Theme: This paper analyses the mechanisms of migration in relation with the European Union’s (EU) upcoming enlargement. It shows that the most effective way to avoid an excess westbound population shift is a thorough commitment to reduce the economic inequalities between the EU’s current members and the ten newcomers.
Summary: Labour migration can be conceived of as a decision motivated by economic incentives at both the points of origin and destination. There are fears that the significant economic differences between Eastern and Western Europe might lead to a large-scale wave of emigration as ten new countries enter the EU in 2004. This analysis confirms that the fears are not unfounded. However, it also shows that large-scale emigration can be avoided if economic differences are reduced.
Analysis: The enlargement of the EU is perhaps the greatest challenge so far to the European concept of free movement, common borders and communitarian migration policies. The ten countries to enter the EU on May 1, 2004, will add a total of approximately 74 million people. The difference in income and in their general economic situation compared with the rest of Europe imply that there is a potential for increased westbound migration as a result of the enlargement. The fear is that this could have repercussions on the western economies which, at the same time, are burdened with the difficult task of jump-starting the newcomers’ economies.
A potential emigration wave from east to west is of course not only a western concern. Since the fall of the Berlin wall, the eastern countries have undergone a huge economic transformation, which has made their economies extremely vulnerable. A massive outward migration wave would most likely be a serious threat to the emerging economic stability of these countries. Labour migration, which is the form I analyse here, implies the loss of human and social capital in one region and its transfer to another. A serious deficit in human capital has far reaching effects on a country’s economic growth potential. Thus, if an emigration wave materializes, it could worsen the prospects for a smooth enlargement, with increased tension between east and west as a result. It could also cast serious doubts about the future of the European concept of common borders and the free movement of people.
It remains unclear whether the worries of a massive westbound emigration are well-founded. The experts have divergent views. The total migration potential to Germany and Austria alone has been set as high as 5 million people. Other less pessimistic predictions (or optimistic, depending on migration preferences) regarding the potential size of westbound migration are also in circulation. The truth is that no one knows what will occur. What we do know is that while some westbound migration is welcome, it would be very difficult politically to justify a massive wave of immigration over a short period of time.
Naturally, these concerns have been the object of discussions at EU summits. The result is that the membership agreements with the applicant countries impose temporary restrictions on the right of free movement of people into and between EU countries. These restrictions are to last for around six years starting from 2004. Whether this will prove to be an effective brake on the potential westbound migration is open to discussion, but at least it should provide western countries with the legal means to prevent a massive migration if and when such a possibility materializes.
Regardless of our view on the potential for east-west migration, it is important to understand the mechanism that triggers the type of labour migration that can be caused by enlargement. A better understanding of this mechanism should be of help when designing policies to counteract an exodus, contribute to a smoother integration of the applicant countries and save the European ideal of the free movement of people from becoming a hollow concept.
Enlargement and migration
To explore the future it is useful to refer to past scenarios. In 1986, the European Community (EC) welcomed Spain and Portugal into the Union. As with the ten eastern European countries that are now knocking on the EU’s door, the two southern countries had sluggish economies, with standards of living far below those of their richer neighbours. Similarly to Eastern Europe, they had a negative net migration, that is, emigration exceeded immigration. A very large share of their emigration was directed towards North-Western Europe, with Germany and France to the fore. Finally, the two countries had had a difficult past, with long spells of authoritarian rule. Similar worries regarding migration and economic feasibility as those discussed today were then voiced: Spain and Portugal were accorded no free movement of people until 1991.
To illustrate the migratory outcome of the 1986 enlargement I will turn to Spanish data on emigration and return migration to and from Europe between 1969 and 2001. After exploring the migration flows I will focus on how economic ‘push and pull’ factors came to shape Spanish migration over the period.
Figure 1 above shows the trends for emigration and return migration to and from Europe as well as net migration (emigration-return migration) of Spanish citizens over the period 1969-2001. To the left of the diagram is the period corresponding to Spain’s political ‘transition’, starting in 1975 with the death of General Franco and ending in 1978 with the first democratic elections. To the right is the period starting in 1985/86 with Spain’s entry into the EC and ending in 1991 when restrictions on the free movement of Spaniards within the EC were lifted. From Figure 1 we can conclude that Spain had a very turbulent migration history up to the start of ‘transition’. During ‘transition’, Spaniards rapidly changed their migration behaviour, leading to sharp drops in both emigration and return migration. Emigration levels started to pick up again in 1980-1985, before Spain’s entry to the EC. However, from 1986 to the present emigration has been on the decline. It is interesting to see that return migration has steadily increased since Spain’s entry into the Community. Since the late 1980s more Spanish nationals are returning than leaving the country. Before entering into the Union, however, Spain had a net outflow of people towards the rest of Europe.
Contrary to the expectations on the effects of enlargement, Spain’s entry into the EC did not result in any extraordinary excess migration to the rest of Europe. Instead, the country’s admission into the Union seems to have had a positive effect on return migration and a negative effect on emigration. The Spanish experience is most likely not a coincidence. It is likely that the increasing return migration is a result of incentives related to the country’s admission to the EC. Pinpointing such incentives, and clarifying how they operate on migration probabilities, is crucial for future policies concerning the integration of the ten applicant countries and the prevention of the undesired large-scale immigration feared by so many of Europe’s policy-makers.
Migration and economic incentives
Labour migration, or economic migration, is often shaped by so-called ‘push and pull’ factors. For example, if a person faces a high risk of unemployment in his country of origin, he may be more inclined to seek employment in a country where opportunities are more abundant. That is, the individual is pushed to migrate. Similarly, people in employment may consider emigration if the economic return is significantly better in another country. The individual is pulled. The concept of ‘push and pull’ can be summarized as follows:
Pull factors =
Push factors =
However, ‘push and pull’ factors do not operate independently. That is, migration decisions hinge on a net assessment of advantages and disadvantages in both the country of origin and the potential destination. The propensity to migrate is higher when the disadvantages exceed the advantages in the country of origin and when the advantages outweigh the disadvantages in the country of destination:
Table 1. Hypothesised migration propensities in response to ‘push and pull’ factors
|
Country of Origin |
|||
|
Advantages>Disadvantages |
Advantages |
||
|
Country of Destination |
Advantages>Disadvantages
|
–/+ |
+/+ |
| Advantages
|
–/– |
+/– |
|
|
+ = High migration propensity. – = Low migration propensity. |
|||
Furthermore, migration is not unidirectional. When and if the incentive structure changes over time, former emigrants may reassess their decision to emigrate. If the incentive structure has changed in favour of their country of origin (a shift in incentives over time from upper right to lower left in Table 1 above), return migration is more likely – and emigration less likely.
Analysis of economic change and Spanish migration behaviour
I shall now assess how economic incentives in the countries of origin and destination affected Spanish migration incentives over time. To measure the combined effects of ‘push and pull’ factors I use GDP per capita. GDP is measured in US dollars using 1995 prices and purchasing power parities (ppp). I focus on the evolution of GDP in Spain and in the nine states which were members of the EC before the southern enlargement.
Figure 2 show how Spain’s GDP has evolved relative to the average GDP of the nine other EC states. After hitting a low in its year of entry into the EC, Spain’s GDP performed better than the EC-9 average, with the exception of the early 1990s. There are good reasons to believe that this was related to Spain’s entry into the Community. Thus, according to the hypotheses explained above, the readiness to emigrate declined as the Spanish economy improved relative to the EC-9, while the attractions of returning increased.
To establish to what extent this statement is true I use simple multiple regression techniques, where the dependent variables are emigration, return migration and net migration. In order to adjust for temporal effects in the data I differentiate the measures. That is, I analyse how changes in Spanish emigration and return migration are affected by changes in GDP in Spain and the nine EC states. Furthermore, I use a dummy variable to control for the exceptional changes in migration patterns during Spain’s democratic ‘transition’. Table 2 below shows the results of the regressions.
Table 2 Regression coefficients predicting Spanish emigration, return migration and net migration
|
Model |
|||
|
(1) |
(2) |
(3) |
|
|
Dependent variable (differenced values) |
Emigration |
Return Migration |
Net Migration |
|
Change in Spain’s GDP/capita (%) |
-13.41 (9.21) |
28.64 (7.64) |
|
|
Change in EC-9’s GDP/capita (%) |
19.77 (11.14) |
-35.49 (9.24) |
|
|
Change in the raw difference between Spain’s GDP/capita and the GDP/capita of EU-9 |
45.01 (10.56) |
||
|
Spanish ‘transition’ (1975-1979) |
-13,846.98 (6,816.59) |
16,752.93 (5,654.18) |
32,355.99 (7,876.27) |
|
Constant |
-5937.54 (3,405.69) |
2,125.20 (2,824.93) |
3,474.80 (2,039.88) |
|
Observations |
31 |
31 |
31 |
|
Adjusted R-squared |
0.16 |
0.47 |
0.54 |
|
Note: a significance test is redundant since the data refer to the population; thus, the estimates reflect the true estimates for the unique event – Spain’s entry into the EC. |
|||
Model one and two are mainly included to establish evidence for the hypotheses described above and to help interpret changes in the two GDP measures explained in table 2. As we can see, the hypotheses are confirmed. For emigration, an increase in Spanish GDP per capita implies that the advantages at origin of not emigrating become greater relative to the disadvantages. The result is less emigration. Similarly, an increase in the EC-9’s GDP per capita implies that the advantages at destination of emigrating become greater than the disadvantages. The result is further emigration. The reverse effect is evident for return migration. Each US dollar increase in Spanish GDP per capita implies 28 more returnees, while each US dollar increase in the EC-9’s GDP implies 35 fewer returnees. The effect of the transition dummy confirms what we already knew from Figure 1: there is lower emigration and lower return migration as the country adapts to democracy.
Model three looks at Spain’s economic development relative to the nine EC countries and how it has affected its net migration with Europe. The results show a clear relation between Spain’s economic performance vis-à-vis the other European countries and its migration patterns. Over the period studied, for every yearly dollar decrease in the difference between the Spanish economy and the EC’s, 45 Spaniards less are to be found abroad.
Judging by model three we can conclude that Spain’s economic development is an important factor in explaining the migration behaviour of its citizens. More than 50% of the variance in migration is explained by this simple model, focusing only on economic factors but controlling for the effects of ‘transition’. We can also conclude that only relatively small changes in the differences between Spain’s GDP and the EC-9’s were necessary to make Spanish net migration become positive. Net migration reached zero once the difference in GDP declined by around 80-90 dollars from its level at the time of Spain’s admission into the EC (considering that net migration, at -3,700, was negative at the time).
The fact that only small changes were necessary was probably due to variations in future expectations about economic improvements as a result of entering the Community. Perceptions about the direction of Spain’s economy would likely have been reversed had they not been endorsed by real changes in Spain’s economic performance of the type shown in Figure 2. In the absence of such changes it is likely that Spain would have continued to be a country with net emigration.
Conclusions: Perhaps the most important lesson to be drawn from the findings reported above is the importance for Spain’s migration pattern of its economic development after admission to the EC. If another more pessimistic scenario had emerged it is not unlikely that Spain would have continued to be a net emigration country for a far longer period.
While data from a single country are not enough to predict migration flows for a group of countries, they are sufficient to make some general recommendations for Europe’s future enlargement. Our analysis shows that economic inequality was one of the most decisive explanatory factors for emigration and return migration in a country that became a member of the European Union (read EC) while it was a source of net emigration.
There are many reasons why a large-scale emigration wave would be unfortunate for both west and east. There are no real shortcuts to prevent this from happening. The only feasible way is to provide the means for a more rapid economic growth in the accession countries than in the rest of the Union, thereby eliminating the differences that encourage labour migration.
This task is likely to be more difficult for this round of enlargement than it was in 1986. First, the accession countries are many more this time and, secondly, the economic differences are greater than in 1986. For instance, the difference between Poland’s GDP per capita and the EU-15 is currently at more than 10,000 dollars, twice the difference between Spain and the nine EC countries in 1985. This is one reason why it is paramount that the six ‘honeymoon’ years, with step-by-step integration, be used as effectively as possible in order to achieve economic growth in the east. In this way, Poles, Hungarians and others can gain confidence in the direction of their countries’ economic development before the free movement of people becomes accessible to them. Failure to do this could oblige the EU to reconsider its ideals of free movement and common borders and this would, no doubt, be a major setback to European unity.
Rickard Sandell
Senior Analyst
Demography, Population and International Migration
Real Instituto Elcano