Examining the role of border closure and post-colonial ties in Caribbean migration

Examining the role of border closure and post-colonial ties in Caribbean migration Abstract This article analyses the role of border regimes and post-colonial ties in Caribbean migration between 1960 and 2010. Over this period, 18 out of 25 countries in the Caribbean region have experienced the closure of borders by their former colonial state, while the remaining seven former colonies have retained open borders with their metropolitan state. In view of the academic and policy debate about the effects of border restrictions, the Caribbean region allows the comparison of emigration volumes and migration destination selection from Caribbean countries with both closed and open borders. Moreover, because the Caribbean region was colonised by Britain, France, the Netherlands and the United States, we examine whether post-colonial ties influence long-term Caribbean emigration, allowing migrants to overcome border regime restrictions. The article finds that countries with closed borders with the former colonial state do not experience a decrease but rather a strong increase in long-term emigration. In fact, emigration gradually regains strength after border closure as migration channels to new destinations develop. Thus, border closure seems to encourage a weakening of ‘post-colonial’ migration patterns through the diversification of emigration towards destinations other than the former colonial state and its former colonies. Conversely, the increase of emigration from countries with open borders has decelerated and generally remained concentrated towards countries within the free-movement colonial sphere. This suggests that the post-colonial migration patterns may be associated with continuous open borders as much as with cultural and linguistic connections. 1. Introduction It is often assumed that closed borders limit migration. As a result, strong border controls are regularly advocated as an imperative measure by which states can prevent the entry of undesirable immigrants, criminals and terrorists (Frontex 2016). However, there is little evidence on the extent to which migration restrictions really impede migration. Previous studies indicate that the various migration policies that make up border regimes tend to produce unexpected ‘substitution effects’, which unintentionally shift migration patterns (de Haas 2011), and growing empirical evidence shows that the border control industry has created a series of ‘migration crises’, thus triggering the very problem it is trying to address (Andersson 2016). This article contributes to this debate by investigating whether we observe differences in the evolution of emigration between 1960 and 2000 when we compare emigration from Caribbean countries whose borders with the former colonial state have remained open and those that have experienced a border closure with the former colonial state. If border controls are effective, closed border countries should show a long-term decrease in emigration and, conversely, the absence of borders should encourage large emigration. Because the Caribbean region was colonised by different states (France, the Netherlands, Britain and the USA1), this analysis also allows us to examine whether post-colonial ties influence long-term migration by making former colonial states preferred destinations and by helping migrants to overcome migration policy restrictions through their migration-facilitating networks, as is often assumed (Beine, Docquier and Ozden 2011). An increasing volume of migration literature has focused on studying the effects of specific migration policies on short- to medium-term migration flows (Hatton 2005; Mayda 2010; Beine, Docquier and Özden 2011; Ortega and Peri 2013; Czaika and de Haas 2016), but the literature on the role of border regimes is much more limited. In this paper, the term ‘border regimes’ refers to the set of policies that aim to regulate the settlement of people by restricting residence and employment rights across national territories. Particularly relevant is the question of what happens to existing migration patterns once a border regime is established between two countries, as in the case of the passage from colony to independence. Through this process, a contiguous political space is divided and the settlement rights of former citizens or former colonial subjects become subjected to specific border regime regulations. We consider that borders are closed in the sense that previously unregulated settlement becomes channelled, and often significantly constrained, through residence and employment restrictions, although entry may not be restricted through travel visa restrictions.2 Naturally, border regimes can also be removed, as for new Member States after joining the European Union. And while a few studies have analysed the effect of the removal of intra-European border regimes (Wallace 2002), research on the establishment of a border regime and its migration consequences are limited to a few country studies (Freeman 1987; van Amersfoort 2011; Vezzoli 2015). The concept of post-colonial ties is examined much more frequently in migration research, with evidence of the general relevance of this historical link in the selection of migration destinations (Hooghe et al. 2008; Constant and Tien 2009; Belot and Hatton 2010). However, the role of post-colonial ties is more ambiguous when analyses focus on specific migration corridors, e.g. former French colonies to France, former British colonies to Britain, with some analyses showing strong post-colonial ties and others no effect (Hooghe et al. 2008; Constant and Tien 2009). The Caribbean and its diverse colonial past offers fertile ground for the analyses of border regimes and post-colonial ties; yet, migration research in the Caribbean region, while rich in case studies, is much more limited in comparative research. In fact, most previous work on regional migration patterns has focused on the links between political status—i.e. independence versus continuous non-sovereignty—of Caribbean countries, economic growth and migration trends. Findings suggest that non-sovereign countries, having experienced stronger economic growth, display increasing immigration, while the independent countries’ lack of economic diversification and slow economic growth makes them countries of net emigration (McElroy and Sanborn 2005; McElroy and Pearce 2006; McElroy 2011; Mitchell and McElroy 2011). Only a few case studies on non-sovereign countries have considered the effects of open borders with the former colonial state and suggested that the freedom of movement leads to high emigration (Audebert 2007), particularly of rural and less educated people (Grosfoguel 1996). However, no comparison has been made of emigration from Caribbean countries where nationals face more limitations to migrate due to the closure of the border. This paper investigates two anticipated effects of closed borders, which have been noted in recent research: First, the implementation of a closed border regime may give rise to a rapid growth of emigration through a ‘now or never’ migration effect, as people anticipate that it will become increasingly difficult to emigrate in the future to the destination countries taking measures to restrict migration (de Haas 2011; Vezzoli 2014c). But in the aftermath of the introduction of border restrictions, emigration intensity often decreases, particularly towards the former colonial state with which the border regime was established (Vezzoli 2014c). Yet, the influence of border closure may be weakened by post-colonial ties and concomitant factors such as language, culture, institutional and educational systems, transport and communication links as well as networks that may facilitate migration to the former colonial state (Thielemann 2006; Hooghe et al. 2008; Beine, Docquier and Özden 2009; Constant and Tien 2009; Belot and Hatton 2010). Second, the closing of the borders by the former colonial state may also influence migration to other destinations, leading to so-called spatial substitution effects (de Haas 2011), diversifying migration to alternative extra- and intra-regional destinations where migration policies are less restrictive than in the former colonial state. By looking at the extent of spatial diversification, we may also learn more about the conditions under which post-colonial ties endure and observe whether there are variations by former colonial state (i.e. former British, Dutch, French colonies or US possessions), or border regimes (open vs. closed). With this article we propose a series of descriptive analyses to highlight the trends and patterns of migration from Caribbean countries and identify visible differences in the way border regimes and post-colonial ties seem to associate with migration trends. We also offer some possible explanations of how border regimes and post-colonial ties may interact with and shape migration. It is not the ambition of this article to provide an analysis of the wide range of factors that might have influenced Caribbean migration over the past 60 years. Rather, we are interested in learning whether, in a region well-known for its high migration volumes (Segal 1987), the implementation of a border regime acts as a barrier to migration and, conversely, open borders encourage large migration, whatever the factors driving migration trends and patterns might be. Similarly, we do not aim to determine the extent to which post-colonial ties drive migration to former colonial states, but whether we observe enduring migration connections between former colonies and former colonial countries despite migration policy barriers. Thus, we set out to examine whether we observe relations between border regimes, post-colonial ties and migration through a descriptive approach. 2. Characteristics of the Caribbean: Migrations and border regimes 2.1 A region rich in migration experiences 2.1.1 Migration before World War Two The Caribbean has been described as the world region most affected by a deep and continuous impact of international migration (Segal 1987). Caribbean societies have their roots in the labour requirements on colonial plantations, which were filled initially with enslaved labour from Africa and after Emancipation with indentured contract labour from Europe (e.g. Madeira Islands) and Asia, primarily from former British India, but also from China and Indonesia. Emancipation, which Britain enacted in 1833, France in 1848 and the Netherlands in 1863, meant that for the first time former enslaved workers gained the freedom to seek alternative occupations. They did so by migrating to other Caribbean islands, including the Spanish Caribbean and Central and South American countries, where they could escape the oppressive working conditions of the plantation system (Thomas-Hope 1978). Caribbean migrants also worked for American companies such as the American Fruit Company, which provided an initial link to later seasonal migrations to the USA (Chaney 1989). With the completion of the Panama Canal in 1914, the onset of the First World War, the crash of sugar prices in 1921 and the advent of the Great Depression, migration decreased and many migrants returned to their origin country (Marshall 1987; Chaney 1989). Starting in the 1920s Panama, and later other Central American countries, began to control the entry of British Caribbean immigrants, which heralded the end of this intra-regional phase of Caribbean migration (Thomas-Hope 1978; Marshall 1987). By the mid-20th century, migration had become an essential part of Caribbean life, particularly in the British colonies. The literature recounts how migration evolved into a social and cultural phenomenon embedded in the Caribbean colonial environment, where expectations of fundamental socio-economic change in the communities of origin were low, while migration ‘offered an alternative to fundamental change’ (Thomas-Hope 1978: 77). In the French and Dutch Caribbean colonies migration was not as prevalent as in the British colonies. Before World War Two migration from the Dutch Caribbean was much more limited and primarily consisted of students from the elite who pursued tertiary studies in the Netherlands, although a small group of working class migrants also participated in migration from Suriname to the Netherlands (Oostindie 2009). Emigration from the French Antilles and French Guiana similarly reflected student migration to pursue secondary and tertiary education in metropolitan France. In the British, French, Dutch and US cases young men were mobilized during World War Two to support the war effort and fight for the colonial state, a process that would make these young men aware of the opportunities available in the metropolitan state and would influence later migrations (Levine 1987; Oostindie 2009; Peach 1991). Generally, the absence of border regimes regulating migration (Hendry 2011) allowed Caribbean migrants to respond to the emergence of opportunities within the Caribbean region and in Central and South America. 2.1.2 The development of extra-regional migrations From the mid-20th century, Caribbean migration patterns underwent an important transformation, primarily through a spatial diversification towards overseas destinations in the Northern hemisphere. As Britain started its post-World War Two reconstruction efforts, British Caribbean migrants contributed to the rebuilding of the British transport and health systems, sometimes through recruitment programmes (i.e. Barbados and Jamaica) but also through spontaneous migration. Concurrently, early recruitment programmes for seasonal workers by the USA and domestic workers by Canada offered regular short-term migration opportunities (Thomas-Hope 2000; Vezzoli 2014a), which strengthened the connections between the Caribbean and North America. Significant migration from the region to France and the Netherlands from their Caribbean colonies started 10 to 15 years after British Caribbean migrations, with peaks in migration in the 1970s (Peach 1991). The rapid growth of migration from the French Antilles, less so from French Guiana, followed the introduction of a state-organised recruitment system which attempted to reduce high unemployment and prevent social unrest in the French Antilles, while filling low-level civil service positions in France, which could only be taken by French citizens (Condon and Ögden 1991). Migration from Dutch Caribbean colonies was not strongly associated with labour. Although small-scale recruitment programmes existed (Vezzoli 2014b), emigration was tied to political anxieties in anticipation of Suriname’s independence and the establishment of its border regime with the Netherlands in 1980 (van Amersfoort 2011). Puerto Rican migration to the USA started many years before World War Two and peaked in the 1950s when on average 40,000 Puerto Ricans were migrating to the USA per year, facilitated by the fact that neither passport nor visa was required. While many factors explain this intense emigration, research has found a high correlation between Puerto Rican emigration and the US business cycles and unemployment rates, linking this migration to migrants’ search for better economic opportunities, higher wages and better quality of life (Levine 1987). The literature acknowledges the importance of migration policies in re-directing (rather than curbing) migration in this period. The US Immigration Act of 1952, also known as the Walter/McCarran Act, restricted immigration in the USA and is believed to have been partly responsible for a spatial substitution of migration from the USA to Britain (Marshall 1987). Conversely, the 1962 British Commonwealth Immigration Act increased restrictions on British Caribbean immigrants, who were required to obtain employment vouchers to be admitted to Britain. In the same period, the liberalization of immigration policies in Canada (in 1962) and in the USA (in 1965) towards non-European countries encouraged migration from the Caribbean as well as secondary migration of Caribbean immigrants already residing in Britain to North America (Palmer 1974; Thomas-Hope 2000; Vezzoli 2014a). 2.2 Evolution of border regimes The decolonisation movement gained strength in the post-World War Two period, but these processes were not homogeneous and reflected national ideologies as well as financial constraints faced by many European countries, which were economically strapped after the substantial war-related losses. While the USA had already established free associated status for Puerto Rico and unincorporated status for the US Virgin Islands by the 1920s, the French government transformed its French Caribbean colonies into integral parts of the French state as Overseas Departments in 1946 and the Dutch government made its Dutch Caribbean colonies countries within the Kingdom of the Netherlands in 1954. Britain, which had the strongest European presence in the Caribbean, was the first colonial power to concede independence to Jamaica and Trinidad and Tobago in 1962 and followed a strategy of decolonization of most of its Caribbean colonies (Christopher 2002; Rabe 2005). Among non-British Caribbean colonies, only Suriname gained independence from the Kingdom of the Netherlands in 1975.3 The establishment of border regimes did not always mirror the changes in political status. We can distinguish three main categories with regards to the evolution of border regimes of Caribbean countries (Figure 1). With the exception of Dutch Caribbean colonies between 1927 and 1951 (Ahmadali and Luk 2015), former Caribbean colonies had an open border regime during colonialism, giving the population full rights to work and reside in the colonial state (Hendry 2011). In Trajectory A, the border regimes have remained open during the entire period before, during and after colonization (the so-called Open–Open–Open Borders, later referred to as open borders). This is the case for the Dutch, French and US colonies that transitioned to a dependent non-sovereign status and that have been guaranteed access to the metropolitan state through full citizenship and freedom of movement and settlement. Trajectory B (Open–Closed–Closed Borders, later referred to as closed borders) represents countries that evolved from open border regimes to closed border regimes. This is applicable to all British former colonies, regardless of political status, as well as Suriname. In the early 1960s, British citizens in the colonies were subjected to partial closure starting with the Commonwealth Immigration Act of 1962 and 1968, which restricted the right of settlement, and the 1971 Immigration Act which restricted all rights to remain. Finally, the 1981 British Nationality Act created a British Dependent Territories citizenship, which did not give the right to abode and essentially gave the Dependent Territories citizens no legal status (Clegg 2005). Because of the immediate restriction on work and residence, all the former British colonies passed from an open to a closed border regime in the 1960s. Similarly, in the case of Suriname where initial immigration barriers were introduced in 1975 and full immigration policies applied in 1980, border closure was recorded in 1975, when free movement and right to residence and work were curtailed. Figure 1. View largeDownload slide Evolution of border regimes of former colonies towards their former colonial state: 3 trajectories among the 25 Caribbean states. Figure 1. View largeDownload slide Evolution of border regimes of former colonies towards their former colonial state: 3 trajectories among the 25 Caribbean states. While the data used in this paper do not allow us to study migration trends beyond 2000, it is important to specify the more recent evolution of Trajectory C (Open–Closed–Open Borders) applicable to the British non-sovereign countries. The Monserrat crisis, triggered by the explosions of the volcano Mount Soufrière starting in 1995 and the departure of two-thirds of the population in the following years, brought to the surface the peculiar lack of a legal status of the inhabitants of British dependencies. By 2002 the British Overseas Territories Bill gave the five British non-sovereign countries the right to British citizenship upon request with matching right of abode in Britain, the right of free movement and residency in the EU and European Economic Area member states (Clegg 2005). The effect of this ‘opening’ of the borders on emigration should be analysed in the future, once post-2000 data are available. These trajectories of border regime evolution are based exclusively on the full set of migration policies established by and with the former colonial state (see Table 1 for country groups). It is meant to provide a lens through which to observe whether such drastic changes in policies are associated with visible shifts in migration intensity and destination choice or, conversely, whether the reinforcement of full mobility and settlement rights in open border countries are associated with any shifts in migration patterns. Hence, the open border category in Trajectory A is intended as freedom of settlement for countries within their own colonial sphere, including their respective former colonial state, not to countries outside of their colonial sphere, which are often subjected to migration restrictions, e.g. French Antilleans migrating to Dutch-sphere Curaçao, Puerto Ricans migrating to the French Antilles or British-sphere Anguillans migrating to Canada. Even open borders countries are affected by migration policies of other potential destination countries. Table 1. Caribbean countries groups by border regime trajectory (and former colonial state) Trajectory A, Closed Borders Trajectory B, Open Borders 1. Anguilla (UK) (Trajectory C since 2002) 2. Antigua and Barbuda (UK) 3. Bahamas (UK) 4. Barbados (UK) 5. Belize (UK) 6. Cayman Islands (UK) (Trajectory C since 2002) 7. Dominica (UK) 8. Grenada (UK) 9. Guyana (UK) 10. Jamaica (UK) 11. Montserrat (UK) (Trajectory C since 2002) 12. Suriname (NL) 13. St Kitts and Nevis (UK) 14. St Lucia (UK) 15. St Vincent and the Grenadines (UK) 16. Trinidad and Tobago (UK) 17. Turks and Caicos (UK) (Trajectory C since 2002) 18. UK Virgin Islands (UK) (Trajectory C since 2002) 19. Aruba (NL) 20. Guadeloupe (FR) 21. French Guiana (FR) 22. Martinique (FR) 23. Netherlands Antilles 24. Puerto Rico(US) 25. US Virgin Islands (US) Trajectory A, Closed Borders Trajectory B, Open Borders 1. Anguilla (UK) (Trajectory C since 2002) 2. Antigua and Barbuda (UK) 3. Bahamas (UK) 4. Barbados (UK) 5. Belize (UK) 6. Cayman Islands (UK) (Trajectory C since 2002) 7. Dominica (UK) 8. Grenada (UK) 9. Guyana (UK) 10. Jamaica (UK) 11. Montserrat (UK) (Trajectory C since 2002) 12. Suriname (NL) 13. St Kitts and Nevis (UK) 14. St Lucia (UK) 15. St Vincent and the Grenadines (UK) 16. Trinidad and Tobago (UK) 17. Turks and Caicos (UK) (Trajectory C since 2002) 18. UK Virgin Islands (UK) (Trajectory C since 2002) 19. Aruba (NL) 20. Guadeloupe (FR) 21. French Guiana (FR) 22. Martinique (FR) 23. Netherlands Antilles 24. Puerto Rico(US) 25. US Virgin Islands (US) Table 1. Caribbean countries groups by border regime trajectory (and former colonial state) Trajectory A, Closed Borders Trajectory B, Open Borders 1. Anguilla (UK) (Trajectory C since 2002) 2. Antigua and Barbuda (UK) 3. Bahamas (UK) 4. Barbados (UK) 5. Belize (UK) 6. Cayman Islands (UK) (Trajectory C since 2002) 7. Dominica (UK) 8. Grenada (UK) 9. Guyana (UK) 10. Jamaica (UK) 11. Montserrat (UK) (Trajectory C since 2002) 12. Suriname (NL) 13. St Kitts and Nevis (UK) 14. St Lucia (UK) 15. St Vincent and the Grenadines (UK) 16. Trinidad and Tobago (UK) 17. Turks and Caicos (UK) (Trajectory C since 2002) 18. UK Virgin Islands (UK) (Trajectory C since 2002) 19. Aruba (NL) 20. Guadeloupe (FR) 21. French Guiana (FR) 22. Martinique (FR) 23. Netherlands Antilles 24. Puerto Rico(US) 25. US Virgin Islands (US) Trajectory A, Closed Borders Trajectory B, Open Borders 1. Anguilla (UK) (Trajectory C since 2002) 2. Antigua and Barbuda (UK) 3. Bahamas (UK) 4. Barbados (UK) 5. Belize (UK) 6. Cayman Islands (UK) (Trajectory C since 2002) 7. Dominica (UK) 8. Grenada (UK) 9. Guyana (UK) 10. Jamaica (UK) 11. Montserrat (UK) (Trajectory C since 2002) 12. Suriname (NL) 13. St Kitts and Nevis (UK) 14. St Lucia (UK) 15. St Vincent and the Grenadines (UK) 16. Trinidad and Tobago (UK) 17. Turks and Caicos (UK) (Trajectory C since 2002) 18. UK Virgin Islands (UK) (Trajectory C since 2002) 19. Aruba (NL) 20. Guadeloupe (FR) 21. French Guiana (FR) 22. Martinique (FR) 23. Netherlands Antilles 24. Puerto Rico(US) 25. US Virgin Islands (US) In particular, North American immigration policy shifts in the 1960s have been deemed important shapers of extra-regional Caribbean migrations: (1) Shifts in Canadian policies to attract skilled migrants provided greater opportunities for Caribbean migrants, who generally possessed secondary and tertiary levels of education (Thomas-Hope 2000); and (2) US policies dismantled racial criteria and opened migration opportunities for Caribbean people with an emphasis on family reunification, allowing Caribbean families to use extended family relations as a migration channel (Maingot 1983). This effect was so great that British Caribbean migration to Britain has been described as being ‘sandwiched between two periods of migration to the Americas’ (Peach 1991: 3), reducing Britain to just one of the many destinations for British Caribbean migrants. However, the effects of North American migration policies on non-British Caribbean migration remains unclear, particularly because many open border countries benefit from freedom of movement and settlement towards their respective metropolitan states, i.e. French Antilles and French Guiana to France, Aruba, the Netherlands Antilles to the Netherlands, and Puerto Rico and the US Virgin Islands to the USA. In fact, the literature suggests that migration from the Dutch and French dependencies remained concentrated towards their related European states (Peach 1991). For Suriname, however, the end of the negotiated preferential migration channels and the introduction of a travel visa requirement to the Netherlands in 1980 (van Amersfoort 2011) made US policies more relevant as migration towards the USA gained some strength (Runs 2006; Vezzoli 2014b). Caribbean countries are also important migration destinations within the region. In fact, while Caribbean migration patterns shifted after the 1940s from intra-regional to extra-regional destinations (Thomas-Hope 1978), intra-regional migration also continues at a steady pace (Thomas-Hope 2000). Within the region, there is ample evidence of various forms of movement, which reflect migrations to work in niche areas such as tourism or off-shore financial services, fluid forms of migration that follow trade and networks of opportunities (Carnegie 1987), circulation and patterns of return for retirement but also as active workers in Caribbean economies (Thomas-Hope 2000). However, it is unclear the extent to which these intra-regional migrations have been affected by border regimes and whether closed borders have in fact reduced overall emigration and open borders within the region have engendered large migrations. 3. Scope, data and methodology This paper proceeds to analyse the evolution of overall, extra-regional and intra-regional emigrations from 1960 to 2000 through a descriptive approach. The objective is to understand whether shifts in emigration intensity and destination may have occurred and how these may be related to the evolution of border regimes with the former colonial state as well as the nature of post-colonial ties. To accomplish this analysis, migrant stock data were taken from the Global Bilateral Migration Database (GBMD) released by the World Bank. This database, which contains bilateral migration population (‘stock’) data for 226 countries, major territories and dependencies for each decade from 1960 to 2000, is based on estimations from census data and population register records (when census data were not available) (Özden et al. 2011). While the release of this database has drastically increased the potential to assess long-term migration trends, it has some limitations. For instance, immigration is likely to be underestimated for countries defining migration on the basis of ‘citizenship’ rather than ‘birth’ because of naturalization. Moreover, missing data was completed with estimated values and, as in most official migration datasets, irregular migrants are generally not taken into account. Because the 1970 data were inconsistent, we decided to drop these data and use only 20-year intervals (i.e. 1960, 1980 and 2000). Other limitations affected particularly data for Britain and its former colonies, whose values seem much inferior to expected values, as well as data for the French Overseas Departments, which are low and erratic compared to data reported in French statistics. Data for the Surinamese in the Netherlands are also much lower than those reported by the Dutch statistical office. Despite these limitations, this database allows us to account for the presence of Caribbean migrants worldwide by origin and destination countries over a 40-year period, during which important changes in border regimes, political status and immigration policies occurred. The low quality of the 2010 migration stock data estimated more recently by the World Bank prevented us from utilizing it, although it would have greatly improved our analytical purpose, as it would have allowed the analysis of the effect of the re-opening of borders for emigration by Britain towards its non-sovereign countries in 2002 (Trajectory C). The United Nations database on international migrant stocks (United Nations 2013) was considered because of the availability of data for 1990, 2000, 2010 (and 2013); however, its total absence of data for residents of the French Overseas Departments and of some other non-sovereign countries made it unsuitable for this analysis. Migrant stock data were utilized in conjunction with UN population estimates to calculate emigration intensities, namely the percentage of migrants among all individuals born in a specific country or group of countries. So, for example, in 2000 the overall emigration intensity of countries which have had an open border over time with their former colonial state was 33 per cent, that is to say 33 per cent of all individuals born in these countries resided abroad. 4. Results 4.1 Overall migrations from Caribbean countries The Caribbean region displays high intensity of emigration, which was already visible in 1960 and has increased in all the countries in the world over time, regardless of which type of border regime they employed. Figure 2 shows how emigration intensity has grown in both those countries that have had constantly open borders (21 per cent of their population lived abroad in 1960 and 33 per cent in 2000) as well as those that have gone from open to closed borders (11 per cent of their population lived abroad in 1960 and 27 per cent in 2000). Countries that have had open borders with their former colonial state and within the colonial sphere experienced stronger growth in emigration intensity between 1960 and 1980, when it grew by 7 per cent, but this growth has since been decelerating. The countries whose borders have closed experienced weaker growth between 1960 and 1980 and a much more rapid growth in the following period, when emigration intensity grew by 12 per cent. This delayed increase may be related to the establishment of border regimes which constrained emigration, but possibly also to the positive appeal of independence which took place in this period and may have encouraged people to stay and take part in the development initiatives in newly independent countries (cf. Vezzoli 2014c). The growth of emigration between 1980 and 2000 has been associated with weak economic growth in many Caribbean countries (McElroy 2011; McElroy and Albuquerque 1988). However, it can be suggested that ‘migration substitution’ effects (de Haas 2011) needed time to take hold. Closing the borders of a country does not necessarily limit emigration, but engenders migration to other destinations (i.e. spatial substitution) or through new channels, e.g. illegally or via family reunification (i.e. categorical substitution). However, time is necessary for migrants to identify new possible locations and/or for new migration channels and new networks to emerge. These trends challenge common assumptions that closing the borders limits emigration, as in fact they seem to slow emigration only in the short-term. Figure 2. View largeDownload slide Evolution of overall emigration intensities by border regime (% of migrants among the population born in the countries), 1960–2000. Source: Global Bilateral Migration Database, World Bank. Figure 2. View largeDownload slide Evolution of overall emigration intensities by border regime (% of migrants among the population born in the countries), 1960–2000. Source: Global Bilateral Migration Database, World Bank. Figure 2 also suggests that migrants from countries that experienced border closure may be pushed into settlement abroad. Stringent entry regulations in destination countries may indeed prevent re-entry, leading to the increase in settlement abroad (i.e. reverse substitution effects) (de Haas 2011), thus increasing the size of migrant stocks. Conversely, the absence of constraints to migration for people in countries with continuous open borders enabled higher levels of circulation and lower settlement patterns, represented by the decreasing levels of permanent settlement abroad. Nevertheless, Figure 2 shows high emigration intensity for open border countries already in 1960, which is primarily due to the high emigration from US dependencies to the US, which alone accounted for over 17 per cent of emigration intensity in 1960. The important contribution of US dependencies to this trend has continued over time, although French dependencies, and to a lesser extent the Dutch dependencies, have also shown some growth in overall emigration intensity. The disaggregation of emigration by extra-regional and intra-regional destinations (Figure 3) shows the dominance of extra-regional emigration, which almost exactly replicates the patterns seen in Figure 2. This shows the important shift in Caribbean migration patterns from intra-regional to extra-regional destinations after the 1940s, which is well reported in the literature (Thomas-Hope 1978). Figure 3. View largeDownload slide Evolution of overall emigration intensities by extra- and intra-regional migration and by border regime (% of migrants among the population born in the countries), 1960–2000. Source: Global Bilateral Migration Database, World Bank. Figure 3. View largeDownload slide Evolution of overall emigration intensities by extra- and intra-regional migration and by border regime (% of migrants among the population born in the countries), 1960–2000. Source: Global Bilateral Migration Database, World Bank. 4.2 Extra-regional migration and the relevance of borders and post-colonial ties When extra-regional emigration intensities are disaggregated to observe migration intensities in the respective former colonial state or the metropolitan state, e.g. Jamaicans towards Britain and Puerto Ricans and US Virgin Islanders towards the USA, we observe (Figure 4) that migration to the metropolitan state has been largely significant for US dependencies and has grown in importance for the French and Dutch sphere countries towards their metropolitan state, which confirm previous evidence that open borders strengthen migrant networks and cumulative migration towards the metropolitan state. Figure 4. View largeDownload slide Emigration intensities in the former colonial states by border regime and former colonial state (% of migrants in their former colonial states among the population born in the countries), 1960–2010. Source: Global Bilateral Migration Database, World Bank. Figure 4. View largeDownload slide Emigration intensities in the former colonial states by border regime and former colonial state (% of migrants in their former colonial states among the population born in the countries), 1960–2010. Source: Global Bilateral Migration Database, World Bank. The countries that have experienced border closure reveal a rapid long-term increase in emigration intensities, but they display diverging patterns of post-colonial effects on migration towards the former colonial states. Suriname, in the Netherlands’ border closure group, shows strong post-colonial ties despite border closure. This is the result of peculiar circumstances leading to independence and the closure of the border regime which encouraged the formation of large Surinamese communities in the Netherlands before border closure. These facilitated migration via family reunification and formation processes after Dutch policies became increasingly restrictive, displaying categorical substitution effects, and stunted the growth of alternative migration destinations (cf. Vezzoli 2014b). Conversely, we observe the rapid weakening of post-colonial effects on emigration from former British colonies towards Britain. Although emigration from British Caribbean countries to extra-regional destinations has grown from just over 7 per cent in 1960 to almost 22 per cent in 2000, it has not been directed to Britain, where the migrant communities have actually shrunk. This offers strong support for claims made in the literature which point to four concurrent patterns: Reduction of immigration due to British immigration policies (Peach 1968, 1995; Marshall 1987). Attraction of North America, where family reunification and skills allowed immigration and the formation of large communities, and where economic prospects were more positive (Palmer 1974; Maingot 1983). The attractiveness of North America leading to step-wise migration of Caribbean people from Britain to North America. Post-retirement return, a trend known to be higher for Caribbean returnees from Britain than North America (Thomas-Hope 2000). A notable exception to this pattern is Montserrat, which saw a great expansion of emigration to Britain in the mid to late 1990s following natural calamities. To understand the importance of the USA in Caribbean migration systems, as it is often viewed, we analysed emigration intensities in North America (Figure 5). We observe that only former British colonies, all of which have experienced border closure with Britain, display the strongest growth in migration to North American destinations, while the other Caribbean countries continue to have negligible emigration intensities in North America. Only US dependencies show high emigration intensity to North America. Thus, although the hegemonic presence of the USA in the western hemisphere is often seen as a strong migration determinant, this does not seem to have affected migration destinations except for its own dependencies and the Anglophone Caribbean. The high degree of spatial substitution of emigration of the former British Caribbean countries to North America seems tied to border closure, the shared English language, but also to the fact that early labour recruitment programmes had established migration networks in the USA and Canada and countries in the British sphere have been historically highly migratory (Thomas-Hope 1978), which may result in high levels of aspiration to find alternative destinations after the introduction of British border controls. Contrarily, the French and Dutch Caribbean were historically less migratory and as migration developed, it did so within the colonial sphere where freedom of movement and settlement were guaranteed. Figure 5. View largeDownload slide Emigration intensities in North America by border regime and former colonial state (% of migrants in North America among the population born in the countries). Source: Global Bilateral Migration Database, World Bank. Figure 5. View largeDownload slide Emigration intensities in North America by border regime and former colonial state (% of migrants in North America among the population born in the countries). Source: Global Bilateral Migration Database, World Bank. Yet there are clear exceptions to these general patterns. For instance, emigration from the Netherlands Antilles was not very high but originally oriented towards North America more than to the Netherlands (see Figure 4, Dutch sphere open border countries did not have a strong presence in the Netherlands) (Oostindie 2009), connections that remain strong even in 2000. This may be associated with the fact that English is commonly spoken in the Netherlands Antilles. Moreover, growing interest in the USA as a migration destination for study reasons is emerging in Suriname, while French Guianese are increasingly interested in Canada, although in both cases the high costs of studying in these countries, particularly in comparison to lower cost or free education in the Netherlands and France, respectively, make the pursuit of these new educational destinations in North America the preserve of the elite (Vezzoli 2015). A growing expansion of ‘worldviews’ beyond the Netherlands and France over the next decade may possibly lead to a greater diversification of non-Anglophone Caribbean emigration towards North America. 4.3 Intra-regional migrations and the relevance of border regimes and colonial spheres within the region Intra-regional migration has remained steadily low (Thomas-Hope 2000), although it is probably higher than the 1 to 3 per cent shown in the results for 2000 (see Figure 3) when we consider the many uncounted forms of movement that occur among Caribbean island nations (Carnegie 1987). They include ‘replacement migration’ of seasonal agricultural labour, particularly in the sugar sector in islands like Barbados, the Dominican Republic, Guadeloupe and Martinique (Segal 1987) and across the Guianas (Vezzoli 2014a, 2014b). Nevertheless, it is interesting to study the intensities and direction of intra-regional emigrations and assess whether border closure with the former colonial states impacted emigration within the region. When we analyse whether open and closed border countries display different patterns of change in intra-regional migration intensities (Figure 6), we find strong evidence of spatial substitution associated with border closure. Caribbean countries whose borders with the former colonial state and with other countries within the same colonial sphere have been closed, display growing emigration towards countries in other colonial spheres. Contrarily, open border countries do not display any shift in destinations since 1960. With regard to the relevance of post-colonial ties, the data suggest that the closing of the borders by the former colonial state may weaken historical migration connections, while continuous open borders encourage migrants to continue relying on cultural and linguistic connections found in countries in the same colonial sphere. However, Suriname is an exception given its strong migration towards the Netherlands. In addition to language, this may be explained by the fact that independence was intensely contested and, concurrently, borders were fully closed only five years after independence (cf. Vezzoli 2014b). Figure 6. View largeDownload slide Evolution of destinations within the Caribbean region, by border regime of Caribbean countries of origin with their former colonial state and by colonial sphere (same or different). Source: Global Bilateral Migration Database, World Bank. Figure 6. View largeDownload slide Evolution of destinations within the Caribbean region, by border regime of Caribbean countries of origin with their former colonial state and by colonial sphere (same or different). Source: Global Bilateral Migration Database, World Bank. The data also suggest that French and US dependencies are strongly oriented towards other destinations within their own colonial sphere (Figure 7). For instance, French Caribbean people migrate to other French Overseas Departments while Puerto Ricans migrate primarily to the US Virgin Islands and vice versa. Interestingly, for US dependency citizens, British-sphere destinations are gaining some relevance, although at an irregular pace, suggesting the facilitating role of language as well as the growing attractiveness of specific British territories. Figure 7. View largeDownload slide Evolution of intra-regional destinations of Caribbean countries, by border regime of Caribbean countries of origin with their former colonial state and by type of sphere and by colonial sphere of destination (same or different). Source: Global Bilateral Migration Database, World Bank. Figure 7. View largeDownload slide Evolution of intra-regional destinations of Caribbean countries, by border regime of Caribbean countries of origin with their former colonial state and by type of sphere and by colonial sphere of destination (same or different). Source: Global Bilateral Migration Database, World Bank. Both British and Dutch sphere countries with border closure show a similar pattern as in Figure 6. For countries in the British sphere this follows the same trend of spatial substitution we see in extra-regional emigration. Conversely, for Suriname, this spatial substitution trend diverges from the continuous concentration of migration in the Netherlands. This is largely due to the strong emigration of Surinamese to French Guiana during the Interior War from 1986 to 1992 and the rapid growth of a Surinamese settled community along the Maroni River on the Suriname–French Guiana border (Léobal 2013; Piantoni 2009). Moreover, Guyana, Suriname’s westerly neighbour, absorbed 15 per cent of Suriname’s intra-regional migration in 2000 (see Table A1 in the Appendix). The only group of countries not conforming to the observed pattern are countries with open borders in the Dutch sphere, i.e. Aruba and the Netherlands Antilles (Figure 7). However, a close examination of the data suggests that this pattern reflects some data inaccuracies, particularly the undercount of population exchanges among the Dutch dependencies in 1960 and 1980, often driven by the growth and later decline of the oil industry. Nevertheless, these countries have also been associated with important historical migrations with other colonial spheres in the region, such as between the Netherlands Antilles and St Kitts and Nevis. Overall, even the open border countries within the Dutch sphere seem to confirm that freedom of movement tends to reinforce migration within the same colonial sphere, although its effect may slowly decrease over time. The French sphere countries have the least spatial substitution, which seems to be linked to high standards of living in the French Overseas Departments, while the highest level of spatial substitution is found among the Dutch sphere countries, with growing emigration towards the French dependencies. An interesting finding is that while the USA was not an important destination in extra-regional emigration from French and Dutch sphere countries, intra-regionally US dependencies clearly exert strong attraction. In fact, in 2000 all countries in the region had an important share of their intra-regional emigration directed towards Puerto Rico and/or the US Virgin Islands (see Table A1 in the Appendix). A number of factors emerge as important in shaping patterns of intra-regional migration. It is clear that a variety of drivers have continued to encourage intra-regional emigration, including political, economic and socio-cultural reasons. Nevertheless, we observe that the colonial sphere continues to be highly relevant for all countries within the region and particularly for countries that maintain open border regimes with former colonial states. However, open border countries have not seen an increase of emigration intensity in countries within the same sphere; rather, any growth in emigration has been to countries outside of the colonial sphere, therefore to countries outside of the free migration area. Concurrently, border closure seems associated not with a decrease in migration but with a substitution of migration destinations and weakening post-colonial migration ties. 5. Conclusions In colonial times, colonial ‘subjects’ largely benefited from rights to free settlement in the metropolitan state and in other colonies. After World War Two, as the decolonization movement gained strength, colonial powers gradually introduced measures to regulate and prevent settlement in the metropolitan state. Britain, in particular, fearing the large arrival of Commonwealth citizens from across the world, adopted a pre-emptive immigration policy by introducing immigration restrictions before many of its Caribbean colonies became independent. The French, Dutch and US governments endorsed more accommodating regulations, retaining open borders, except for the Dutch closure of the border with Suriname starting in 1975. This article set out to understand whether long-term migration patterns between former colonies and their former colonial states showed visible signs of the shift from open to closed border regimes. In particular, we were looking for shifts in emigration intensity over the course of the years studied, any evidence of post-colonial ties engendering strong continuous migration between former colonies and former colonial states and possible spatial migration substitution effects of migration to new destinations as a result of border closure. Contrary to common expectations that closed borders impede migration, our analysis showed that emigration intensity in closed border countries is not only high, but has grown exponentially despite the closure of borders by the former colonial state and other extra-regional destinations through the introduction of migration policies in main destinations in Europe and North America. Border closure enacted by former colonial states has also encouraged step-wise migration from former colonial states towards alternative extra-regional and Caribbean destinations. Countries that have retained open borders and have thus not gone through the ‘shock’ of border closure, have not experienced a similar acceleration of emigration and have rather seen decelerating emigration. This may be related to the fact that several of these countries show economic growth and that the presence of the state may ensure high levels of living conditions and opportunities across the dependencies. Nevertheless, wage differentials exist between these open border countries and their metropolitan states. Thus, this finding contradicts the general assumptions that border regimes are needed to prevent high emigration. Moreover, we show that when borders remain open, there is generally much lower diversification of destinations either extra-regionally or intra-Caribbean. Nevertheless, the results suggest that from the British perspective the border closure has been effective. However, state restrictions have not stopped emigration, but rather deflected it to other destinations. This deflection reflects the establishment of the border regime, although it is clear that the border regime was not in itself the driver of migration as political and economic reasons within the region, as well as a ‘culture of migration’ contributed to continuing migration. However, border regimes and migration policies generally played an important role in shaping the observed emigration patterns. As Britain closed its borders, new labour and family migration opportunities opened in Canada (1962) and the USA (1965). These permitted entry and settlement for several categories of migrants and marked the beginning of long-term patterns of family reunification from the Caribbean to the USA. In addition, Britain was experiencing an economic downturn, while economic growth in the USA offered better opportunities. From a Dutch perspective, however, the closure of the border with Suriname was not effective. Surinamese migration continued to concentrate strongly in the Netherlands, undoubtedly because of language and cultural proximities, but also because the full enforcement of a border regime occurred five years after independence, which led to the formation of large migrant networks before the borders were closed. This paper has elaborated an innovative approach to categorize Caribbean countries by closed and open border regimes. It has included a unique analysis of the associations between types of border regime with changes in extra- and intra-regional Caribbean migration patterns and examined the conditions under which we observe the weakening and strengthening of post-colonial ties. This approach may be useful to analyse the evolution of emigration from other regions where decolonization and border regimes were established, particularly the Pacific region, where open border non-sovereign countries exist, as well as in the Asian and African context. Data limitations narrowed the scope of the analysis and better flow and stock data would enhance this analysis. In particular, the availability of good stock data for 2010 would allow us to test the effects of borders for those countries which experienced border closure in the 1960s and border re-opening in the 2000s (trajectory C). Such an analysis would make an important contribution to our understanding of the migratory shifts not only in association with the closing but also with the re-opening of borders. Acknowledgements The research leading to these results is part of the DEMIG (The Determinants of International Migration) project and has received funding from the European Research Council under the European Community’s Seventh Framework Programme (FP7/2007-2013)/ERC Grant Agreement 240940. The authors are grateful to Hein de Haas, María Villares-Varela, Mathias Czaika, Katharina Natter and Edo Mahendra for their valuable comments on earlier drafts of this article. Footnotes 1. Although the USA is generally not considered to have a colonial past, we classify it as a former colonial state in this article because (1) Puerto Rico held a colonial status until its establishment as an associated free state in 1950 and (2) the US is the sovereign state of the US Virgin Islands. 2. We acknowledge that a closed border regime may be more (e.g. bans on specific nationalities or specific types of migrants such as refugees) or less restrictive (e.g. constraining primarily the entry of low-skilled workers), which could make comparisons of closed border regimes problematic. However, the countries in this study largely adopted similar policies (i.e. labour migration restrictions and constraints to family reunification such as income thresholds) in roughly the same period. Thus, the comparison of these countries over 20-year intervals largely captures the effects of similar ‘closed border regime’ policy packages. 3. The Kingdom of the Netherlands is now composed of four countries: (1) the Netherlands, which includes the European mainland and the three Caribbean municipalities of Bonaire, Saba and Sint Eustatius; (2) Aruba; (3) Curaçao; and (4) Sint Maarten (all in the Caribbean). In this analysis, Curaçao and Sint Maarten are included in the Netherlands Antilles because of data limitations. References Ahmadali H. , Luk N. C. ( 2015 ) ‘ Report on Citizenship Law: Suriname ’, Robert Schuman Centre for Advanced Studies, Edinburgh University Law School . Andersson R. ( 2016 ) ‘Europe’s Failed “Fight”Against Irregular Migration: Ethnographic Notes on a Counterproductive Industry’, Journal of Ethnic and Migration Studies , 42 / 7 : 1055 – 75 . Google Scholar CrossRef Search ADS Audebert C. ( 2007 ) ‘Le cadre politico-institutionel des migrations antillaises: des dynamiques différenciées dams un contexte géopolitique segmenté’, Études caribéennes (online) 8. <http://etudescaribeennes.revues.org/922> accessed 20 Apr 2017. Beine M. , Docquier F. , Özden C. ( 2009 ) ‘Diasporas’, CESifo Working Paper Series No. 2607. Beine M. , Docquier F. , Özden C. ( 2011 ) ‘ Diasporas ’, Journal of Development Economics , 95 / 1 : 30 – 41 . Google Scholar CrossRef Search ADS Belot M. , Hatton T. J. ( 2010 ) ‘Immigrant Selection in the OECD’, Discussion Paper No. 571, The Australian National University. Carnegie C. V. ( 1987 ) ‘A Social Psychology of Caribbean Migrations: Strategic Flexibility in the West Indies’, in Levine B. B. (ed.) The Caribbean Exodus . Westport, CT : Praeger . Chaney E. M. ( 1989 ) ‘ The Context of Caribbean Migration’, Center for Migration Studies (special issues) 7 / 1 : 1 – 14 . Google Scholar CrossRef Search ADS Christopher A. J. ( 2002 ) ‘ Decolonisation without independence ’, GeoJournal , 56 : 213 – 24 . Google Scholar CrossRef Search ADS Clegg P. ( 2005 ) ‘The UK Caribbean Overseas Territories: Extended Statehood and the Process of Policy Convergence’, in de Jong L. , Kruijt D. (eds) Extended Statehood in the Caribbean: Paradoxes of Quasi Colonialism, Local Autonomy and Extended Statehood in the USA, French, Dutch and British Caribbean . Amsterdam : Rozenberg Publishers . Condon S. A. , Ogden P. E. ( 1991 ) ‘ Emigration from the French Caribbean: the Origins of an Organized Migration’, International Journal of Urban and Regional Research , 15 / 4 : 505 – 23 . Google Scholar CrossRef Search ADS Constant A. F. , Tien B. N. ( 2009 ) Brainy Africans to Fortress Europe: For Money or Colonial Vestiges? <http://www.iza.org/en/webcontent/publications/papers/viewAbstract?dp_id=4615> accessed 15 Sep 2015. Czaika M. , de Haas H. ( 2016 ) ‘ The Effect of Visas on Migration Processes ’, International Migration Review , DOI: 10.1111/imre.12261. de Haas H. ( 2011 ) ‘The Determinants of International Migration: Conceptualizing Policy, Origin and Destination Effects’, in IMI Working Papers No. 32. Oxford: International Migration Institute. Freeman G. P. ( 1987 ) ‘Caribbean Migration to Britain and France: From Assimilation to Selection’, in Levine B. B. (ed.) The Caribbean Exodus . Westport, CT and London : Praeger . Frontex ( 2016 ) Risk Analysis for 2016. European Agency for the Management of Operational Cooperation at the External Borders of the Member States of the European Union. Grosfoguel R. ( 1996 ) ‘ Colonial Caribbean Migrations to the Metropoles in Comparative Perspective ’, Journal of Social Sciences , 3 / 1–2 ): 20 – 39 . Hatton T. J. ( 2005 ) ‘ Explaining Trends in UK Immigration’, Journal of Population Economics , 18 / 4 : 719 – 40 . Google Scholar CrossRef Search ADS Hendry I. ( 2011 ) British Overseas Territories Law . Oxford : Hart Publishing . Hooghe M. et al. ( 2008 ) ‘ Migration to European Countries: A Structural Explanation of Patterns, 1980–2004 ’, International Migration Review , 42 / 2 : 476 – 504 . Google Scholar CrossRef Search ADS Léobal C. ( 2013 ) Saint-Laurent-du-Maroni: une porte sur le fleuve . Matoury, Guyane : Ibis Rouge Editions . Levine B. B. ( 1987 ) ‘The Puerto Rican Exodus: Development of the Puerto Rican Circuit’, in Levine B. B. (ed.) The Caribbean Exodus . Westport, CT and London : Praeger . Maingot A. P. ( 1983 ) ‘ Caribbean Migration as a Structural Reality ’, Latin American and Caribbean Center, Florida International University . Marshall D. ( 1987 ) ‘A History of West Indian Migrations: Overseas Opportunities and “Safety-Valve” Policies’, in Levine B. B. (ed.) The Caribbean Exodus . Westport, CT : Praeger . Mayda A. M. ( 2010 ) ‘ International Migration: A Panel Data Analysis of the Determinants of Bilateral Flows’, Journal of Population Economics , 23 / 4 : 1249 – 74 . Google Scholar CrossRef Search ADS McElroy J. L. ( 2011 ) ‘ Island Demography: A Review of Selected Caribbean Contributions ’, Island Studies Journal , 6 / 2 : 245 – 60 . McElroy J. L. , de Albuquerque K. ( 1988 ) ‘ Migration Transition in Small Northern and Eastern Caribbean States ’, International Migration Review , 22 / 3 : 30 – 58 . Google Scholar CrossRef Search ADS PubMed McElroy J. L. , Pearce K. B. ( 2006 ) ‘ The Advantages of Political Affiliation: Dependent and Independent Small-Island Profiles ’, The Round Table , 95 / 386 : 529 – 39 . Google Scholar CrossRef Search ADS McElroy J. L. , Sanborn K. ( 2005 ) ‘ The Propensity for Dependence in Small Caribbean and Pacific Islands ’, Bank of Valletta Review [Malta], 31 :10–16 (Spring). <http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.502.174&rep=rep1&type=pdf> accessed 20 Apr 2017. Mitchell S. A. , McElroy J. L. ( 2011 ) ‘ Emigrant and Immigrant Small Island Profiles ’, Bank of Valletta Review [Malta], 43: 1–11(Spring) . <https://www.google.nl/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&ved=0ahUKEwivy5CCmLPTAhVMI8AKHXvgBWEQFggjMAA&url=https%3A%2F%2Fwww.bov.com%2Fdocuments%2Fbov-review-43-paper-1&usg=AFQjCNGMP5hOQ8T7SVWp8iC6j3Qymo-DyA&sig2=oetFqc2ZsPjQii5-saEUSw&cad=rja> accessed 20 Apr 2017. Oostindie G. ( 2009 ) ‘Migration Paradoxes of Non-Sovereignty: A Comparative Perspective on the Dutch Caribbean’, in Clegg P. , Pantojas-Garcia E. (eds) Governance in non-independent Caribbean: Challenges and Opportunities in the 21st Century . Kingston and Miami : Ian Randle Publishers . Ortega F. , Peri G. ( 2013 ) ‘ The Effect of Income and Immigration Policies on International Migration’, Migration Studies , 1 / 1 : 47 – 74 . Google Scholar CrossRef Search ADS Özden Ç. et al. ( 2011 ) ‘ Where on Earth is Everybody? The Evolution of Global Bilateral Migration 1960–2000 ’, The World Bank Economic Review , 25 / 1 : 12 – 56 . Google Scholar CrossRef Search ADS Palmer R. W. ( 1974 ) ‘ A Decade of West Indian Migration to the United States, 1962–1972: An Economic Analysis ’, Social and Economic Studies , 23 / 4 : 571 . Peach C. ( 1968 ) West Indian Migration to Britain: A Social Geography . London : Oxford University Press . Peach C. ( 1991 ) ‘The Caribbean in Europe: Contrasting patterns of migration and settlement in Britain, France and the Netherlands’, in Research Paper in Ethnic Relations . Coventry, UK : Centre for Research in Ethnic Relations, University of Warwick . Peach C. ( 1995 ) ‘Anglophone Caribbean Migration to the USA and Canada’, in Cohen R. (ed.) The Cambridge Survey of World Migration . Cambridge : Cambridge University Press . Piantoni F. ( 2009 ) ‘La question migratoire en Guyane française: Histoire, société et territoires’, Hommes et migrations, 1278/March–April: 198–216. Rabe S. G. ( 2005 ) U.S. Intervention in British Guiana: A Cold War Story . Chapel Hill : The University of North Carolina Press . Runs A. ( 2006 ) ‘Surinamese Immigrants in the United States of America: A Quest for Identity?’, Master thesis, Latin American Studies, CEDLA, Universiteit van Amsterdam. Segal A. ( 1987 ) ‘The Caribbean Exodus in a Global Context: Comparative Migration Experiences’, in Levine B. B. (ed.) The Caribbean Exodus , pp. 44 – 64 . Westport, CT : Praeger . Thielemann E. R. ( 2006 ) ‘The Effectiveness of Governments’ Attempts to Control Unwanted Migration’, in Parsons C. and Smeeding T. (eds) Immigration and the Transformation of Europe . Cambridge : Cambridge University Press . Thomas-Hope E. ( 1978 ) ‘ The Establishment of a Migration Tradition: British West Indian Movements to the Hispanic Caribbean in the Century after Emancipation ’, Centre for Latin-American Studies, The University of Liverpool . Thomas-Hope E. ( 2000 ) Trends and Patterns of Migration to and from Caribbean Countries. Economic Commission for Latin America and the Caribbean (ECLAC/CEPAL) <www.cepal.org/celade/proyectos/migracion/thomashope.doc> accessed 20 April 2017. United Nations ( 2013 ) Trends in International Migrant Stock: Migrants by Destination and Origin . United Nations Department of Economic and Social Affairs, Population Division . van Amersfoort H. ( 2011 ) ‘How the Dutch Government Stimulated the Unwanted Immigration from Suriname’, in IMI Working Papers, No. 47. Oxford: International Migration Institute. Vezzoli S. ( 2014a ) ‘The Effects of Independence, State Formation and Migration Policies in Guyanese Migration’, in IMI Working Papers, Series 94/DEMIG Project Papers 20. Oxford: International Migration Institute. Vezzoli S. ( 2014b ) ‘The Evolution of Surinamese Emigration Across and Beyond Independence: The Role of Origin and Destination States’, in IMI Working Papers, Series 106/DEMIG Project Papers 28 . Oxford: International Migration Institute. Vezzoli S. ( 2014c ) ‘The Role of the State in International Migration: Exploring the Transition from Colony to Independence’, in IMI Working Papers, Series 102/DEMIG Project Papers 26. Oxford: International Migration Institute. Vezzoli S. ( 2015 ) ‘ Borders, Independence and Post-Colonial Ties: The Role of the State in Caribbean Migration ’, Maastricht Graduate School of Governance . Wallace C. ( 2002 ) ‘ Opening and Closing Borders: Migration and Mobility in East-Central Europe ’, Journal of Ethnic and Migration Studies , 28 / 4 : 603 – 25 . Google Scholar CrossRef Search ADS Appendix Table A.1. Proportion of intra-Caribbean migrants by country of destination in the Caribbean in 2000, for each Caribbean country of origin sorted by closed and open borders Table A.1. Proportion of intra-Caribbean migrants by country of destination in the Caribbean in 2000, for each Caribbean country of origin sorted by closed and open borders © The Authors 2017. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.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) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Migration Studies Oxford University Press

Examining the role of border closure and post-colonial ties in Caribbean migration

Migration Studies , Volume Advance Article (2) – May 17, 2017

Loading next page...
 
/lp/ou_press/examining-the-role-of-border-closure-and-post-colonial-ties-in-wCAriJ60pV
Publisher
Oxford University Press
Copyright
© The Authors 2017. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com
ISSN
2049-5838
eISSN
2049-5846
D.O.I.
10.1093/migration/mnx034
Publisher site
See Article on Publisher Site

Abstract

Abstract This article analyses the role of border regimes and post-colonial ties in Caribbean migration between 1960 and 2010. Over this period, 18 out of 25 countries in the Caribbean region have experienced the closure of borders by their former colonial state, while the remaining seven former colonies have retained open borders with their metropolitan state. In view of the academic and policy debate about the effects of border restrictions, the Caribbean region allows the comparison of emigration volumes and migration destination selection from Caribbean countries with both closed and open borders. Moreover, because the Caribbean region was colonised by Britain, France, the Netherlands and the United States, we examine whether post-colonial ties influence long-term Caribbean emigration, allowing migrants to overcome border regime restrictions. The article finds that countries with closed borders with the former colonial state do not experience a decrease but rather a strong increase in long-term emigration. In fact, emigration gradually regains strength after border closure as migration channels to new destinations develop. Thus, border closure seems to encourage a weakening of ‘post-colonial’ migration patterns through the diversification of emigration towards destinations other than the former colonial state and its former colonies. Conversely, the increase of emigration from countries with open borders has decelerated and generally remained concentrated towards countries within the free-movement colonial sphere. This suggests that the post-colonial migration patterns may be associated with continuous open borders as much as with cultural and linguistic connections. 1. Introduction It is often assumed that closed borders limit migration. As a result, strong border controls are regularly advocated as an imperative measure by which states can prevent the entry of undesirable immigrants, criminals and terrorists (Frontex 2016). However, there is little evidence on the extent to which migration restrictions really impede migration. Previous studies indicate that the various migration policies that make up border regimes tend to produce unexpected ‘substitution effects’, which unintentionally shift migration patterns (de Haas 2011), and growing empirical evidence shows that the border control industry has created a series of ‘migration crises’, thus triggering the very problem it is trying to address (Andersson 2016). This article contributes to this debate by investigating whether we observe differences in the evolution of emigration between 1960 and 2000 when we compare emigration from Caribbean countries whose borders with the former colonial state have remained open and those that have experienced a border closure with the former colonial state. If border controls are effective, closed border countries should show a long-term decrease in emigration and, conversely, the absence of borders should encourage large emigration. Because the Caribbean region was colonised by different states (France, the Netherlands, Britain and the USA1), this analysis also allows us to examine whether post-colonial ties influence long-term migration by making former colonial states preferred destinations and by helping migrants to overcome migration policy restrictions through their migration-facilitating networks, as is often assumed (Beine, Docquier and Ozden 2011). An increasing volume of migration literature has focused on studying the effects of specific migration policies on short- to medium-term migration flows (Hatton 2005; Mayda 2010; Beine, Docquier and Özden 2011; Ortega and Peri 2013; Czaika and de Haas 2016), but the literature on the role of border regimes is much more limited. In this paper, the term ‘border regimes’ refers to the set of policies that aim to regulate the settlement of people by restricting residence and employment rights across national territories. Particularly relevant is the question of what happens to existing migration patterns once a border regime is established between two countries, as in the case of the passage from colony to independence. Through this process, a contiguous political space is divided and the settlement rights of former citizens or former colonial subjects become subjected to specific border regime regulations. We consider that borders are closed in the sense that previously unregulated settlement becomes channelled, and often significantly constrained, through residence and employment restrictions, although entry may not be restricted through travel visa restrictions.2 Naturally, border regimes can also be removed, as for new Member States after joining the European Union. And while a few studies have analysed the effect of the removal of intra-European border regimes (Wallace 2002), research on the establishment of a border regime and its migration consequences are limited to a few country studies (Freeman 1987; van Amersfoort 2011; Vezzoli 2015). The concept of post-colonial ties is examined much more frequently in migration research, with evidence of the general relevance of this historical link in the selection of migration destinations (Hooghe et al. 2008; Constant and Tien 2009; Belot and Hatton 2010). However, the role of post-colonial ties is more ambiguous when analyses focus on specific migration corridors, e.g. former French colonies to France, former British colonies to Britain, with some analyses showing strong post-colonial ties and others no effect (Hooghe et al. 2008; Constant and Tien 2009). The Caribbean and its diverse colonial past offers fertile ground for the analyses of border regimes and post-colonial ties; yet, migration research in the Caribbean region, while rich in case studies, is much more limited in comparative research. In fact, most previous work on regional migration patterns has focused on the links between political status—i.e. independence versus continuous non-sovereignty—of Caribbean countries, economic growth and migration trends. Findings suggest that non-sovereign countries, having experienced stronger economic growth, display increasing immigration, while the independent countries’ lack of economic diversification and slow economic growth makes them countries of net emigration (McElroy and Sanborn 2005; McElroy and Pearce 2006; McElroy 2011; Mitchell and McElroy 2011). Only a few case studies on non-sovereign countries have considered the effects of open borders with the former colonial state and suggested that the freedom of movement leads to high emigration (Audebert 2007), particularly of rural and less educated people (Grosfoguel 1996). However, no comparison has been made of emigration from Caribbean countries where nationals face more limitations to migrate due to the closure of the border. This paper investigates two anticipated effects of closed borders, which have been noted in recent research: First, the implementation of a closed border regime may give rise to a rapid growth of emigration through a ‘now or never’ migration effect, as people anticipate that it will become increasingly difficult to emigrate in the future to the destination countries taking measures to restrict migration (de Haas 2011; Vezzoli 2014c). But in the aftermath of the introduction of border restrictions, emigration intensity often decreases, particularly towards the former colonial state with which the border regime was established (Vezzoli 2014c). Yet, the influence of border closure may be weakened by post-colonial ties and concomitant factors such as language, culture, institutional and educational systems, transport and communication links as well as networks that may facilitate migration to the former colonial state (Thielemann 2006; Hooghe et al. 2008; Beine, Docquier and Özden 2009; Constant and Tien 2009; Belot and Hatton 2010). Second, the closing of the borders by the former colonial state may also influence migration to other destinations, leading to so-called spatial substitution effects (de Haas 2011), diversifying migration to alternative extra- and intra-regional destinations where migration policies are less restrictive than in the former colonial state. By looking at the extent of spatial diversification, we may also learn more about the conditions under which post-colonial ties endure and observe whether there are variations by former colonial state (i.e. former British, Dutch, French colonies or US possessions), or border regimes (open vs. closed). With this article we propose a series of descriptive analyses to highlight the trends and patterns of migration from Caribbean countries and identify visible differences in the way border regimes and post-colonial ties seem to associate with migration trends. We also offer some possible explanations of how border regimes and post-colonial ties may interact with and shape migration. It is not the ambition of this article to provide an analysis of the wide range of factors that might have influenced Caribbean migration over the past 60 years. Rather, we are interested in learning whether, in a region well-known for its high migration volumes (Segal 1987), the implementation of a border regime acts as a barrier to migration and, conversely, open borders encourage large migration, whatever the factors driving migration trends and patterns might be. Similarly, we do not aim to determine the extent to which post-colonial ties drive migration to former colonial states, but whether we observe enduring migration connections between former colonies and former colonial countries despite migration policy barriers. Thus, we set out to examine whether we observe relations between border regimes, post-colonial ties and migration through a descriptive approach. 2. Characteristics of the Caribbean: Migrations and border regimes 2.1 A region rich in migration experiences 2.1.1 Migration before World War Two The Caribbean has been described as the world region most affected by a deep and continuous impact of international migration (Segal 1987). Caribbean societies have their roots in the labour requirements on colonial plantations, which were filled initially with enslaved labour from Africa and after Emancipation with indentured contract labour from Europe (e.g. Madeira Islands) and Asia, primarily from former British India, but also from China and Indonesia. Emancipation, which Britain enacted in 1833, France in 1848 and the Netherlands in 1863, meant that for the first time former enslaved workers gained the freedom to seek alternative occupations. They did so by migrating to other Caribbean islands, including the Spanish Caribbean and Central and South American countries, where they could escape the oppressive working conditions of the plantation system (Thomas-Hope 1978). Caribbean migrants also worked for American companies such as the American Fruit Company, which provided an initial link to later seasonal migrations to the USA (Chaney 1989). With the completion of the Panama Canal in 1914, the onset of the First World War, the crash of sugar prices in 1921 and the advent of the Great Depression, migration decreased and many migrants returned to their origin country (Marshall 1987; Chaney 1989). Starting in the 1920s Panama, and later other Central American countries, began to control the entry of British Caribbean immigrants, which heralded the end of this intra-regional phase of Caribbean migration (Thomas-Hope 1978; Marshall 1987). By the mid-20th century, migration had become an essential part of Caribbean life, particularly in the British colonies. The literature recounts how migration evolved into a social and cultural phenomenon embedded in the Caribbean colonial environment, where expectations of fundamental socio-economic change in the communities of origin were low, while migration ‘offered an alternative to fundamental change’ (Thomas-Hope 1978: 77). In the French and Dutch Caribbean colonies migration was not as prevalent as in the British colonies. Before World War Two migration from the Dutch Caribbean was much more limited and primarily consisted of students from the elite who pursued tertiary studies in the Netherlands, although a small group of working class migrants also participated in migration from Suriname to the Netherlands (Oostindie 2009). Emigration from the French Antilles and French Guiana similarly reflected student migration to pursue secondary and tertiary education in metropolitan France. In the British, French, Dutch and US cases young men were mobilized during World War Two to support the war effort and fight for the colonial state, a process that would make these young men aware of the opportunities available in the metropolitan state and would influence later migrations (Levine 1987; Oostindie 2009; Peach 1991). Generally, the absence of border regimes regulating migration (Hendry 2011) allowed Caribbean migrants to respond to the emergence of opportunities within the Caribbean region and in Central and South America. 2.1.2 The development of extra-regional migrations From the mid-20th century, Caribbean migration patterns underwent an important transformation, primarily through a spatial diversification towards overseas destinations in the Northern hemisphere. As Britain started its post-World War Two reconstruction efforts, British Caribbean migrants contributed to the rebuilding of the British transport and health systems, sometimes through recruitment programmes (i.e. Barbados and Jamaica) but also through spontaneous migration. Concurrently, early recruitment programmes for seasonal workers by the USA and domestic workers by Canada offered regular short-term migration opportunities (Thomas-Hope 2000; Vezzoli 2014a), which strengthened the connections between the Caribbean and North America. Significant migration from the region to France and the Netherlands from their Caribbean colonies started 10 to 15 years after British Caribbean migrations, with peaks in migration in the 1970s (Peach 1991). The rapid growth of migration from the French Antilles, less so from French Guiana, followed the introduction of a state-organised recruitment system which attempted to reduce high unemployment and prevent social unrest in the French Antilles, while filling low-level civil service positions in France, which could only be taken by French citizens (Condon and Ögden 1991). Migration from Dutch Caribbean colonies was not strongly associated with labour. Although small-scale recruitment programmes existed (Vezzoli 2014b), emigration was tied to political anxieties in anticipation of Suriname’s independence and the establishment of its border regime with the Netherlands in 1980 (van Amersfoort 2011). Puerto Rican migration to the USA started many years before World War Two and peaked in the 1950s when on average 40,000 Puerto Ricans were migrating to the USA per year, facilitated by the fact that neither passport nor visa was required. While many factors explain this intense emigration, research has found a high correlation between Puerto Rican emigration and the US business cycles and unemployment rates, linking this migration to migrants’ search for better economic opportunities, higher wages and better quality of life (Levine 1987). The literature acknowledges the importance of migration policies in re-directing (rather than curbing) migration in this period. The US Immigration Act of 1952, also known as the Walter/McCarran Act, restricted immigration in the USA and is believed to have been partly responsible for a spatial substitution of migration from the USA to Britain (Marshall 1987). Conversely, the 1962 British Commonwealth Immigration Act increased restrictions on British Caribbean immigrants, who were required to obtain employment vouchers to be admitted to Britain. In the same period, the liberalization of immigration policies in Canada (in 1962) and in the USA (in 1965) towards non-European countries encouraged migration from the Caribbean as well as secondary migration of Caribbean immigrants already residing in Britain to North America (Palmer 1974; Thomas-Hope 2000; Vezzoli 2014a). 2.2 Evolution of border regimes The decolonisation movement gained strength in the post-World War Two period, but these processes were not homogeneous and reflected national ideologies as well as financial constraints faced by many European countries, which were economically strapped after the substantial war-related losses. While the USA had already established free associated status for Puerto Rico and unincorporated status for the US Virgin Islands by the 1920s, the French government transformed its French Caribbean colonies into integral parts of the French state as Overseas Departments in 1946 and the Dutch government made its Dutch Caribbean colonies countries within the Kingdom of the Netherlands in 1954. Britain, which had the strongest European presence in the Caribbean, was the first colonial power to concede independence to Jamaica and Trinidad and Tobago in 1962 and followed a strategy of decolonization of most of its Caribbean colonies (Christopher 2002; Rabe 2005). Among non-British Caribbean colonies, only Suriname gained independence from the Kingdom of the Netherlands in 1975.3 The establishment of border regimes did not always mirror the changes in political status. We can distinguish three main categories with regards to the evolution of border regimes of Caribbean countries (Figure 1). With the exception of Dutch Caribbean colonies between 1927 and 1951 (Ahmadali and Luk 2015), former Caribbean colonies had an open border regime during colonialism, giving the population full rights to work and reside in the colonial state (Hendry 2011). In Trajectory A, the border regimes have remained open during the entire period before, during and after colonization (the so-called Open–Open–Open Borders, later referred to as open borders). This is the case for the Dutch, French and US colonies that transitioned to a dependent non-sovereign status and that have been guaranteed access to the metropolitan state through full citizenship and freedom of movement and settlement. Trajectory B (Open–Closed–Closed Borders, later referred to as closed borders) represents countries that evolved from open border regimes to closed border regimes. This is applicable to all British former colonies, regardless of political status, as well as Suriname. In the early 1960s, British citizens in the colonies were subjected to partial closure starting with the Commonwealth Immigration Act of 1962 and 1968, which restricted the right of settlement, and the 1971 Immigration Act which restricted all rights to remain. Finally, the 1981 British Nationality Act created a British Dependent Territories citizenship, which did not give the right to abode and essentially gave the Dependent Territories citizens no legal status (Clegg 2005). Because of the immediate restriction on work and residence, all the former British colonies passed from an open to a closed border regime in the 1960s. Similarly, in the case of Suriname where initial immigration barriers were introduced in 1975 and full immigration policies applied in 1980, border closure was recorded in 1975, when free movement and right to residence and work were curtailed. Figure 1. View largeDownload slide Evolution of border regimes of former colonies towards their former colonial state: 3 trajectories among the 25 Caribbean states. Figure 1. View largeDownload slide Evolution of border regimes of former colonies towards their former colonial state: 3 trajectories among the 25 Caribbean states. While the data used in this paper do not allow us to study migration trends beyond 2000, it is important to specify the more recent evolution of Trajectory C (Open–Closed–Open Borders) applicable to the British non-sovereign countries. The Monserrat crisis, triggered by the explosions of the volcano Mount Soufrière starting in 1995 and the departure of two-thirds of the population in the following years, brought to the surface the peculiar lack of a legal status of the inhabitants of British dependencies. By 2002 the British Overseas Territories Bill gave the five British non-sovereign countries the right to British citizenship upon request with matching right of abode in Britain, the right of free movement and residency in the EU and European Economic Area member states (Clegg 2005). The effect of this ‘opening’ of the borders on emigration should be analysed in the future, once post-2000 data are available. These trajectories of border regime evolution are based exclusively on the full set of migration policies established by and with the former colonial state (see Table 1 for country groups). It is meant to provide a lens through which to observe whether such drastic changes in policies are associated with visible shifts in migration intensity and destination choice or, conversely, whether the reinforcement of full mobility and settlement rights in open border countries are associated with any shifts in migration patterns. Hence, the open border category in Trajectory A is intended as freedom of settlement for countries within their own colonial sphere, including their respective former colonial state, not to countries outside of their colonial sphere, which are often subjected to migration restrictions, e.g. French Antilleans migrating to Dutch-sphere Curaçao, Puerto Ricans migrating to the French Antilles or British-sphere Anguillans migrating to Canada. Even open borders countries are affected by migration policies of other potential destination countries. Table 1. Caribbean countries groups by border regime trajectory (and former colonial state) Trajectory A, Closed Borders Trajectory B, Open Borders 1. Anguilla (UK) (Trajectory C since 2002) 2. Antigua and Barbuda (UK) 3. Bahamas (UK) 4. Barbados (UK) 5. Belize (UK) 6. Cayman Islands (UK) (Trajectory C since 2002) 7. Dominica (UK) 8. Grenada (UK) 9. Guyana (UK) 10. Jamaica (UK) 11. Montserrat (UK) (Trajectory C since 2002) 12. Suriname (NL) 13. St Kitts and Nevis (UK) 14. St Lucia (UK) 15. St Vincent and the Grenadines (UK) 16. Trinidad and Tobago (UK) 17. Turks and Caicos (UK) (Trajectory C since 2002) 18. UK Virgin Islands (UK) (Trajectory C since 2002) 19. Aruba (NL) 20. Guadeloupe (FR) 21. French Guiana (FR) 22. Martinique (FR) 23. Netherlands Antilles 24. Puerto Rico(US) 25. US Virgin Islands (US) Trajectory A, Closed Borders Trajectory B, Open Borders 1. Anguilla (UK) (Trajectory C since 2002) 2. Antigua and Barbuda (UK) 3. Bahamas (UK) 4. Barbados (UK) 5. Belize (UK) 6. Cayman Islands (UK) (Trajectory C since 2002) 7. Dominica (UK) 8. Grenada (UK) 9. Guyana (UK) 10. Jamaica (UK) 11. Montserrat (UK) (Trajectory C since 2002) 12. Suriname (NL) 13. St Kitts and Nevis (UK) 14. St Lucia (UK) 15. St Vincent and the Grenadines (UK) 16. Trinidad and Tobago (UK) 17. Turks and Caicos (UK) (Trajectory C since 2002) 18. UK Virgin Islands (UK) (Trajectory C since 2002) 19. Aruba (NL) 20. Guadeloupe (FR) 21. French Guiana (FR) 22. Martinique (FR) 23. Netherlands Antilles 24. Puerto Rico(US) 25. US Virgin Islands (US) Table 1. Caribbean countries groups by border regime trajectory (and former colonial state) Trajectory A, Closed Borders Trajectory B, Open Borders 1. Anguilla (UK) (Trajectory C since 2002) 2. Antigua and Barbuda (UK) 3. Bahamas (UK) 4. Barbados (UK) 5. Belize (UK) 6. Cayman Islands (UK) (Trajectory C since 2002) 7. Dominica (UK) 8. Grenada (UK) 9. Guyana (UK) 10. Jamaica (UK) 11. Montserrat (UK) (Trajectory C since 2002) 12. Suriname (NL) 13. St Kitts and Nevis (UK) 14. St Lucia (UK) 15. St Vincent and the Grenadines (UK) 16. Trinidad and Tobago (UK) 17. Turks and Caicos (UK) (Trajectory C since 2002) 18. UK Virgin Islands (UK) (Trajectory C since 2002) 19. Aruba (NL) 20. Guadeloupe (FR) 21. French Guiana (FR) 22. Martinique (FR) 23. Netherlands Antilles 24. Puerto Rico(US) 25. US Virgin Islands (US) Trajectory A, Closed Borders Trajectory B, Open Borders 1. Anguilla (UK) (Trajectory C since 2002) 2. Antigua and Barbuda (UK) 3. Bahamas (UK) 4. Barbados (UK) 5. Belize (UK) 6. Cayman Islands (UK) (Trajectory C since 2002) 7. Dominica (UK) 8. Grenada (UK) 9. Guyana (UK) 10. Jamaica (UK) 11. Montserrat (UK) (Trajectory C since 2002) 12. Suriname (NL) 13. St Kitts and Nevis (UK) 14. St Lucia (UK) 15. St Vincent and the Grenadines (UK) 16. Trinidad and Tobago (UK) 17. Turks and Caicos (UK) (Trajectory C since 2002) 18. UK Virgin Islands (UK) (Trajectory C since 2002) 19. Aruba (NL) 20. Guadeloupe (FR) 21. French Guiana (FR) 22. Martinique (FR) 23. Netherlands Antilles 24. Puerto Rico(US) 25. US Virgin Islands (US) In particular, North American immigration policy shifts in the 1960s have been deemed important shapers of extra-regional Caribbean migrations: (1) Shifts in Canadian policies to attract skilled migrants provided greater opportunities for Caribbean migrants, who generally possessed secondary and tertiary levels of education (Thomas-Hope 2000); and (2) US policies dismantled racial criteria and opened migration opportunities for Caribbean people with an emphasis on family reunification, allowing Caribbean families to use extended family relations as a migration channel (Maingot 1983). This effect was so great that British Caribbean migration to Britain has been described as being ‘sandwiched between two periods of migration to the Americas’ (Peach 1991: 3), reducing Britain to just one of the many destinations for British Caribbean migrants. However, the effects of North American migration policies on non-British Caribbean migration remains unclear, particularly because many open border countries benefit from freedom of movement and settlement towards their respective metropolitan states, i.e. French Antilles and French Guiana to France, Aruba, the Netherlands Antilles to the Netherlands, and Puerto Rico and the US Virgin Islands to the USA. In fact, the literature suggests that migration from the Dutch and French dependencies remained concentrated towards their related European states (Peach 1991). For Suriname, however, the end of the negotiated preferential migration channels and the introduction of a travel visa requirement to the Netherlands in 1980 (van Amersfoort 2011) made US policies more relevant as migration towards the USA gained some strength (Runs 2006; Vezzoli 2014b). Caribbean countries are also important migration destinations within the region. In fact, while Caribbean migration patterns shifted after the 1940s from intra-regional to extra-regional destinations (Thomas-Hope 1978), intra-regional migration also continues at a steady pace (Thomas-Hope 2000). Within the region, there is ample evidence of various forms of movement, which reflect migrations to work in niche areas such as tourism or off-shore financial services, fluid forms of migration that follow trade and networks of opportunities (Carnegie 1987), circulation and patterns of return for retirement but also as active workers in Caribbean economies (Thomas-Hope 2000). However, it is unclear the extent to which these intra-regional migrations have been affected by border regimes and whether closed borders have in fact reduced overall emigration and open borders within the region have engendered large migrations. 3. Scope, data and methodology This paper proceeds to analyse the evolution of overall, extra-regional and intra-regional emigrations from 1960 to 2000 through a descriptive approach. The objective is to understand whether shifts in emigration intensity and destination may have occurred and how these may be related to the evolution of border regimes with the former colonial state as well as the nature of post-colonial ties. To accomplish this analysis, migrant stock data were taken from the Global Bilateral Migration Database (GBMD) released by the World Bank. This database, which contains bilateral migration population (‘stock’) data for 226 countries, major territories and dependencies for each decade from 1960 to 2000, is based on estimations from census data and population register records (when census data were not available) (Özden et al. 2011). While the release of this database has drastically increased the potential to assess long-term migration trends, it has some limitations. For instance, immigration is likely to be underestimated for countries defining migration on the basis of ‘citizenship’ rather than ‘birth’ because of naturalization. Moreover, missing data was completed with estimated values and, as in most official migration datasets, irregular migrants are generally not taken into account. Because the 1970 data were inconsistent, we decided to drop these data and use only 20-year intervals (i.e. 1960, 1980 and 2000). Other limitations affected particularly data for Britain and its former colonies, whose values seem much inferior to expected values, as well as data for the French Overseas Departments, which are low and erratic compared to data reported in French statistics. Data for the Surinamese in the Netherlands are also much lower than those reported by the Dutch statistical office. Despite these limitations, this database allows us to account for the presence of Caribbean migrants worldwide by origin and destination countries over a 40-year period, during which important changes in border regimes, political status and immigration policies occurred. The low quality of the 2010 migration stock data estimated more recently by the World Bank prevented us from utilizing it, although it would have greatly improved our analytical purpose, as it would have allowed the analysis of the effect of the re-opening of borders for emigration by Britain towards its non-sovereign countries in 2002 (Trajectory C). The United Nations database on international migrant stocks (United Nations 2013) was considered because of the availability of data for 1990, 2000, 2010 (and 2013); however, its total absence of data for residents of the French Overseas Departments and of some other non-sovereign countries made it unsuitable for this analysis. Migrant stock data were utilized in conjunction with UN population estimates to calculate emigration intensities, namely the percentage of migrants among all individuals born in a specific country or group of countries. So, for example, in 2000 the overall emigration intensity of countries which have had an open border over time with their former colonial state was 33 per cent, that is to say 33 per cent of all individuals born in these countries resided abroad. 4. Results 4.1 Overall migrations from Caribbean countries The Caribbean region displays high intensity of emigration, which was already visible in 1960 and has increased in all the countries in the world over time, regardless of which type of border regime they employed. Figure 2 shows how emigration intensity has grown in both those countries that have had constantly open borders (21 per cent of their population lived abroad in 1960 and 33 per cent in 2000) as well as those that have gone from open to closed borders (11 per cent of their population lived abroad in 1960 and 27 per cent in 2000). Countries that have had open borders with their former colonial state and within the colonial sphere experienced stronger growth in emigration intensity between 1960 and 1980, when it grew by 7 per cent, but this growth has since been decelerating. The countries whose borders have closed experienced weaker growth between 1960 and 1980 and a much more rapid growth in the following period, when emigration intensity grew by 12 per cent. This delayed increase may be related to the establishment of border regimes which constrained emigration, but possibly also to the positive appeal of independence which took place in this period and may have encouraged people to stay and take part in the development initiatives in newly independent countries (cf. Vezzoli 2014c). The growth of emigration between 1980 and 2000 has been associated with weak economic growth in many Caribbean countries (McElroy 2011; McElroy and Albuquerque 1988). However, it can be suggested that ‘migration substitution’ effects (de Haas 2011) needed time to take hold. Closing the borders of a country does not necessarily limit emigration, but engenders migration to other destinations (i.e. spatial substitution) or through new channels, e.g. illegally or via family reunification (i.e. categorical substitution). However, time is necessary for migrants to identify new possible locations and/or for new migration channels and new networks to emerge. These trends challenge common assumptions that closing the borders limits emigration, as in fact they seem to slow emigration only in the short-term. Figure 2. View largeDownload slide Evolution of overall emigration intensities by border regime (% of migrants among the population born in the countries), 1960–2000. Source: Global Bilateral Migration Database, World Bank. Figure 2. View largeDownload slide Evolution of overall emigration intensities by border regime (% of migrants among the population born in the countries), 1960–2000. Source: Global Bilateral Migration Database, World Bank. Figure 2 also suggests that migrants from countries that experienced border closure may be pushed into settlement abroad. Stringent entry regulations in destination countries may indeed prevent re-entry, leading to the increase in settlement abroad (i.e. reverse substitution effects) (de Haas 2011), thus increasing the size of migrant stocks. Conversely, the absence of constraints to migration for people in countries with continuous open borders enabled higher levels of circulation and lower settlement patterns, represented by the decreasing levels of permanent settlement abroad. Nevertheless, Figure 2 shows high emigration intensity for open border countries already in 1960, which is primarily due to the high emigration from US dependencies to the US, which alone accounted for over 17 per cent of emigration intensity in 1960. The important contribution of US dependencies to this trend has continued over time, although French dependencies, and to a lesser extent the Dutch dependencies, have also shown some growth in overall emigration intensity. The disaggregation of emigration by extra-regional and intra-regional destinations (Figure 3) shows the dominance of extra-regional emigration, which almost exactly replicates the patterns seen in Figure 2. This shows the important shift in Caribbean migration patterns from intra-regional to extra-regional destinations after the 1940s, which is well reported in the literature (Thomas-Hope 1978). Figure 3. View largeDownload slide Evolution of overall emigration intensities by extra- and intra-regional migration and by border regime (% of migrants among the population born in the countries), 1960–2000. Source: Global Bilateral Migration Database, World Bank. Figure 3. View largeDownload slide Evolution of overall emigration intensities by extra- and intra-regional migration and by border regime (% of migrants among the population born in the countries), 1960–2000. Source: Global Bilateral Migration Database, World Bank. 4.2 Extra-regional migration and the relevance of borders and post-colonial ties When extra-regional emigration intensities are disaggregated to observe migration intensities in the respective former colonial state or the metropolitan state, e.g. Jamaicans towards Britain and Puerto Ricans and US Virgin Islanders towards the USA, we observe (Figure 4) that migration to the metropolitan state has been largely significant for US dependencies and has grown in importance for the French and Dutch sphere countries towards their metropolitan state, which confirm previous evidence that open borders strengthen migrant networks and cumulative migration towards the metropolitan state. Figure 4. View largeDownload slide Emigration intensities in the former colonial states by border regime and former colonial state (% of migrants in their former colonial states among the population born in the countries), 1960–2010. Source: Global Bilateral Migration Database, World Bank. Figure 4. View largeDownload slide Emigration intensities in the former colonial states by border regime and former colonial state (% of migrants in their former colonial states among the population born in the countries), 1960–2010. Source: Global Bilateral Migration Database, World Bank. The countries that have experienced border closure reveal a rapid long-term increase in emigration intensities, but they display diverging patterns of post-colonial effects on migration towards the former colonial states. Suriname, in the Netherlands’ border closure group, shows strong post-colonial ties despite border closure. This is the result of peculiar circumstances leading to independence and the closure of the border regime which encouraged the formation of large Surinamese communities in the Netherlands before border closure. These facilitated migration via family reunification and formation processes after Dutch policies became increasingly restrictive, displaying categorical substitution effects, and stunted the growth of alternative migration destinations (cf. Vezzoli 2014b). Conversely, we observe the rapid weakening of post-colonial effects on emigration from former British colonies towards Britain. Although emigration from British Caribbean countries to extra-regional destinations has grown from just over 7 per cent in 1960 to almost 22 per cent in 2000, it has not been directed to Britain, where the migrant communities have actually shrunk. This offers strong support for claims made in the literature which point to four concurrent patterns: Reduction of immigration due to British immigration policies (Peach 1968, 1995; Marshall 1987). Attraction of North America, where family reunification and skills allowed immigration and the formation of large communities, and where economic prospects were more positive (Palmer 1974; Maingot 1983). The attractiveness of North America leading to step-wise migration of Caribbean people from Britain to North America. Post-retirement return, a trend known to be higher for Caribbean returnees from Britain than North America (Thomas-Hope 2000). A notable exception to this pattern is Montserrat, which saw a great expansion of emigration to Britain in the mid to late 1990s following natural calamities. To understand the importance of the USA in Caribbean migration systems, as it is often viewed, we analysed emigration intensities in North America (Figure 5). We observe that only former British colonies, all of which have experienced border closure with Britain, display the strongest growth in migration to North American destinations, while the other Caribbean countries continue to have negligible emigration intensities in North America. Only US dependencies show high emigration intensity to North America. Thus, although the hegemonic presence of the USA in the western hemisphere is often seen as a strong migration determinant, this does not seem to have affected migration destinations except for its own dependencies and the Anglophone Caribbean. The high degree of spatial substitution of emigration of the former British Caribbean countries to North America seems tied to border closure, the shared English language, but also to the fact that early labour recruitment programmes had established migration networks in the USA and Canada and countries in the British sphere have been historically highly migratory (Thomas-Hope 1978), which may result in high levels of aspiration to find alternative destinations after the introduction of British border controls. Contrarily, the French and Dutch Caribbean were historically less migratory and as migration developed, it did so within the colonial sphere where freedom of movement and settlement were guaranteed. Figure 5. View largeDownload slide Emigration intensities in North America by border regime and former colonial state (% of migrants in North America among the population born in the countries). Source: Global Bilateral Migration Database, World Bank. Figure 5. View largeDownload slide Emigration intensities in North America by border regime and former colonial state (% of migrants in North America among the population born in the countries). Source: Global Bilateral Migration Database, World Bank. Yet there are clear exceptions to these general patterns. For instance, emigration from the Netherlands Antilles was not very high but originally oriented towards North America more than to the Netherlands (see Figure 4, Dutch sphere open border countries did not have a strong presence in the Netherlands) (Oostindie 2009), connections that remain strong even in 2000. This may be associated with the fact that English is commonly spoken in the Netherlands Antilles. Moreover, growing interest in the USA as a migration destination for study reasons is emerging in Suriname, while French Guianese are increasingly interested in Canada, although in both cases the high costs of studying in these countries, particularly in comparison to lower cost or free education in the Netherlands and France, respectively, make the pursuit of these new educational destinations in North America the preserve of the elite (Vezzoli 2015). A growing expansion of ‘worldviews’ beyond the Netherlands and France over the next decade may possibly lead to a greater diversification of non-Anglophone Caribbean emigration towards North America. 4.3 Intra-regional migrations and the relevance of border regimes and colonial spheres within the region Intra-regional migration has remained steadily low (Thomas-Hope 2000), although it is probably higher than the 1 to 3 per cent shown in the results for 2000 (see Figure 3) when we consider the many uncounted forms of movement that occur among Caribbean island nations (Carnegie 1987). They include ‘replacement migration’ of seasonal agricultural labour, particularly in the sugar sector in islands like Barbados, the Dominican Republic, Guadeloupe and Martinique (Segal 1987) and across the Guianas (Vezzoli 2014a, 2014b). Nevertheless, it is interesting to study the intensities and direction of intra-regional emigrations and assess whether border closure with the former colonial states impacted emigration within the region. When we analyse whether open and closed border countries display different patterns of change in intra-regional migration intensities (Figure 6), we find strong evidence of spatial substitution associated with border closure. Caribbean countries whose borders with the former colonial state and with other countries within the same colonial sphere have been closed, display growing emigration towards countries in other colonial spheres. Contrarily, open border countries do not display any shift in destinations since 1960. With regard to the relevance of post-colonial ties, the data suggest that the closing of the borders by the former colonial state may weaken historical migration connections, while continuous open borders encourage migrants to continue relying on cultural and linguistic connections found in countries in the same colonial sphere. However, Suriname is an exception given its strong migration towards the Netherlands. In addition to language, this may be explained by the fact that independence was intensely contested and, concurrently, borders were fully closed only five years after independence (cf. Vezzoli 2014b). Figure 6. View largeDownload slide Evolution of destinations within the Caribbean region, by border regime of Caribbean countries of origin with their former colonial state and by colonial sphere (same or different). Source: Global Bilateral Migration Database, World Bank. Figure 6. View largeDownload slide Evolution of destinations within the Caribbean region, by border regime of Caribbean countries of origin with their former colonial state and by colonial sphere (same or different). Source: Global Bilateral Migration Database, World Bank. The data also suggest that French and US dependencies are strongly oriented towards other destinations within their own colonial sphere (Figure 7). For instance, French Caribbean people migrate to other French Overseas Departments while Puerto Ricans migrate primarily to the US Virgin Islands and vice versa. Interestingly, for US dependency citizens, British-sphere destinations are gaining some relevance, although at an irregular pace, suggesting the facilitating role of language as well as the growing attractiveness of specific British territories. Figure 7. View largeDownload slide Evolution of intra-regional destinations of Caribbean countries, by border regime of Caribbean countries of origin with their former colonial state and by type of sphere and by colonial sphere of destination (same or different). Source: Global Bilateral Migration Database, World Bank. Figure 7. View largeDownload slide Evolution of intra-regional destinations of Caribbean countries, by border regime of Caribbean countries of origin with their former colonial state and by type of sphere and by colonial sphere of destination (same or different). Source: Global Bilateral Migration Database, World Bank. Both British and Dutch sphere countries with border closure show a similar pattern as in Figure 6. For countries in the British sphere this follows the same trend of spatial substitution we see in extra-regional emigration. Conversely, for Suriname, this spatial substitution trend diverges from the continuous concentration of migration in the Netherlands. This is largely due to the strong emigration of Surinamese to French Guiana during the Interior War from 1986 to 1992 and the rapid growth of a Surinamese settled community along the Maroni River on the Suriname–French Guiana border (Léobal 2013; Piantoni 2009). Moreover, Guyana, Suriname’s westerly neighbour, absorbed 15 per cent of Suriname’s intra-regional migration in 2000 (see Table A1 in the Appendix). The only group of countries not conforming to the observed pattern are countries with open borders in the Dutch sphere, i.e. Aruba and the Netherlands Antilles (Figure 7). However, a close examination of the data suggests that this pattern reflects some data inaccuracies, particularly the undercount of population exchanges among the Dutch dependencies in 1960 and 1980, often driven by the growth and later decline of the oil industry. Nevertheless, these countries have also been associated with important historical migrations with other colonial spheres in the region, such as between the Netherlands Antilles and St Kitts and Nevis. Overall, even the open border countries within the Dutch sphere seem to confirm that freedom of movement tends to reinforce migration within the same colonial sphere, although its effect may slowly decrease over time. The French sphere countries have the least spatial substitution, which seems to be linked to high standards of living in the French Overseas Departments, while the highest level of spatial substitution is found among the Dutch sphere countries, with growing emigration towards the French dependencies. An interesting finding is that while the USA was not an important destination in extra-regional emigration from French and Dutch sphere countries, intra-regionally US dependencies clearly exert strong attraction. In fact, in 2000 all countries in the region had an important share of their intra-regional emigration directed towards Puerto Rico and/or the US Virgin Islands (see Table A1 in the Appendix). A number of factors emerge as important in shaping patterns of intra-regional migration. It is clear that a variety of drivers have continued to encourage intra-regional emigration, including political, economic and socio-cultural reasons. Nevertheless, we observe that the colonial sphere continues to be highly relevant for all countries within the region and particularly for countries that maintain open border regimes with former colonial states. However, open border countries have not seen an increase of emigration intensity in countries within the same sphere; rather, any growth in emigration has been to countries outside of the colonial sphere, therefore to countries outside of the free migration area. Concurrently, border closure seems associated not with a decrease in migration but with a substitution of migration destinations and weakening post-colonial migration ties. 5. Conclusions In colonial times, colonial ‘subjects’ largely benefited from rights to free settlement in the metropolitan state and in other colonies. After World War Two, as the decolonization movement gained strength, colonial powers gradually introduced measures to regulate and prevent settlement in the metropolitan state. Britain, in particular, fearing the large arrival of Commonwealth citizens from across the world, adopted a pre-emptive immigration policy by introducing immigration restrictions before many of its Caribbean colonies became independent. The French, Dutch and US governments endorsed more accommodating regulations, retaining open borders, except for the Dutch closure of the border with Suriname starting in 1975. This article set out to understand whether long-term migration patterns between former colonies and their former colonial states showed visible signs of the shift from open to closed border regimes. In particular, we were looking for shifts in emigration intensity over the course of the years studied, any evidence of post-colonial ties engendering strong continuous migration between former colonies and former colonial states and possible spatial migration substitution effects of migration to new destinations as a result of border closure. Contrary to common expectations that closed borders impede migration, our analysis showed that emigration intensity in closed border countries is not only high, but has grown exponentially despite the closure of borders by the former colonial state and other extra-regional destinations through the introduction of migration policies in main destinations in Europe and North America. Border closure enacted by former colonial states has also encouraged step-wise migration from former colonial states towards alternative extra-regional and Caribbean destinations. Countries that have retained open borders and have thus not gone through the ‘shock’ of border closure, have not experienced a similar acceleration of emigration and have rather seen decelerating emigration. This may be related to the fact that several of these countries show economic growth and that the presence of the state may ensure high levels of living conditions and opportunities across the dependencies. Nevertheless, wage differentials exist between these open border countries and their metropolitan states. Thus, this finding contradicts the general assumptions that border regimes are needed to prevent high emigration. Moreover, we show that when borders remain open, there is generally much lower diversification of destinations either extra-regionally or intra-Caribbean. Nevertheless, the results suggest that from the British perspective the border closure has been effective. However, state restrictions have not stopped emigration, but rather deflected it to other destinations. This deflection reflects the establishment of the border regime, although it is clear that the border regime was not in itself the driver of migration as political and economic reasons within the region, as well as a ‘culture of migration’ contributed to continuing migration. However, border regimes and migration policies generally played an important role in shaping the observed emigration patterns. As Britain closed its borders, new labour and family migration opportunities opened in Canada (1962) and the USA (1965). These permitted entry and settlement for several categories of migrants and marked the beginning of long-term patterns of family reunification from the Caribbean to the USA. In addition, Britain was experiencing an economic downturn, while economic growth in the USA offered better opportunities. From a Dutch perspective, however, the closure of the border with Suriname was not effective. Surinamese migration continued to concentrate strongly in the Netherlands, undoubtedly because of language and cultural proximities, but also because the full enforcement of a border regime occurred five years after independence, which led to the formation of large migrant networks before the borders were closed. This paper has elaborated an innovative approach to categorize Caribbean countries by closed and open border regimes. It has included a unique analysis of the associations between types of border regime with changes in extra- and intra-regional Caribbean migration patterns and examined the conditions under which we observe the weakening and strengthening of post-colonial ties. This approach may be useful to analyse the evolution of emigration from other regions where decolonization and border regimes were established, particularly the Pacific region, where open border non-sovereign countries exist, as well as in the Asian and African context. Data limitations narrowed the scope of the analysis and better flow and stock data would enhance this analysis. In particular, the availability of good stock data for 2010 would allow us to test the effects of borders for those countries which experienced border closure in the 1960s and border re-opening in the 2000s (trajectory C). Such an analysis would make an important contribution to our understanding of the migratory shifts not only in association with the closing but also with the re-opening of borders. Acknowledgements The research leading to these results is part of the DEMIG (The Determinants of International Migration) project and has received funding from the European Research Council under the European Community’s Seventh Framework Programme (FP7/2007-2013)/ERC Grant Agreement 240940. The authors are grateful to Hein de Haas, María Villares-Varela, Mathias Czaika, Katharina Natter and Edo Mahendra for their valuable comments on earlier drafts of this article. Footnotes 1. Although the USA is generally not considered to have a colonial past, we classify it as a former colonial state in this article because (1) Puerto Rico held a colonial status until its establishment as an associated free state in 1950 and (2) the US is the sovereign state of the US Virgin Islands. 2. We acknowledge that a closed border regime may be more (e.g. bans on specific nationalities or specific types of migrants such as refugees) or less restrictive (e.g. constraining primarily the entry of low-skilled workers), which could make comparisons of closed border regimes problematic. However, the countries in this study largely adopted similar policies (i.e. labour migration restrictions and constraints to family reunification such as income thresholds) in roughly the same period. Thus, the comparison of these countries over 20-year intervals largely captures the effects of similar ‘closed border regime’ policy packages. 3. The Kingdom of the Netherlands is now composed of four countries: (1) the Netherlands, which includes the European mainland and the three Caribbean municipalities of Bonaire, Saba and Sint Eustatius; (2) Aruba; (3) Curaçao; and (4) Sint Maarten (all in the Caribbean). In this analysis, Curaçao and Sint Maarten are included in the Netherlands Antilles because of data limitations. References Ahmadali H. , Luk N. C. ( 2015 ) ‘ Report on Citizenship Law: Suriname ’, Robert Schuman Centre for Advanced Studies, Edinburgh University Law School . Andersson R. ( 2016 ) ‘Europe’s Failed “Fight”Against Irregular Migration: Ethnographic Notes on a Counterproductive Industry’, Journal of Ethnic and Migration Studies , 42 / 7 : 1055 – 75 . Google Scholar CrossRef Search ADS Audebert C. ( 2007 ) ‘Le cadre politico-institutionel des migrations antillaises: des dynamiques différenciées dams un contexte géopolitique segmenté’, Études caribéennes (online) 8. <http://etudescaribeennes.revues.org/922> accessed 20 Apr 2017. Beine M. , Docquier F. , Özden C. ( 2009 ) ‘Diasporas’, CESifo Working Paper Series No. 2607. Beine M. , Docquier F. , Özden C. ( 2011 ) ‘ Diasporas ’, Journal of Development Economics , 95 / 1 : 30 – 41 . Google Scholar CrossRef Search ADS Belot M. , Hatton T. J. ( 2010 ) ‘Immigrant Selection in the OECD’, Discussion Paper No. 571, The Australian National University. Carnegie C. V. ( 1987 ) ‘A Social Psychology of Caribbean Migrations: Strategic Flexibility in the West Indies’, in Levine B. B. (ed.) The Caribbean Exodus . Westport, CT : Praeger . Chaney E. M. ( 1989 ) ‘ The Context of Caribbean Migration’, Center for Migration Studies (special issues) 7 / 1 : 1 – 14 . Google Scholar CrossRef Search ADS Christopher A. J. ( 2002 ) ‘ Decolonisation without independence ’, GeoJournal , 56 : 213 – 24 . Google Scholar CrossRef Search ADS Clegg P. ( 2005 ) ‘The UK Caribbean Overseas Territories: Extended Statehood and the Process of Policy Convergence’, in de Jong L. , Kruijt D. (eds) Extended Statehood in the Caribbean: Paradoxes of Quasi Colonialism, Local Autonomy and Extended Statehood in the USA, French, Dutch and British Caribbean . Amsterdam : Rozenberg Publishers . Condon S. A. , Ogden P. E. ( 1991 ) ‘ Emigration from the French Caribbean: the Origins of an Organized Migration’, International Journal of Urban and Regional Research , 15 / 4 : 505 – 23 . Google Scholar CrossRef Search ADS Constant A. F. , Tien B. N. ( 2009 ) Brainy Africans to Fortress Europe: For Money or Colonial Vestiges? <http://www.iza.org/en/webcontent/publications/papers/viewAbstract?dp_id=4615> accessed 15 Sep 2015. Czaika M. , de Haas H. ( 2016 ) ‘ The Effect of Visas on Migration Processes ’, International Migration Review , DOI: 10.1111/imre.12261. de Haas H. ( 2011 ) ‘The Determinants of International Migration: Conceptualizing Policy, Origin and Destination Effects’, in IMI Working Papers No. 32. Oxford: International Migration Institute. Freeman G. P. ( 1987 ) ‘Caribbean Migration to Britain and France: From Assimilation to Selection’, in Levine B. B. (ed.) The Caribbean Exodus . Westport, CT and London : Praeger . Frontex ( 2016 ) Risk Analysis for 2016. European Agency for the Management of Operational Cooperation at the External Borders of the Member States of the European Union. Grosfoguel R. ( 1996 ) ‘ Colonial Caribbean Migrations to the Metropoles in Comparative Perspective ’, Journal of Social Sciences , 3 / 1–2 ): 20 – 39 . Hatton T. J. ( 2005 ) ‘ Explaining Trends in UK Immigration’, Journal of Population Economics , 18 / 4 : 719 – 40 . Google Scholar CrossRef Search ADS Hendry I. ( 2011 ) British Overseas Territories Law . Oxford : Hart Publishing . Hooghe M. et al. ( 2008 ) ‘ Migration to European Countries: A Structural Explanation of Patterns, 1980–2004 ’, International Migration Review , 42 / 2 : 476 – 504 . Google Scholar CrossRef Search ADS Léobal C. ( 2013 ) Saint-Laurent-du-Maroni: une porte sur le fleuve . Matoury, Guyane : Ibis Rouge Editions . Levine B. B. ( 1987 ) ‘The Puerto Rican Exodus: Development of the Puerto Rican Circuit’, in Levine B. B. (ed.) The Caribbean Exodus . Westport, CT and London : Praeger . Maingot A. P. ( 1983 ) ‘ Caribbean Migration as a Structural Reality ’, Latin American and Caribbean Center, Florida International University . Marshall D. ( 1987 ) ‘A History of West Indian Migrations: Overseas Opportunities and “Safety-Valve” Policies’, in Levine B. B. (ed.) The Caribbean Exodus . Westport, CT : Praeger . Mayda A. M. ( 2010 ) ‘ International Migration: A Panel Data Analysis of the Determinants of Bilateral Flows’, Journal of Population Economics , 23 / 4 : 1249 – 74 . Google Scholar CrossRef Search ADS McElroy J. L. ( 2011 ) ‘ Island Demography: A Review of Selected Caribbean Contributions ’, Island Studies Journal , 6 / 2 : 245 – 60 . McElroy J. L. , de Albuquerque K. ( 1988 ) ‘ Migration Transition in Small Northern and Eastern Caribbean States ’, International Migration Review , 22 / 3 : 30 – 58 . Google Scholar CrossRef Search ADS PubMed McElroy J. L. , Pearce K. B. ( 2006 ) ‘ The Advantages of Political Affiliation: Dependent and Independent Small-Island Profiles ’, The Round Table , 95 / 386 : 529 – 39 . Google Scholar CrossRef Search ADS McElroy J. L. , Sanborn K. ( 2005 ) ‘ The Propensity for Dependence in Small Caribbean and Pacific Islands ’, Bank of Valletta Review [Malta], 31 :10–16 (Spring). <http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.502.174&rep=rep1&type=pdf> accessed 20 Apr 2017. Mitchell S. A. , McElroy J. L. ( 2011 ) ‘ Emigrant and Immigrant Small Island Profiles ’, Bank of Valletta Review [Malta], 43: 1–11(Spring) . <https://www.google.nl/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&ved=0ahUKEwivy5CCmLPTAhVMI8AKHXvgBWEQFggjMAA&url=https%3A%2F%2Fwww.bov.com%2Fdocuments%2Fbov-review-43-paper-1&usg=AFQjCNGMP5hOQ8T7SVWp8iC6j3Qymo-DyA&sig2=oetFqc2ZsPjQii5-saEUSw&cad=rja> accessed 20 Apr 2017. Oostindie G. ( 2009 ) ‘Migration Paradoxes of Non-Sovereignty: A Comparative Perspective on the Dutch Caribbean’, in Clegg P. , Pantojas-Garcia E. (eds) Governance in non-independent Caribbean: Challenges and Opportunities in the 21st Century . Kingston and Miami : Ian Randle Publishers . Ortega F. , Peri G. ( 2013 ) ‘ The Effect of Income and Immigration Policies on International Migration’, Migration Studies , 1 / 1 : 47 – 74 . Google Scholar CrossRef Search ADS Özden Ç. et al. ( 2011 ) ‘ Where on Earth is Everybody? The Evolution of Global Bilateral Migration 1960–2000 ’, The World Bank Economic Review , 25 / 1 : 12 – 56 . Google Scholar CrossRef Search ADS Palmer R. W. ( 1974 ) ‘ A Decade of West Indian Migration to the United States, 1962–1972: An Economic Analysis ’, Social and Economic Studies , 23 / 4 : 571 . Peach C. ( 1968 ) West Indian Migration to Britain: A Social Geography . London : Oxford University Press . Peach C. ( 1991 ) ‘The Caribbean in Europe: Contrasting patterns of migration and settlement in Britain, France and the Netherlands’, in Research Paper in Ethnic Relations . Coventry, UK : Centre for Research in Ethnic Relations, University of Warwick . Peach C. ( 1995 ) ‘Anglophone Caribbean Migration to the USA and Canada’, in Cohen R. (ed.) The Cambridge Survey of World Migration . Cambridge : Cambridge University Press . Piantoni F. ( 2009 ) ‘La question migratoire en Guyane française: Histoire, société et territoires’, Hommes et migrations, 1278/March–April: 198–216. Rabe S. G. ( 2005 ) U.S. Intervention in British Guiana: A Cold War Story . Chapel Hill : The University of North Carolina Press . Runs A. ( 2006 ) ‘Surinamese Immigrants in the United States of America: A Quest for Identity?’, Master thesis, Latin American Studies, CEDLA, Universiteit van Amsterdam. Segal A. ( 1987 ) ‘The Caribbean Exodus in a Global Context: Comparative Migration Experiences’, in Levine B. B. (ed.) The Caribbean Exodus , pp. 44 – 64 . Westport, CT : Praeger . Thielemann E. R. ( 2006 ) ‘The Effectiveness of Governments’ Attempts to Control Unwanted Migration’, in Parsons C. and Smeeding T. (eds) Immigration and the Transformation of Europe . Cambridge : Cambridge University Press . Thomas-Hope E. ( 1978 ) ‘ The Establishment of a Migration Tradition: British West Indian Movements to the Hispanic Caribbean in the Century after Emancipation ’, Centre for Latin-American Studies, The University of Liverpool . Thomas-Hope E. ( 2000 ) Trends and Patterns of Migration to and from Caribbean Countries. Economic Commission for Latin America and the Caribbean (ECLAC/CEPAL) <www.cepal.org/celade/proyectos/migracion/thomashope.doc> accessed 20 April 2017. United Nations ( 2013 ) Trends in International Migrant Stock: Migrants by Destination and Origin . United Nations Department of Economic and Social Affairs, Population Division . van Amersfoort H. ( 2011 ) ‘How the Dutch Government Stimulated the Unwanted Immigration from Suriname’, in IMI Working Papers, No. 47. Oxford: International Migration Institute. Vezzoli S. ( 2014a ) ‘The Effects of Independence, State Formation and Migration Policies in Guyanese Migration’, in IMI Working Papers, Series 94/DEMIG Project Papers 20. Oxford: International Migration Institute. Vezzoli S. ( 2014b ) ‘The Evolution of Surinamese Emigration Across and Beyond Independence: The Role of Origin and Destination States’, in IMI Working Papers, Series 106/DEMIG Project Papers 28 . Oxford: International Migration Institute. Vezzoli S. ( 2014c ) ‘The Role of the State in International Migration: Exploring the Transition from Colony to Independence’, in IMI Working Papers, Series 102/DEMIG Project Papers 26. Oxford: International Migration Institute. Vezzoli S. ( 2015 ) ‘ Borders, Independence and Post-Colonial Ties: The Role of the State in Caribbean Migration ’, Maastricht Graduate School of Governance . Wallace C. ( 2002 ) ‘ Opening and Closing Borders: Migration and Mobility in East-Central Europe ’, Journal of Ethnic and Migration Studies , 28 / 4 : 603 – 25 . Google Scholar CrossRef Search ADS Appendix Table A.1. Proportion of intra-Caribbean migrants by country of destination in the Caribbean in 2000, for each Caribbean country of origin sorted by closed and open borders Table A.1. Proportion of intra-Caribbean migrants by country of destination in the Caribbean in 2000, for each Caribbean country of origin sorted by closed and open borders © The Authors 2017. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.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)

Journal

Migration StudiesOxford University Press

Published: May 17, 2017

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

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

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

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.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off