At a Brussels joint lunchtime meeting with the European Centre for Advanced Research in Economics on 24 March, Klaus F Zimmermann presented results of his recent research on prospects policies for European migration. Zimmermann is Professor of Economics and Director of the Seminar for Labor and Population Economics at the Universität München and Co-Director of CEPR's Human Resources programme. His remarks were based on his CEPR Discussion Paper No. 641, `Towards a European Migration Policy', written jointly with Thomas Straubhaar. Financial support for the research presented in this paper and for the meeting were provided by the Commission of the European Communities under its SPES programme. The views expressed by Professor Zimmermann were his own, however, not those of the Commission of the European Communities nor of CEPR, which takes no institutional policy positions.

Zimmermann noted that several EC member countries have had substantial experience of immigration since World War II, and it has often been beneficial in economic terms. Current concerns over East-West migration are exaggerated: the size of the potential migration is unclear, while a controlled inflow of migrants is desirable to offset population decline and labour force ageing in Western Europe. The immense political pressures against immigration may induce policy-makers to attach greater weight to political rather than economic factors, however, while the current debate confuses the political issue of asylum with the economic issue of labour migration. It is migrants' inability to apply for permission to work legally that is currently driving the abuse of asylum rights in some EC member countries, which is further complicated by the Community's failure to harmonize its national asylum laws.

Zimmermann noted that the current debate focuses mainly on migration from Eastern Europe. Projections of East-West permanent or temporary migration during 1991-2000 range from serious academic estimates of some 5-15 million (some 3-4% of Eastern Europe's current population) to speculative estimates by newspapers and politicians of some 20-40 million. Austria, Benelux, France, Germany and Italy are the most likely recipient countries, to which these 5-40 million migrants would represent an average annual inflow of some 0.3-2.4% of their 1990 population (comparable to the US peak figure of 1.16% in 1900- 10). The Community also faces major migration pressure from Turkey, whose population is expected to grow from 34.9 million in 1990 to some 53 million in 2010.

Zimmermann stressed that labour migration can be welfare improving so long as the marginal productivity of labour differs across countries. Simulation studies indicate substantial welfare gains from free labour mobility; as a rule of thumb, a country should be open to foreigners whose marginal productivity in that country is higher than their marginal costs of adaptation. Despite large economic differentials within Western Europe, intra-EC migration has never been large and is not expected to increase substantially with the completion of the single-market programme. While regional policy may save considerable adjustment costs, migration from non-EC countries can also increase the speed of adjustment and avoid adjustment costs to natives.

The threat of immigration usually increases with the substitutability of foreign for domestic workers, since increased immigration leads to a decline in native wages or increased native unemployment. Immigrants are often complements to native workers in production, however, and can therefore raise native productivity and wages. They also provide different goods in the service sector, and they generate demands for goods and services produced by natives, with a consequent multiplier effect. Empirical studies indicate that most immigration is beneficial to receiving countries. Most such findings are largely based on experience in the US, however, where migrants tend to have higher rates of labour force participation, self-employment and savings and greater motivation than natives, so they must be applied to Europe with caution.

Zimmermann argued that these substantial potential gains support a policy of openness towards labour migration, in which labour markets should determine the size of migrant inflows. Active migration policy is desirable, however, for several reasons. Social frictions associated with migration require an active integration policy, and its benefits can only be reaped if adjustment costs are kept under control. Receiving countries that bear substantial costs of immigration should also be compensated by entrance fees and migration taxes: together with country and qualification quotas, these can control and smooth immigration flows. Free labour and product markets within the Community imply that independent national immigration policies may harm other member states: migration into one country can affect its partners' labour market equilibria if immigrants' inability to move freely throughout the Community causes native migration instead.

Zimmermann concluded that the Community should adopt a common migration policy, which should ideally be embedded in a world- wide General Agreement on Migration Policy. Analogous to the GATT, with a remit to liberalize international migration flows in order to equalize factor prices, this could establish an international framework for rules of entry, migration taxes, treatment of foreign labour, quotas, remittances, social rights, pension transfers, etc. in order to promote an internationally efficient reallocation of labour. Even without such an agreement, the Community should establish a system of country quotas for both sending and receiving countries, based on the medium-term needs of EC labour markets. Zimmermann proposed an initial maximum of 1 million visas per year (including family members). Applicants for such work-permits should automatically forgo all asylum rights, and vice versa, to bring the current abuse of such rights to an end.