Lessons from the EU Eastern Enlargement: Chances And
Total Page:16
File Type:pdf, Size:1020Kb
Forum LESSONS FROM THE EU A8 migrants to the UK was about 304,000 and A8 migrants accounted for about 25 percent of all net- EASTERN ENLARGEMENT: migration to the UK during that period.2 CHANCES AND CHALLENGES Poland is the demographic giant among the A8 coun- FOR POLICY MAKERS1 tries. Hence, as expected, the large majority of the A8 migrants to the UK (around 69 percent) were Po- lish citizens. In fact, between the year ending Decem- CARLOS VARGAS-SILVA* ber 2003 and the year ending September 2010 the Polish-born population of the UK increased from Introduction 75,000 to 521,000. On 1 May 2004 eight Eastern European countries Migration in the UK has always been a controversial (Czech Republic, Estonia, Hungary, Latvia, Lithuania, political issue and a large majority of the British pub- Poland, Slovakia and Slovenia) joined the European lic has been opposed to more immigration since at Union (EU). These countries as a group are common- least the 1960s (Blinder 2011). But the surge in immi- ly known as the Accession 8 or simply A8 countries. gration after the accession of the Eastern European As members of the EU, citizens of the A8 have the countries led to migration becoming one of the main right of mobility within the EU system. However, the political issues during the 2010 general election and accession agreements allowed the 15 pre-existing EU to David Cameron becoming prime minister, in part, member states to impose restrictions on the employ- by promising to control immigration and reducing ment of citizens from the A8 countries for a maximum net-migration from “hundreds of thousands” to “tens of seven years. of thousands”. In contrast to Cyprus and Malta (two other countries The restrictions for movement of A8 workers to other that joined the EU in 2004), the A8 countries had per EU countries terminated on 1 May 2011, including capita incomes well below EU levels. As a result, any transitional arrangements such as the UK’s WRS. there was speculation of large migration movements At the time of ending the restrictions, only two coun- after accession from the eight new member states to tries, Germany and Austria, had important restric- the old member states. It was not surprising, there- tions in place. Nowadays, A8 workers have the same fore, that most existing members chose to impose re- rights to mobility as any of the citizens of the 15 pre- strictions on the movement of these workers. Only existing EU member states. Therefore, this is an ideal Ireland, Sweden and the UK opened their labour time to reflect on the overall experience of enlarge- markets to workers from the A8 countries immedi- ment in regards to migration and identify potential ately upon EU enlargement. lessons for the future. This article draws mainly from the UK experience after the EU enlargement to in- In the UK, A8 workers have been able to take up dentify several lessons from this process. This is ever employment freely and legally since May 2004 as more relevant because Romania and Bulgaria, two long as they registered with the Worker Registration countries with relatively low per capita incomes Scheme (WRS), a relatively simple procedure. The joined the EU in 2007, and restrictions for the move- opening of the UK labour market to workers from ment of workers from these countries to all EU coun- these countries led to a surge of immigration. During tries should be lifted by 2014 at the latest. Moreover, 2004–09, net-migration (inflow minus outflows) of 2 All figures in this article are based on estimations from data of the UK Office for National Statistics (ONS). The main sources of data 1 Thanks to Martin Ruhs and Agnieszka Kubal for useful comments are the Labour Force Survey (LFS), the Annual Population Survey on this article. (APS) and the Long-Term International Migration (LTIM) esti- * COMPAS, University of Oxford. mates from the International Passenger Survey (IPS). 9 CESifo DICE Report 4/2011 Forum there is already vast speculation about the immigra- namics of previous EU accessions. The previous en- tion impact of a possible accession of Turkey to the largement (1 January 1995) saw the accession of Aus- EU (Home Affairs Committee 2011). tria, Finland and Sweden to the EU (i.e., fourth en- largement) and resulted in no significant migration movement to the other member states. However, Decisions under uncertainty: the need for caution these countries had per capita income levels similar and gradualism to those of the existing EU members and enjoyed freedom of movement with the EU as members of Before the entrance of the A8 countries to the EU, the European Economic Area (EEA). In order to there was a high level of uncertainty on the future find a somewhat better comparison it was necessary level of immigration from these countries after ac- to look at the third enlargement of the EU that cession. The UK government had to make a decision involved the accession of Spain and Portugal to the about whether to grant A8 workers full rights in the EU (1 January 1986). Given the considerable existing labour market using very limited information. What income per capita (and unemployment) differentials little information was available came from forecast- between these two countries and the EU at the time, ing (i.e. econometric) exercises predicting the future transitional arrangements were agreed in order to level of migration from the A8 to the UK and a re- minimise the possibility of mass immigration imme- view of the experiences of previous EU accessions. diately after accession. Yet, even after the end of the transitional period there were no major immigration The major econometric analysis commissioned by movements from these two countries to the rest of the UK government (e.g., Dustmann et al. 2003) sug- the EU, in contrast to the predictions of some acade- gested that flows were going to be much smaller than mic work at the time (e.g., Straubhaar 1984). As a in reality, in the order of 5,000 to 13,000 net-immi- result, the experience from previous accessions sug- grants per year and suggested that “even in the worst gested low levels of migration after accession. case scenario, migration to the UK as a result of east- ern enlargement of the EU is not likely to be overly The problem is that even in the case of Spain and large”. Part of the problem with the projections was Portugal, income differentials were small compared that due to a lack of historical data on migration to the income differentials between the EU and the from A8 countries to the UK, the projections for A8 countries. Moreover, contrary to the A8 countries, post-enlargement immigration were based on a mo- Spain and Portugal had well developed market eco- del whose parameters had to be estimated using data nomies at the time of accession. As a result, the expe- from other countries.The Cold War had severely lim- riences from the previous EU enlargements provided ited immigration from the A8 countries to Western a poor guide to possible immigration patterns after Europe for decades. As a result the model parame- the entrance of the A8 countries to the EU. ters assumed invariance across time and across coun- tries, a very strong assumption. The authors suggest- The UK government was, therefore, forced to make ed that the estimations should be evaluated with a decision on A8 immigration restrictions under con- caution given the methodological caveats. However, siderable uncertainty, due to a lack of a pre-acces- these technical limitations do not always translate sion history of immigration movements and compa- clearly into the political discussion. rable precedents and limited information on the in- tentions of other key countries. Under situations of It was also not completely clear at the time what other uncertainty, it is reasonable for decision makers to countries were going to do in order to restrict their adopt a cautious approach, especially, if it is possible labour markets to A8 workers. A scenario where to gradually adjust the strategy later on. some key labour markets, such as Germany, where open to A8 workers was likely to result in minimal For instance, the UK could have adopted a relative- immigration from the A8 to the UK (i.e., diversion ly restrictive approach with regards to A8 immigra- effect). Meanwhile, a scenario in which just a few tion and then liberalise the restrictions gradually. countries opened their labour markets to A8 workers The transitional arrangements of the accession trea- could result in major immigration movements. ty of 2004 facilitated this process as the restrictions were divided in periods of two, three and two years. The large-scale immigration that resulted from the For the first two years countries could impose their A8 entrance to the EU also contrasted with the dy- national laws and policies to A8 immigration. These CESifo DICE Report 4/2011 10 Forum restrictions could then be extended for three more evidence also suggests that the geographical disper- years. Finally, those countries whose labour markets sion of A8 migrants is very high in comparison to were affected by A8 immigration could impose re- other migrants that tend to concentrate in London strictions for two additional years. and other urban centres (Sumption and Somerville 2010). In fact, it seems that in proportion to the size Instead of following the cautious approach of impos- of the local labour force these workers are concen- ing serious restrictions for at least the first two years, trated in rural areas (Commission for Rural Com- the UK decided to open labour markets to A8 work- munities 2007).