WPS5049 Policy Research Working Paper 5049 The Euro-Mediterranean Partnership Trade in Services as an Alternative to Migration? Bernard Hoekman Çalar Özden The World Bank Poverty Reduction and Economic Management Network International Trade Department September 2009 Policy Research Working Paper 5049 Abstract This paper discusses options to facilitate movement of potential for such "embodied" trade in services will workers between high-income and developing countries not be anywhere near what would be associated with within the framework of trade agreements, focusing substantial liberalization of migration regimes, furthering on the European Union's partnership agreements the services trade dimension in the European Union's with neighboring countries. Existing frameworks for trade agreements offers significant potential Pareto gains. cooperation offer the possibility of expanding temporary For the partner countries these gains from temporary rather than longer-term or permanent movement of movement of service providers are both direct--through workers since extant trade agreements provide scope greater employment in/revenue from providing services for negotiating specific market access commitments for in the European Union--and indirect--by helping to services, including those delivered through the cross- increase and sustain higher growth at home. border movement of natural persons. Even though the This paper--a product of International Trade Department, Poverty Reduction and Economic Management Network--is part of a larger effort in the department to assess the development implications of regional trade and integration agreements. Policy Research Working Papers are also posted on the Web at http://econ.worldbank.org. The authors may be contacted at Bhoekman@worldbank.org and Cozden@worldbank.org. The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Produced by the Research Support Team The Euro-Mediterranean Partnership: Trade in Services as an Alternative to Migration? Bernard Hoekman and Çalar Özden* Keywords: Mode 4, trade in services, trade agreements, migration, EU, Mediterranean. JEL codes: F15, F16, F22. * Hoekman: World Bank and CEPR; Özden: World Bank. E-mails: bhoekman@worldbank.org and cozden@worldbank.org. Mailing address: Development Research Group, The World Bank, MC3-303, 1818 H Street, NW, Washington, DC 20433. We are grateful to Ricardo Faini, Ahmed Galal, Bachir Hamdouch, Christophe Kiener, Giorgio Navaretti, Maurice Schiff and to participants at the 2006 annual meetings of the Economic Research Forum for the Arab countries, Iran and Turkey (ERF), Kuwait, December 16-18, the 2005 Euro-Mediterranean Economic Transition conference, Brussels and the CEPR Conference on Migration, Rome, June 2006 for helpful comments on earlier drafts. The views expressed in this paper are personal and should not be attributed to the World Bank. The Euro-Mediterranean Partnership: Trade in Services as an Alternative to Migration? 1. Introduction The expansion of the European Union (EU) since 2004 has redefined the European "neighborhood" to span a diverse group of countries in Eastern Europe, the Middle East, the Mediterranean and Central Asia. One consequence of the EU expansion was the formulation of a new approach by the EU towards these countries ­ the European Neighborhood Policy (ENP). For Mediterranean countries, the ENP complements the treaty instruments used since 1995 under the Barcelona Process and the associated Euro- Mediterranean Partnership (EMP) agreements. The basic objectives of the EMP agreements are to achieve reciprocal free trade between the EU and Mediterranean countries in manufactured goods, improve access for agricultural products; establish conditions for gradual liberalization of trade in services and capital and encourage the economic integration of partner countries with the EU. These objectives are pursued through binding bilateral trade agreements and various forms of development cooperation. A major innovation of the ENP relative to the status quo ante is that it offers the partner countries "a stake" in the EU internal market (Dodini and Fantini, 2006, Hoekman, 2005). That is, the partner countries are offered an opportunity to integrate into specific elements of existing and evolving EU structures, law and systems on an à la carte basis. From an economic perspective these agreements also need to address labor market needs on both sides of the Mediterranean in order to maximize the potential welfare gains for both the EU and its neighbors. Given current demographic trends, the EU will face a severe labor shortage over the next decade while Mediterranean economies will continue to experience significant growth in their labor force. These trends will create pressures for labor flows into the EU countries, even if high domestic employment growth rates are sustained in Mediterranean countries. These diverging labor market dynamics provide an opportunity for both the EU and partners to complement their economic and institutional integration efforts with a strategy 2 to enhance South-North migration. However, while Europe's need for foreign workers will be growing over the coming decades, political winds across the continent are blowing strongly against more migration, and the financial crisis/recession of 2007-09 has further reduced the prospects for liberalization of movement of people originating outside the EU. European policymakers therefore face a daunting challenge of reconciling labor market demands with political realities at home. This paper discusses an alternative instrument that would facilitate movement of workers between the EU and Mediterranean countries within the EMP/ENP framework while addressing political concerns in the EU. Existing frameworks for cooperation offer the possibility of expanding temporary rather than longer-term or permanent movement of workers since EMP agreements provide scope for negotiating specific market access commitments for services, including those delivered through the cross-border movement of natural persons. Even though the potential for such "embodied" trade in services will not be nearly large enough to address the demographic challenges facing either side, furthering the services trade dimension in EMP context offers significant potential Pareto gains. These gains are both direct ­ through greater employment in/revenue from providing services in the EU ­ and indirect by helping to increase and sustain higher growth in the partner countries. 2. Demographic changes shape labor market reality Relative wage differences across countries are the primary determinants of global migration patterns (Grogger and Hanson, 2008). These wage gaps can be very high and vary over time and by skill level. But overall, the largest relative wage gaps between the North and the South are in the unskilled segment of the labor market, reflecting varying supply and demand conditions as well as productivity differences. In the EU, the demographic outlook of much of Western and Eastern Europe suggests that the Union will face severe shortages in labor markets over the next decades. The UN forecasts that the population of Europe1 will decline to 653 million (the medium projection) in 2050 from the current level of 728 million (UN, 2009). The size of the 15- 1 Europe includes EU25 as well as all Eastern European, (including Russia), Balkan and Nordic countries that are not EU members. Turkey is included in Western Asia. 3 59 age group will decline by 0.7% annually during this period, while the 60+ age group will increase by 0.9% annually. The dependency ratio will increase from 46.5 to 74.2 in 2050, when the elderly will form 65% of the dependents, up from 50% in 2005. The demographic trends in the newly-acceded countries are no different. Meanwhile, developing and least developed countries' share in world population will rapidly increase with the fastest population growth countries being concentrated in Africa and the Middle East. The total population of Northern Africa and Western Asia (which includes Armenia, Azerbaijan, Georgia, and Cyprus but excludes Iran) will increase from 400 million in 2005 to 700 million in 2050 under the medium projection. The size of the 15-64 age group will increase from 250 million to 450 million while the 60+ age group will grow from 26 to 128 million. Figure 1 Population Distribution by Age Group 40% 30% 20% 10% 0% 0-20 20-40 40-60 60-80 80+ EU in 2005 EU in 2020 Maghreb in 2005 Maghreb in 2020 Source: UN (2009) Figure 1 presents the current (2005) and projected (2020) population distribution in the EU-15 and the three North African countries with significant migration flows to the EU ­ Algeria, Morocco and Tunisia. The first two bars for each age group represent the EU, the second set of bars the average for the three Maghreb countries. The dark bars are for 2005, while the lighter bars are 2020 figures projected by the UN. The current population in the 0-20 age group is significantly larger in the three Maghreb countries ­ 4 40% of the current population ­ whereas they are only 22% of the population of the EU- 15. Over the next 15 years, the UN is projecting a sharp decline in the fertility rates in the Maghreb countries. This will result in a decline in the share of the 0-20 age group to 34% of the population. In comparison, the sharpest decline in the EU will occur in the 20-40 age group, which will decline from 28% to 24% of the population. The sharpest increase in the three Maghreb countries will take place in the 40-60 group ­ from 18% to 24%. The 20-40 age group ­ the prime working age cohort, and the one in greatest need of employment ­ will be the largest one in the Maghreb countries in 2020. This is the group that can take advantage of migration policies that are the subject of this paper. Figure2: Total Job Opennings in the EU due to Replacement and Expansion Demand 10 Expansion 9 Replacement 8 7 6 millions 5 4 3 2 1 0 Grem any UK Italy France Spain Poland Holland Denm ark Portugal Czeck Greece Belgium Hugary Sw eden Republic Source: Ozden, Sewadeh and Wahba (2009) Occupations with the greatest demand for workers in the EU are reported in Figure 2. This shows the number of projected job openings in the EU generated by both replacement demand (arising from exit of workers because of retirement and other reasons) and expansion demand. Between 2005 and 2016, 61 million jobs are expected to become available in the EU25. The largest number of job openings will arise in the EU's major economies: UK, Germany, Italy and France. But smaller economies and EU members with lower per capita incomes will also generate an increasing number of jobs 5 over the next few years. Hungary, Poland and the Czech Republic will together generate more than 7 million job openings between 2006 and 2015. Figure 3 plots job openings for the five EU members with the highest demand for specific occupations. Most of the demand is concentrated in technical/professional occupations, services and elementary (unskilled) activities. Denmark, for example, is expected to have over 1.7 million job openings for technicians, second only to Italy. Similarly, Poland will have over 1 million job openings for professionals and another 1 million openings in craft related trades by 2015. Figure 3 Total Job Openings by Country 2006-15 2500 2000 T ho u san ds 1500 1000 500 0 Italy Italy Italy Italy Italy Italy Italy P ortugal Franc e P oland P oland Franc e S pain Franc e S pain Franc e A us tria Greec e S weden P oland Franc e Franc e P oland Franc e S pain Denm ark Germ any Netherlands Denm ark Netherlands Germ any Germ any Germ any Germ any Cz ec k Germ any UK UK UK UK UK UK UK UK UK Legislators, Professionals Technicians Clerks Sevices Skilled Craft and Plant machine Elementary senior officials and associate workers agriculture related operators occupations and managers professionals and fishery trades Source: Ozden, Sewadeh and Wahba (2009) Compelling benefits for both sides The current demographic trends in the EU and future labor shortages will make migration an economic necessity for EU. Moreover, the demographic transformation will place significant pressures on social programs in member countries. Admitting more migrant workers would help alleviate the pressure on these services at least in the short term. In addition, the 60+ age group, which will grow to 28% of the population in 2020, will be 6 the main beneficiaries of migration as they are the main consumers of the services provided such as healthcare and household services. Significant migration has already been occurring between the Mediterranean countries and the EU, driven by the forces discussed previously. Figure 4 presents data on the number of migrants in OECD members from the countries in the region. The largest source countries are Turkey (close to 2 million as of 2000), Morocco (slightly over 1 million), Algeria and Iran (around 500,000), and Egypt, Iraq, Lebanon and Tunisia (around 250,000).2 Note that the data are plotted on a logarithmic scale, as otherwise the large source countries would visually dwarf the smaller countries. All countries have experienced an increase in migration during 1990-2000, the overall number of migrants increasing by some 40%. However, this includes the children of the migrants who were born in the destination countries and who are considered migrants according to the laws in most of the EU countries.3 Smaller countries have experienced more rapid increases in migration. The impact of the post-2000 conflicts in the region is not reflected in the data. If migration is measured as a share of the native labor force, Lebanese migration stands out, at 15% of the native labor force. This reflects decades of political instability and a long history of social and economic relations with Europe. Among other high outflow countries, Moroccan migration is partly explained by the relatively low cost of movement to France and Spain, reflecting historical links and geographic proximity. Other high emigration countries include Turkey, Tunisia and Algeria. As mentioned, these data underestimate total migration flows, as they do not include the significant migration to the Persian Gulf countries from other Arab countries, such as Egypt and Jordan. 2 The data used in what follows are from Docquier and Marfouk (2006). Their dataset is based on the national censuses of the receiving countries, and includes bilateral migration stocks in OECD countries from over 170 source countries for the years 1990 and 2000, by three education levels. Data is also compiled to allow a comparison of migrants' educational attainment relative to the native populations. Although a major advance, the dataset has several shortcomings. First, it is limited to OECD members as destination countries. Thus it ignores several major destination labor markets such as the Persian Gulf, Singapore, and South Africa. Second, the dataset only includes employed migrants. Students are not captured, even though students, especially in post-graduate programs, are a major source of brain drain. 3 This is not the case in the US or Canada where the children of migrants receive automatic citizenship if they are born in the destination country. 7 Figure 4 M id d le E a s t M ig r a t io n t o O E C D b y d e s t io n a t io n , 2 0 0 0 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% O m an D jibouti Is rael T urk ey Egy pt Iran Iraq M oroc c o T unis ia U AE J ordan Q atar Bahrain Lebanon Liby a Sy ria Yem en Algeria S. Arabia Kuw ait US A / C AN / AUS FR A UK G e r m a n y / N e t h e r la n d s S p a in Figure 5 M id d le E a s t M ig ra tio n to O E C D C o u n trie s 1 0 ,0 0 0 ,0 0 0 1 ,0 0 0 ,0 0 0 1 0 0 ,0 0 0 1 0 ,0 0 0 1 ,0 0 0 100 Is ra e l E gypt M o ro c c o D jib o u ti T u rk e y UAE K u w a it A lg e r ia Ir a n Ir a q J o rd a n L ib y a S . A r a b ia S y r ia Y em en Q a ta r B a h r a in Lebanon O m an T u n is ia 1990 2000 Source: Docquier and Marfouk (2006) There is substantial variation in the destination of migrants (Figure 5), partially reflecting geography and partially historical linkages. For example, the majority of migrants from Algeria, Morocco and Tunisia go to France while Turkish migrants have gone to Germany, the Netherlands and Belgium. Migrants from ex-British colonies such as Egypt, Iraq and Jordan often migrate to the UK as well as the other English speaking "new world" countries such as the US, Canada and Australia. There is also substantial migration to Spain from Morocco and to the UK from the oil-rich countries Gulf 8 countries. Finally, there are a significant number of migrants from Iraq, Iran and Syria in Turkey (this is the "missing" portion of each bar in Figure 5). This migration reflects geographic proximity and is often a transit stop en route to Europe.4 The geographic destination choices for the educated migrants from the region are similar to those of total migration overall. Thus, the majority of educated Turkish migrants are in Germany, while North African migrants are concentrated in France. However, the proportion of migrants with tertiary education going to the US, Canada and Australia is much higher for every source country. Only 5% of Turkish and Moroccan migrants go to these destinations, as compared to over 25% of all university educated Turks and Moroccans who migrate. Reasons for this may include perceptions that educated migrants perform better in these labor markets (have more opportunities) and may find it easier to be legally and socially accepted. Notwithstanding the benefits of migration for developing countries, there are growing concerns about "brain drain". Even though migrants from most developing countries are less educated than the native population in destination OECD countries, they are more educated than their fellow citizens in their source countries (Docquier and Marfouk 2006). There are many reasons for this positive selection effect. Among these are the migration policies of the receiving countries that are biased towards educated migrants as well as the financial and assimilation constraints faced by unskilled migrants. The share of tertiary educated migrants in the total number of migrants increased between 1990 and 2000. As of 2000, between 10-15% of migrants from Turkey, Tunisia, Morocco and Algeria to the EU have tertiary education. On the other hand, more than 50% of migrants to OECD countries from other countries in the region, including Egypt, Lebanon and Jordan, have tertiary education. One reason for these differences is that unskilled migrants from the latter countries have gone to the oil-rich Persian Gulf countries. As a result, the figures to the EU capture high levels of educated migrants from Persian Gulf countries, as well as Egypt, Jordan, Lebanon. Among the countries in the region, Turkish migrants are most similar to the native population in terms of their 4 In considering these data, it is important to recognize that overall migration flows from MENA countries are much higher, reflecting the significant numbers of migrants from countries such as Egypt in the GCC countries. The EU accounts for about 50% of total cumulative migrant flows out of the region; other MENA countries account for another 40%. 9 education profile ­ around 10% of natives and migrants have tertiary education. Similarly, the bias is relatively small for Morocco and Tunisia due to the significant number of unskilled migrants. The share of the educated population that migrates is of central importance from an economic growth perspective since human capital is one of its main determinants. Large migrant-sending countries ­ with the exception of Turkey ­ and countries experiencing conflict are the main victims of brain drain in the Mediterranean region in terms of absolute numbers. Poorer countries have low shares of college educated workers in the labor force. As a result, even migration of a small number of the skilled might imply a significant loss of the existing stock of human capital. On the other hand, for the wealthier countries, the total number of migrants relative to the population is much smaller and a smaller portion of the highly educated chooses to migrate ­ hence the lower levels for these countries. Mid dle Ea s t Sk ille d M ig ra tio n to F ra n c e , b y A ge 2 0 0 0 100% 80% 60% 40% 20% 0% Oman UAE Turkey Mali Iraq Jordan Djibouti Israel Egypt Qatar Iran Malta Morocco Tunisia Bahrain Yemen Lebanon Libya Syria Algeria S. Arabia Kuwait les s than 11 y rs . old betw een 12 and 17 y rs . old betw een 18 and 21 y rs . old abov e 22 y rs . Old Figure 6 A neglected factor in the literature on the brain drain is where the migrants obtain their education. In the case of the EU, a large portion of those people considered migrants actually received or completed their education in the destination countries after they either migrated as children with their families or migrated as young adults to complete their education and stayed to enter the labor force. In the case of France, only around 10 30% of Moroccan, Tunisian and Algerian educated migrants arrived after the age of 22 (Figure 6). Around 60% arrived in France before age 18, indicating they came as children and grew up there. Similarly, 40% of Turkish and Lebanese migrants ­ the other two large groups ­ also arrived as children. Thus, a large number of educated migrants in France have not consumed fiscal resources of their native countries for their education. 3. Reconciling economic needs with the political reality While the economic reality in the EU and its Mediterranean neighbors is creating medium to long-term pressures for increased migration, the political winds across Europe are blowing strongly in favor of stemming the flow of immigrants into the continent. The current stock of Middle Eastern migrants contributed to rising resistance to migration according to public opinion surveys and political debates. The increase in negative views is the outcome of a mix of security, cultural and more recently economic concerns Evidence provided by Mayda (2005) suggests that cultural assimilation and security/crime concerns are relatively more important in shaping attitudes than labor market variables. The opinion surveys5 reveal seemingly conflicting attitudes toward migration and the economic interests of respondents. For example, older people tend to be more anti-immigration even though they are the ones likely to benefit the most from migrants whether via increased supply of domestic and health service providers or lower dependency ratios. The growing hostility towards immigrants has made the political environment in many EU countries much less welcoming of immigrants. Ironically, the increased restrictions on immigration have not done much to stem the flow of immigrants. Instead it has lead to an increase in illegal entry, with a consequent change in the composition of migration towards less educated people (Boeri and Bruckner 2005). 6 Resistance to further migration amidst the rising demand for workers in the EU suggests that policy makers should consider alternative mechanisms that can partially substitute for permanent movement of workers across borders while at the same time encourage more two-way flows. In this context, a temporary movement scheme might be seen as less "threatening." In a recent EU consultation on development and in response to 5 Among these are the World Values Survey, the International Social Survey Program and others. 6 See also Kessler and Freeman (2005) and Mayda (2004) for analyses of the determinants of European attitudes to migration. 11 a question about the objective of EU migration policy, 56% of respondents said migration should benefit source countries, 38% worried about negative effects of brain drain, and 27% thought that temporary movement possibilities needed to be enhanced.7 While the views expressed by respondents to the survey may not reflect the beliefs of the EU public at large, they do suggest it may be possible to find support for effective temporary labor movement schemes. A temporary migration program has many advantages. First, temporary labor inflows will have little impact on the demographic profile of the EU, thus easing political resistance while partially addressing labor market needs. Second, it would be consistent with the temporary nature of the excess supply of labor in Mediterranean countries. As noted previously, the demographic profiles in these countries are such that, in the next 20 years or so they will experience a reduction in the supply of workers. Third, temporary migration flows are likely to reduce brain drain losses for source countries, while still generating significant remittance flows.8 Fourth, returnees help raise productivity in home countries since they bring back newly acquired human and physical capital. Fifth the incentives for undocumented migration by facilitating legal and recurrent cross- border movement of workers are reduced. The aggregate potential gains associated with liberalization of temporary movement can be significant. Walmsley and Winters (2005) analyze the potential global welfare gains associated with a relaxation of barriers to temporary movement of workers and conclude that aggregate gains to the world can easily exceed the complete elimination of remaining barriers to trade in goods. Tani (2005) analyzes the potential growth effects of temporary movement of people, focusing specifically on business travelers, and finds that such movement is positively correlated with economic growth. 9 7 EU Public Consultation Report, 2005: http://ec.europa.eu/comm/development/body/theme/consultation/doc/Stat_IPM_EN.pdf. 8 In principle remittances associated with temporary movement may be higher than that associated with longer-term movement, as links with source country household members are likely to be much stronger. More generally, however, Philippe Fargues has argued that as migration comes to involve younger workers who are not encumbered by families and have less need to take care of extended families that remain in source countries, the relative importance of remittances will decline. Instead, income earned will be devoted more to consumption and investment in the human capital of the migrants themselves. 9 Note that migration ­ temporary or permanent ­ is not a solution to the fiscal pressures created by an aging population. This will require changes in the level of social benefits and the way they are financed. An example is shifting from defined benefit pension programs to defined contribution based schemes. 12 The obvious challenge is to ensure that temporary movement is indeed temporary. A common view of many EU voters and policymakers ­ based on past experience ­ is that nothing is more permanent than a temporary "guest" worker. This widely held view is a major factor impeding any serious effort to negotiate and put in place effective temporary movement schemes that would substantially enhance access to the EU market. These concerns are well-founded, as incentives for both employers and temporary entrants to ensure temporariness can be rather weak. Amin and Mattoo (2005) analyze the incentives for host and source countries to prefer temporary to long-term or permanent migration. They show that unilateral migration policies are globally inefficient in that they lead to excessive permanent migration relative to temporary movement, and too little overall movement of workers. Their analysis suggests that the challenge is to design self-enforcing cooperation agreements that make temporary entry feasible (desirable) ex post. This could be done by increasing the cost to host firms to retain workers by taxing them or by requiring them to deposit lump-sum payments into an escrow account that is refunded on return of workers, and by increasing worker's incentives to return by refunding social security taxes (see e.g., Martin, 2006; Schiff, 2004; Holzmann and Münz, 2004). Chaudhuri, Mattoo and Self (2004) and Winters et al. (2003) identify a number of desirable components of trade agreements covering Mode 4. A central element is that the focus should be on "contract-based" movement of service suppliers, not employment- based movement (and associated visas/work permits). Contract-based movement focuses not on people (workers or individuals) but on tasks (services) that have been contracted by a firm in the EU. Examples would be medical or home care services. The EU has demonstrated a willingness to consider reciprocal negotiations on such movement in the WTO context (Commission of the European Communities, 2003). As discussed below, the EU has begun to move down this track in recent agreements with Chile and the 2008 Economic Partnership Agreement with the Caribbean. Suppose there is a specific demand for certain services by a firm or government entity in the EU, and that these services are covered by the agreement so that they may be provided by either EU or partner-based firms. A firm that obtains (wins) a contract would 13 be free to source the required inputs from EU or providers at home. If they choose the latter, the service providers (workers) would be granted a visa/permit for a limited period to be negotiated. These workers would be employees of the Mediterranean firm, although they might also be employed by an affiliate of an EU firm that has established in the partner country. Indeed, the affiliate may be connected to the EU firm that is buying or supplying the services (in this case the sub-contracting of service inputs would be in- house, not arms-length).10 The advantage of contract-based movement as opposed to employment-based movement is that it helps to make temporariness more credible as contracts are time-bound. Although the contracts may be recurrent, once the worker reaches the maximum length of stay, he/she must be rotated. Contract-based movement also allows internalization of some objectives that otherwise require regulation. For example, a common constraint on international trade in services is non-recognition of licenses and professional qualifications. Rather than require re-certification or the negotiation of (mutual) recognition agreements ­ a cumbersome and time consuming process because the focus of attention is on national conditions, laws and institutions ­ a contract-based approach leaves it to the buying firm or entity to establish whether foreign suppliers satisfy prevailing quality and related performance standards. Contract-based approaches can also more easily be designed to generate incentives to encourage workers to return. As the contracts are between firms, the government can hold them accountable for performance, further helping to internalize incentives. An explicit focus on expanding the scope for cross-border movement of service providers can provide a framework for cooperation on policies that are needed to maximize the joint benefits of such trade. A major negative dimension of current restrictive migration regimes is that they encourage illegal movement, with its associated risks, costs, uncertainty, and potential exploitation of migrants by intermediaries and employers. But legal movement is also affected by uncertainty and problems of 10 There is some experience with such types of schemes. Winters et al. (2003) note that in the early 1990s Germany used sub-contracting agreements to attract foreign labor on a temporary basis. Under this arrangement a Central European firm would contract to provide specific services for a German firm on a sub-contracting basis. These were subject to government approval, and required satisfaction of wage parity conditions and acceptable working conditions. The German firm was liable for noncompliance of the foreign sub-contractors (through fines), and noncomplying foreign firms could be blacklisted. This scheme was subject to a countrywide quota of 100,000 foreign workers, as well as industry-specific caps. 14 asymmetric information, ex post contract enforcement, excess costs of repatriation of savings, and so forth. Migrants may confront discrimination and be denied basic rights. These factors can give rise to selection bias and serious mismatches between the skills required for jobs and the qualifications/capacity of the migrants (Amin and Mattoo, 2006). Mode 4 agreements can be used as a focal point for reforms in complementary policies in both host and home countries to address many of these well known problems. For example, if mode 4 type movement is contract-based, problems of skill mismatch are less likely to occur as the firms involved have incentives not to "under employ" workers ­ more educated service providers are less likely to be forced to drive taxis, for example, as is currently too often the case in some OECD countries. On a sector-by-sector basis, they can allow a focus on improving educational policies in source countries so as to make graduates of programs fit better the needs of the market ­ in the EU and at home ­ while increasing the EU share of financing for this educational investment, as part of it benefits consumers in the EU. They can also help move Mediterranean countries in the direction of emulating policies of the type that have been pursued by successful temporary emigration countries such as the Philippines (Wahba, 2006). This includes contracts that bind workers and employers/recruiters and that specify the rights of workers. These are enforceable under Philippine law. The Philippine government also provides a variety of complementary inputs and products, including briefing workers on the social and work conditions in the host country, life insurance and pension plans, medical coverage and access to credit. As stressed by Wahba (2006), Mediterranean countries have not pursued such policies to support outward temporary migration. A contract-based services approach can constitute a coordination mechanism to put in place such supporting policies, especially if complemented by development assistance (technical and financial) from the EU. 4. Euro-Med services trade liberalization: A partial solution Despite the rather extensive development cooperation programs and the implementation of bilateral free trade agreements, economic growth in many Mediterranean countries has not generated enough jobs to absorb the younger generations entering the labor force. 15 This is mostly due to the pace and depth of economic reforms. It is also partially an outcome of important holes in the edifice of the EU's trade agreements. For example, it makes no sense from an economic perspective to limit international cooperation (trade agreements) to products and capital flows (investment). A concerted effort to increase the potential for regional services trade could both generate greater employment for workers in service sectors both in partner countries and in the EU, and improve economic policies in the Mediterranean countries.` Trade in many services requires provider and consumer to be located in the same place, implying that either provider or the consumer needs to move to the other's location. Thus, by definition, liberalization of trade in services will often lead to movement of people. They may be managers of affiliates of multinational firms that establish in a foreign market to supply services locally, employees of foreign firms or independent contractors who move temporarily across borders to provide a service to clients, or consumers who travel to the location of a service provider. The latter "mode" of trade comprises activities such as tourism, education, and medical services. Current status of Euro-Med partnership agreements As part of the Euro-Med partnership, the EU has concluded seven Association Agreements between 1998 and 2005 with Algeria, Egypt Israel, Jordan, Lebanon, Morocco, and Tunisia. As mentioned earlier, the agreements provide a framework for commitments on trade in goods and services between the EU and these countries. They also aim to provide a basis for gradual liberalization of trade in the Mediterranean area and set out general conditions for economic, social and cultural cooperation between the EU and those countries. With relation to labor mobility, most of these agreements only reaffirm each Party's obligations under the WTO's General Agreement on Trade in Services (GATS). Some of the agreements provide for a dialogue between parties to "achieve progress" with relation to movement of workers. The EU Agreements with Morocco and Tunisia also include commitments on non-discrimination with respect to working conditions, remuneration dismissal and social security for nationals of these countries that are legally employed in the EU. Both agreements also call for "regular dialogue" on the living and 16 working conditions of migrant communities; migration in general (not only with connection to services) and illegal immigration. The EU Agreements with Algeria and Jordan provide for limited movement of key personnel. Jordanian companies established in the EU may employ Jordanian nationals that are "key personnel", defined as "persons working in a senior position within an organization" or "persons working within an organization who possess uncommon knowledge essential to the establishment's service." The EU agreement with Israel does not include specific commitments on labor movement. The Parties only agree to cooperate on preventing and increasing the effectiveness of measures aimed at curbing illegal migration. Other EU trade agreements The 2000 EU-Mexico FTA contains provisions on movement of natural persons in connection with supply of financial services only. Negotiations on services were to take place within three years of the entry into force of the FTA, but to date no agreements have been reached. A noteworthy element of the EU-Mexico Agreement is the linkage between GATS Mode 3 (commercial presence) and GATS Mode 4 (temporary movement of natural persons supplying services): Parties may not require that managerial or other key personnel be of a particular nationality, or that more than a simple majority of boards of directors of financial service suppliers of the other Party be composed of nationals and/or residents of the host country. The 2002 EU-Chile Association Agreement covers services and contains provisions on the temporary movement of service suppliers. Unlike the EU-Mexico FTA, the agreement provides for movement of three categories of service suppliers in a large number of services sub-sectors. Article 99 of the Agreement provides for movement of (i) intra-corporate transferees, (ii) business service sellers and, for the first time in a bilateral agreement, (iii) contractual service suppliers operating in some 30 service sectors. 11 The 11 The covered sectors are legal; accounting; auditing; tax advisory; architectural, urban planning and landscape design services; engineering; medical, dental, midwife, veterinary, nursing, physiotherapy and paramedical services; computer and related services; research and development services; advertising; market research and opinion polling; management consulting and services related to management consulting; technical testing and analysis services; related scientific and consulting services; advisory and consulting services in the fields of agriculture, hunting and forestry; fishing, and mining; maintenance and 17 entry and stay of such contractual service suppliers is subject to duration of stay restrictions in the EU (6 months), limits on the overall length of service contracts and constraints on how contracts can be obtained. The agreement only applies to movement of (highly) skilled persons. The conditions and restrictions applying to movement of natural persons are similar those maintained by the EU under the GATS. The 2008 Economic Partnership Agreement (EPA) with the CARIFORUM countries (Antigua and Barbuda, Bahamas, Barbados, Belize, Dominica, the Dominican Republic, Grenada, Guyana, Haiti, Jamaica, Saint Lucia, Saint Vincent and the Grenadines, Saint Kitts and Nevis, Suriname, and Trinidad and Tobago) and has the broadest coverage of labor mobility of any agreement between the EU and developing countries. Five categories of natural service suppliers are covered: (i) key personnel (including business visitors and intra-corporate transferees); (ii) graduate trainees; (iii) business service sellers; (iv) contractual service suppliers; and (v) independent professionals. Movement of natural persons from the first three groups is not linked to the supply of services, but movement of contractual service suppliers and independent professionals is allowed only in connection to supply of specific services. CARIFORUM companies with contracts to provide services in any of 29 sectors covered by the agreement may send their employees to the EU to provide these services for up to six months at a time. Similar provisions apply to skilled self-employed service suppliers for a much smaller number of activities (eleven) (e.g. computer services, management consulting).12 None of the included services are ones where developing countries have a repair of equipment; photographic services; convention services; translation services; construction services and site investigation work; environmental services; higher education services; adult education services; Travel agencies and tour operator services; tour guide services; entertainment services; news agency services; and services related to the sale of equipment or to the assignment of a patent. 12 The covered sectors for contractual services are legal; accounting; tax advisory; architectural, urban planning and landscape design services; engineering; medical, dental, midwife, veterinary, nursing, physiotherapy and paramedical services; computer and related services; research and development services; advertising; market research and opinion polling; management consulting and services related to management consulting; technical testing and analysis services; related scientific and consulting services; maintenance and repair of equipment; chef de cuisine and fashion model services; translation services; site investigation work; environmental services; private higher education services; travel agencies and tour operator services; tour guide services; entertainment services. 18 strong (potential) comparative advantage. EU and national working conditions apply, including minimum wage requirements, collective wage agreements and licensing requirements. Specific criteria must be satisfied before contractual services suppliers or independent professionals can apply for entry into the EU, including the following.13 Contractual suppliers must be employees of the juridical person supplying the services for at least one year immediately preceding the date of submission of an application for entry. They must have at least three years professional experience in the sector of activity concerned; Suppliers must possess (i) a university degree or a qualification demonstrating an equivalent level of knowledge and (ii) the professional qualifications required by law or regulation to exercise an activity in an EU Member; The temporary entry and stay is limited to a cumulative period of not more than six months in any 12-month period or for the duration of the contract, whichever is less. In addition to these conditions, a significant number of Member States have maintained the right to apply economic needs tests and to limit the scope of intra- corporate transferees, directors or auditors. For example, 16 of the EU-25 have indicated that they will apply such a test in the case of medical and dental services, 21 member states did so for veterinary services, and 22 apply a test in the case of provision of nursing services. The EPA is the only EU agreement to date to include firm commitments on movement of contractual service suppliers and independent professionals. However, it does not include disciplines on regulatory conditions or requirements and does not include mutual recognition of academic and professional qualifications. The criteria relating to educational attainment and the continued application of economic needs tests, The sectors that are open to self-employed providers are: legal advisory services in respect of international public law and foreign law (i.e. non-EU law); architectural services; urban planning and landscape architecture services; engineering services; integrated engineering services; computer and related services; research and development services; market research and opinion polling; management consulting and services related to management consulting; and translation and interpretation. 13 These conditions apply to service suppliers from both parties 19 limit the extent of liberalization. Very few of the covered sectors involve activities in which Mediterranean countries have a comparative advantage, reducing the scope for agreements along the lines of the EPA to help address demographic pressures in a significant way in the coming decades. The problem is not the bias against independent professionals relative to contracts obtained by service companies in CARIFORUM countries, as this is consistent with what economic analysis suggests is a necessary condition for making temporariness operationally feasible (incentive compatible). It is the limited sectoral coverage and the rather open-ended acceptance of economic needs tests that is the major factor limiting the potential for large joint gains. Improving on the EPA At just noted, the EPA with the CARIFORUM constitutes progress relative to the EMP/ENP, by including specific commitments on mode 4 liberalization. The EPA and the agreement with Chile illustrate both the potential that exists for exploiting trade agreements to facilitate movement of services providers and the challenges confronting EU neighboring countries in negotiating meaningful access to the EU market. These agreements show that progress on mode 4 can be made, but that for this to be meaningful it is important to minimize the restrictions and uncertainty that is implied by EU member states maintaining economic needs tests and limiting liberalization to high-skill activities. Both the Chile agreement and the EPA also demonstrate that negotiations with the EU will have to be sector-specific as opposed to horizontal or across-the-board. This suggests that it will be important to build coalitions with EU firms operating in specific sectors that confront serious worker shortages, as well as with government entities and enterprises that have an interest in both expanding the supply of services in specific areas and reducing the costs of service provision. Rather than accept open-ended economic needs tests and the associated uncertainty, a better approach might be to focus on putting place effective safeguard instruments that can be invoked if imports of services grow "too fast" following liberalization. Harnessing support inside the EU for liberalization in specific tasks/sectors that are of interest to EU constituencies is a necessary condition for putting less highly skill intensive services on the negotiating table. Moving down the skill gradient is critical both 20 for there to be more significant gains for Mediterranean and other neighboring economies and for the EU. It will also be necessary to negotiate longer duration of stay provisions, as the 6 month limit is likely to be too short to allow the needed training and other start- up/fixed costs to be recouped. Either a longer period is required, or relaxation of the 12 month overall constraint. In practice a number of revolving (recurrent) 6 months stays that are interspersed with a periodic return to the home country (of say 3 months) will achieve both the temporariness constraint, allow workers to maintain a strong link with the families and allow fixed costs of training etc. to be recovered over a longer period of time. 5. Concluding remarks Movement of people is a positive sum gain for both the EU and neighboring countries. Both have political and economic reasons to prefer more temporary movement in addition to permanent migration. The economic rationales are particularly compelling from the perspective of source countries, given lower brain drain losses and the temporary nature of the demographic bulge of young workers that are entering the labor market. The problem is that the policy debate and bilateral cooperation tends to be focused primarily on control of (illegal) migration and not on the appropriate design of temporary movement policies. Migration must be part of the longer-term response to the demographic trends in the EU countries. However, it is also clear that many EU citizens do not support a policy that leads to significant further opening of the European labor markets to foreign workers. Prospects for liberalization of migration flows in the short to medium run seem low, especially in light of the recent accessions to the EU. Therefore, negotiating additional temporary entry opportunities for service providers through the vehicle of EMP/ENP trade agreements is an option that can be implemented now. While such trade agreements have inherent limitations in that access must be restricted to services, much of the demand for workers is in fact in service activities. That said, negotiating additional temporary access is clearly not a panacea. The potential for a significant expansion of access to service jobs through mode 4 is limited. One reason for this is that the EU is not (yet) a customs union when it comes to services 21 policies, implying that national policies cannot be ignored. This has also been a factor impeding progress in liberalizing mode 4 trade in the ASEAN context (Bhatnagar and Manning, 2005). To make temporariness credible, specific measures need to be put in place that address the incentives of EU firms and MENA workers to resist the turnover costs associated with regular mobility. An effective expansion of temporary movement of people supplying services will imply lower entry costs and may reduce adverse selection effects for jobs that can be provided on a temporary but rolling basis. There will be extra costs for firms associated with transport and recurrent training. To make temporariness feasible is also likely to require financial transfers in addition to more access to EU markets. One reason is to induce South Mediterranean countries to manage illegal migration pressures, including transit migrants. Against this there is a presumptive "social" payoff of not having lower skilled people move permanently ­ something that appears to be becoming part of the objective function of many governments in Europe. Less ambiguous are the likely economic payoffs to the exporting countries ­ an income flow/remittance transfer and a brain drain reduction benefit associated with temporary movement. Finally, it is important to note that the diverging demographic trends are relatively short term, spanning only the next 20 years or so. Therefore the opportunity to take advantage of demographic differences to ease social and economic pressures will not be there for very long. References Amin, Mohammad and Aaditya Mattoo. 2005. "Does Temporary Migration Have to be Permanent?," World Bank, mimeo. Amin, Mohammad and Aaditya Mattoo. 2006. "Can Guest Worker Schemes Reduce Illegal Migration?," World Bank, mimeo. Bhatnagar, Pradip and Chris Manning. 2005. "Regional arrangements for mode 4 in the services trade: lessons from the ASEAN experience," World Trade Review, 4(2): 171-99. Blouin, Chantal. 2005. "Liberalizing the Movement of Services Suppliers: Lessons from the Canadian Experience with Temporary Worker Programs," Journal of World Trade 39(5): 881-94. Boeri, Tito and Herbert Brückner. 2005. "Why are Europeans so Tough on Migrants?" Economic Policy, 20(44): 629-703. Cassarino, Jean Pierre. 2005. "Web Site Guide on Europe's Migration Policy in the Mediterranean: An Overview," Analytic and Synthetic Notes 2005/10, Euro-Mediterranean Consortium for Applied Research in International Migration (CARIM). 22 Chaudhuri, Sumanta, Aaditya Mattoo and Richard Self. 2004. "Moving people to deliver services: how can the WTO help?" Journal of World Trade 38(3): 363-93. Commission of the European Communities. 2003. "GATS Offer," mimeo. Commission of the European Communities. 2005. "Framework protocol of Istanbul for the liberalization of trade in services among Euro-Mediterranean partners," mimeo. Docquier, Frederic and A. Marfouk. 2006. "International migration by educational attainment (1990-2000)," in C. Ozden and M. Schiff (eds). International Migration, Remittances and Development, Palgrave: New York. Available at: http://perso.uclouvain.be/frederic.docquier/oxlight.htm. Dodini, Michaela, and Marco Fantini. 2006. "The European Neighborhood Policy: Implications for Economic Growth and Stability," Journal of Common Market Studies 44:507-32. Fargues, Philippe. 2005. "Temporary Migration: Matching Demand in the EU with Supply from the MENA," Analytic and Synthetic Notes 2005/11, Euro-Mediterranean Consortium for Applied Research in International Migration (CARIM). Fargues, Philippe. 2006. "The Demographic Benefit of International Migration: Hypothesis and Application to Middle Eastern and North African Contexts." Policy Research Working Paper 4050, Washington, D.C.: The World Bank. Grogger, Jeffrey and Gordon Hanson. 2008. "Income Maximization and the Selection and Sorting of International Migrants," mimeo. Hatton, T. and Jeffrey Williamson. 2004. "International Migration in the Long Run: Positive Selection, Negative Selection and Policy," NBER Working Paper 10529. Hoekman, Bernard. 2005. "From Euro-Med Partnership to European Neighborhood: Deeper Integration à la Carte and Economic Development," ECES Working Paper No. 103, Cairo. Holzmann, Robert and Rainer Münz. 2004. "Challenges and Opportunities of International Migration for the EU, Neighboring Countries and Regions: A Policy Note," Social Protection Discussion Paper 0411, The World Bank. Kessler, Alan and Gary Freeman. 2005. "Public Opinion in the EU on Immigration from Outside the Community," Journal of Common Market Studies 43(4): 825-50. Kox, Henk and Arjan Lejour. 2006. "The Effects of the Services Directive on Intra-EU Trade and FDI," Revue Economique, 57(4): 747-69. Kremer, Michael and Stanley Watt. 2006. "The Globalization of Household production," Harvard University, mimeo. Langhammer, Rolf. 2005. "The EU Offer of Service Trade Liberalization in the Doha Round: Evidence of a Not-Yet-Perfect Customs Union," Journal of Common Market Studies 43(2), 311-25. Martin, Philip. 2003. "Managing Labor Migration: Temporary Worker Programs for the 21st Century," International Institute for Labour Studies, Geneva, mimeo. Martin, Philip. 2006. "Population and Migration," in Bjorn Lomborg (ed.), How to Spend $50 Billion to Make the World a Better Place. Cambridge: Cambridge University Press. Mattoo, Aaditya and Randeep Rathindran. 2006. "Does Health Insurance Impede Trade in Health Care Services?" Health Affairs 25(2), 358-68 Mayda, Anna Maria. 2004. "Who is Against Migration? A Cross-Country Investigation of Individual Attitudes toward Immigrants," IZA Discussion Paper 1115. Ozden, Caglar, Mirwat Sewadeh and J. Wahba. 2009. "Temporary Labor Migration from Egypt to the EU: Economic and Legal Background," World Bank, mimeo. Schiff, Maurice. 2004. "When Migrants Overstay their Legal Welcome: A Proposed Solution to the Guest Worker Program," World Bank, mimeo. Tani, Massimiliano. 2005. "Head-content or Headcount? Temporary Labour Movement as a Source of Growth," University of New South Wales, mimeo. United Nations. 2009. World Population Prospects. At http://esa.un.org/unpp/index.asp. 23 Wahba, Jacqueline. 2006. "International Migration, Education and Market Failure in MENA", paper prepared for a World Bank Regional report on Education, mimeo. Walmsley, Terrie and L. Alan Winters. 2005. "Relaxing the Restrictions on the Temporary Movement of Natural Persons: A simulation Analysis," Journal of Economic Integration, 20(4): 688-726. Winters, L. Alan. 2005. "Developing Country Proposals for the Liberalization of Movements of Natural Service Suppliers," Working Paper T8, Sussex Centre for Migration Research. Winters, L. Alan, Terrie Walmsley, Zhen Kun Wang and Roman Grynberg. 2003. "Liberalising Temporary Movement of Natural Persons: An Agenda for the Development Round," World Economy 26(8): 1137-61. 24