CGD Brief October 2005 The Global Migration of Talent: What Does it Mean for Developing Countries? Devesh Kapur and John McHale*

Summary: Human capital flows from poor countries to rich countries are large and growing. A leading cause is the increasing skill-focus of immigration policy in a number of leading industrialized countries—a trend that is likely to intensify as rich countries age and competitive pressures build in knowledge-intensive sectors. The implications for development are complex and poorly understood. While fears of the “brain drain” were overwrought in earlier decades, the recent celebration of “brain gain” is also overdone, especially as highly selective migration policies deprive poor countries of scarce innovators and institution builders. We explore available policy responses to improve the net effect on development without making the international migration system even more illiberal than it is today.

Cross-border financial capital flows have transformed the global economic and political landscape over the last half-century. Over the next few decades, global migration, driven by demographic and technological factors and the inevitable persistence of large income gaps across countries, is likely to play an equally influential role in shaping the nature of politics and economics internationally. The bulk of migration will take place within developing countries themselves, specifically regarding the movement from rural to urban settings in the giants—China and India. The second largest migration flows will occur internationally, among developing countries, thus continuing the trend of recent years. A third body of migrants will travel from developing to developed countries.

The consequences of substantial immigrant inflows have prompted much debate and analysis about their effects on advanced industrial countries. As these inflows continue to swell over time, the welfare of many migrants, both intra- and international, will require the attention of policymakers the world over, irrespective of the specific circumstances—income gaps, ethnic cleansing, economic instability, or human trafficking—that provoked their departure from their country of origin. However, another reality has received short shrift. What will be the consequences of large cross-border flows of people on the countries of origin? While the majority of international migrants will be low-skill workers, a critical number of them will be highly skilled. What will be the impact of these flows on development, and how substantial will these effects be? The issue demands greater attention because competition for developing country talent will intensify in the years ahead.

Brain-Drain Trends

Skilled emigration rates substantially increased during the 1990s (Table 1). For many poor countries, the share of skilled nationals residing in rich countries is staggeringly high—more

*Devesh Kapur ([email protected]) is Associate Professor of Government and Asian Studies, University of Texas, Austin, and a non-resident fellow at the Center for Global Development. John McHale ([email protected]) is Associate Professor of Economics at Queens School of Business, Kingston, Canada. This brief is based on their new book Give Us Your Best and Brightest: The Global Hunt for Talent and Its Impact on the Developing World (Washington, DC:Center for Global Development, 2005).

www.cgdev.org © Center for Global Development. All Rights Reserved. The Global Migration of Talent: What Does it Mean for Developing Countries? 2 Source forTable1and Figure 1:Docquierand Marfouk (2004)andauthors'calculations . trends: long-term by threepowerful of thehighlyskilledfrompoorcountrieswillcontinue,driven barrierstoimmigration selective dismantlingofrichcountry arecastingashadow,heightened securityconcerns the Although theburstingoftechnologybubbleaswell formanyAfricancountries(Tableis morethanaquarter 2). educationliveinOECD countries,whilethefigure a tertiary than two-fifthsofnationalstheCaribbean(Figure1)with Table 1. uii . 23429.6 5.6 4.2 0.1 27.5 2.0 4.6 1.9 12.3 0.1 4.2 5.0 4.6 3.3 4.7 4.6 0.2 0.0 0.1 0.1 6.3 24.8 9.2 6.2 6.3 0.8 0.7 0.3 * 5.3 0.1 4.2 1.7 2.3 21.5 Tunisia 11.0 0.2 3.5 5.6 82.5 Sudan 0.1 0.1 6.1 0.8 Lanka Sri 4.2 8.3 0.3 Indonesia 4.2 0.2 21.7 31.5 0.1 Turkey 18.2 9.2 0.1 5.8 Bangladesh 11.2 84.1 2.1 Egypt 14.8 14.3 0.5 Brazil 3.1 11.0 1.4 9.5 Pakistan 17.9 2.6 32.9 Peru 0.1 3.8 8.2 Guatemala 0.1 Colombia 12.8 10.4 Jamaica 1.1 6.5 Republic Dom. Salvador El China Tertiary India Primary Philippines Tertiary Mexico Primary ofbirth Country Percent ofpopulationlivinginOECDcountries,byeducation level service providersforanolderpopulation.Butthehugefiscal service labor marketside,thistrendislikelytoincreasedemandfor populations.Onthe Second, theagingofrich-country gains ofhumancapitalinflowsareprobablymostmanifest. thatthelong-term is here,ratherthaninlabormarketeffects, growthisaboutinnovationandit based industries.Modern have acompetitiveadvantageinemergingknowledge- desireto technological advancementwithgovernments’ First, thecombinationofskill-biasmuchrecent Better-educated Workers AreMore Likely toEmigrate:* 902000 1990 Figure 1. Western Africa flexibility tohireforeignworkersdomestically. sometimes usethethreatofmovingjobstowinmore betweenlocations,and the flexibilitytomovetheirstaff together. Forexample,multinationalcompaniesdesire labor marketintegration,inpracticetheytendtoevolve capital marketintegrationcansubstituteforinternational productand international trade. Althoughintheory Finally, thebroaderglobalizationofproductionand ipr”txaesat taxpayers “import” population, there will bestrongpressuresto retired the for on theworking population. W pay forpensionandhealthcarebenefitsthedomestic foreignworkerstohelp attempts toattracthigher-earning costs ofpopulationagingarealsolikelytodrivetargeted Dom. Republic Eastern Africa Eastern Bangladesh El Salvador Guatemala Caribbean Philippines Colombia Sri Lanka Pakistan Jamaica Mexico Sudan China Brazil India Peru in OECDCountries,(1990-2000) Percentage ofSkilledWorkers Living 204060801000 ith the alternatives beinggreatertaxincreases ith thealternatives population ormoresubstantialbenefitcuts the margin. However, the welfare effects on TRBs are less well understood 3 Table 2. Estimates of the Brain Drain and the multiple channels strongly suggest that the implications October 2005 from Africa: Emigration Rates for international skilled migration are complex and defy any for Tertiary Educated, 2000 simple-minded, facile bottom lines.

Percentage of nationals with university The prospect channel captures the way in which a prospect or education living abroad, 2000 an option of emigration affects the decision-making of people in sending countries, whether or not they actually end up > 50 Cape Verde, Gambia, Seychelles, Somalia emigrating. In particular the prospect of emigration increases the inducement to get more education as well as the type of 25–50 Angola, Equatorial Guinea, Eritrea, Ghana, education. The massive increase in nursing education in the Guinea Bissau, Kenya, Liberia, Madagascar, Philippines illustrates this phenomenon. This channel affects Mauritius, Mozambique, Nigeria, Sao Tome and Principe, Sierra Leone decisions related to the types of skill acquisition, the supply of entrepreneurship, the development of relationships with 5–25 Algeria, Benin, Burundi, Côte d’Ivoire, Cameroon, co-located individuals, and the extent and form of savings. Chad, Comoros, Congo, DRC (formerly Zaire), Djibouti, Ethiopia, Gabon, Guinea, Malawi, Mali, The absence channel focuses on the effects on TRBs Mauritania, Niger, Morocco, Rwanda, South when skilled individuals actually leave. At the simplest level, Africa, Senegal, Sudan, Swaziland, Tanzania, Togo, Tunisia, Uganda, Zambia, Zimbabwe the country loses an “emigration surplus”—the difference between the values an emigrant was adding to the economy < 5 Botswana, Lesotho, Burkina Faso, Central African and that which the emigrant was being paid. This is a direct Republic, Egypt, Libya, Namibia money measure of the economic loss of the emigrant’s absence. However, the absence of skilled workers on the domestic econ- omy also results in larger skill premiums, fiscal losses, diminished scale economies, and changed comparative advantage. Looking forward, the evolving liberalization of trade in services is likely to blur the lines between trade and migration. Most importantly, it might affect a country’s capacity to build The intensification of product market competition will increase domestic institutions, increasingly recognized as the critical the pressure on governments to ease immigration restrictions so variable for development. How does the absence of highly as to provide domestic firms with a source of competitive talented individuals affect institutional development? We find it advantage through improved and cheaper access to a diverse useful to distinguish between the impacts of emigration on the set of skills. When it comes to the innovation-intensive sectors supply of institution-builders and the demand for better of the economy, governments of rich countries will be more and institutions. The supply side is the more straightforward. more sensitive to claims that other countries are providing more Countries have limited supplies of people willing and able to conducive competitive environments to their firms. take on entrenched interests to reform schools, establish clinics, and fight for the rule of law. The dilemma is that potential Effects on Developing Countries institution-builders are most likely to leave where institutional quality is worst. And the very individuals most likely to be How will this increased competition for poor countries’ talent institution-builders by talent and temperament, be it professionals affect their development prospects? There are four key channels or managers, are the most likely to be internationally through which international skilled migration affects sending marketable. If people of talent and drive are essential for countries. Our analysis of each of the four channels—prospect, building institutions, then their loss can have severe absence, diaspora, and return—focuses on effects of skilled consequences. Added to this, the incomes of non-corrupt but emigration on the welfare of “those remaining behind” (TRBs) talented individuals working in public institutions with in the home country. This is not to say that the welfare of TRBs compressed wage scales are likely to compare unfavorably is all that matters. Emigration obviously has important welfare with foreign alternatives. Such people are highly vulnerable to effects on the emigrants themselves. Generally, however, the giving up and starting over where they are more valued. effect will be positive—otherwise they would not leave. This potential loss of an institution building middle-class

Source for Table 2: Docquier and Marfouk (2004) and authors' calculations. The Global Migration of Talent: What Does it Mean for Developing Countries? 4 Table 3. Kuwait, andGermany). are usedwheredataonnativitynot available(Japan,Iran,Jordan, Estimatesofthenumbernoncitizens of thenumberforeign-born. Country Thousands Country total population ukn ao ,2 9.7 3.0 1,108 57.9 1,124 Source: 1,303 Venezuela 1,006 4.2 Kuwait Burkina Faso 1,284 Belarus South Africa 12.6 Singapore 1,352 33.6 Malaysia 1,392 73.8 6.3 Argentina 1,419 3.8 Turkey 1,503 9.9 14.6 Netherlands 1,576 2.3 Japan 1,620 39.4 Italy 1,634 1,922 1.3 25.1 Switzerland 1,801 6.8 United Arab Emirates2.8 2,336 Jordan 1,945 25.8 Moldova 2,088 5.4 39.6 2,701 Israel 2,256 Iran 2,321 37.5 4,029 Côte d’Ivoire China Hong Kong, 3.3 5,255 18.7 Kazakhstan 3,028 18.9 6,27710.6 United Kingdom Pakistan 4,243 12.4 3.0 Australia 4,705 14.0 24.6 Saudi Arabia Canada 5,826 India 6,271 France 34,988 0.6 Ukraine 6,947 Germany 7,349 9.0 Russia 13,259 United States 9.1 Million in2000 Countries withMigrantStocksofMorethanOne a. Formostcountries,themigrantstocknumberisamidyearestimate United NationsPopulationDivision(UNPD)(2002). Where theMigrantsAre: a Percent of The thirdchannelistheroleof strongest interestinseeingthattheseinstitutionsarebuilt. in highqualityinstitutionalenvironmentsthathavethe productive individualswhoarecapableofbeingsuccessful institutions whenthemostproductiveleave.Itis the developmentproblems gular mistaketobelievethatfinancialremittancescanaddress augmenting andsmootheningconsumption,itwouldbeasin- of overseasresidentemigrantsonTRBs.These has giventhischannel well (Table 3).Recently, theupsurge infinancialremittances asthesourceoflargeemigrantcommunities while serving grant communites, host tolargeimmi- South Africaare Iran, Turkey and countries suchas some developing countries and rich andpoor resident inboth may becompoundedbythereduced connections (Table 4). four broadcategories:control, compensation,creation,and fraction ofthestock.Thepolicy optionscanbeclassifiedin sending countriesinwhichthe flowsarearelativelymodest migrants themselvesandthereceivingcountries,tothose humancapitalflowsaccruetothe advantages ofinternational Policy responsesshouldkeepinmindthatthemaximum shouldbedone? So what connections to international supplies,customersandfinanciers. connections tointernational experiences andchangedexpectations;newideas may havegreatereducationandfinancialwealth;different thaniftheyhadneverleft.They domestic economydifferently the augmented capital—financialandhuman—canaffect Finally, the and actastransnationalentrepreneurs. projects, their accumulatedwealthtoinvestinhome-country socialnetworks,use also linkdomesticresidentstointernational consequences.Emigrantscan have moresignificantlong-term for economicdynamismandinstitutionalch gin. Theflowsofideasandbusinessnetworksthatarecritical ofori- often itselftheresultofseriousproblemsincountry or Liberia,giventhattheir return return channel looks at how emigrants returning with channel looksathowemigrantsreturning to emigrate. likely those most arebuilders to beinstitution- likely most those that is dilemma The the mostsalience. source—inter of countriessuch diasporas, demand national While important for While important which is the effect which istheeffect ange arelikelyto as Albania, Haiti diasporas migration—is for improved are Control related policies seek to curb the flow of skilled compensation is paid by the rich-country governments, 5

immigrants or emigrants. Although there will be occasions rich-country employers, or the emigrants themselves. The October 2005 when the damage being done by selective recruitment (or mechanisms also differ in the agency that enforces the “poaching”) of talent is so damaging—not least the compensation. The options include: tying development aid recruitment of doctors and nurses from countries experiencing to human capital recruitment; arranging for replacement of health crises—that curbs are justified, in general we rich-country personnel; sharing payroll and income recommend against going down this policy route, for both tax revenues with poor-country providers of human capital; normative and pragmatic reasons. Rather than removing U.S.-style continuing post-emigration tax obligations to emigration options for those lucky enough to have them, it is countries of origin (the “Bhagwati tax”); conditional generally better to look for ways to make sure that everyone education grants that are repayable on emigration; shares in the spoils when those options are exercised. sharing the proceeds of visa fees or the revenues of visa auctions. All of these mechanisms face practical This leads us to consider ways of compensating the difficulties. And almost all require the cooperation of society of “those remaining behind.” There are numerous rich-country governments. But given the skill-biased trends conceivable ways that such compensation could be paid. and the development stakes, the issue of compensation These mechanisms differ in the degree to which the needs to be addressed.

Table 4. Policy Responses to Skilled Migration

Instruments Policies Rich countries Poor countries International organizations

Control Shift balance toward Curb illegal migration Promote economic development unskilled immigration Improve economic and Curb skill-poaching political stability programs unless compensation schemes are in place

Creation Avoid shortages in Higher education reforms Increase support for higher areas such as health and Liberalize skilled education education due to poor immigration human capital planning Transparent mechanisms for recognition of foreign credentials

Compensation Share social security taxes Exit tax Improve migration-related data Tie development aid Tax foreign income to skilled emigration Firms pay headhunter fees to source country

Connection Encourage circulatory Develop systems of IRAs Develop network infrastructure migration for migrants Strengthen temporary migration programs Dual citizenship 6 An important implication is that policy makers need to pay come from having much greater heed to both rich and poor country human a well-connected Countries need talent to capital creation policies that address the skill imbalances diaspora and the that are both cause and consequence of emigration. On the ensure innovation, build capital-augmented rich country side, systematic underinvestment in sectors such institutions and implement return. Receiving as health care and education (especially for public systems) programs—the key pillars of and sending coun- has led to almost permanent skill shortages, and ongoing long-term development. tries can ease “crisis” recruitment from poorer countries. When such crises barriers to travel, reoccur in sectors where poor countries are adversely affected to sending remit- by emigration, rich countries should increase their own tances, and to human capital investments, as well as invest in cheaper making investments. One way to increase the probability of human capital creation institutions in developing countries. return is to make visas temporary without the possibility of The problem is that the costs of investing in the necessary transitioning to permanent status. The idea is that young peo- manpower must often be borne years before the benefits are ple build skills, savings, and social networks while abroad, realized, leading to systematic neglect by governments and and then return to use their accumulated human, financial and individuals with short horizons and financial constraints. social capital to the benefit of the home country.

Also, rich country governments must do more to use One could argue that such policies are fair if the migrants well what human capital they do recruit. Too often when understand the terms of their visa from the outset. And professionals such indeed we think that such time-limitations are perfectly as doctors actually legitimate for short stays. But recognizing the way many Poor and rich countries arrive, they find it people quickly put down roots in their adopted countries, we alike must invest in human difficult to have think that medium- to long-term temporary and non-convert- capital creation. their credentials ible visas are not a humane policy. A better approach is to recognized. While create incentives to return as opposed to prohibitions on The Global Migration of Talent: What Does it Mean for Developing Countries? The Global Migration of Talent: all countries have staying. Such policies include making it possible to re-return an obligation to if things do not work out, making social security entitlements regulate quality in the professions, they sometimes serve as a portable, having rich-country governments facilitate the screen to protect domestic competitors, and other times the return of people with badly needed skills, putting money in result of too few resources being devoted to immigrant inte- special accounts during the migrant’s stay that can only be gration. Temporary skilled migration programs often bar accessed on return, and having origin country governments spouses from working, again unnecessarily wasting talent. provide information on opportunities at home. Poor countries—and the development community—need to place much greater attention on reforms in tertiary education, Conclusion not least because weak institutions themselves drive out the talented educators on whom successful domestic skill creation It is foolhardy to make simple judgments about the desir- depends. Needless to say this begins with a macroeconomic ability of talent flows from poor to rich countries. For small, and political environment that does not drive out talent. poor countries, the migration of a significant fraction of their And where substantial skill outflows are foreseeable, govern- best and brightest risks being harmful to those remaining ments need to “over-invest” in skill creation and implement behind. The fundamental reality is that countries need talent tuition fees and loans in public institutions that put a greater to ensure innovation, build institutions, and implement burden of the costs of those investments on the recipients of the programs—the key pillars of long-term development. education—especially if they ultimately decide to leave. With large income gaps across countries likely to persist over A less controversial policy approach is to ensure that the foreseeable future, flows of talent from poor to rich emigrants remain economically and socially connected to their countries are more likely to increase than decrease, as the rich former homes, which include policies that affect the probability country demand for talent appears more powerful than any of return. This approach seeks to maximize the benefits that possible reduction in poor country supply from the 7 October 2005 . Working (Washington, DC: Center (Washington, . , migration component. Available at , migration component. Available www.cgdev.org themselves fundamentally improve the development prospects of the development fundamentally improve themselves they are while financial remittances are helpful, And a country. human capital. not a substitute for institution-building a to understand that remittances are not Rich countries need a brain-drain. The broad elements of quid pro quo for the policy agenda are clear: more development-focused on of less skilled; greater emphasis balanced recruitment temporary incentives to return; recruitment with enhancement connections with the diaspora; and of financial and other the institutional failures that drive emphasis on remedying in place, migration the most talented out. With these elements chance of being from poor to rich countries stands a better beneficial for development. Commitment to Development Index Give Us Your Best and Brightest: The Global Hunt for Talent and Its Impact on the (Washington, DC: Center for Global Development, 2005). (Washington, Global Demographic Change: Economic Impacts and Policy Challenges Boom Towns and Ghost Countries: Geography,Agglomeration, and Population Mobility

www.cgdev.org/rankingtherich. Developing World policies, including migration, see the Paper 36 (Washington, DC: Center for Global Development, 2004). Paper 36 (Washington, for Global Development, 2006).

However, when society’s most skilled people leave at a high most when society’s However, in signal of deep and significant problems rate, it is invariably a simply such cases, tackling the brain drain would In a country. of the problem rather than its root be treating the symptoms historycauses. The checkered of foreign aid clearly illustrates the the extent that the outsiders can do. To severe limitations of what problems of a country are also are endogenous, the solutions aid, diasporas can likely to lie largely within. Like foreign facilitate (and sometimes harm) development, but they cannot by narrowing of income gaps. Given the substantial, complex, Given the substantial, of income gaps. narrowing and often negative effects on those remaining behind, the inter- community needs to pay much greater national development phenomenon. attention to this neglected Lant Pritchett, Devesh Kapur and John McHale, Nancy Birdsall, Forthcoming Titles Forthcoming

Migration is one of the Center’s core areas of policy-based research. Our work on migration concentrates on the areas of policy-based research. Our work on migration concentrates on core Migration is one of the Center’s challenges and opportunities presents to global development and poverty that large-scale migration reduction, this labor mobility across national boundaries. For related CGD materials on including issues of “brain drain” and which are available online at issue, please refer to the following works, Related CGD Works CGD Related For a complete ranking of 21 of the world’s richest countries according to the “development friendliness” of their richest countries according For a complete ranking of 21 of the world’s The Center for Global Development is an independent, non-partisan, non-profit think tank

dedicated to reducing global poverty and inequality through policy oriented research and active

engagement on development issues with the policy community and the public. A principal focus of

the Center’s work is the policies of the United States and other industrialized countries that affect

development prospects in poor countries. The Center’s research assesses the impact on poor people

of globalization and of the policies of governments and multilateral institutions. In collaboration with civil

society groups, the Center seeks to identify policy alternatives that will promote equitable growth

and participatory development in low-income and transitional economies. The Center works with other

institutions to improve public understanding in industrialized countries of the economic, political,

and strategic benefits of promoting improved living standards and governance in developing countries.

1776 Massachusetts Ave., NW Third Floor Washington, D.C. 20036 www.cgdev.org CGD Brief The Global Migration of Talent: What Does it Mean for Developing Countries?

Devesh Kapur and John McHale

October 2005