Dr Evita Schmieg is Associate in the EU / Europe Division at SWP. This study was prepared in the context of a project on trade and development in relation to the sustainable development goals, funded by the German Federal Ministry of Economic Cooperation and Development.
■ Trade agreements can contribute to long term development – and thus to addressing the causes of flight and migration – as long as they consistently pursue sustainable development and real market opening.
■ The latest theoretical and empirical findings highlight the enormous complexity of the relationship between trade and migration. Other factors (such as war, economic crisis etc.) often play a larger role as triggers of migration than trade policy and trade agreements.
■ One aspect is incontestable: Migration always has a positive effect on trade flows.
■ Conversely, the effect of trade agreements on migration is sometimes positive, sometimes negative: If they lead to increasing per capita income they may temporarily stimulate migration. A certain level of income is required before people are able to emigrate at all.
■ Using trade agreements to create legal migration opportunities in the area of services reduces the incentive for irregular migration. This question is especially relevant for the EU, as it faces the looming problem of labour shortages in the ageing societies of its member states.
■ Ecological and social aspects of investment and trade should be better integrated in all free trade agreements. South Africa and the countries of North Africa offer the greatest potential to expand market access. The EU has already completely opened its markets to most sub-Saharan countries.
■ The respective policy instruments for trade and migration need to be better coordinated in order to reduce the contradictions between them and to address justified concerns about uncontrolled immigration.
Table of contents
Issues and Recommendations
The political debate about migration and flight touches on all spheres of policy. To what extent can policy contribute to stemming migration flows? Or does it function as a catalyst or even a trigger? These questions also apply to trade policy, even if there has been almost no research to date into the connection between trade policy and migration with respect to “combating the causes of refugee movements”. One reason for this is that neoclassical international trade theory posits a clear and unambiguous relationship between the two: trade leads to wage convergence and this lessens the incentive to migrate. However, more recent theories and empirical observations both suggest that the relationship is a great deal more complex. Another hindrance to research is that the effects of trade policy instruments are almost impossible to isolate from other factors, such as internal and external circumstances, wars, climate change, economic crises etc.
One relationship is uncontested: Migration always has a positive effect on trade flows, increasing exports and imports especially in the receiving country. A second is more ambivalent: it is assumed that international trade reduces the causes of migration in the long term, by contributing to economic growth and thus to sustainable development. This does not mean, however, that trade and growth necessarily and automatically reduce migration movements. Research into the “migration hump” shows that growth in poor countries initially spurs migration, with a reduction coming only after a minimum income threshold has been crossed. Moreover, trade liberalisation only generates economic growth under specific conditions. Whether trade liberalisation encourages or deters migration thus depends largely on the specific details of trade policy and free trade agreements: Are they appropriate to the country’s level of development and economic structure? Do they rapidly increase exports while setting limits to negative structural change? What are their effects on internal distribution issues? Do they contribute to raising social and ecological standards in the long term? Are there accompanying reforms in other policy areas?
At the same time, trade policy has itself become an instrument for enabling legal migration. The cumbersome designation for this is “supply of services by natural persons”, introduced in 1995 when the member states of the World Trade Organisation agreed on this as one of four forms of international trade in services. The European Union has a growing interest in this area, where it hopes that concessions can win it better market access outside of Europe (for goods as well as services). Other motives are also involved: the expectation that possibilities for legal migration will help to make irregular migration less attractive, as well as the EU’s vital interest in securing the long-term labour resources needed by the ageing societies of its member states. Trade policy offers a channel for pursuing these objectives. One reason why it has rarely been used to date is that trade agreements lack instruments for dealing satisfactorily with the non-economic and societal concerns of destination countries in connection with migration questions. The instruments of migration policy devote much greater attention to these. It is therefore obvious to more closely link together the instruments of trade and migration policy. But migration should not be subject to conditionality in trade agreements, such as obligations to take back irregular migrants. That would fundamentally undermine the actual objectives of trade agreements.
If one examines the European Union’s trade policy instruments for Africa against that background, several conclusions can be drawn. Firstly, it is important to grant African states access to the EU market for products that they are actually capable of exporting. Although most countries in Sub-Saharan Africa already enjoy free access to the EU market, there is room for improvement in relation to South Africa and the North African states. Secondly, trade agreements must seek to promote sustainable economic growth in the partner countries. Whether the EU’s economic partnership agreements (EPAs) with African states offer adequate and sufficiently flexible responses to negative effects of trade liberalisation remains to be seen in the course of their implementation. Aid for trade can contribute to increasing exports and supporting their development effects. Thirdly, development goals need to be addressed more earnestly in trade agreements with North African states; the EPAs can serve as a model in some respects. Fourthly, future deepenings of the agreements offer opportunities to foreground the social and ecological aspects of trade and production. Fifthly, legal possibilities for migration can be created through trade agreements in the area of services. A meaningful linkage of the instruments of migration and trade policy would be helpful, as would closer cooperation with the private sector.
Migration and Trade: Theoretical Advances and Empirical Observations
The so-called “refugee crisis” of 2015 pushed the topic of migration to the top of European political agenda. Since then all policy areas have been up for scrutiny. To what extent do they contribute to addressing the causes of forced displacement or, conversely, encourage people to leave their homeland? The underlying question here is that of coherence between migration policy and other policy areas. Discussion is especially intense around trade policy, as a sphere of globalisation that is known to produce losers as well as winners.1 Fundamentally, the relationships between international trade and cross-border labour migration are elementary to macro-economic theory.
Trade theory makes no distinction between different forms of migration. When discussing the connections with trade, this study uses the International Organisation for Migration’s definition of migration: “The movement of persons away from their place of usual residence, either across an international border or within a State.”2 Here, however, the discussion is confined to international migration. More specific manifestations are introduced where relevant and necessary, in particular legal labour migration under international agreements. To address the deficits of international trade theory – whose narrow assumptions exclude important aspects of the possible relationships between trade and migration – the investigation is enriched with analytical thoughts and empirical observations.
Trade and Migration in Economic Theory
Wage differentials between countries create an economic incentive for labour migration: the greater the gap, for example between developing and industrialised countries, the stronger the incentive. Neoclassical international trade theory, established by David Ricardo at the beginning of the nineteenth century, assumes that trade between countries reduces wage differentials.3 According to Ricardo, trade is based on national productivity differences that narrow as the respective economies specialise in line with their comparative cost advantages. Paul Samuelson and Ronald Jones expanded the model in 1971 to take account of multiple production factors: strong exporting sectors gain through trade, while sectors that are competing with imports lose. The Heckscher-Ohlin model, developed in the mid-twentieth century by Bertil Ohlin and Eli Heckscher, on the other hand, treats countries’ different production factor endowments as the sole source of international trade. In their understanding trade brings gains to the owners of the factors in which the country is rich, while the owners of scarce factors lose out. The difference to Samuelson and Jones is that the specificity of a factor is generally temporary in nature, while the fundamental endowment – of labour, land, capital – is permanent.
In all the models international trade leads to changes in the distribution of production factors, enables growth and causes wage convergence between trade partners. These effects occur because growth in trade sharpens competition, leading the more competitive enterprises to reduce their production costs by operating more efficiently and improving their productivity. And this wage convergence reduces the incentive to emigrate to a wealthier country in search of a better living. There is therefore an element of substitution between trade and migration: without trade wage differences create an incentive to move to countries where they are higher; trade reduces the incentive to migrate by encouraging wage convergence, growth and poverty reduction.
Yet these theoretical observations provide only a broad-brush description of the actual effects of trade policy. In fact some of the fundamental premises of the neo-classical model (the “perfect market”) do not really apply in reality and this distorts the findings. One case in point is the idea that the adjustment processes caused by heightened competition are immediate and costless (absence of transaction costs). For example, a person who becomes unemployed immediately finds a new job. Another assumption is perfect market transparency: All economic actors are always fully informed about all relevant factors (prices, costs, new jobs). In reality, adjustment processes are associated with significant social and economic costs.
Between 1980 and the mid-1990s many countries unilaterally liberalised their trade under World Bank structural adjustment programmes, and by the mid-1990s it was time to take stock. In the mid-1990s, two studies on the relationship between trade and growth rates concluded that countries with open economies (including poor, resource-dependent countries) demonstrated considerably stronger growth than countries with closed economies and protectionist economic policies.4 Fundamentally both pairs of authors sought to prove that it always made political sense for a country to integrate into the global market. Firstly, they argued, this allowed them to exploit comparative cost advantages and realise economies of scale by participating in a larger market. Secondly, open economies were less susceptible to lobbying and rent-seeking (in the sense of market actors seeking to acquire privileges to boost their own income, such as tax breaks or protective tariffs).5 Thirdly, they said, international trade improved the availability of technology. In fact, however, these studies only demonstrated that a country’s external orientation is one important factor for sustainable success and economic growth. But to conclude that it the sole cause of success is just as false as the opposite line of argument, the idea that the faults of liberalisation prove that protectionism leads to growth.
In many cases the disjoint between neoclassical model and reality was glaring. Trade liberalisation is undeniably associated with short-term negative effects ranging from unemployment to the collapse of entire sectors. Workers are not infinitely flexible, and for most it is not an option to shift seamlessly from dying industries to new dynamic sectors. Developments contradicting the assumptions of economic theory were in fact observed in Kenya, Tanzania and Zimbabwe, where firms responded to import competition by curtailing their output rather than seeking to improve their efficiency.6 Even in rich countries, integration in global markets certainly cannot be said to have led automatically to a reduction in poverty. The fear of trade critics is therefore that those who find their economic perspectives destroyed by trade liberalisation will set off to seek opportunities in other countries.7
The New Trade Theory in the 1980s/90s expanded the neoclassical theory, drawing attention to increasing economies of scale, imperfect markets, external effects and the existence of different technologies in different countries.8 This also cast new light on the relationship between trade and migration. For example, if technological differences – rather than differences in factor endowment – are seen as the basis of trade, trade and migration can have complementary effects.9
Trade means contact, which facilitates migration.
New Institutional Economics investigates the role of information and transaction costs that are neglected by the neoclassical theory.10 Institutions with formal rules, informal restrictions and relevant implementation mechanisms are established to reduce the costs of transactions. Institutions are themselves associated with costs. According to Douglass C. North, the success of an economy stands and falls with how it deals with transaction costs.11 This supplies another new starting point for examining the relationship between trade and migration.
Empirical Evidence and Econometric Models: Does Trade Increase or Decrease Migration?
Trade between countries implies contact between people. And being better informed about a country makes it easier to decide to move there, especially if personal ties already exist. Both these aspects – contact and ties – can be intensified by expanding of trade relations. Trade also leads to economic change: new sectors appear, others are lost, people feel compelled to follow their sources of employment. If people are having to relocate anyway, the inhibition to moving abroad is already lower.12 It is therefore plausible that closer trade relations also encourage migration, in the sense of a correlation between the two.
Conversely, a correlation is observed between trade protectionism and a reduction in migration, as demonstrated by the comprehensive model put forward by Rosmaiza Abdul Ghani and colleagues. Although they emphasise that correlation does not prove causation, they arrive at a conclusion that trade protectionism leads to a decrease in migration.13
It is, however, a great challenge to move beyond correlation in empirical investigations of the relationship between migration and trade: adequate migration data is lacking and isolating trade liberalisation from other influencing factors is methodologically tricky.14 The few deeper analyses of the effects of trade flows on migration therefore say relatively little about the underlying causalities. An investigation of the interaction between trade and migration in the scope of the free trade agreements concluded between the European Union and the Mediterranean states between 1970 and 2000 concludes that increasing exports led to growing migration.15 The author therefore rejects the assumption of the neoclassical theory (which is also found in politics), that liberalisation and expansion of external trade reduces migration. But the study says nothing about long-term effects that might potentially reverse the effect. The reason for the observed relationship is thought to be that international trade strengthens the connections between states and thus facilitates migration. But the model used in the study also demonstrates that the relationship is weaker if the effects of factors as disparate as distance, GDP of countries of origin and destination, population, exchange rate or the existence of historical colonial ties are controlled. In other words: all these factors also influence the outcome.
The United Nations Conference on Trade and Development (UNCTAD) has also noted the existence of a correlation between increasing trade and increasing migration, in a study analysing intra-African food imports in the scope of the regional integration communities SADC (Southern African Development Community), CEN-SAD (Communauté des États Sahálo-Sahariens) and COMESA (Common Market for Eastern and Southern Africa).16 But UNCTAD does not see the relationship between trade and migration as causal; instead the two develop in parallel on account of other factors (for example closer economic relations between countries or regional integration decisions such as easing movement of persons).
Trade liberalisation stimulates migration in the short term, but reduces it in the long term.
Other studies concluded that trade liberalisation encouraged migration in the short term but reduced it in the longer term through the emergence of new trade flows and growth. This so-called “migration hump” was already discussed in 1986 by the Commission for the Study of International Migration and Co-operative Economic Development, which investigated these questions on behalf of the US government.17
Later studies confirmed that the migration hump was an effect of trade liberalisation. For instance, the establishment of the North American Free Trade Area was followed by about fifteen years of increasing migration from Mexico to the United States – but a decrease after that. The length of time before the downturn in migration depends on the technological “head start” of the industrialised country (for example in relation to advantages of mass production), on the extent of income differences, and on the adjustment costs to the developing country associated with the changing circumstances caused by a trade agreement. One study points out that different ex-ante estimates and ex-post calculations on the same topic arrive at contradictory findings.18
In an analysis based on econometric studies, the European Commission identifies the existence of a diaspora as the most important driver of outward migration from Africa (see Figure). “Trade” and “Share of urban population” share second place, “Income differential” follows. According to this research, migration flows are most strongly inhibited when per capita GDP and population are both growing.19 But this relativises the findings suggesting that international trade is of great importance for migration and instead places the latter in relation to the trend of GDP per capita. As already demonstrated in 1997 by a comprehensive gravity model of trade flows, trade changes proportionally to GDP where population size is constant and is inhibited where an increase in GDP is attributable to population growth alone. The explanation for this is that a country with population growth will tend to look inward because it can realise economies of scale within its own markets and is therefore less reliant on trade.20 Absolute and per capita GDP are thus important explanatory factors for both trade and migration.
Within Africa, according to UNCTAD, the strong demand for labour in growth sectors is in most cases the main cause of migration. Rwanda has achieved strong economic growth, partly through foreign direct investment, and has attracted highly qualified workers from elsewhere in East Africa (labour migration being normal within Africa). Ethiopia’s growing formal economy is an important driving force for migration, with special economic zones contributing to industrialisation and creating employment in the footwear, textiles and clothing sectors. These new industries work overwhelmingly for the export economy. In the Horn of Africa informal trade also plays an important role.
Migrants from the West African state of Burkina Faso migrate to work in agriculture and forestry in Gabon. According to UNCTAD, this type of labour migration in export sectors is frequently associated with the use of unqualified child labour in agriculture, in domestic service and in the informal sector. This also applies to migration originating in Burkina Faso, from where children travel above all to Côte d’Ivoire to work in the cocoa sector. In the aforementioned cases there is a connection to the topic of trade, but not to specific instruments such as free trade agreements. Instead migration is incentivised by earning opportunities in specific sectors, in these cases export sectors.21
The correlation between trade and migration is weaker within Africa than for Africa’s relationships with Asia.22 This is another indication that there must be other important explanatory factors for migration, which could potentially be more important than the trade flows.
Income and Migration
Both the theories and the empirical evidence suggest that change in standard of living is an important factor in explaining migration flows. But closer examination reveals that the relationship is by no means as clear-cut as one might assume. As a rule, very little migration occurs where per-capita income is very low: people simply cannot afford to relocate. But when annual per-capita income rises to $7,000 to $13,000 more choose to migrate; above that level the propensity declines again.23 This produces the aforementioned migration hump, whose peak is identified with different income levels depending on the source.24
However, when individual countries are tracked over multiple decades the findings on the relationship between per-capita income and migration diverge enormously. One reason for this could be that the periods investigated – generally fifteen to twenty-five years – are too short to reflect the entire transition. If one believes that the migration hump can also explain developments within an individual country, then one must expect very long transitional periods before the peak is reached:25 Assuming 3 percent annual growth in per-capita income, a country at the level of development of Niger or Burundi will need forty-two years; for countries at the level of Cambodia or Zambia the figure would be sixty-three years. And it must be remembered that 3 percent is a relatively high growth rate for per-capita income. In 2017 it was achieved by only thirteen countries in Sub-Saharan Africa (Benin, Burkina Faso, Côte d’Ivoire, Eritrea, Ethiopia, Ghana, Guinea, Guinea-Bissau, Mauritius, Rwanda, Senegal, Seychelles, Tanzania); the mean for the region as a whole was –0.2 percent, because steady population growth means that good economic performance is not reflected in significant increases in per-capita GDP.26
According to Michael A. Clemens, the causes for rising migration rates before the peak of the hump are declining child mortality (which increases population pressure), growing financial flexibility (because a minimum level of resources are required to be able to migrate), structural change and internal labour migration, increasing inequality and awareness thereof, and the expansion of legal immigration opportunities in receiving countries.27 A special role is played by migration networks, in other words the existence of a diaspora. Most migrants move to regions where they already have family and/or friends and prospects of employment. This connection between country of origin and receiving country can play a role in further consolidating migration flows.28
Thu Hien Dao and colleagues on the other hand focus on the positive correlations between per-capita GDP and growth in the low-income phase, and find that the level of qualifications in the population and (otherwise unspecified) macroeconomic explanations outweigh microeconomic factors based on existing individual contacts to the diaspora. Their explanation for the correlation is that growth leads to broader tertiary education and thus expands the most mobile group.29
Migration deepens trade flows between countries.
The econometric models tell us little of substance about causalities between per-capita GDP and migration. As for the correlation between trade and migration, few general conclusions can be drawn – aside from identifying a correlation between the two factors. The more detailed the models the more diverse the findings. Nor does identification of the migration hump as a phenomenon in itself say anything about the development of a specific country.
Does Migration Lead to More or Less Trade?
While it is unclear whether expanding trade encourages or discourages migration, it is uncontested that migration deepens trade relations between the countries involved.30 Empirical studies show that migration stimulates exports from the receiving country more strongly where the migrants are highly qualified and where the countries of origin and destination are linguistically and/or culturally very different. Migrants from countries where a very rare language is spoken or where corruption, legal insecurity and weak institutions predominate can make an important contribution in the receiving country, by improving the understanding of how their country of origin functions and consolidating the connection to their complicated home markets. In economic terms this makes it easier to overcome informal trade barriers and thus lowers trade costs.31 Migrants are, for instance, capable of accessing and supplying information about customers and procurement opportunities. Trade and productivity in the receiving country may thus profit from their cultural and technological abilities. A quantification of this phenomenon for the OECD states found that a 1 percent increase in migration was equivalent to lowering import tariffs by 3.7 percent, in the case of highly qualified labour migrants and strongly differentiated products the figure can be as high as 21.7 percent.32 UNCTAD investigated the same phenomenon for migration within Africa and also came to a positive finding.33 But this export-boosting effect in the receiving country only occurs if the immigrants are integrated in its economy and labour market. The effect is, as intimated, greater the better qualified the migrant workers are. And especially highly qualified workers are especially mobile.
This equation leaves the countries of origin bearing the burden of brain drain. As well as losing their investment in training the emigrants, they may also forfeit tax revenues and face a general deterioration of perspectives. Sub-Saharan Africa, as the region containing most of the world’s poorest countries, is especially severely affected. More than 20 percent of those who have received a tertiary education there today live in OECD countries.34 The problem is exacerbated by the fact that higher-qualified workers are generally better able to bear the costs of migration and more likely to be admitted by the OECD countries. Especially many have migrated to industrialised countries experiencing labour shortages, notably in the information technology and communications sectors.35 The WTO did, however, point out at the end of the 1990s that certain countries trained more workers than their labour markets were able to absorb,36 which in these cases relativised the brain drain issue.
Immigration also tends to stimulate imports in the receiving countries, because many immigrants retain their consumer preferences and are able to arrange corresponding imports through their contacts in their country of origin. For instance UNCTAD found that the growing diasporas created by intra-African migration have led above all to a rise in food imports from the countries of origin.37 For countries of origin in turn, the diaspora frequently becomes an important source of capital (via remittances and investments) as well as know-how and technology.38 The volume of remittances has increased significantly since 2000. In 2010 they represented 42 percent of private capital inflows to Africa; by 2016 this had increased to 51 percent.39 These inflows – which exceed total private investment – can in turn invigorate trade flows.
It is empirically clear that migration has a stimulating effect on imports and exports. And, as Ghani and colleagues demonstrate, restricting migration has inhibiting effects on international trade.40
The Role of Migration in Trade Policy
Given that neoclassical international trade theory automatically assumes that expanding trade reduces migration, a specific examination of the role of migration in relation to trade policy instruments would appear superfluous. Indeed, the term “migration” is absent from the WTO rules and one searches in vain in WTO texts for references to the work of other institutions dealing with migration, such as the International Organisation for Migration (IOM). Yet in fact the issue of labour migration has been a topic in trade negotiations for more than two decades – but under a different label.
GATS – Market Opening for Supply of Services by Natural Persons
The General Agreement on Trade in Services (GATS) came into effect in 1995, as one outcome of the very comprehensive last round of negotiations on liberalising world trade (the Uruguay Round). GATS is explicitly not about labour migration. Its object is movement of people across borders to supply services. In practice, however, this also creates legal possibilities for labour migration. The Agreement defines four Modes of cross-border trade in services:
Mode 1: Cross-border trade (example: insurance services offered abroad)
Mode 2: Consumption abroad (example: tourism)
Mode 3: Commercial presence (service supplier establishes a branch or subsidiary abroad; example: car rental abroad)
Mode 4: Presence of natural persons (example: care workers).
An Annex specifies that Mode 4 applies to “measures affecting natural persons who are service suppliers of a Member, and natural persons of a Member who are employed by a service supplier of a Member, in respect of the supply of a service”.
The mainstream interpretation within the WTO is that GATS does not serve to open the possibility of labour migration to persons employed in the importing country (although certain countries do precisely that, including the United States with the so-called H‑1B visa).41 The Annex explicitly states that: “The Agreement shall not apply to measures affecting natural persons seeking access to the employment market of a Member.”42 It also clarifies that: “The Agreement shall not prevent a Member from applying measures to regulate the entry of natural persons into, or their temporary stay in, its territory.”
It is, however, almost impossible to make a sharp distinction between “migration” and “Mode 4”, as underlined by a statement published jointly by the IOM, the World Bank and the WTO: “GATS Mode 4 is not a migration agreement, and it was not created with direct regard to the policies, practices and administrative mechanisms utilized by states in managing temporary labour migration. Nonetheless, that is the context within which implementation of existing Mode 4 commitments, and consideration of potential new Mode 4 commitments, takes place.”43 The European Commission notes that “[T]rade agreements, and in particular those negotiated by the EU, aim to steer clear of migration policies, by adopting a different vocabulary (professionals vs. workers, mobility vs. migration) and by underlining the temporary nature and specific purpose of stays.” Yet, as the document goes on to point out: “[I]t is also clear that the liberalisation agreed in those trade agreements cannot have any effect as regards entry and temporary stay of natural persons for business purposes if no adequate admission policies are put in place in the host countries.”44 In other words, efforts to draw a clear distinction between trade liberalisation under Mode 4 and arrangements for migration must be understood primarily as a political declaration of intent. It has little basis in reality.
Negotiating Interests in Mode 4
Countries of origin and destination each have their own economic interests in connection with trade in services supplied by natural persons (Mode 4). States call on other states to open their markets to service exports, and make offers – in effect “concessions” – on opening their own markets to precisely defined services. The term “concessions” springs from the same logic as that followed by the GATS negotiations.
Above and beyond the narrow economic interest in trade in services, there is another important reason for including labour migration in the scope of trade talks: Expanding the scope covered by the negotiations makes it easier to achieve an outcome because interests can be balanced across a larger set of issues. If one side – let’s say the EU – agrees to a trade partner’s demand to grant market access to service suppliers, it improves its own chances of receiving concessions in other areas such as market access for goods or intellectual property protections. For example in the WTO’s abortive Doha Round the question of Mode 4 was a major issue for India, which wanted the United States to grant 250,000 additional visas.45
Receiving countries in particular are concerned to avoid losing control over their own immigration policy in the course of liberalisation of trade in services. Opening markets to foreign service suppliers certainly does not mean the same as conceding unrestricted migration. A range of labour market and migration policy instruments is available to ensure this. In the negotiations states have the right to precisely define the professions, periods and conditions for which they open their markets. One example from the EU’s Economic Partnership Agreement with Caribbean states is limited access for fashion models, chefs and music groups.46 States are free to stipulate how many service providers may enter within a specified period and for how long they may stay; this may be limited to a few months. Importing states generally employ additional instruments such as needs assessments to protect their own service sectors.47 This allows them to counter the widespread fears that concessions on Mode 4 would lead to the displacement of local workers. While this may be politically expedient, in practice such provisions frequently undermine the agreed concessions. Very rarely are the criteria underlying the needs assessments defined, creating great leeway for inconsistency in their application. Requiring licences and formal qualifications also has inhibiting effects. This applies in particular to the condition (which is included in most schedules) that persons entering under Mode 4 possess a valid employment contract in their country of origin. This naturally subverts the incentive to migrate in the first place.
Investment (largely from industrialised countries) accounts for 60 percent of trade in services; natural persons account for 5 percent.
Despite all these restrictions, market opening under Mode 4 is a matter of great political sensitivity. The WTO has concluded that most member states have avoided comprehensive commitments to opening for natural persons under GATS: Not all states have offered commitments under Mode 4, most have added numerous exceptions, and market access bound in the WTO therefore often lags behind the actual market access situation.
The Actual Role of Mode 4 in World Trade
Although poorer countries in particular have a great interest in exporting services supplied by natural persons, Mode 4 plays only a marginal role in global trade in services with a share of less than 5 percent. Mode 3, commercial presence in the context of foreign direct investment, accounts for 55 to 60 percent.48 These trade statistics reflect the power imbalance between industrialised countries and developing countries. Services traded are usually those that are exported principally by industrialised countries that possess a highly qualified workforce and capital seeking investment opportunities. Additionally, the option of temporary migration under Mode 4 is frequently only granted in connection with the trade partner opening their economy to outside investment; many of the concessions apply to movement of staff within corporations. Obviously, poorer countries without significant investments abroad cannot make use of these possibilities.49 In addition, the vagueness of the terms used in relation to Mode 4 (business visitors, senior executives) offers greater leeway for diverging national interpretations than is the case with other modes.
Motives for export and import of services under Mode 4
Migration countries of origin
Migration receiving countries
Source: Author, using data from IOM, World Bank and World Trade
Market opening for natural persons in the scope of trade agreements requires national agencies in the participating countries to modify their visa regulations accordingly. This creates great problems in practice, because visa regulations are often not brought into line with new free trade agreements. This further devalues the small number of market access offers granted to date for natural persons. India proposed a solution for this problem in the WTO: a special GATS visa, which would be given administrative priority ahead of other visa applications.50 Although this does not yet exist at WTO level, some bilateral trade agreements with the United States do provide for such visas. Certain bilateral migration agreements ease access through occupational shortage lists, which obviate the need for case-by-case labour market tests.51 The EU is currently working to more closely connect trade and migration policy.52
Even unilateral preferences for the Least Developed Countries (LDCs), on the basis the so-called LDC waiver introduced by the WTO in 2011,53 have failed to significantly stimulate Mode 4 imports from these countries. A waiver is necessary because the WTO rules require all countries to be treated equally. The privileges permitted by the waiver are not negotiated between trading nations but granted unilaterally, in this case twenty-five industrialised countries including the EU grant the LDCs time-limited preferences until 2030. However they apply largely to Mode 2, where consumers make use of services abroad. This is an area where there are already few restrictions, so the preference schemes do little more than codify a pre-existing level of liberalisation.54 According to the United Nations Committee for Development Policy these arrangements have almost no effect on service exports from this group of countries – least of all in relation to supply of services by natural persons. Moreover, because the preferences are granted unilaterally they can also be taken away again. This uncertainty means they are of less value than the results of trade talks, which are unrestricted.
In the greater scheme, labour migration – in the sense of provision of services by natural persons – occupies a marginal place in world trade. This is a function of the domestic political fears that the issue provokes almost everywhere. However the negotiating structure of GATS itself may also contribute to the slowness of progress on concessions on Mode 4. The approach of scrupulously avoiding any impression that this is about labour migration means that the societal risks associated with cross-border movement of persons are neither addressed nor curtailed. There are also technical deficits: the categories of service suppliers are not clearly defined and the structure of the schedules is complicated.
Trade talks and terms also need to be better coordinated with the needs and requirements of national immigration policy. Categorisation under “trade in services” generally creates incoherence with national immigration rules.55 Visa regulations, exemptions and other questions of movement of persons still remain explicitly outside the scope of GATS. Action on multiple fronts would be required to bring this sphere into line with national immigration frameworks. The countries of origin would like to see the structure of Mode 4 schedules modified to address their concerns about brain drain; the receiving countries want to ensure that foreign migrants do not overstay their work permits. The researcher Marion Panizzon sees a shared commitment to “ensure the timely and voluntary return of workers” as a precondition for achieving more concessions on Mode 4 within GATS.56 That would mean including elements normally found in migration agreements.
Better coordinate trade and migration instruments.
The idea of expanding the scope touches on the current fundamental understanding of free trade negotiations, which treats Mode 4 purely as a question of trade. Yet WTO rules and trade agreements do in fact have a bearing on internal regulatory matters, for example in relation to trade facilitation, intellectual property protections and non-tariff barriers. So the proposed changes would not be completely alien to the system. In view of the labour shortages in numerous industrialised countries it might be worth considering combining aspects of trade and migration agreements in order to better address the justified concerns of the receiving countries.
Bilateral migration agreements are still regarded as the primary instrument of labour market management. The International Labour Organisation put the total number of bilateral labour agreements in 2015 at 358.57 These agreements permit temporary migration to be controlled within a narrowly defined framework. Since around year 2000 certain countries have expanded the possibilities for temporary migration and sought to improve its effectiveness through a multitude of individual laws and instruments. According to the political scientist and migration researcher Steffen Angenendt, agreements concluded since the global economic crisis of 2008/09 prioritise the goal of restricting irregular migration and ensuring that rejected migrants are repatriated.58 Demographic ageing and the growing shortages of skilled labour affecting most industrialised countries will also play a role in future.
The ILO’s Model Agreement on Temporary and Permanent Migration for Employment of 1949 has in the meantime been revised and expanded.59 It contains comprehensive texts covering all aspects of dealing with labour migration, from working conditions and social insurance to double taxation and return. But the main thrust of the Model Agreement is defining the framework for labour migration, rather than describing possible jobs. There is therefore little overlap between it and the GATS rules for Mode 4, even though they would actually complement each other very well: although trade concessions in the sphere of services and movement of persons precisely define the possible areas of employment for labour migrants, they exclude the aspects that the Model Agreement proposes to regulate.
Finding the correct balance between regulation and liberalisation is a great challenge. A comprehensive study by the ILO reviewed more than 151 bilateral agreements and memoranda of understanding on migration of low skilled workers and laid out recommendations. Some of these relate to mechanisms that are already common in free trade agreements, and in some cases have already been tested in practice. For instance “a system for regular monitoring and periodic evaluation” is proposed, principally through joint committees.60 EU trade agreements already contain such instruments, which could be supplemented with migration aspects in connection with Mode 4. The ILO also advises “[u]sing multilateral and regional forums and regional integration areas to arrive at consensus on mutually beneficial improvements to agreements and minimum standards”. This would also suggest using regional trade negotiations on services for the issue of migration/ Mode 4, for example in connection with economic partnership agreements with the European Union. Other proposals made in the ILO study resemble those that have already been put forward in connection with free trade agreements. Thus it would be important to support implementation of trade agreements by initiating reforms in other areas. For the area of migration that would mean for example minimum wages, social insurance regimes and reintegration measures.
From Theory to Practice: Can Trade Policy Influence Migration?
Findings of the Analytical Section
Beyond the observed correlations, an unambiguous relation between trade liberalisation and migration cannot be demonstrated empirically. Investigations to date arrive at strongly diverging findings depending on which additional factors are included: trade can sometimes contribute to increasing migration, sometimes decreasing it, or other factors may affect both.
It turns out that a series of factors associated with trade flows are also important explanatory factors for migration flows. Per-capita GDP is especially important. But this also raises the fundamental question of the extent to which international trade can contribute to sustainable economic development.
Combining the discussion around New Trade Theory and New Institutional Economics with empirical observations produces differentiated policy recommendations for leveraging trade policy to promote growth and development as the necessary long-term basis for disincentivising migration. The prevailing opinion until the early 1990s – that the free market with minimum state intervention was the ideal basis for economic development – has in the meantime been rejected. More recent research suggests that much more complex policy approaches are required to do justice to the diversity of situations of LDCs:
The globalisation process produces winners and losers. Which countries these are depends on multiple factors: historical happenstance, concrete economic structure, level of development, geographical proximity to flourishing economic centres.61 Successful trade policy must therefore take each country’s specific situation into account: trade does not automatically lead to convergence. The new models therefore stress “the necessity of autonomous development in the periphery”.62
Trade reforms must be embedded in broader reform strategies. Although all countries with sustained growth have lowered trade barriers, but trade reforms could only prove successful where other restrictions were also addressed at the same time through prudent economic policy, expansion of trade-related infrastructure and institutions, and investment in human capital.63 Efficiency can be enhanced by reducing internal and external trade costs, including those for infrastructure, as well as market opening for intermediate goods.
Almost all successful countries have implicitly or explicitly promoted their exports. But that worked only where the administration is efficient and well-functioning. Especially in poorer countries this fundamental precondition is not fulfilled.64
A country will only be able to profit from free flows of trade and capital if it possesses a minimum of institutional capacity and human capital. Otherwise trade liberalisation will lead to instability. Institutional capacities are tied to preconditions: innovation, education and training are among the most important for success under glob-
Relationships between trade and migration according to theoretical, econometric and analytical research
Type of trade/migration relationship
Trade leads to wage convergence and increase in GDP, which lowers incentive to migrate.
Other motivations than wage differences lead to trade and factor price equalisation (New Trade Theory: economies of scale, technological differences, external effects), trade can stimulate or reduce migration.
First more migration, then less
Trade liberalisation initially causes adjustment costs that drive migration. In the long term GDP increases, migration decreases (migration hump).
Contradicting the predominant assumption of the 1980s, trade liberalisation can exacerbate inequality and poverty. Market integration should therefore be backed by social and distribution policies addressing those problems.67
The role of trade policy should not be overestimated as many factors affect migration.
Many influences and underlying circumstances play a role in deciding whether trade policy helps in the long term to reduce the incentive to migrate as per-capita income rises. Weakening import protections, and even granting easier access to markets, do not automatically lead to rising standards of living. The development and migration policy expectations placed on trade agreements should therefore be kept moderate. The defining factors for a country’s development are above all its internal conditions and strategic support for trade reforms through reforms in other policy areas, through rule of law and good governance. The outcomes of trade agreements thus depend crucially on their implementation. In developing countries this is hampered by inadequate financial, administrative and entrepreneurial capacities.
Against this background, a trade policy that seeks to take into account the connection between trade and migration must make a long-term contribution to sustainable growth perspectives. That requires the scope and shape of trade policy instruments to satisfy certain conditions.68 The following elements play a role:
In order to avoid negative effects such as unemployment, sustainability impact assessments must be used to identify the chances and risks for the trading partners before trade agreements are concluded.
Positive effects of agreements should occur as rapidly as possible. Agreements therefore need to include real market opening for significant export products.
This demands prudent sequencing of the liberalisation steps foreseen in the agreement: short-term negative effects attributable to the process of adjustment to heightened competition must be avoided where possible through longer transitional periods for lowering import tariffs in endangered sectors. Sectors classified as important for internal development need to be protected.
On the other hand, tariffs on important production inputs – which may also include services – should be lowered immediately.
Agreements must be flexible enough to satisfy requirements such as food security and domestic industrial development (protection of infant industries).
In the case of problematic developments following trade liberalisation it is important to be able to respond with flexible safeguard mechanisms.
Countries must actually be able to make use of the new market access granted by agreements. But especially in developing countries the capacities and financial prerequisites are frequently lacking. Support for poorer countries through aid for trade is therefore of central importance for successful implementation.
Aspects of social and ecological sustainability must also be addressed in order to prevent negative effects that cause displacement (such as environmental harm associated with manufacturing).
Checklist Trade and Migration
The following checklist on the relationship between trade and migration serves to summarise and assess the most important possible impacts (also of individual trade instruments). But it can only approximate the complex connections between trade instruments and migration flows, and makes no claim to completeness. Nevertheless, it does identify the factors and indicators that are known and must be taken into account in impact analyses.
Effectiveness of the trade instrument
Are the participating actors/importers in Europe and the producers in the partner country properly informed about the terms of the agreement?
Do the producers possess the capacities required to increase production?
Do the internal circumstances – including infrastructure – permit production to be increased? Or do other factors such as access to credit and other production inputs stand in the way of expanding production?
Does the government pursue export promotion?
Are other forms of trade-related support available to resolve the aforementioned problems?
Framework condition: trade agreement oriented on sustainable development
Has real market access been granted on a scale large enough that it cannot be negated by other market factors (general market development, volumes and prices)?
Have limits to negative effects of trade liberalisation on people and sectors been set through broader reform strategies (social security, education/training policy, labour market policy)? Are long term positive effects on social and ecological standards expected?
Framework condition: internal distribution issues
Do growing exports actually lead to a reduction in poverty (through wage rises or growing profits for small producers)? Is there freedom of association or are other forms of worker participation available?
Is the overall distribution of the effects of the agreement socially equitable? Which groups in society profit?
Distribution issues (local/regional): How does the situation vary in different regions? Are structurally weak regions especially affected (positively and/or negatively)?
Can short- and long-term effects be differentiated, and if so, to what extent?
How does the agreement affect those willing and able to migrate?
What attitudes to migration are found among the social groups that gain or lose through the agreements?
Are there other factors affecting migration? What is the security situation for life and property, in terms of the internal political, social and ecological circumstances?
Is there a diaspora? Are there links to the local population?
Does the agreement create possibilities for legal migration?
General framework conditions
What national political, social and ecological factors affect the country’s development (natural disasters, changing agricultural conditions, new taxes etc.)?
What international factors might influence economic and social development more strongly than the trade agreement (for example economic and social crisis)?
Overview 2 illustrates how the checklist can be applied, using the example of a putative increase in Tunisia’s olive oil quota in a trade agreement with the European Union.
Perspectives for the Topic of Migration in EU Trade Policy
There are several options for including the topic of migration in trade policy. First it can be introduced directly into free trade agreements by creating possibilities for legal migration, as described above. Free trade agreements usually contain a dispute settlement system. In order to strengthen the incentive to comply with the provision of an agreement, migration issues may even be linked with the possibility of trade sanctions. Trade agreements can also formulate conditionalities with regard to migration issues or preconditions that have to be fulfilled before ratification. Another possibility consists in partner countries identifying forms of cooperation.
In future the EU will integrate migration aspects more closely into its trade policy. That is what the European Commission sets out in its Trade for All strategy, where it names concrete areas for action:69 Mobility provisions are to supplement trade negotiations for example, and the conditions for intra-corporate transfers are to be relaxed reciprocally in the scope of market access talks. The strategy also calls for better coordination of immigration and visa conditions with trade policy and easier recognition of professional qualifications. Support for exchange and training should “facilitate the effective use of mobility provisions in FTAs”.
Example of application of the checklist for the theoretical case of “increased olive oil quota for Tunisia”
1. Effectiveness of the trade instrument: increase in olive oil quota
2. Framework condition: trade agreement orientated on sustainable development
3. Internal framework condition distribution issues
4. How does the agreement affect those willing and able to migrate?
5. Broader circumstances
Easier access to the European market for natural persons supplying services (Mode 4), will play a larger role in future free trade agreements. Services are an important export sector for the EU, lending it an offensive interest in further liberalisation in this area. The EU accounts for 22 percent of global trade in services (United States 15 percent, China 8 percent) and has a positive balance of trade exceeding €150 billion/year.70 If it wants to expand its service exports, the EU will certainly have to demonstrate more flexibility on Mode 4. This is because presence of natural persons to supply services in the EU is the most important offensive trade interest in the area of services for many partner countries, at least on a temporary basis. The European Commission assumes that this issue will remain on the agenda. Additionally, the EU’s immigration rules will have to be coordinated more closely in order to ensure that the market access granted to trading partners under free trade agreements can actually be realised,71 given that the EU’s administrative and visa conditions are decisive for whether those supplying services abroad can actually make use of trade preferences granted. Caribbean states have complained about this in no uncertain terms in relation to their economic partnership agreement with the EU.72 For instance if a businessperson requires separate visas for different EU countries, this not only increases the time and cost involved, but also complicates scheduling. “[D]evising schemes to fulfil trade commitments and facilitating the admission of third-country nationals for business purposes” is therefore high up the Commission’s agenda.73 The language makes it clear that the EU is talking principally about facilitating access for highly skilled workers (“professionals”).
So which aspects will need to be considered if the EU wishes to expand market access under Mode 4 in order to satisfy partner countries’ demands in trade talks? First of all, there are important arguments for addressing the topic of Mode 4 in bilateral or regional contexts rather than multilaterally in the WTO. Market opening at WTO level is comparatively difficult to achieve, because it must apply to all trading partners under the most-favoured nation principle (although exceptions are possible). The context of free trade agreements is more conducive to defining groups of service providers for which access is to be eased in the economic and social interest. Additionally, the smaller number of parties means that the scope of application is more clearly identifiable.
Because of the brain drain issue, which represents a considerable drag on the development opportunities of poorer countries, as well as public resistance in receiving countries, it is recommended – at least at first – to enable migration on a temporary or circular basis. “Circular migration” – where (labour) migrants return at least once to their country of origin before relocating again to the destination country – could make a contribution to know-how transfer to developing countries without causing brain drain and at the same time accommodate the labour needs of ageing societies in the EU. The development idea – that circular migration is in the interests of country of origin, destination country and of the migrants themselves – corresponds to the basic understanding of circular migration.74
For their part, developing countries are advised to link the access they concede for foreign direct investment (an import under Mode 3), to their own demand for market access under Mode 4. The countries of origin of investments could for example commit to enabling temporary migration for the purpose of in-house training. In particular in relation to the demands of the global labour market, working for a period at the headquarters of a multinational cooperation would grant employees from developing countries the opportunity to expand their knowledge and experience.75 Although such conditions are prohibited under the WTO Agreement on Trade Related Investment Measures (TRIMS), GATS, whose Mode 3 applies to commercial presence, does permit national restrictions in the schedules. In their own self-interest developing countries should make more use of these flexibilities.
The connection between investment and temporary migration of employees could also be strengthened by closer cooperation between governments in host countries and the private sector in countries of origin. UNCTAD proposes establishing platforms for dialogue and information exchange between private sector and government,76 in order to generate employment creation projects and counter the negative perceptions of migration. German business has launched an initiative (“Afrika kommt!” or “Africa is coming!”) to bring successful young Africans to work in German companies for a limited time.77 Such initiatives could be productively linked to the implementation of free trade agreements, where cooperation is also required with the businesses that are supposed to be making use of the new opportunities.
It would also be conceivable for the EU to include the topic of migration in future trade agreements, with a conditionality tying restrictions on migration to incentives for trade. Signs of this are found in the Fitness Check Migration, which states: “Trade policy should take into account the policy framework for the return and readmission of irregular migrants,”78 without further concretising the point.
Conditionality already features in the EU’s trade policy (although not in relation to migration) in the Generalised System of Preferences Plus (GPS+), which was designed as an incentive system for sustainable development and good governance. It reduces import tariffs to zero for “vulnerable” low and lower middle income countries that implement twenty-seven international conventions in the areas of human rights, labour rights, environmental protection and good governance. Unilateral trade preferences are thus used to reward countries that make special efforts to improve social and ecological sustainability.
The United States also uses negative conditionality in its preference system, to enforce its own foreign policy objectives. The US African Growth and Opportunity Act (Washington’s system of preferences for states in sub-Saharan Africa) thus grants privileges only to countries that refrain from engaging in “activities that undermine United States national security or foreign policy interests”.79 In practice the instrument is used to threaten countries with withdrawal of AGOA trade preferences if they move to apply tariffs that would also affect American exports. This occurred in 2018 for example, when East African countries proposed banning imports of used clothing.80
Including sanctions against irregular migration would undermine the fundamental concept of trade agreements.
Fundamentally it would be possible to apply migration-related conditionalities to trade agreements, for example in relation to returning irregular migrants. Conditionality could even be linked to trade sanctions, such as the reinstatement of tariffs, via the dispute settlement system that is part of every EU trade agreement. However, instrumentalising trade policy for purposes of migration policy would undermine the stability and predictability of the international trade framework, and as such contradict the central objective of trade agreements. It would also mean the EU abandoning its line of not using sanction instruments in trade agreements to enforce foreign policy goals.
Migration in Trade Arrangements between the EU and Africa
The question of the extent to which trade policy is connected to migration is especially relevant for Europe in relation to Africa. Although the European Union remains Africa’s most important trade partner, its predominance has waned. In 1995 the members of today’s EU still accounted for more than 40 percent of Africa’s exports; by 2018 the figure had fallen below 25 percent.81 Africa is not a major trading region for the EU, with a share of 7.1 percent of EU imports and 7.5 percent of exports (2018) – even if the absolute volume (exports and imports) has roughly tripled since year 2000 (to about $350 billion, €307 billion).82 More than half of the imports came from just four countries (South Africa, Algeria, Nigeria and Morocco). Sub-Saharan Africa accounts for just 3.7 percent of the EU’s international trade. To this day the EU principally imports raw materials from Africa (with the exception of Morocco) and exports principally manufactured and processed goods. In other words, trade between the EU and African states is asymmetrical. And trade provisions differ depending on country and region (see map).
Migration from Africa, on the other hand, has become an increasingly important issue for the EU. Legal immigration from Africa to Europe was in fact falling until 2012, since when it has remained broadly constant (2016: 288,000). But the numbers of asylum-seekers migrating via irregular channels have spiked, from less than 100,000 per year (until 2013) to more than 200,000 in 2016, largely from Sub-Saharan Africa.83
On the basis of the above discussion, two questions are especially relevant for assessing EU trade instruments for Africa and their effects on migration: (1) Does the question of migration play a role in EU trade instruments? This may relate to both irregular migration and associated issues such as repatriation, as well as questions of legal migration and market opening for service providers. (2) Do the details of the trade provisions promise a contribution to sustainable development and to creating future prospects?
Association Agreements with North African states
The EU is linked to the states of North Africa by association agreements concluded under the Euro-Mediterranean Partnership (indicated in yellow on the map). In the scope of the Partnership the EU has free trade agreements – largely restricted to trade in manufactured goods – with all countries apart from Syria and Libya.84 The EU has also discussed with Morocco and Tunisia the possibility of incorporating wider issues and negotiating deep free trade agreements, including steps towards market opening for services and agriculture. Talks Tunisia are ongoing, but those with Morocco are on hold (as of autumn 2019).85
The agreements to date contain no market opening for services and Mode 4, although that is being negotiated with Tunisia. In parallel, and supplementing negotiations for trade liberalisation, talks are held about easier access to short-term visas, and “procedures for the readmission of irregular migrants”.86 As such the talks hew to the objective of the EU’s Trade Strategy and its Fitness Check Migration: to improve coherence between the policy areas of trade and migration.
To assess the extent to which the agreements with North African states can contribute to long-term development perspectives, the first issue is whether the market access granted to trading partners does in fact correspond to their comparative advantages (in other words, offering real opportunities). Because the opening for agriculture in existing association agreements is very limited, important export products remain excluded. North African countries would welcome a significant market opening for products such as olive oil and tomatoes for example, but the burden of such concessions would fall disproportionately on the EU’s Mediterranean member states, which are already struggling economically. Such a move would therefore demand a quid pro quo in the scope of European solidarity.87 Services too, where Tunisia has an interest in Mode 4, offers potential for expansion.
The agreements with Mediterranean countries and regions were concluded between 1995 (Israel and Tunisia) and 2002 (Lebanon and Algeria).88 As such they originate from an era when sustainability-related aspects such as workers’ rights, environmental standards and civil society participation did not yet play the role in trade talks that they do today. The current negotiations with Tunisia create an opportunity not only to correct this but also, through asymmetry and flexible safeguard and review clauses, to strengthen the role of sustainable development aspects in the agreements.
Economic Partnership Agreements with African ACP States
The EU has concluded Economic Partnership Agreements with African states and regions from the ACP group (the African, Caribbean and Pacific states), which currently comprises seventy-nine states (indicated in light green in the map, p. 27).89 In these asymmetrical agreements the partner countries open up to 80 percent of their markets for EU products while the EU grants them completely tariff- and duty-free access for all goods. The Economic Partnership Agreements were intended to replace the previous unilateral EU trade preferences with a system that conforms to WTO rules, and to serve the topline goals of “sustainable development” and “regional integration of partner countries”. Market access conditions for South Africa differ, even though it belongs to the EPA with the members of the SADC. This special status is due to its higher level of development, and it does not receive completely free access to the EU market.
The agreements say nothing specific on the topic of migration. The question of services is also excluded to date. The European Union had originally proposed concluding deep trade agreements whose scope extended beyond trade in goods. This was rejected by the African states, partly out of concern that their limited negotiating capacities would leave them unable to keep track of the possible repercussions. The EU’s EPAs with African countries therefore remain restricted to trade in goods; services, and thus also Mode 4, are excluded. But the possibility does exist to negotiate additional topics at a later date, at which point the partner countries could raise demands on Mode 4. The EPA region of eastern and southern Africa (implemented by Madagascar, Mauritius, Seychelles and Zimbabwe) has in the interim started talks with the EU about potentially deepening the agreement to include services.90
The EPAs are free trade agreements whose provisions largely satisfy the requirements of sustainable development. At least in theory they can thus contribute in the long term to addressing the causes of migration. Through tariff- and duty-free access to the EU market they offer partner countries the opportunity to market all their export products – including processed agricultural products – to Europe.
The asymmetry of the EPAs permits partner countries to exclude sensitive products from liberalisation. They include monitoring mechanisms and dialogue forums which also bring in civil society actors. A series of precautions were instituted to tackle problematic import trends that could lead to unemployment and to promote food security and industrial development.91 No other North-South agreement contains comparable flexibilities for achieving its development objectives.
However, these agreements also need to prove in practice that they can satisfy the needs of the participating countries. The proposed monitoring mechanisms need to be rapidly established, capable of detecting problems quickly and equipped to initiate any protective measures deemed necessary. Moreover, labour and environmental provisions in these EPAs have remained rudimentary (at the insistence of the African partners). These topics should be put on the table again in subsequent negotiating rounds – including the current talks with the countries of eastern and southern Africa. In order to ensure that the partners can actually make use of the opportunities created by the agreements, despite their scarcity of funds and human resources, they need to be backed by aid for trade measures.
Unilateral Trade Preferences
The EU’s Generalised System of Preferences (GSP) grants privileged market access to all low- and lower-middle-income countries. Under this system their exports in about two-thirds of tariff lines are subject to reduced or zero tariffs.92 For the African middle-income countries the GSP is economically irrelevant because it offers much worse market access than the Economic Partnership Agreements. For this reason most countries in this income group implement an EPA (the few exceptions including Nigeria).93 Imports from the poorest countries – the LDCs – are completely exempt from tariffs and quotas under the Everything But Arms initiative of the GSP (shaded red on the map, p. 27). The EU is to date the only region to offer the poorest countries completely free market access – even though WTO member states agreed at the 2005 WTO Ministerial Conference in Hong Kong it should be granted by all industrialised countries and emerging economies.
The GSP leaves the topic of migration to one side. It makes no mention of services, and thus has no bearing their provision by natural persons; all it does is grant trade preferences for goods. Nor does the European GSP presently contain other migration-related provisions or conditionalities.
The EU does however grant the LDCs unilateral preferences under the WTO’s services waiver, also for Mode 4.94 It enables, for example, contractual services suppliers in twenty-six fields – including accounting, architecture and environmental services – as well as independent professionals in eleven areas to pursue limited activity within the EU. Relevant activities are subject to a string of conditions. For example a firm that wishes to send staff to the EU must have existed for at least one year; the worker in question must have been employed by the firm for at least one year and must be able to demonstrate three years’ professional experience. Independent professionals can operate for a maximum of six months in the EU (25 weeks in Luxembourg). There are also additional restrictions such as needs assessments. The qualification requirements are mostly tied to the respective national equivalents in the EU member states and access is restricted to degree-holders. Altogether this is likely to mean that the actual utility of the scheme for the LDCs will remain extremely limited. The same also applies to this instrument’s relevance to migration.
Trade policy instruments play virtually no direct role in limiting migration flows. Empirical observations reveal that increasing trade stimulates migration in the short and medium term. To that extent it is important to distinguish between flight and migration. In the long term trade can contribute to combating causes of flight whose grounds lie in poverty. To that end trade agreements should be shaped in such a way as to contribute to sustainable development. All existing agreements with African states and regions leave room for improvement in this respect. Talks are currently under way with Tunisia, and all the Economic Partnership Agreements permit the inclusion of new topics. But the final shape of any agreement depends on the partner countries as well as the EU.
Opening the EU market to products that can actually be exported from the partner countries is central if an agreement is to promote sustainable development and help to combat causes of flight in the long term. Because the EU has already completely opened its market to Sub-Saharan Africa, this is relevant only to South Africa and the countries of North Africa.
But increasing trade will probably at the same time encourage migration, above all by the better-educated. Possibilities of legal migration can reduce the incentive for people to migrate illegally in search of a better life. It lies in the interests of all participating countries to manage these processes in ways that ensure that their effects are positive for all involved. Developing countries want better access to the markets of the EU and other industrialised countries for their service providers, above all in association with the prospect that these individuals will return with new skills and make a contribution to development in their country of origin. Industrialised countries, especially those with ageing societies, have tangible labour needs in multiple service sectors. Active use of Mode 4 in negotiations on services is therefore suitable for creating legal possibilities for migration and balancing the interests of countries of origin and destination. To this end a determined linkage of migration and trade policy instruments is recommended. Closer cooperation with the private sector would also be helpful. The point is not only to agree which service sectors should be opened. The private sector can also be involved in longer-term processes of circular migration, for example by participating in local vocational training and offering internships and training in the parent company.
Numerous factors lead people to leave their homeland. Some of them, such as the ability to obtain information about possible destination countries, have steadily improved during recent years with the progress in information and communications technology. Additionally, globalisation processes and prior migration flows mean that many people already live outside their country of origin, which makes it easier for those who follow. Trade policy is only one influencing factor among many, and its possible role in migration management must not be overestimated. Nevertheless, there is every reason to take the topic of migration into account when formulating trade policy, especially with regard to legal possibilities of migration and their utilisation in the mutual interest. This will be the EU’s task in the coming years, since the topic of Mode 4 is only going to become more important.
African, Caribbean and Pacific Group of States
Communauté des États Sahálo-Sahariens/
Common Market for Eastern and Southern Africa
Economic Partnership Agreement (EU)
General Agreement on Trade in Services
Gross domestic product
Generalised System of Preferences
International Labour Organisation
International Organisation for Migration
Least developed country
Organisation for Economic Cooperation and Development
Southern African Development Community
Trade-Related Investment Measures
United Nations Conference on Trade and Development
World Trade Organisation
See Evita Schmieg, “Europäische Handelspolitik: Fördert oder verhindert sie Migrationsbewegungen?” in Fluchtursachen “Made in Europe”: Über europäische Politik und ihren Zusammenhang mit Migration und Flucht, ed. Felix Braunsdorf (Berlin: Friedrich-Ebert-Stiftung, November 2016), 33–38.
International Organization for Migration (IOM), Key Migration Terms, https://www.iom.int/key-migration-terms#
Migration (accessed 8 September 2019).
On the development of international trade theory, see Paul R. Krugman and Maurice Obstfeld, International Economics: Theory and Policy (Reading, MA, 1997).
Jeffrey D. Sachs and Andrew Warner, Economic Reform and the Process of Global Integration, Brookings Papers on Economic Activity 1/1995 (Washington, D.C.: Brookings, 1995), https://
bpea_sachs_warner_aslund_fischer.pdf (accessed 29 January 2019), and Jeffrey A. Frankel and David Romer, Trade and Growth: An Empirical Investigation, NBER Working Paper 5476 (Cambridge, MA: National Bureau of Economic Research [NBER], March 1996).
The term “rent-seeking” was coined by Anne Krueger in her renowned essay, “The Political Economy of the Rent-Seeking Society”, American Economic Review 64, no. 3 (1974): 291–303.
World Bank, Economic Growth in the 1990s: Learning from a Decade of Reform (Washington, D.C.: World Bank Group, PREM Network, 2005), chapter 5: “Trade Liberalization: Why so Much Controversy?” 133–55 (149).
For one such critique, see Francisco J. Marí, “Fischerei-, Agrar-, Wirtschaftspolitik: Wie die EU Hunger und Armut in Afrika schafft”, in Fluchtursachen “Made in Europe”, ed. Braunsdorf (see note 1), 27–32.
See Krugman and Obstfeld, International Economics: Theory and Policy (see note 3).
See, for example, Edo Mahendra, Trade Liberalisation and Migration Hump: NAFTA as a Quasi-Natural Experiment, IMI Working Paper 98 (Oxford: University of Oxford, International Migration Institute [IMI], August 2014).
Horst Feldmann, Eine institutionalistische Revolution? Zur dogmenhistorischen Bedeutung der modernen Institutionenökonomik (Berlin, 1995).
See Douglass C. North, “Economic Performance through Time”, American Economic Review 84, no. 3 (1994): 359–68, and idem., Institutionen, institutioneller Wandel und Wirtschaftsleistung (Tübingen, 1992).
Michael A. Clemens, Does Development Reduce Migration? IZA Discussion Paper 8592 (Bonn: Center for Global Development, Institute for the Study of Labor [IZA], 2014), 12, http://
ftp.iza.org/dp8592.pdf (accessed 15 August 2019).
Rosmaiza Abdul Ghani, Michael P. Cameron, William Chochrane and Mathew Roskruge, A Gravity Model Estimation of the Bi-Directional Relationship between International Trade and Migration, Working Paper in Economics 19/2 (Hamilton, NZ: University of Waikato, 2019).
Mahendra, Trade Liberalisation and Migration Hump (see note 9), 10f.
Nadia Campaniello, “The Causal Effect of Trade on Migration: Evidence from Countries of the Euro-Mediterranean Partnership”, Labour Economics 30 (October 2014): 223–33, https://pdfs.semanticscholar.org/c996/b34013540
647df5ea6f159807ee7536df691.pdf (accessed 28 January 2019).
United Nations Conference on Trade and Development (UNCTAD), Economic Development in Africa: Report 2018: Migration for Structural Transformation (New York and Geneva, 2018).
Philip L. Martin and J. Edward Taylor, “The Anatomy of a Migration Hump”, in Development Strategy, Employment and Migration: Insights from Models, ed. J. Edward Taylor (Paris: OECD, 1996), 43–61.
Mahendra, Trade Liberalisation and Migration Hump (see note 9).
European Commission, Many More to Come? Migration from and within Africa (Luxembourg: Joint Research Centre, 2018), 21, https://ec.europa.eu/jrc/sites/jrcsh/files/africa_
policy_report_2018_final.pdf (accessed 27 September 2019). Unfortunately the sources upon which the study is based are not stated.
Jeffrey A. Frankel, Regional Trading Blocs in the World Economic System (Washington, D.C., 1997).
UNCTAD, Economic Development in Africa (see note 16).
European Commission, Many More to Come? (see note 19), 21.
Ibid. See also Clemens, Does Development Reduce Migration? (see note 12).
Clemens identifies the peak at “an income per capita of roughly PPP$5,000–6,000 (today’s Jordan or Jamaica)”, in Does Development Reduce Migration? (see note 12), 5.
Wilbur Zelinsky (1971) coined the term “mobility transition” to describe this, quoted in ibid. (see note 12), 10.
World Bank, GDP per Capita Growth (Annual %), https://
data.worldbank.org/indicator/NY.GDP.PCAP.KD.ZG?view=chart (accessed 24 June 2019).
Clemens, Does Development Reduce Migration? (see note 12), 9ff.
Martin and Taylor, “The Anatomy of a Migration Hump” (see note 17).
Thu Hien Dao, Frédéric Docquier, Chris Parsons and Giovanni Peri, “Migration and Development: Dissecting the Anatomy of the Mobility Transition”, Journal of Development Economics 132 (2018): 88–101.
In a major study analysing the period 1970 to 2000, Rodolfo Metulini and colleagues find significant indirect effects of migration on trade flows. See Rodolfo Metulini, Paolo Sgrignoli, Stefano Schiavo and Massimo Riccaboni, “The Network of Migrants and International Trade”, Economia Politica 35, no. 3 (2018): 763–87, https://link.springer.com/
article/10.1007/s40888-018-0106-6 (accessed 25 September 2019).
See Barry R. Chiswick and Paul W. Miller, eds., Handbook of the Economics of International Migration, vol. 1B: The Impact and Regional Studies (Oxford and Amsterdam, 2015), http://www.
sciencedirect.com/science/handbooks/22120092/1 (accessed 11 February 2019).
Gabriel J. Felbermayr and Farid Toubal, “Revisiting the Trade-Migration Nexus: Evidence from New OECD Data”, World Development 40, no. 5 (2012): 928–37 (935).
UNCTAD, Migration for Structural Transformation (see note 16), 102.
Marion Panizzon, Trade and Labor Migration: GATS Mode 4 and Migration Agreements, Dialogue on Globalization, Occasional Papers 47 (Geneva: Friedrich-Ebert-Stiftung, January 2010).
IOM, World Bank and World Trade Organization (WTO), Background Paper, Trade and Migration Seminar (Geneva,
4–5 October 2004, https://www.wto.org/english/tratop_e/
serv_e/sem_oct04_e/background_paper_e.pdf (accessed 9 July 2018).
WTO, Council for Trade in Services, Presence of Natural Persons (Mode 4), Background Note by the Secretariat, S/C/W/75 (8 December 1998), http://bit.ly/31DZSu8 (accessed 16 August 2019).
UNCTAD, Migration for Structural Transformation (see note 16), 98.
Mina Mashayekhi, Contribution of Migrants to Development: Trade, Investment and Development Linkages (Geneva: UNCTAD, 29 July 2009), http://unctad.org/en/docs/emditctncd_01_en.
pdf (accessed 18 June 2018).
UNCTAD, Migration for Structural Transformation (see note 16), 135.
Ghani et al., A Gravity Model Estimation (see note 13).
Johanna Jacobsson, “Liberalisation of Service Mobility in the EU’s International Trade Agreements: As External as It Gets”, European Journal of Migration and Law 15, no. 3 (2013): 245–61 (247).
WTO, Annex on Movement of Natural Persons Supplying Services under the Agreement, https://www.wto.org/english/
tratop_e/serv_e/8-anmvnt_e.htm (accessed 4 February 2019).
IOM, World Bank and WTO, Background Paper (see note 35), 2.
European Commission, Fitness Check on EU Legislation on Legal Migration, Commission Staff Working Document, SWD(2019)1056 final (Brussels, 29 March 2019), 151, https://
documents/policies/legal-migration/swd_2019-1055-staff-working-part2.pdf (accessed 24 September 2019).
Alexander Betts and Kalypso Nicolaidis, The Trade-Migration Linkage: GATS Mode IV (Oxford: University of Oxford, 19 February 2009), http://citeseerx.ist.psu.edu/viewdoc/
download?doi=10.1.1.508.3249&rep=rep1&type=pdf (accessed 9 July 2018).
“Economic Partnership Agreement between the CARIFORUM States, of the one part, and the European Community and its Member States, of the other part”, Official Journal of the European Union, L 289/I/3 (30 October 2008), https://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=
OJ:L:2008:289:0003:1955:EN:PDF (accessed 20 June 2019).
On the following, see WTO, Presence of Natural Persons (see note 36), Para. 40.
Panizzon, Trade and Labor Migration (see note 34).
WTO, Presence of Natural Persons (see note 36).
WTO, Council for Trade in Services, Communication from India: Proposed Liberalisation of Movement of Professionals under GATS, S/CSS/W/12 (Geneva, 24 November 2000), http://bit.ly/
2P975Rw (accessed 4 February 2019). The document provides a very good overview of the restrictions to Mode 4 concessions.
See Panizzon, Trade and Labor Migration (see note 34), 34.
Ibid.; see also the next section, pp. 17f.
WTO, Preferential Treatment to Services and Service Suppliers of Least-Developed Countries, WT/L/847 (Geneva, 19 December 2011).
United Nations Committee for Development Policy, What Benefit? Preferential Treatment to Services and Services Suppliers of LDCs, https://www.un.org/ldcportal/what-benefit-preferential-treatment-to-services-and-services-suppliers-of-ldcs/ (accessed 23 November 2018).
See Panizzon, Trade and Labor Migration (see note 34), 9.
Ibid., 15, 24.
Aurelia Segatti, “Bilateral Labour Migration Agreements: Trends and Examples of Good Practice”, presentation at ILO Labour Migration Information-sharing Session, Kadoma Hotel, Zimbabwe, 16–17 July 2015.
See Steffen Angenendt, Entwicklungspolitische Perspektiven temporärer und zirkulärer Migration, SWP-Studie 13/2014 (Berlin: Stiftung Wissenschaft und Politik, August 2014).
ILO, Model Agreement on Temporary and Permanent Migration for Employment, including Migration of Refugees and Displaced Persons, Annex to R 086 Migration for Employment Recommendation, (Revised) 1949 (no. 86) (Geneva, 1949), https:/www.
P12100_INSTRUMENT_ID:312424 (accessed 28 January 2019).
ILO and Global Knowledge Partnership on Migration and Development (KNOMAD), Bilateral Agreements and Memoranda of Understanding on Migration of Low-Skilled Workers: A Review (Geneva, 2015), https://www.ilo.org/wcmsp5/groups/
genericdocument/wcms_357389.pdf (accessed 4 February 2019).
For a summary of the New Trade Theory see Evita Schmieg, Regionale Integration: die Europäische Union und die Karibik, Dissertation (Leipzig, 2006).
Original: “Notwendigkeit eigenständiger Entwicklungen in der Peripherie”; Robert Kappel, “Paul R. Krugman (1953–) – Die neue Außenhandelstheorie und die Ungleichheit der Nationen”, Entwicklung und Zusammenarbeit 40, no. 1 (1999): 179–82.
See World Bank, Economic Growth in the 1990s (see note 6), 133–55.
Jean-Yves Huwart and Loïc Verdier, Economic Globalisation: Origins and Consequences, OECD Insights (Paris: OECD Publishing, 2013), https://read.oecd-ilibrary.org/economics/
economic-globalisation_9789264111905-en#page3 (accessed 19 June 2019).
Martin and Taylor, “The Anatomy of a Migration Hump” (see note 17).
Neil McCulloch, L. Alan Winters and Xavier Cirera, Trade Liberalization and Poverty – A Handbook (London: Centre for Economic Policy Research [CEPR], 2001), https://vi.unctad.org/
trade_liberalization_poverty.pdf (accessed 18 September 2019).
See Evita Schmieg, Trade and Investment Agreements for Sustainable Development? Lessons from the EU’s Economic Partnership Agreement with the Caribbean, SWP Research Paper 6/2015 (Berlin: Stiftung Wissenschaft und Politik, July 2015).
European Commission, Trade for All: Towards a More Responsible Trade and Investment Policy (Luxembourg, 2014), 12, http://trade.ec.europa.eu/doclib/docs/2015/october/tradoc_
153846.pdf (accessed 26 June 2019).
European Commission, Fitness Check on EU Legislation on Legal Migration (see note 44), 188.
See Schmieg, Trade and Investment Agreements for Sustainable Development? (see note 68).
European Commission, Fitness Check on EU Legislation on Legal Migration (see note 44), 188.
European Commission, Fitness Check on EU Legislation on Legal Migration (see note 44).
Panizzon, Trade and Labor Migration (see note 34), 26.
UNCTAD, Migration for Structural Transformation (see note 16), 154.
See the initiative’s website: https://afrika-kommt.de/ (accessed 23 January 2019).
European Commission, Fitness Check on EU Legislation on Legal Migration (see note 44), 150.
AGOA Country Eligibility, AGOA.info/tralac, https://agoa.
info/about-agoa/country-eligibility.html (accessed 2 July 2019).
Susanne Maria Kraus, “Ostafrika kämpft weiter gegen Second-Hand-Kleidung”, Deutsche Welle, 26 March 2018, https://www.dw.com/de/ostafrika-k%C3%A4mpft-weiter-gegen-second-hand-kleidung/a-42744807 (accessed 23 September 2019).
Eurostat, Africa-EU – Key Statistical Indicators (November 2018), https://ec.europa.eu/eurostat/statistics-explained/
International_trade (accessed 12 February 2019).
European Commission, Many More to Come? (see note 19), 15f.
European Commission, Euro-Mediterranean Partnership (last update: 7 May 2019), http://ec.europa.eu/trade/policy/
countries-and-regions/regions/euro-mediterranean-partnership/ (accessed 12 February 2019).
European Commission, Overview of FTA and Other Trade Negotiations (last update: July 2019), http://trade.ec.europa.eu/
doclib/docs/2006/december/tradoc_118238.pdf (accessed 16 August 2019).
European Commission, “The EU and Tunisia Start Negotiations on Visa Facilitation and Readmission”, press release (Brussels, 12 October 2016), http://europa.eu/rapid/press-release_IP-16-3394_en.htm (accessed 20 June 2019).
See also Schmieg, “Europäische Handelspolitik” (see note 1), 38.
European Commission, Euro-Mediterranean Partnership (see note 84).
European Commission, Economic Partnerships (last update: 21 February 2019, http://ec.europa.eu/trade/policy/countries-and-regions/development/economic-partnerships/ (accessed 20 June 2019).
European Commission, Overview of Economic Partnership Agreements (last update: July 2019), http://trade.ec.europa.eu/
doclib/docs/2009/september/tradoc_144912.pdf (accessed 25 September 2019).
Bettina Rudloff and Evita Schmieg, “European Chicken Drumsticks for West Africa – A Threat to Local Markets?” Rural21 51, no. 1 (2017): 15–17, https://www.rural21.com/
fileadmin/downloads/2017/en-01/Rural21_1_2017_v11.pdf (accessed 24 June 2019).
European Commission, Generalised Scheme of Preferences (GSP) (last update: 24 May 2019), http://ec.europa.eu/trade/
policy/countries-and-regions/development/generalised-scheme-of-preferences/index_en.htm (accessed 28 June 2019).
Explaining the complicated history of the EPA negotiations is beyond the scope of this study, as is the question of why particular countries participate while others do not. See, for example, Evita Schmieg, “Außenhandel für nachhaltige Entwicklung? Freihandelsabkommen zwischen der EU und dem globalen Süden”, Aus Politik und Zeitgeschichte 68, no. 4–5 (2018): 40–46.
See WTO, Council for Trade in Services, Notification from the European Union, S/C/N840 (18 November 2015), http://trade.
ec.europa.eu/doclib/docs/2015/november/tradoc_153997.pdf (accessed 12 February 2019).
All rights reserved.
© Stiftung Wissenschaft und Politik, 2019
SWP Research Papers are peer reviewed by senior researchers and the executive board of the Institute. They are also subject to fact-checking and copy-editing. For further information on our quality control procedures, please visit the SWP website: https:// www.swp-berlin.org/en/ about-swp/quality-management-for-swp-publications/.
SWP Research Papers reflect the views of the author(s).
Stiftung Wissenschaft und Politik
German Institute for International and Security Affairs
Phone +49 30 880 07-0
Fax +49 30 880 07-200
Translation by Meredith Dale
(English version of SWP‑Studie 22/2019)