International Migration Regimes and Economic Development
13 May 2004
This is a summary of the seminar on International Migration Regimes and Economic Development, arranged by the Swedish Ministry for Foreign Affairs and the Expert Group on Development Issues (EGDI), 13 May 2004 in Stockholm.
The purpose of the seminar was to present and discuss an ongoing EGDI-study authored by Robert E.B. Lucas, Boston University. Submitted to the seminar were the first seven (out of ten) chapters. The seminar was chaired by Mr. Ola Henrikson, Director-General for Migration and Asylum Policy (morning session) and Ms. Ruth Jacoby, Director-General for International Development Cooperation (afternoon session).
Rapporteurs: Ms. Lisa Arrehag and Ms. Mirja Sjöblom, Stockholm School of Economics.
Introduction and welcome
Annika Söder, State Secretary for International Development Cooperation, gave an introduction to EGDI and the Swedish policy of global development. This newly adopted policy is unique in the sense that it covers all policy areas of relevance for global development, also migration. Ms. Söder stressed that international migration must be included in the policy discussion of how to make globalization work for all. Moreover, she noted that although migration is on everybody‘s lips, there is a great need for better data. The two starting points for the study conducted by Professor Lucas, she explained, were to make a contribution to the research of regional variations in migration regimes and to understand how the policy setting affects the development impact of migration in the migrant source nations.
Causes of migration,
Consequences for Economic Development in the Sending Countries I: Labour market responses, Remittances
Initiating the seminar, Professor Lucas gave a brief introduction to his study on labour migration and its effects on the economic development in the sending countries. The study focuses on four major migration regimes; migration to the European Union, particularly from Eastern Europe and the Maghreb region, contract workers in the Persian Gulf from South and Southeast Asia, the brain drain to North America, and the migration transition in East Asia.
Professor Lucas presented the chapters on, respectively, the determinants and patterns of migration, labour market responses to migration in the countries of origin, and remittances. The reason for including the determinants of migration was because of the two-way link between development and migration. It is not only the controls imposed by the immigration country that shape the outcome of migration, but also migration pressures in the country of origin. Professor Lucas stated that the conventional wisdom in literature is that economic development does little to reduce migration pressures. Even a counter-effect has been observed, i.e. that international migration stems from development itself, the so-called migration hump. Contemporary evidence suggests, however, that if there is a turning point causing a ‗hump‘, it comes at very low-income
levels, and that its mere existence is far from clear, being sensitive to the way in which migration is measured.
Professor Lucas claimed that what is known is that tighter labour markets at home reduce emigration. The migration transition in East Asia is one of the examples illustrating this. He concluded that economic development is not the only factor shaping migration pressure. However, economic development at origin typically reduces migration pressure and especially when development tightens labour markets.
The second area of focus was domestic labour market responses to emigration. Although there is virtually no systematic evidence regarding this matter, Professor Lucas argued that the impact is likely to depend on factors such as the skills, gender and origin of the emigrants. Also the labour market conditions in the sending country, in terms of unemployment rate and flexibility, were mentioned as important factors. Professor Lucas presented a few cases of different labour market responses to emigration. He concluded that stayers, on the one hand, could lose with the departure of the irreplaceable skilled, if the consequence would be a reduction of demand for unskilled workers. This is however less likely in the long run, and in a context where the educated are unemployed and training costs are low. On the other hand, stayers can benefit from emigration. They either gain by replacing the departure worker, like in Albania, Bangladesh and possibly India and Indonesia, or through higher wages, which has been observed in the case of Pakistan and possibly in the Philippines.
Thirdly, on remittances, Professor Lucas presented data showing how remittances to the developing countries have grown during the past 15 years. The remittances from the four Gulf States that report remittances, alone equal the total net remittances coming from industrialised countries. Thereafter, he identified remittances‘ effects on GDP, noting that the positive effects are that remittances stimulate investments and function as a domestic spending multiplier. The negative consequences mentioned, were that recipient families
could drop out of the labour force, and the so-called Dutch disease effect, i.e. that an inflow of remittances results in appreciation of the exchange rate which in turn reduces exports. Professor Lucas further suggested that efforts should be focused on creating an environment favourable for investments rather than redirecting remittances. Many countries are today trying to adopt policies to maximize the inflow of remittances, but systematic information regarding the success of such policies is non-existent. There is a huge network of people sending money through informal channels. For fear of lacking tax revenues, as well as a concern that these channels are used for financing terrorism, efforts have recently been directed to regulate these financial flows. Professor Lucas raised the issue that the net effect of such regulations could reduce total remittances to the developing countries. Professor Lucas concluded that although remittances‘ effect on
GDP and inequality remain controversial, many home governments seek to maximize the inflow. Further, there is evidence that even if the poorest do not directly receive the money transfers, there are side-effects from the spending of the remittances that have an indirect positive effect on poverty alleviation.
The first discussant, Professor Ernesto Pernia, University of the Philippines School of Economics, commented that economic development, if characterized by tighter labour markets, serves to attenuate pressure to migrate not solely for unskilled workers but also for skilled workers and professionals. Further, Professor Pernia missed one factor, here and in other studies, i.e. how greater trade openness in developed countries effect international migration. Moreover, he stressed that cross-country regressions need to be complemented or validated by country-specific studies.
Regarding labour market responses, Professor Pernia suggested that the conclusion concerning labour market gains for those left behind could undergo two refinements, firstly by looking at two time periods, pre and post 1985, which was the year of the Plaza Accord, a watershed particularly for Asian developing countries, implying tightened labour markets. Secondly, to examine the effect of departing workers on the quality of goods and services, that in turn reflects the quality of replacement workers, since there have been indications of deterioration in quality. Professor Pernia questioned what the saw as a somewhat simplistic way of explaining the determinants of remittances. He stressed that remittances should be regarded as the returns to migration, viewed as investment in human capital on behalf of the migrant and his/her family. In terms of macro determinants, he added that social and political stability in the home country would probably favour the rise of formal remittances and the corresponding fall of informal remittances.
Professor Pernia concluded on remittances by stating that there is a need to search for more efficient and attractive modes of remitting through formal channels, as well as finding ways to widen and deepen their impact on economic growth and poverty reduction in the sending countries, perhaps in combination with development assistance.
The second discussant, Mr. Dilip Ratha, Senior Economist at the World Bank, proposed that definitions concerning different types of migration could be included in an introductory chapter. He also questioned why migration is expected to have a great impact on world growth and development given that it solely concerns three per cent of the world‘s population.
In his remaining comments Mr. Ratha put emphasis on remittances. He stressed that in comparison with other financial flows, remittances are characterized by stability, in the sense that they are not as volatile as official flows and do not vary substantially over time. These are characteristics that make remittances contribute significantly to poverty reduction. In terms of remittance data he raised caution about cohering remittances flows
to other kinds of financial flows. He also stressed the importance of distinguishing between gross and net flows of remittances and lastly to not disregard intra EU-flows of remittances. Furthermore, he praised Professor Lucas‘ summary of the determinants of remittances. He added that the amount of remittances is not exclusively dependent on the magnitude of emigration, but also on the income level of both the emigrant and the recipient family, i.e. everything else being equal, an emigrant from a poor family is likely to remit more than an emigrant from a wealthy family. With the last point he indicated that remittances do not raise inequality but rather move recipient families up one or two income levels. Moreover, he stated that the exchange rate factor, the high cost of using formal channels, and language, are important reasons why informal channels are used.
Mr. Ratha stated that since remittances in many cases represent a relatively small part of GDP, it is difficult to find a relationship between remittances and growth. However, if remittances make up a large share of the trade deficit, a strong relationship between remittances and the exchange rate can indeed be traced. Mr. Ratha also remarked that the cross-country regressions do not capture the regional differences of how remittances affect development.
Regarding a suggestion that remittances may reduce domestic labour supply, Mr. Ratha raised two questions: firstly, if the recipients do not work, what do they do instead, and does it necessarily has a negative affect on GDP? Secondly, if there is a high rate of unemployment, which is prevalent in many sending countries, does the drop out of labour affect the aggregate employment level? He further stressed that the reason why Professor Lucas has difficulties in finding that migration affect real wages is because of high rates of unemployment in many sending countries. To conclude, Mr. Ratha expressed that remittances and migration should be considered as personal decisions. He recommended policy makers to focus on improving the conditions for remittances markets and for remittances users, rather then interfering in the migrants‘ personal decisions regarding migration and remittances.
In responding to the comments by the first two discussants, Professors Lucas said that he planned to include the complex trade-migration-linkages in the concluding chapter. On the industrialised countries‘ policies of imposing protecting trade policies on agricultural
and garment industries, he however commented that they result in employment of unskilled people at the same time as immigration of unskilled labour is refused and brings about large amount of irregular migration. Moreover, he encouraged the World Bank and IMF to join efforts to solve the problems of the poor quality of remittance data. Finally, he presented the idea of publicising cost margins of money transfer companies on a public website to increase competition in the currently cartel like money transfer market.
A general discussion followed: Professor Florian Alburo, University of the Philippines School of Economics, called for something more insightful in the sense of relevance to policy. In particular he inquired for this part in the discussion about the migration hump
and the determinants of migration. Professor Binod Khadria, Jawaharlal Nehru University, pointed to a new tendency of return remittance flows from India in the form of tuitions that are part of the national income in terms of Research and Development, but not of GDP. He further suggested that the focus on the number of migrants moving from different countries, i.e. the ability to move, could be complemented by considering the willingness to move, arguing that the ability depends on the receiving countries‘ policies,
while willingness is not necessarily affected by policies.
Mr. Maurice Schiff, Lead Economist, the World Bank, firstly asked for clarification as regards the definition of labour market tightness and wondered what other issues, than wage, were relevant in discerning the relationship between labour market responses to migration and international migration. Secondly, he made reference to the presented case of Sri Lanka, where wages went down despite huge migration, noting that the causality between the two was unclear. Thirdly, he highlighted the case of Mexico, explaining that fluctuations in the exchange rate are likely to affect incentives to remit. Mr. Schiff also objected to the income effect of remittances and proposed that leisure should be discounted for. He also stressed that it is very hard to measure remittances‘ effect on reforms, arguing that a high inflow of remittances enable a country to maintain unsustainable macro economic policies. Finally, he stressed the importance of distinguishing between the household level and the nation level. It is typically not the poorest that migrate and receive remittances. Hence, even if the poorest countries receive remittances it does not necessarily mean that the poorest part of the population is included among the recipients.
Professor Louka Katseli, the OECD Development Center, suggested that Professor Lucas should consider coherent policies affecting migration, for instance by reflecting on the links between trade and migration, ODA and migration, FDI and migration, etc. Professor Katseli also questioned why any efforts by the government to create incentives for more productive savings of remittances should be discouraged. Both diaspora and traditional emigration countries would benefit from the implementation of financial instruments that have a beneficial effect on the communities of the sending countries.
In Professor Lucas reply to these comments, he started by explaining that the lack of policy recommendation is due to the difficulties in saying anything firmly from the existing evidence. Moreover, he responded to Mr. Schiff‘s comment on the tightness of labour market by pointing out the difficulties in measuring wages especially in poorly developed countries. Regarding the case of Sri Lanka, he agreed that there is indeed a need to disentangle what is cause and what is effect. Furthermore, he broadened Mr. Schiff‘s point on remittances and government‘s willingness to reform, arguing that migration itself is taking pressure off the government to provide jobs at home.
Notwithstanding the criticism of migration‘s positive impact on poverty alleviation Professor Lucas stood by his opinion that it might not be the poorest that emigrate but certainly people below the poverty line. Lastly, Professor Lucas and Professor Katseli entered a discussion regarding financial instruments. Professor Lucas claimed that development of financial instruments should not in particular target remittances receiving families and Professor Katseli asserted that efforts should be directed towards the diaspora in order to create incentives for remitting.
During the lunch break, Mr. Jan O. Karlsson, Co-chair of the Global Commission on International Migration (GCIM), gave a short presentation of the Commission, which is mandated to address the question of a normative framework to regulate global migration, and will present its report in July 2005. Besides, Mr. Karlsson noted that there are both positive and negative circles of migration; he exemplified the first with the return of skilled migrants from the US to India and the related IT-revolution in Bangalore and the latter with the emigration of health workers in South Saharan Africa, a region highly affected by HIV/AIDS. He further called for the need to abandon the concepts of sending and receiving countries in view of a much more complex reality. Many countries, such as South Africa, Korea and India are experiencing both emigration and immigration. Finally he stressed the importance of finding a multilateral framework for migration in order to avoid the ‗blame game‘ between countries that is prevalent in the contemporary international debate.
Consequences for Economic Development in the Sending Countries II: Brain Drain, The Diaspora and Transnational Networks, Return Migration
Professor Lucas started by discussing the brain drain and brain gain. Concerning the brain drain, there is, as he phrased it, the bad, the good and the ugly. The ‗bad‘ are externalities
and fiscal losses, the ‗good‘ is induced education and brain gain, and the ‗ugly‘ is the data. Professor Lucas claimed that the only relevant data that exist at this point concerns the migration of highly skilled people to the US. He stressed that the US has the third largest fraction of college educated migrants and that the number of migrants are larger in the US than anywhere else, thus, the US is still in the dominant end of brain drain. Further, he pointed out that there exists a negative association between income per capita and the amount of brain drain to the US. The lowest income countries are the ones having the highest proportional rates of brain drain to the US.
The ‗bads‘ concerned negative externalities on the people left behind and the high
costs to the sending country when highly educated people leave. Professor Lucas made a conjecture that in some contexts, like in Southern Africa, the departure of highly skilled people matters more than in others, which is closely related to the problem of HIV. Moreover, he underlined the fiscal losses when the tax base of highly skilled people disappears, but he also argued that the departure of the highly skilled implies lower government expenditure on this group. In this context, Professor Lucas stressed that one has to look at the situation in each individual country. The other component of the fiscal cost mentioned was the high public costs of education. He showed that, while the costs of tertiary education in the lowest income countries are exceptionally high, the social returns are low.
Thereafter, he turned to the good side of brain drain by bringing up the discussion on the concept of brain gain, seen in the last ten years of literature. Professor Lucas referred to a recent paper by Riccardo Faini, in which a negative relationship between brain drain and tertiary education in the sending country was observed. However, the paper did find a
positive association between brain drain and secondary education. Hence, one of the conclusions was that the brain drain is encouraging overseas education, but whether that is good or bad is far from clear. Professor Lucas concluded that the lower income countries had the highest staying rates of students with a completed PhD in science and engineering in the US. Thus, the brain drain encourages people to go abroad to study, but in the low income countries the migrants do not return. As an exception to this he mentioned Indonesia.
He continued by arguing that especially highly skilled migrants can help stimulate three things; international trade with their country of origin, capital flows and technology transfers. According to Professor Lucas this is because the migrants have better information about the opportunities in their home country, since they transmit information through network connections with those at home and since they can transcend reputation barriers whereby firms in the home countries begin to export. He exemplified by the so-called Bamboo network in South East Asia.
Summing up on the brain drain, Professor Lucas mentioned that virtually all OECD countries have programs to facilitate the entry of highly skilled people, that this is becoming a competitive tool where the US is the dominant outlet. He further accentuated that since it is impossible to quantify the bad effects that the brain drain has on the developing countries, it is neither possible to quantify the effects it has on the receiving countries. Still, most of the high-income countries are assuming that they have gained of the process. He also stressed that the brain drain is unlikely to decline, and that this raises a complex issue of what policies should be in place. If the big concern is the loss of money invested in students, he suggested that the countries affected should reevaluate higher education financing, because higher education in low-income countries is extremely expensive and, to a large extent, it is the wealthy children and families that are the beneficiaries.
The second topic was temporary and return migration. Professor Lucas expressed a few pros and cons of return migration for development. He argued that the links with the diaspora are stronger, and so also the remittance flows, if the migrants intend to return. Another possible positive consequence is that migrants return with fresh skills and experiences from abroad. However, the stories about mismatches of the returning migrants are plentiful. An example of the negative sides of return migration is unemployment upon return. To solve this, a number of countries have put in place policies for re-assimilation, but growing evidence suggests that the people who actually return tend to be the less successful in the destination countries. Moreover, policies inducing people to go back are very costly.
Professor Lucas stated that despite poor measurements, the data shows that the remittances originating from the temporary and unskilled migrants of the Gulf are enormous. One specific reason mentioned was that migrants tend to stay for long periods of time. A too rapid turnover of migrants imposes costs for both the employers in the host countries, who have to replace and retrain people, and on the sending countries, by limiting pay increments. Professor Lucas claimed that the alternative to guest workers is irregular migrants and that it is not clear that they give a higher turnover than the regular migrants. He continued by pointing out that, with temporary migration programs, it is easier to export people when there is a shock to the receiving country. One such example is the Gulf countries, from where immigrants have been exported during periods of shock.
This imposes severe costs on some of the sending countries, but typically the consequences have been temporary. These results led Professor Lucas to question if temporary programs can be cut back on a long-term basis. Further, he pointed out that he had noticed a contrast in the debate regarding if guest workers and irregular migrants are substitutes or complements, i.e. whether guest worker programs diminish irregular migration or ultimately serve to promote further irregular migration. While the US recommends against the guest working programs, reasoning that the earlier experience with the so called bracero program caused irregular migration to the US, in Europe, there has been a discussion about the establishment of guest workers programs, often through bilateral agreements, as a way to reduce irregular migration.
In his final point, Professor Lucas stated that the long term demand for low-skilled workers depends on two things, whether the trade protection and agricultural subsidies are going to remain in place, and on future technologies. He concluded that it is most likely that the demand for low skilled workers in the high-income countries will remain.
Discussant Professor Devesh Kapur, Harvard University, chose to highlight the complexity of the migration field and how little that is known, by stressing that even though a remittances receiving family in which the mother has emigrated appears better off economically, the social consequences for the children that are raised without their mother remain unknown. Further, he pointed out that it is important to distinguish who are the real beneficiaries of migration - the nation or the individual. That is, people from poor countries might be better off, while the sending countries might be worse off. He stressed that when considering the impact of the skilled migration, it does not only depend on the students who leave but also on their type of education. Professor Kapur further stressed the effect on public institution building. He considered especially that certain groups such as the middle class have a higher propensity to emigrate. The loss of the middle class can be particularly severe, since this is an extremely important group for building public institutions. While financial capital can return quite fast, the middle class often leave on a permanent basis, representing an important loss to their country.
Furthermore, he raised the example of China arguing that sending students abroad has become an important way of money laundry. With regard to the diaspora, he said that its influence has to do with social remittances, i.e. the flows of ideas. For example, surveys in Mexico have shown that migrant households in the US have greater market orientation. He also mentioned that the social consequences of diaspora are country specific and depend on the reasons for leaving and who is leaving. For example in Colombia and Venezuela where a large part of the elite has left, the diaspora has a strong political influence. Other consequences are network effects. Professor Kapur claimed that mafias are an example of international networks made possible through the migrant diaspora.
In his last comment he reflected upon real effects on return. He claimed that these depend on the sending countries‘ policies, but it is also of relevance where migrants go and to what context they return. The less positive effects mentioned were for instance the rise of criminality in Honduras and Guatemala, because US prisons spread more sophisticated knowledge on how to commit crimes. Finally, Professor Kapur emphasised that the effect of the diaspora depends on the mechanism through which one enters the work force in the destination country, i.e. whether one studies or joins a firm will affect the sending country in different ways.
The second discussant, Dr. Nyberg Sörensen, Danish Institute for International Studies, initiated her comment asking for a clarification of the definitions of diaspora and transnational networks, stating that most recent theoretisations of diaspora have been marked by the ambiguities of the term diaspora itself. Dr. Nyberg Sörensen further commented on theory and concepts fearing that the present emphasis on diaspora, remittances and brain-circulation in various European development agencies introduce a specific rationality or donor agenda through which some relations are revealed and others concealed. She emphasised that one should not forget that migration above all is a human phenomenon and that there is more to development than pure economics. Furthermore she remarked that most diasporas are highly heterogeneous and therefore often develop different transnational relationships to their homelands.
Dr. Nyberg Sörensen continued by congratulating Professor Lucas for rightly pointing to class differences in transnational networks between unskilled workers, middle class migrants, and highly skilled migrants. She however expressed that the study had overlooked the form and conditions of movements, the nature of host societies, and the status and standing of diasporas within host countries. Moreover, she expressed the constraints that diasporas face, exemplifying that the link with family and kinship can be a source of draining obligations through the demand for remittances. She urged development agencies not to place additional stress on already vulnerable groups through their renewed focus on diasporas and remittances as a development tool.
In relation to trade she commented that while it is difficult to discern whether migration enhances trade or trade enhances migration, she did not see a reason for why this role should be limited to highly skilled migrants. She claimed that low skilled migrants contribute financially to the development in their home countries through their demand and purchasing power for local goods and services. Concerning the statement that highly skilled migrants have the potential for establishing knowledge networks that transmit new ideas and technologies back to their home country, she added that the other side of the coin is that a large quantity of migrants and refugees only find employment below their skills. Reflecting upon both the positive and the negative role that diasporas play in shaping politics and the state of security within their home countries, she proposed a few different strategies. A first policy priority must be to secure the rights of migrants and refugees in the receiving societies, maintenance of flexible asylum and resettlement policies. Migrants and refugees also need a stronger voice in shaping development in their home countries. Her final comment was on the subject of gender, pushing for the present and similar forums to consistently examine the degree to which diaspora practices and transnational networks are gendered, and if so, how this should be incorporated in forming development policies.
Ms. Marion Eeckhout, The Netherlands Ministry for Foreign Affairs, called for a more dynamic perspective of the migration process. She pointed out that it would be interesting to see how Professor Lucas‘ cross-country analysis could be blended with more evidence
on the micro level, and asked if the author saw opportunities of linking a micro perspective to the country analysis. Professor Katseli then identified four brain drain regimes. Firstly, the brain drain in East Asia which has to do with excess demand for and access to education which is highly related to education policy. Secondly, the brain drain
to meet short run demands in the high income countries of for instance nurses and teachers. Thirdly, she considered the second-generation migrants, particularly in the US. The fourth regime has to do with the stock of migrants and the externalities related to trade and services. She urged the author to separate these four categories and thereafter analyse and form policy recommendations based on this classification.
Professor Katseli‘s second point concerned the highly debated question whether there should be a shift in the incentives towards more temporary migration policies. She claimed that this issue depends on whether the sending and receiving countries have integrated labour markets. It would be one thing to talk about Spain, Italy and Greece towards Albania, and quite another to consider Philippine workers coming into selected European countries.
Professor Khadria stressed the importance of the consequences of policies. He argued that if there is more permanent or temporary migration ultimately depends on the reason behind the policy priority. He said that there are three prime motivating factors for the receiving countries encouraging migration to Europe; age, wage and vintage. It is typically the younger generation that leave, which affect the whole age structure of the population in the receiving country. Wages, he argued, could be kept low if the youth replaced the senior managers and workers. Thirdly, he mentioned the importance of the vintage of technology. Mr. Hans Werner Mundt, GTZ Germany, called for a comment from the academia on how to deal with the fact that training measures create incentives to brain drain.
Professor Lucas replied to the comments from the discussants and the floor by stressing that the most worrying part with regard to brain drain is the data. Regarding the diaspora, he mentioned how vast it is for certain countries such as the Philippines and China, especially when irregular migrants are included. He stated that the big concern for governments today is to keep in contact with the diaspora in order to not loose remittances. Further, he argued that some transnational networks, such as the India-Bangalore network, are better organised than others that are not more than a website. Concerning the consequences of brain drain, Professor Lucas held that it is hard to predict what would have been the outcome, for instance in health care in Sub-Saharan Africa, if the most competent health workers had not left. Professor Lucas further expressed that he liked Professor Katseli‘s structure of brain drain regimes and that he thought the commentators added many interesting points.
Policy implications for receiving and sending countries: panel discussion
The afternoon panel discussion was initiated by Jean-Pierre Garson, OECD, who stated that in order to understand the situation of migration today it is important to draw conclusions from the past by considering how countries such as Portugal, Greece and Italy went from sending to receiving countries. At the same time he expressed scepticism toward the global discussion of migration since migration is very context specific. He stressed that receiving countries have different migration policies with respect to the immigrants‘ social status, human rights, etc. He further pointed out the importance of identifying where the real problems arise, to avoid having the same discussion with