1. INTERNATIONAL MIGRATION DATA as input for population projections Anne HERM and Michel POULAIN Estonian Interuniversity Population Research Centre, Estonia Université catholique de Louvain, Belgium

2. CONTENTS International migration data needed for national population projection Availability of data Reliability and international comparability of data that are key problems with observed international migration data. Caution : this presentation takes a international perspective and some conclusions do not necessarily apply when considering population projections in a specific country.

3. WHICH DATA ARE NEEDED ? Immigration and emigration flows versus net migration? How long time series are needed? Age and sex – which level of disaggregation? Citizenship or origin and destination of migrants?

4. Immigration and emigration flows versus net migration There are no 'net migrants'; there are, rather, people who are arriving at places and leaving them. Why they are doing so is central to understanding the dynamics of... growth and decline (Morrison 1977). In order to understand the possible future developments in the international migration, the data on both immigration and emigration flows should be analysed distinctly and not just net migration.

5. PRO’s and CON’s NET MIGRATIONMIGRATION FLOWS Can be more easily available (census results, calculated as difference of population and natural change) + - Ad hoc data collection system is needed More reliable even if flows are under or over counted with the same error or if the two censuses are fully comparable. ( + ) - Difficult to ensure the complete coverage of all groups of migrants More volatile and more difficult to predict as it is a difference of two flows - + Trends can be identified even if fluctuation takes often place Rates cannot be properly calculated (no at risk population) - + Target and at risk population groups can be identified No models are available for age distribution of net migrants - + Age distribution of immigrants or emigrants can be modelled Such data are not possible to collect but can only be obtained through calculation - + Data can be collected by statistical or administrative systems

6. TIME SERIES How long should be time series in order to identify a trend that can be extrapolated? A too short time series (e.g. less than 10 years) reflects only temporary fluctuations and global trends may be hidden. A very long time series (e.g. 30 years) can not be useful because it would include too many historical changes influencing trends that will not probably not occur anymore in future.

7. ICELAND total immigration, total emigration and net migration (Source: Iceland Statistics and Eurostat database)

8. TIME SERIES Between 10 – 15 year most appropriate while only few years is clearly not enough and longer than 20 years will not neseccesary give additional value or help to do more clear assumtion for future trend.

9. AGE and SEX Disaggregation by sex is compulsory in all demographic data collection and in population projection. Usually there is no problem to find the data needed. Disaggregation by age is also compulsory but even is the projection is done by single year of age up to 110+, such detailed level of disaggregation is not recommended for international migration flows.

10. Example: Immigration of foreign EU citizens in Germany, 2006 (Eurostat database)

11. Single-year age or larger age groups? Most appropriate for migration flows: 5-years groups up to 80+ these data are usually more reliable and less volatile. if needed, single year age distribution up to 110+ can be calculated on this base by using migration schedule models.

12. ‘CITIZENSHIP’ versus ‘ORIGIN and DESTINATION’ In addition to age and sex, which other characteristics of migrants should be considered as essential? Total numbers of immigrants and emigrants are not sufficient for predicting future trends for international migration to be included in population projections. Group of country of citizenship or country of origin and destination are needed.

13. ‘CITIZENSHIP’ versus ‘ORIGIN and DESTINATION’ Country of citizenship is the most policy-relevant characteristic of international migration. There are important differences in conditions for immigrating or emigrating depending the citizenship of migrants. The most important difference is observed between migrants with free or with restricted movement rights.

14. Recommended groups of citizenships NATIONALS FOREIGNERS of which : EU or EEA Non-EU or non-EEA + most important partner countries if needed (e.g. neighbouring countries)

15. Lithuania : immigration by groups of citizenship

16. The Netherlands: immigration by groups of citizenship

17. Immigration to Germany in 2006, EU (left) and non-EU (right) foreigners, males and females

18. Other characteristics of migrants or groups of countries Country of birth could be an interesting characteristic of migrant to be considered in population projection but country of citizenship is more relevant for policy support. Country of previous or next residence does not bring any additional information except that such data could be useful for checking the reliability of data and when considering intra-EEA flows (under conditions of free movement)

19. AVAILABILITY OF DATA The first obstacle for considering international migration flows in population projection is the lack of data. For the projection of international migration flows and their inclusion in population projection, we need relatively long time series of observed migration flows characterised by age, sex and group of country of citizenship of migrants.

20. AVAILABILITY OF IMMIGRATION DATA

21. RELIABILITY of (available) DATA For the projection of international migration flows, we need observations characterised by : Correct level Correct trend Such achievements should be reached separately for immigration and emigration flows considering that different reasons of errors may exist.

22. TOWARDS RELIABLE LEVEL Are all categories of migrants considered (included or excluded)? Asylum seekers Students Illegal migrants Are the internationally recommended definitions used? Not only permanent migrants Time criterion for the duration of stay or absence (one year is recommended but other choices coexist – ad hoc working group of UNECE-EUROSTAT). Strength of the data collection system based on administrative rules and considering advantages of registration for individuals and local communities.

23. How to estimate the real LEVEL of an international migration flow ? By sharing and analyzing data collected by sending and receiving countries (UNECE working group on the use of data on immigrations produced by receiving countries – DATA SHARING). By using MIMOSA adjusted factors (developed within a EUROSTAT research project but only for intra-EU migration flows). By considering census results on net migration between successive censuses or data obtained through the question on the country of residence 1 year prior census.

24. TOWARDS RELIABLE TRENDS Break in time series that do not reflect real changes in behaviour Changes in law or administrative rules with an impact on administrative and statistical data that is either temporary or could stay longer. Changes in the method for statistical data collection. Application of EU Regulation on migration statistics.

25. Adoption of new law or change in administrative rules (Sweden 1998 – 2009)

26. Adoption of new law or change in administrative rules (Italy 1998 – 2005)

27. Change in production of statistical data (1) Spain added new migrant groups in total immigration flow in 2004 YEARS National immigrations Foreign immigrations with well known country of origin Foreign immigrations with unknown country of origin Total number of immigrations as available in official statistics 2 000 31,587 330,881 Not included 362,468 2 001 20,724 394,048 Not included 414,772 2 002 40,175 443,085 Not included 483,260 2 003 40,486 429,524 Not included 470,010 2 004 38,717 520,152 125,692 684,561 2 005 36,573 578,736 103,975 719,284 2 006 37,873 679,840 123,131 840,844

28. YEARS Nationals emigrations Foreign emigrations with well known country of destination Foreign emigrations with unknown country of destination Foreign emigrations with unknown country of destination due to the ‘expiry procedure’ Total number of emigrations as available in official statistics 2 002 29,674 6,931 Not included 36,605 2 003 15,990 9,969 Not included 25,959 2 004 13,156 13,589 28,347 Not included 55,092 2 005 19,290 17,756 30,965 Not included 68,011 2 006 22,042 23,223 40,429 56,602 142,296 Change in production of statistical data (2) Spain added new groups in total emigration flow in 2004 and 2006

29. On-going improvements linked to the implementation of the EU Regulation : immigration in Poland (2002-2008)

30. On-going improvements linked to the implementation of the EU Regulation : immigration in Estonia (2004-2008)

31. On-going improvements linked to the implementation of the EU Regulation : change of definition of immigrant in Denmark (2008)

32. CONCLUSIONS Time series of observed data on international immigration and emigration flows exist. In most countries details can be found by sex, age groups and group of citizenship. Considering separately groups of citizenship is essential to predict future trends. Investigations should be carried out to identify real levels and real trends.