This contribution argues that the EU should open up to skilled immigrants through a points system via a European "Blue Card" granting access to its entire labour market. This European version of the Green Card could become a powerful complement to any national effort to attract top talent. In addition, students graduating with a Masters degree or equivalent from European universities or from top universities abroad should be automatically eligible for a Blue Card. This "Blue Diploma" would help attract young talent early. Finally, in future rounds of EU enlargement, higher-skilled workers should be welcome immediately, provided they reach an earnings threshold : the "External Minimum Wage".

To motivate the discussion, a tentative explanation why countries like Germany and France are lagging behind in the global competition for talent is provided. In Section 2, the basic facts of migration, its skill content, and the increasing supply of skills worldwide are examined. In Section 3 the basic efficiency and distribution arguments for and against high and low-skilled migration are analysed. The impact of emigration - “brawn drain” and “brain drain” - on developing source countries is also discussed. Finally, Section 4 proposes potential policy options for Europe.

One explanation for this might be the popular but flawed idea that the capacity for absorbing immigrants is essentially fixed, say at 100 000 per year, and the only question is : How should these immigration slots be allocated ? If the number of slots were fixed, a moral case could be made to give the available slots to the most deserving : those suffering from political persecution, from abject poverty, from family separation. And an economic case could be made that the slots should be given those who would benefit the local economy the most : highly skilled immigrants. Thus, there would be a head-on conflict between what is morally right and what is in the national interest.

But reality is different. The number of migrants that a country is willing to absorb is not fixed. In particular, the immigration of high-skilled workers need not reduce the capacity to absorb less skilled and perhaps morally more deserving migration. This is illustrated in Table 1. Canada, Switzerland, and Australia, the countries with a points system for immigration, have a strong bias in favour of highly skilled immigrants. But this does not appear to have come at the expense of less skilled immigrants. Foreign born without tertiary education make up only 8 percent of the population in France, the Netherlands and Belgium, and 11 percent of the population in Germany. By contrast, in the three countries with a points system, the foreign born without tertiary education amount to 12 percent of the population in Canada, 14 percent of the population in Australia, and 17 percent of the population in Switzerland.

Table 1
International comparison of the extent and skill composition of migration

Thus, decisions on high-skilled immigration and on low-skilled immigration can to a large extent be treated separately. But should they be ? Economic effects of high-skilled immigration are generally positive for the receiving country while low-skilled migration has more ambiguous effects, as shown further below. Hence, decisions about attracting more high-skilled immigration will tend to be relatively easy. By contrast, decisions regarding low-skilled migration often prove highly complex and controversial for both economic and non-economic reasons. By bundling the discussion of high and low-skilled migration together, many European countries, including France and Germany, are loosing valuable time in the global competition for talent.

Therefore, there would appear to be a strong case to treat the issues of high-skilled and low-skilled migration separately. This is reinforced by the potentially benign effect that successful high-skilled immigration policies might have on the politics of low-skilled immigration. Greater numbers of immigration success stories in the economic and scientific arena could help to reduce any exaggerated anxieties associated with immigration and immigrants overall.

In order to attract more highly skilled migrants, this article proposes the introduction of a “Blue Card”, a European version of the US Green Card that would provide highly skilled third country nationals with instant access to the entire European labour market. The Blue Card would be allocated on the basis of skill through a Europe wide points system. Such an EU wide system would be more attractive than any national system from the perspective of high-skilled immigrants. Also, a European solution would provide greater visibility, predictability, and transparency than 25 different national systems.

Chart 1
Recent increase in immigration to the EU and US

Over the coming years, migration rates and migration pressures might well increase further. Globalisation is rapidly “shrinking” the world without shrinking worldwide income differences quite as fast. There are concerns in Europe over this influx of immigrants, and low-skilled immigrants in particular. At the same time, to become a competitive and dynamic knowledge-driven economy as spelled out in the Lisbon agenda, Europe will need to become much better at attracting talent from the rest of the world. The European Commission has accordingly become active in this area (Box 1).

However, progress has been slow. Some of the reasons for this have already been mentioned in the previous section. In addition, many relevant stakeholders still use problematic economic concepts to discuss migration, most importantly the “lump-of-labour” fallacy according to which the number of jobs in an economy is fixed. This policy brief argues that the issue of economic migration should instead be framed in terms of the skill level of immigrants.

The differences in both the extent and the skill composition of migration, among developed countries, are striking. In particular, the percentage of highly-skilled foreign-born in the entire population varies widely, as highlighted by Table 1.

Australia, Canada, and Switzerland have been phenomenally successful in attracting large numbers of migrants with a strong bias towards high-skilled immigration. All three countries have a points-based system for attracting high-skilled immigrants (Box 2).

In mid-range immigration countries, the picture is much more varied. Germany, for example, has attracted disproportionate numbers of low-skilled immigrants, consistent with its historically large guest worker program. In other European countries, migration from former colonies, often in the aftermath of independence, has played a more important role. In particular, it is worth pointing out that Spain, a country with a relatively low stock of immigrants, is presently experiencing a massive surge in immigration, not least from South America almost 200 years after most countries of that continent gained independence.

However, it seems that English-speaking countries generally have a distinct advantage in attracting high-skilled migrants, because English is the most widespread second language in the world. This also makes universities in English speaking countries more attractive internationally. This English language advantage sometimes leads other countries to conclude that they shouldn’t even try to engage in the global competition for talent. How about trying even harder instead ?

Contrary to popular perception, the US does not seem to have this high-skilled bias in its migration. The reason is that the high skill bias of migration to the US from many countries of the world cancels out with the low skill migration bias of the large influx of immigrants from Mexico.

But what if many more countries followed the examples of Canada, Switzerland, and Australia ? Would those other countries simply be competing for the same scarce international supply of skilled labour ? Chart 2 helps to dispel this concern. Over the last 15 years, the number of students in tertiary education has increased dramatically.

Chart 2
Number of students in tertiary education

Today, the share of students in the population is in fact lower in the old EU member states (EU-15) than in the new member states (EU-10) or the wider EU neighbourhood (EU Neighbourhood Policy Countries + Russia). Turkey is also catching up rapidly.

Even more striking is the increasing supply of skill among the Pop-10 [1], the 10 most populous economies outside the US and Europe. Over the last 15 years, the Pop-10 have collectively increased their numbers of students by 156%. As a result, they now have more students than the enlarged EU and the US combined.

Overall, tertiary education rates have been converging much faster than incomes. As a result, no shortage of young and high-skilled migrants is to be expected any time soon. If Europe wants to welcome more high-skilled migrants, it can.

But most people would prefer to stay at home if it wasn’t for the money. Therefore, why not upgrade productivity where the people currently are instead of having people chase more productive jobs abroad ? International trade and cross-border movement of capital are helping to do just that. According to the classic factor price equalisation theorem of trade theory, wages might in principle be equalised internationally through the trade of goods alone ! However, there are important reasons why migration pressures are likely to persist even under free trade, full mobility of capital, and flexible labour markets domestically.

First, many poor countries suffer from an inferior “production function” because of poor institutions. Despite recent development success stories, upgrading poor institutions is a slow process. In the meantime, workers in many developing countries will continue to suffer from inferior wages. Migration can short-circuit this development problem by allowing workers move to locations with a better “production function” immediately.

Second, agglomeration effects are an important rationale for migration. For example, France and the UK are large countries with fairly uniform institutions, free trade and free movement of capital. Nevertheless, workers continue to migrate to extremely expensive and crowded places like London or Paris. The reason is that people become more productive by virtue of geographic concentration. By moving to a large agglomeration, often in a foreign country, they can also hope to greatly improve the match between their skill and their job, thus boosting their productivity.

In summary, important factors driving migration today include good institutions and agglomeration effects. In both areas, Europe is well positioned. In view of these important efficiency arguments [2] for migration, why is free migration such a remote prospect ? Besides non-economic factors, distributional concerns are the main reason.

The basic argument is easily understood : Almost by definition, the migrant himself or herself derives benefits from a higher wage abroad. In the host country, wages of workers with labour market characteristics similar to those of the migrant can be expected to marginally decline while the income of those with different skills increases. Beneficiaries typically include people with different skills and owners of capital and land. In the source country, the wage impact will be a mirror image : the wage prospects of workers similar to the emigrant are set to improve while those with complementary factors of production would tend to suffer somewhat.

On that basis, low-skilled immigrants will tend to increase income inequality among the native population in the host country as the already below average wages of low-skilled natives will come under additional pressure. By analogous arguments, high-skilled migration has a benign distributional impact in the host country and an adverse distributional impact in the source country. Ultimately, these opposing effects between different skill groups and different countries are likely to be at the core of any economic controversy over migration.

However, while this theoretical argument is simple and compelling, it has been surprisingly difficult to find convincing empirical evidence to support it. In their analysis of empirical surveys, Longhi et al. (2005, 2006) find only a minute “consensus estimate” of the distributional impact : a one per cent increase in immigration only leads to a 0.12% decline in wages within the relevant skill segment and a 0.024% decline in employment.

If this were true, it would be wonderful news. Essentially, one could stop worrying about the distributional implications of migration altogether. However, as Borjas (2003) has pointed out, most of the empirical studies that fail to find a significant distributional impact of migration focus on the impact of immigration on wages in small geographic areas. But such an approach fails to control for the endogeneity of migration. Migrants tend to be attracted to locations that have the most vibrant local economies and therefore typically the most attractive wages. Hence, any negative wage impact of immigration might be hidden by above-average wages in areas that manage to attract the largest numbers of migrants.

By applying an econometric approach immune to this particular concern, Borjas (2003) and Aydemir and Borjas (2006) obtain substantially higher estimates of the wage impact of migration for the US, Canada, and Mexico. According to these studies, immigration of 1% reduces wages at the respective skill level by between 0.3 and 0.4% and migration could explain up to one third of the increase in the wage gap between low-skilled and high-skilled wages in the US over recent decades. Furthermore, Borjas finds in a simulation that any efficiency gains may well be tiny compared to these adverse distribution effects.

But those findings are unlikely to mark the end of the empirical debate. Bonin (2005) applies Borjas’ methodology to German data and finds much smaller effects. Also, questions remain as to why the wage effects of classic natural experiments like the Miami Boatlift and the mass emigration from the Commonwealth of Independent States (CIS) countries have not been more marked. Finally, Ottaviano and Peri (2006) find significant complementarity of native and foreign workers within the same skill group and they argue that only the least skilled group of natives in the US are likely to experience a negative wage impact due to migration.

In summary, the empirical literature has not been able to comprehensively dispel the distributional concerns that come with low-skilled migration. Therefore, such concerns cannot be entirely dismissed at this stage. At the same time, it seems already clear that any adverse distributional effects of low-skilled migration have not been the most important factor in rising income inequality over recent years. Hence, attempts to lay much of the blame for the economic stagnation of low-skilled natives in recent years on immigrants are not only politically but also factually misguided.

Assuming a welfare function that is inequality averse, the efficiency and distributional findings can now be brought together. High-skilled immigration is likely to increase welfare among the host country population since both efficiency and equity are likely to be improved. By contrast, low-skilled immigration has an ambiguous welfare effect in the host country. It increases efficiency from the perspective of the native population provided that there is sufficient adjustment in the capital stock while it is probably somewhat widening the income gap between rich and poor in the host country.

The welfare argument in favour of high-skilled immigration and the welfare ambiguity of low-skilled immigration are reinforced by a number of additional aspects that have so far been neglected :

Chart 3
Unemployment in Europe by skill-level

For the source country, low-skilled emigration or “brawn drain” typically improves welfare as it improves both efficiency and redistribution. This positive impact of low-skilled emigration is reinforced by remittances. Low-skilled emigrants will often help to support poor relatives in the source country with their higher earnings abroad.

By contrast, the welfare impact of high-skilled emigration or “brain drain” is ambiguous. The source country may suffer from an adverse efficiency and distributional impact as a result of the brain drain. There will be fiscal loss since high-skilled emigrants will no longer pay taxes in their home country. And just as high-skilled migrants help to uplift their ethnic communities abroad, they could have made notable contributions to public life had they stayed at home.

But a brain drain is not all bad for the source country. The option to emigrate may substantially increase the expected returns to education, thereby improving education incentive. Finally, if migrants return to their country of origin, and many of them do, the skills and savings they have acquired abroad become a powerful force of development. Therefore, moderate levels of brain drain may actually be beneficial for the source country as, for example, argued in Beine et al. (2003).

The findings of the previous sections are summarised in Figure 1 : high-skilled migration tends to improve the welfare of the host country while the welfare impact of migration on the source country is ambiguous. By contrast, low-skilled migration has an ambiguous welfare impact on the host country while generally improving welfare of the source country. Hence, there need not be a conflict of interest between source and host country but there may well be. This raises the question how could the positions of host and source country be reconciled, if indeed there were a conflict ?

Figure 1
Theoretical migration preferences

Unlike trade, migration will generally require international compensating transfers instead of purely national ones if efficiency gains are in part to be used to compensate the losers. In particular, it will typically be necessary for a rich host country to share some of the gains from high-skilled migration, including the gains from the immigrants, with the poor source country. Perhaps the most prominent proposal in this respect is the Bhagwati tax, a special income tax on the high-skilled immigrant the proceeds of which are to be transferred back to the source country in compensation for the migratory move.

This tax may be difficult to implement in practice because of international cooperation, and some forms of tax discrimination based on the place of birth may not be constitutional under all circumstances, but Europe could certainly go a long way towards gaining the moral high ground simply by fulfilling last year’s commitments to significantly increase development aid as a percentage of GDP. More specifically, the EU could make a point of subsidising education systems as an increasing function of the net inflow from any particular source country. Alternatively, rich countries could allocate more work permits for low-skilled workers, since a mixed strategy between high-skilled and low-skilled migration could also overcome the potential conflict of interest depicted in Figure 1. Additional possibilities are discussed in the next section.

Perhaps the most obvious definition of skill would be based on formal qualifications. This also makes some economic sense since formal qualifications tend to be a fairly good predictor of future earnings potential. This is relevant since most migrants move in their 20s or early 30s when their current earnings are only a relatively poor predictor of future earnings potential.

However, from an economic perspective, a definition of skill ought to go beyond formal qualifications. As long as an immigrant is likely to achieve high earnings in the host country, he could be argued to have sufficiently rare talent so as to be regarded as highly skilled. This is the salary-based definition of skill. It is a highly flexible definition, capturing all kinds of professional excellence, including sports such as football and creative professions, which are difficult to standardise.

This raises the question whether there is any room for European involvement regarding high-skilled migration. High-skilled migrants could give rise to positive cross-border externalities within the EU similar to research and development. In principle, this argument might even be used to justify subsidies for attracting high-skilled migrants, perhaps in the form of a centrally financed Erasmus style programme to attract third country nationals.

However, before going down the route of explicit subsidies, the attractiveness of Europe could be increased for free by providing third country nationals immediate access to the entire EU labour market. This will be more valuable from the perspective of the migrant than access to any national labour market due to the option value of the additional markets. Also, such an EU wide immigration regime would provide much greater visibility, predictability, and transparency than 25 different national systems.

 

Therefore, it is recommended to introduce a “Blue Card”, a European Green Card that would provide highly skilled third country nationals with instant access to the entire European labour market. This Blue Card would be allocated on the basis of skill through a Europe wide points system. Overall, such a system should make it significantly easier for Europe to compete for top talent with countries like the US or Canada.

However, national initiatives to attract more high-skilled immigrants need not wait until a European Blue Card is introduced. The reason is that for most EU member states is would appear to be in their narrow self-interest to attract greater numbers of high-skilled migrants. EU level coordination would merely help countries to get there faster, and to go even further.

Sometimes, the concern is voiced that EU level coordination of high-skilled migration would make it difficult for countries to take needs into account that are specific to any member state in particular. However, it is not obvious what the economic basis for this concern might be. If the rationale for EU level coordination rests indeed on the argument of a positive externality of high-skilled migrants, any coordination of high-skilled migration policies would typically lead to more not less high-skilled migration than countries would be willing to accommodate individually. And since it would be subject to unanimous decision making, all member states are set to benefit either directly or indirectly as part of a larger package.

Even from a development perspective, the Blue Diploma could turn out to be beneficial. By providing guaranteed access to the European labour market without requiring a permanent presence, circular migration in the spirit of the proposal by Weil (2006) would be encouraged. The Blue Card would in effect act as an insurance policy for graduates from developing countries in case they would like to take the risk of going back home. They could always return to Europe for a second chance.

As with immigration from third countries, the old member states will typically be more readily persuaded to open up their labour markets for high-skilled than for low-skilled workers from new member states. Furthermore, due to the strictly limited duration of the transition process to full labour mobility, the dangers of abuse are less pronounced than they would otherwise be. Hence, a simple salary-based approach can be used to introduce full mobility for high-skilled workers while delaying access for low-skilled workers.

Citizens of new EU member states would be allowed to enter work contracts that pay above an external minimum wage in all old member states. This wage floor could initially be set at the median wage in each old member state. With time, the external minimum wage could be lowered at a jointly agreed minimum pace to reach full free mobility. Of course, any old member state would be allowed to open its labour market faster, or even lift all restrictions immediately, if it chose to do so.

Germany and other countries who have not yet introduced full mobility from the 2002 round of enlargement would be well advised to apply this approach immediately to attract high-skilled workers from the present new member states in Eastern Europe. Germany could introduce an external minimum wage of €30,000 per year for citizens of the new member states in Eastern Europe. For young workers, this threshold could even be set somewhat lower, at €24,000. As a result, the low wage sector in Germany would continue to be protected for the time being while Germany could start enjoying the benefits of skilled migration immediately. Over the next 3 to 5 years, this external minimum wage could then be progressively lowered in order to assure a smooth transition to full worker mobility in 2009 or 2011.

While Europe would clearly benefit by attracting more talent from abroad, continued low-skilled migration will not necessarily be harmful. In this brief, it has merely been argued that the complex issue of low-skilled migration should not hold us back in finding a better approach to skilled migration.

By the same token, the introduction of better integration policies is a no-brainer that must not be delayed by the continuing complex discussion of how restrictive or liberal migration policies for low-skilled migrants should be.