplomacy. Egyptian migrant labor in Jordan has a long history spanning back decades, when thousands of Egyptian workers travelled through Jordanian territory in order to reach Saudi Arabia and other oil-producing countries in the Gulf that had been recruiting Arab labor in the aftermath of the 1973 Arab-Israeli War (Amin and Awni 1986). Over the past few decades, Jordan also became a host state for unskilled Egyptian migrant labor. This has been due both to its fairly developed economy as well as to its position; many Egyptians that were either unable to enter the Gulf states, or did not wish to return to Egypt when their employment there was terminated, chose to pursue employment in Jordan. A key characteristic of these migrant laborers is that thousands of them lack legal documentation to validate their employment status in Jordan. In 2012, it was reported that only 167,000 possessed a legal working permit (Egypt Daily News, December 19).
Jordan employed a strategy of restriction and, following Egyptian noncompliance, displacement in its efforts to have Egypt shift its natural gas provisions back to the 2004 agreement's stipulations. In early December, Jordanian media and government sources began hinting at the need to restructure the domestic labor market. Jordanian Minister of Labor Nidal Qatamin declared "the necessity of launching a campaign in Jordan to regulate the labor market," (al-Ahram, December 9). The announcement focused particularly on Egyptian workers that were employed in the Kingdom without a permit. Did Jordan's decision to regulate the domestic labor market by targeting illegal Egyptian migrants constitute a strategy of coercion against Egypt? Or, could it merely have been driven by other factors, such as the deterioration of the post-Arab Spring Jordanian economy (for contrast Yom 2015)? Qatamin publicly stated that the regulation campaign was driven by a "security," not "political" rationale (Middle East News Agency, December 9). Yet, two pieces of evidence strongly suggest that this was a clear attempt at using migration as interstate coercion. First, targeted Egyptians had been pursuing employment within Jordan for years, if not decades, making the timing of Jordan's policy shift suspect. Second, numerous sources confirmed that Jordan's policy shift formed part of its broader strategy on securing Egyptian natural gas. Al-Jazeera reported an early December 2012 meeting that Abdullah held at the house of former minister Rajaii al-Maashar. The piece, entitled "Jordan Swaps Egyptian Labor for Gas," states the following: According to the sources that spoke to Al-Jazeera.net, the king revealed that his country is in possession of some cards that it is using to send messages to Cairo. Such cards include the presence of … Egyptian workers in Jordan and the fact that Jordan is the only passageway for the exported Egyptian vegetables towards Iraq … One of the meeting participants quoted the king as saying that the weak flow of the Egyptian gas and it being cut off Jordan "destroyed the Jordanian economy." He also indicated that the rise in debts over the past two years was mainly due to the losses incurred by the power plant that had to use heavy fuel instead of gas. (al-Jazeera, December 11) Al-Quds al-Arabi confirmed the meeting, adding that Abdullah mentioned how "tens of thousands of Egyptians working in the Gulf states go through [Jordanian town] Aqaba in their movements," which the Kingdom could further use this as a bargaining chip against Cairo (al-Quds al-Arabi, December 12). Given Egyptian inaction, Jordan adopted a strategy of displacement, as the government announced that any Egyptian workers found working illegally within the Kingdom would be subject to deportation. At the same time, the Ministry of Labor began targeting Egyptians who possessed valid permits but had been working in professions not included in their permits (Middle East News Agency, December 13). Al-Hayat quotes a number of sources that affirm how: The officials in Amman are very upset with their counterparts in Cairo, and this is why they have sent a signal to the Egyptians in order to express the dissatisfaction of the Jordanian leadership. Amman is upset about a number of issues, one of which being the fact that the gas pipeline from Egypt to Jordan is no longer functioning normally. Other official sources told al-Hayat that "thousands of Egyptian workers were arrested in Jordan during the last two weeks due to irregularities in their papers and sent back to their country." (al-Hayat, December 12) Deportations of Egyptian workers formally started on December 8, when approximately two hundred migrants lacking necessary documentation were expelled (al-Ahram, December 9 and 11). By December 14, it was revealed that Jordan had arrested hundreds of Egyptian workers, who were found in violation of their residency permits and were facing deportation. Following Abdullah, other Jordanian institutions underlined how the potential expulsion of Egyptian migrants was linked to Egypt's policy on natural gas. The Amman Center for Human Rights Studies (2012, 1) issued a telling declaration: "sadly the recent procedures of deporting Egyptian workers came as a result of poor political relationship between the two countries. While we stress the relationship with our neighbor Egypt," the report continued, "it is unacceptable for the new Egyptian leadership to stop the implementation of bilateral economic agreements and sacrifice [a] profound relationship." Elites also confirmed this state strategy: "official authorities in Amman were leaking information about the presence of [a number of] Egyptians working in Jordan without a license and in an illegal way," al-Quds al-Arabi reported. "The collective deportation of the latter would address a harsh message to the Egyptian government, and the Jordanian government chose the shorter path and arrested hundreds of Egyptian workers who were in violation of their residency conditions and indeed deported them." The newspaper's report concluded with a message from an unnamed, "prominent" Jordanian politician who wished to send a message to Cairo: "We beg of you. Our situation cannot tolerate having you tamper with the gas issue any further" (al-Quds al-Arabi, December 13).
As per the study's third claim, when faced with Jordan's strategy of displacement, Egypt complied with the host-state demands. Morsi called Abdullah from London, where he was on an official visit on December 9. This was the first time the two leaders engaged in conversation since Morsi's electoral victory. The Egyptian president reportedly guaranteed an increase in the supply of Jordan with Egyptian gas and agreed to create a joint bilateral committee on the issue, in order to appease Abdullah. Simultaneously, Morsi also dispatched Egyptian Prime Minister Hisham Qandeel to Amman, in an effort to solve this "silent crisis" (al-Quds al-Arabi, December 13). Jordanian Prime Minister Abdullah al-Ensour later reported to the media that Qandeel had reiterated Morsi's natural gas assurances to Abdullah and confirmed that "Egypt will always keep its pledges and has no intention to renege on any agreements" (al-Gomhuriya, January 19, 2013). According to al-Hayat, [t]he diplomatic contacts launched at the highest level between Jordan and Egypt succeeded in containing the crisis that had erupted between the two sides. In this respect, Egypt pledged to continue supplying Jordan with gas and Amman promised to stop deporting the thousands of Egyptian workers present in the kingdom … Cairo promised to keep supplying Jordan with gas after the misunderstanding that had erupted lately. The two states had signed agreements in this regard and President Morsi pledged to respect them. Jordan for its part said that it would no longer expulse the Egyptian workers whose papers are not in order. Following these contacts and the agreement that was reached, thousands of Egyptian workers were released, and [1,900] people were even allowed to go back to Jordan after being deported to Egypt. (al-Hayat, December 21) The Jordanian Ministry of Labor announced the formal halting of Egyptian workers' deportation, and the Egyptian Minister of Migration was dispatched to Jordan to resolve Egyptian migrants' remaining administrative problems. A few days later, Morsi publicly thanked Abdullah for "his decision to stop these measures" (al-Ahram, December 13). "Egypt and Jordan overcame a labor problem that could have strained bilateral relations," the Middle East News Agency wrote, "thanks to direct high-level contacts that were crowned by a personal phone call Egyptian President Mohamed Morsy had with King Abdullah II on Monday evening" (Middle East News Agency, December 13). A minor glitch in negotiations appeared in late January 2013, when Qandeel was visiting Jordan to sign a bilateral agreement on the legalization of Egyptian migrant workers. As was reported across Egyptian media, host-state elites identified "a delay of pumping Egyptian natural gas to Jordan," which presented a "new obstacle" in any bilateral migration agreement. Khaled Azhari, the Egyptian Minister of Manpower and Immigration, went on record to state that his Ministry will coordinate with the Egyptian Ministry of Petroleum "to remove any obstacles facing Egyptian laborers in Jordan" (quoted in al-Ahram, January 29, 2013).

Libyan-Egyptian Relations and the Extradition of Gaddafi-Era Elites 6
Early the following year, Libya also successfully employed restriction and displacement strategies in its coercive migration diplomacy against Egypt. The ousting of the Gaddafi regime in 2011 had created tremendous political turmoil in Libya. Once the first phase of the civil war ended, the General National Congress emerged as the country's legislative authority and attempted to consolidate the transitional process in Libya. The Congress identified that, since 2011, a number of ancien régime members of the Gaddafi years had fled Libya seeking refuge in Egypt. Egypt in the immediate outbreak of the Libyan civil war had become a safe haven for elites affiliated with the Gaddafi regime, who continued their economic and political activities from Cairo. The Congress believed that their extradition and trial in Libya was necessary for Libya's post-Gaddafi political transition. But Egyptian authorities resisted such extradition requests, citing a number of legal obstacles. For instance, Egyptian courts had forbidden the extradition of Gaddafi's cousin, Ahmed Qaddaf Alddam, because he claimed to have Egyptian citizenship. Similarly, other Libyan elites had already filed applications for Egyptian asylum, making a potential extradition to Libya legally difficult (see details in Reuters, April 16).
In contrast to Jordan, Libya had been a preferred destination for Egyptian workers seeking employment opportunities abroad since the early 1970s (Tsourapas 2015b ;Feiler 2003, 101). The proximity to Egypt, the porous border between the countries, and the ample wealth that flowed into the resource-rich but labor-poor country, which contains the largest oil reserves of any African state, allowed the country to become a major destination of Egyptian migrant labor. In the later years of Gaddafi's rule, immigration controls for Egyptians became particularly relaxed (Tsourapas 2017). Yet, Libya in 2012 contained only a small fraction of its earlier migrant labor force, mostly unskilled and low-skilled workers; in the post-2011 era, most Egyptians fled due to the brutal Libyan civil war that, coupled with the NATO-led military intervention, had contributed to the collapse of the oil industry.
In the second half of 2012, Libya let it be known publicly that remaining Egyptian migrants would be employed as a strategy against Egypt in case of noncompliance with the extradition requests. "A high-level source in [Egypt's] Ministry of Manpower and Migration has confirmed that Libya's threats to expel workers is not just a rumor," reported the al-Zaman daily in August 2012. "[The source] confirmed that there has been deafening silence on the part of officials, as concerns this issue and its being wrapped up with the critical and private political matter about the handover of senior members of the Gaddafi regime" (al-Zaman, August 28, 2012). Egypt downplayed the matter, while Farouk el Dessouky, head of the Union of Egyptians Abroad, asserted that "there is no abuse against Egyptians in Libya because of the crisis on the remnants of the Gaddafi regime" (al-Watan, October 19, 2012).
In early 2013, Libya engaged in a strategy of restriction seeking to shift Egypt's extradition policy. The General National Congress initiated the implementation of tougher measures against Egyptians working in the country and identified that many Egyptians lacked proper immigration or health certificates, were holding expired residency permits, or had gained entrance into Libya via fraudulent documents. At the same time, members of the Egyptian Christian Coptic community in Libya were harassed and arrested on illegal emigration charges (Daily News Egypt, March 2). In mid-March, an Egyptian Coptic Church in Benghazi was set on fire for the second time in 2013, while 55 Egyptian Copts were arrested on charges of proselytization and reportedly tortured. One of them, Ezzat Attallah, eventually died while in custody in a Tripoli prison (see full coverage in Middle East News Agency, March 14). The Egyptian government did not respond to the Libyan strategy of restriction, leading Libya to employ a strategy of displacement from late February onward; Egyptians found without proper documentation were declared subject to immediate deportation, as the so-called bilateral "visa crisis" escalated (al-Ahram, February 25). The assistant Egyptian foreign minister officially denied that Egyptians were facing any discrimination, although news broke that more than one hundred Egyptians had been deported on March 18, all of whom had reportedly paid 3,000 LE each (or $338, a sizeable sum) to purchase employment visas for Libya (al-Jumhuriya, March 18). Similar to the Jordanian case, Libya gradually intensified its displacement strategy; in late February, it formally closed the border to Egyptian migrants while, by late March, Libyan authorities were deporting more than four hundred workers daily on a variety of charges, even accusing some of being infected with hepatitis (Middle East News Agency, March 29).
Similar to Jordan, Libya's adaptation of displacement coerced Egypt into compliance. In late March, Egyptian authorities duly arrested the ex-Gaddafi Libyan officials, including Ali Maria, the former ambassador to Egypt, and Mohammed Ibrahim, the brother of senior Gaddafi-era official Ahmed Ibrahim. Ahmed Qaddaf Alddam himself was also arrested (AFP, 2013). The latter "handed himself in to the security forces after [ten] vehicles surrounded his house in the Zamalek neighborhood of the Egyptian capital," effectively nullifying the Egyptian courts' decision of nonextradition (BBC News, 2013). Twenty-three other Libyan citizens were immediately arrested and duly extradited. Al-Ahram cited officials employed in the Egyptian Interpol, who declared that these extraditions were conducted in return for Egyptian migrant workers jailed in Libya being released (al-Ahram, March 28). The timing of these arrests strengthens the claim that Egypt complied with Libyan demands because of the latter's use of displacement; Prosecutor General Tala'at Abdallah issued the extradition decision two days after a Libyan delegation arrived in Egypt on March 19, bearing an updated list of eighty-eight names of "Gaddafiera fugitives" (al-Ahram, March 22). They were returning a visit to Libya by a delegation representing Egyptian prosecution, which had been dispatched to investigate the arrests of Egyptian migrants and the death of Ezzat Attalah (Daily News Egypt, March 25). According to al-Hayat, "the Egyptian authorities wanted to send a positive message to their neighbors and that this was the main reason behind their decision to act now." The report continued: "it must be noted that in the last few months, bilateral relations between the two neighbors had greatly deteriorated" March 20).
Egyptian compliance with Libyan demands led to the restoration of the status quo ante in terms of migration interdependence, as had occurred in Jordan a few months earlier. The arrest of the Gaddafi-era elites in Cairo led Libya to reopen the border between the two countries, after keeping it closed for more than two months (al-Ahram, April 26 2016). Providing further proof of the connection between Egyptian labor migration and the forced repatriation of Libyan elites, "the Libyan authorities also promised to open their market to the Egyptian workers in case all the figures taking refuge in Egypt are handed over [emphasis added]" March 20). By June, the Egyptian Ministry of Manpower and Migration extolled the excellent bilateral relations between the two countries, going as far as to claim that Libya was now ready to receive "millions of Egyptians" (al-Ahram, June 7). The Libyan Deputy Central Bank Governor Ali Salem Hibri summed up the relationship between Libya and Egypt: "there are times when your upstairs neighbor has a leaking pipe and you need to help him fix it for your own peace of mind" (Reuters, April 16).

Explaining Egyptian Compliance to Host-State Coercion
The previous section identified how two host states leveraged their position in order to induce sending-state compliance via two mechanisms, restriction and displacement. Confirming the study's first claim, both Jordan and Libya employed the restriction and displacement of Egyptian migrant labor within their territory in order to induce Egyptian compliance on the issue of natural gas and the extradition of Libyan elites, respectively. Why would a more powerful state yield to the demands of two weaker states? Why was Egypt, a state that numbered more than three million labor migrants abroad in 2012, affected by the expulsion of a few thousand Egyptians? According to the study's second claim, a sending state is likely to comply with a strategy of displacement when it demonstrates vulnerable migration interdependence. Indeed, Egypt under Morsi demonstrated migration interdependence vulnerability for it satisfied the two conditions theorized above: firstly, it faced a lack of alternative states able to shoulder the political economy costs of Jordanian and Libyan strategies, and, secondly, its economy was not in a position to absorb these costs domestically.
With regard to the first condition for establishing vulnerability, the ruling Muslim Brotherhood was viewed with distrust, to say the least, by the vast majority of the conservative Arab monarchies due to their fear of a "domino effect" after the Egyptian and Tunisian revolutions and their historically hostile views on the Muslim Brotherhood itself (Ryan 2014). While Egypt had received diverse forms of aid from oil-producing Arab states, particularly Saudi Arabia, under Sadat and Mubarak, its regional isolation under the Morsi presidency strongly suggested that the post-2011 situation was completely different. As a result, it was highly unlikely that displaced Egyptian workers from Libya or Jordan would be redispatched to other oil-producing Arab states that had supported Egypt economically before the rise of the Muslim Brotherhood. In fact, not only were Arab oil-producing states unwilling to support Morsi by absorbing any costs, but Saudi Arabia, Kuwait, and the United Arab Emirates-the three main host states in 2012-were themselves engaging in a strategy of restriction against Egypt, by arresting Egyptian labor migrants with suspected ties to the Muslim Brotherhood (for more details, see National 2013).
With regard to the second condition for establishing vulnerability, the deteriorating condition of the Egyptian economy implied that the state was unable to absorb the burden of host states' displacement strategies. In November 2012, a few weeks before the Jordanian crisis, Morsi had turned to the International Monetary Fund (IMF) for a $4.8 billion loan in order to help Egypt stay afloat. But Egypt's inability to meet the Fund's loan requirements delayed the signing of the agreement until after Morsi was ousted from power. As Egypt's foreign reserves plummeted, economic remittances from Arab host states came to form a major part of the national economy, while Egypt's high unemployment rate, growing from 9 percent to 12.7 percent in 2012, implied that Egypt was increasingly reliant on continuing labor emigration as a socioeconomic "safety valve" and was unable to absorb any displaced workers domestically. While she refused to disclose any details on the Jordanian and Libyan cases, Nahed Ashry, Minister of Manpower and Migration, stated that "Egyptians [working] in Arab sister-countries are more important to the Egyptian state than ever before" (personal interview, February 24, 2014). It bears noting that Egyptian labor in both post-2011 Libya and Jordan has been typically employed in the unskilled and semiskilled sectors, which made reintegration into the Egyptian economy highly problematic. Retail, tourism, and the unskilled sector were already overburdened in Egypt, which indicates that the economic impact of displacement was amplified given the spiking unemployment rates, the increased demand on government services, and broader strains on infrastructure.
Two historical examples of earlier attempts at exploiting Egyptian migration interdependence help shed light on the importance of these two conditions in establishing vulnerability. Gaddafi would deport thousands of Egyptian workers out of Libya in the mid-1970s, whenever his relations with the Egyptian President Anwar Sadat deteriorated (Tsourapas 2017). When, in 1976, Gaddafi expelled thousands of Egyptian migrants, Der Spiegel interviewed Sadat on his counterstrategy. "Well … we don't take the situation so seriously," Sadat responded, sketching out the two conditions establishing migration interdependence vulnerability. "We are ready to absorb our workers; they can find work here, or in other Arab countries" (quoted in Der Spiegel, 1976). Indeed, Egypt at the time did not meet either of the two vulnerability conditions. For one, the close relationship between Egypt and the Gulf states meant that the latter were willing to absorb any political economy costs by recruiting those displaced Egyptians (Feiler 2003, 98-116). "[Egypt had] zero worry," recalled former Prime Minister (1974)(1975) Abd El Aziz Hegazi. "We relied on the help of the Saudis when we dealt with … Qaddafi's actions" (personal interview, April 27, 2014). At the same time, elites' ability to continue expanding the Egyptian economy's public sector also allowed for the partial absorption of displaced migrants' costs domestically, as the economy profited from the 1975 reopening of the Suez Canal, increased revenues from tourism and oil exports, remittances from Egyptians in the Gulf, as well as American and Arab financial aid. Despite the fact that Libya constituted the largest regional host state for Egyptian migrants at the time, Egypt did not comply with Gaddafi's demands, for it was not vulnerable to Libyan strategies.
In 1990, President Saddam Hussein ordered the displacement of Egyptians from Iraq (at the time hosting more than 44 percent of all Egyptian regional migrants), partly to compel Egypt to withdraw from Operation Desert Storm (Farouk-Sluglett and Sluglett 2001, 288). "The Iraqis, they were our brothers, but now after two weeks everything has changed," one returning migrant told the New York Times in August 1990, reporting harrowing stories of rape and theft. Egyptian deportees exiting Iraq recounted how they "had to sleep at night in desert camps, and Iraqi soldiers, they came in the night and searched our pockets" (New York Times, August 22, 1990). Despite the forced return of more than seven hundred thousand migrants (Feiler 2003, 245), Egypt did not shift its policy on Iraq. While Egypt was not able to absorb these costs domestically (only a year later, in 1991, Mubarak had to resort to the IMF and initiate an arduous process of economic restructuring), it was able to locate alternative host states willing to shoulder the cost. Saudi Arabia and Libya, with whom Egypt had friendly relations (Vandewalle 2012, 179), absorbed the displaced migrants out of Iraq and minimized the socioeconomic cost to Egypt. In terms of Saudi Arabia, Feiler writes the following: Fortunately for Egypt, Saudi Arabia offered [Egypt] various economic compensations for its cooperation with the anti-Iraqi coalition, and by the end of April 1991, the number of Egyptian workers in Saudi Arabia had reached 1.25 million-double their prewar number. (Feiler 2003, 244) But Libya also came to the help of Egypt: "it is interesting to note that the moment relations between Egypt and neighboring Libya improved … [more than fifty thousand] Egyptians moved into Libya seeking employment," Vatikiotis wrote in his authoritative history of modern Egypt.  Source: Farrag (1995, 73) "So long as Egyptian migrant labor found ready employment in these Arab countries … things went well" (Vatikiotis 1991, 432-33). While Saudi Arabia and Libya do not release detailed migration statistics for security reasons, Farrag has estimated the geographical distribution of Egyptian regional migrants before and after Operation Desert Storm (see Table 2). Her findings exemplify Egypt's strategy of shifting the cost of host-state coercion to alternative host states. Overall, the fact that the Egyptian economy was not in a position to absorb the cost of Iraq's strategy of displacement did not render Egypt vulnerable to it: given that oil-producing Arab states were willing to shoulder this cost, Egypt did not satisfy one of the two conditions establishing migration interdependence vulnerability and was able to resist Iraqi coercion (for contrast, see footnote 3). In sum, unlike earlier coercive attempts against Egypt under Sadat or Mubarak, the two cases examined in this study's empirical section occurred at a time when Egypt was both regionally isolated and economically weak. As such, the article demonstrated how a sending state that is neither able to shift the cost of host-state strategies to other states nor able to absorb this cost domestically exhibits migration interdependence vulnerability and, thus, is likely to comply. Egyptian noncompliance to Libya and Jordan would have likely been an option in two counterfactual cases, as per the article's theoretical framework: firstly, had Morsi been able to shift the cost to another regional state (as per Mubarak's strategy in 1990), and, secondly, if the Egyptian economy had been able to absorb these costs domestically (as per part of Sadat's strategy in 1976).

Conclusion
This article developed a framework for understanding how labor migration features in coercive interstate relations. It analyzed the mechanisms through which host states attempt to leverage their position against sending states, as well as the conditions under which such a strategy succeeds. I argued for viewing these dynamics through the prism of migration interdependence: reciprocal political economy effects created through cross-border population mobility between sending and host states. A host state may attempt to leverage its position via restriction, namely the imposition of costs via curbing remittances, immigration, or both. It may also attempt to do so via displacement, namely the forced expulsion of a sending state's migrant group. The effectiveness of restriction or displacement depends on whether a sending state is vulnerable to migration interdependence. Vulnerable sending states are more likely to comply with host-state demands, particularly if a host state exercises a strategy of displacement rather than one of mere restriction. I detailed an unexplored aspect of migration diplomacy by drawing on Middle East cases to highlight the potential potency of labor migration as a weapon of weak host states against stronger sending states.
Do my arguments travel beyond the two cases discussed here? Some examples highlight frequent uses of restriction and displacement by host states; in Afghanistan, authorities have not hesitated to use both restriction and displacement against Pakistani labor migrants entering the country since mid-2016. Afghanistan recently adopted a law that requires passports, rather than mere identity cards, and frequently deported Pakistani labor migrants, in an effort to force Pakistan to reverse its decision to unilaterally close the Chaman border crossing (Times of India 2016); in response, in March 2017, Pakistan's prime minister decided to open the crossing "as a gesture of goodwill" (BBC News 2017). Elsewhere, bilateral disputes between Egypt and Sudan, including the contested border of the Halayeb Triangle, led Khartoum to adopt a strategy of restriction in April 2017, by barring entry to Egyptian men aged between sixteen and fifty without visas (Atef 2017). In both cases, my framework should help make sense of the nature of these strategies and the mechanisms that account for their effects.
This article challenges long-held understandings of crossborder migration as necessarily conducive to interstate cooperation. In doing so, the article avoids reproducing the conventional wisdom that stronger states induce weaker ones into compliance; rather, it affirms Keohane and Nye's (2012) insight that, under certain conditions, weaker states can successfully employ a nonmilitary coercive strategy against more powerful states. In explaining the dynamics of such processes, the article also intervenes in ongoing debates within the growing literature on migrant remittances by putting them firmly into conversation with work on economic sanctions. Future work should test, and modify, my framework via medium-and large-N studies of host-state leverage strategies against sending states. We also need additional single-case or small-N studies to uncover the political importance of sending states' sensitivity to migration interdependence. Even if this study suggests that host-state attempts at manipulating sensitivity interdependence are unlikely to succeed, this does not mean that such efforts are politically unimportant. Finally, we should conduct research on host-state leverage within a cooperative, rather than coercive, migration diplomacy framework: What are the mechanisms through which migration interdependence might provide host states with "carrots" rather than "sticks" in the management of bilateral relations? Under what conditions would preferential access to a host-state labor market, or the reduction of taxes imposed on migrant remittances, contribute to policy changes in sending states? The mechanisms and dynamics of host-state use of labor migration against sending states constitute an important, underexplored field of inquiry in the study of world politics.