Leaving, returning: reconstructing trends in international migration with five questions in household surveys

Transcription

1 Leaving, returning: reconstructing trends in international migration with five questions in household surveys Bruno Schoumaker (UCL), Cris Beauchemin (INED) 1. Background and objectives Data to study trends of international migration flows are crucially lacking. This is especially true in developing countries but also to some extent in developed nations. Census data allow estimating bilateral stocks of migrants for many countries (Parsons et al., 2007), but they give no direct information on migration flows. Administrative statistics on immigration flows are mainly limited to developed countries, and suffer from various imperfections (Poulain et al., 2006) 1. Statistics on outmigration flows are even less frequent, and are also seriously deficient (OECD, 2008). As a consequence, in most countries, reconstructing trends in departures and return is not possible with existing data. The lack of basic information on migration is in sharp contrast with the increasing importance of migration in the policy agenda of both sending and receiving countries. Demographic surveys offer unique opportunities to collect original data on international migration. The reconstruction of internal migration trends with survey data is relatively common (Piché, Gregory et al. 1984; Beauchemin, 2011). However, measuring trends in international migration with survey data is less frequent. The Mexican Migration Project (MMP) was a pioneer in this regard (Massey, 1987; Donato, 1993), but focused on flows between Mexico and the USA. We build on the experience of the MMP to estimate trends in international migration from African countries to various destinations, as well as trends in return migrations from all destinations. The basic idea is that a few simple questions in relatively small scale surveys can be used to reconstruct migration trends with a reasonable degree of precision, and can greatly improve the knowledge of levels, trends, and patterns of international migration. The availability of socio demographic data (age, gender, education ) also allows richer descriptions of migration than possible with other data sources. By including such simple questions in existing surveys (DHS, living standard surveys ), data on international migration could be collected a relatively low cost. The objectives of this paper are threefold: (1) To describe the data and method used to reconstruct trends in first departure and returns. (2) To use data collected in Ghana, Senegal, and DR Congo as part of the MAFE project 2 to reconstruct trends in departures and returns over more than 30 years. (3) To assess the quality of the estimates by discussing the assumptions and the limitations of the data and the methods, and by comparing results of the method to other data sources. 1 Only legal migrations are recorded in migration statistics. Moreover, data published in some countries only refer to permanent migration (e.g. in the US), or exclude asylum seekers from migration statistics (e.g. in Belgium). Definitions of migration also vary across countries (e.g. 3 months in Belgium, 12 months in France). 2 MAFE : Migration between Africa and Europe (

2 2. Data The data used in this paper come from the MAFE project (Migration between Africa and Europe). The MAFE project is a multi site project on international migration. Its objectives and questionnaires were inspired by the Mexican Migration Project (Beauchemin, 2012). The objectives of the MAFE project are to measure trends and patterns of migration, causes of departures and returns, and consequences of international migration on economic and family outcomes. The MAFE project includes both household and individual data, collected in cities of three sub Saharan countries (Accra and Kumasi in Ghana, Dakar in Senegal and Kinshasa in DR Congo) and in six destination countries (Belgium, France, Italy, Netherlands, Spain, UK). The same questionnaires were used in all settings, making data entirely comparable across countries. Household surveys were conducted in sending countries (in 2008/2009) among representative samples of households (1,187 in Accra/Kumasi; 1,141 in Dakar; 1,576 in Kinshasa). Data on international migration was collected for each current member of the household, as well as on all the spouse(s) and children of the head of household, regardless of his/her place of residence at the time of the survey (in the household, elsewhere in the country, abroad), and including deceased children. The data on migration consists in five questions (see Figure 1) and includes the year of the first departure for at least 12 months to another country, the name of the country, the year of the first return, and the date of the last departure for people currently living abroad. In addition, a series of socio economic characteristics were collected for all household members and for all children of the head of household (age, gender, education). These data are used to reconstruct migration trends in these countries. Figure 1. Questions on first migration and first return from the MAFE household questionnaire Biographic data was also collected among individuals aged 25 and over in the three origin countries (non migrants and return migrants, around 1500 individuals per country) and in the six European destination countries (current migrants, around 200 migrants per destination country). Full migration

3 histories were collected, along with employment histories, family histories. These data will be used to assess the robustness of some assumptions of the method used to reconstruct migration trends with household data. 3. Methods Using household survey data, discrete time event history analysis is used to compute (retrospective) probabilities of first migration by age and period for three categories of people: head of household, partner(s)/spouse(s) of the head of household, and children of the head of household. A person year data file is constructed, starting from age 18 to the first international migration or the time of the survey. Logistic regression is used to model the conditional probability of first migration by age and year. A period indicator of migration (lifetime probability of migration) is computed from the regression coefficients (Donato, 1993). It measures the probability that an individual would do at least one international migration between if he/she experienced the period age specific probabilities of migration of that period. 5 year moving averages are used to smooth migration probabilities. A similar approach is used for first returns. Migrants enter the risk set at their first departure. A person year data file is constructed, starting from the time of the first departure to the first return or the time of the survey. Coefficients of the logistic regression (duration of migration and period) are transformed into a probability of doing at least one return within 10 years of first departure. Probabilities are also smoothed using moving averages. 4. Preliminary results Preliminary results on migration trends (departures and returns) by destination are produced with the MAFE data. Although such results are descriptive, no similar trends are available from existing data sources. Trends in departure by destinations Figures 1a to 1c show trends in first departure by destination (Africa, Europe, and, for Ghanaians, North America) reconstructed with the method just described. These trends show results that are consistent with existing fragmented knowledge on migration from these countries. For instance, migration from DR Congo to Europe increased at the end of the 1980s when political and economic troubles increased (Figure 1a). A second wave of migration started in the second part of the 1990s, at the end of the Mobutu regime. Migration from Ghana to African countries greatly decreased in the early 1980s, when Ghanaians were expelled from Nigeria (Figure 1b). Migrations to Europe and North America followed very similar trends, with a first increase in the late 1980s, and a second increase in the late 1990s. Results for Senegal (Figure 1c) show that migration to Europe has steadily increased, and that Europe has become the major destination of Senegalese migrants.

4 Figure 2. Lifetime probability of migration (18 70) from DR Congo, Ghana and Senegal, by destination. Annual estimates smoothed by 5 year moving averages. (a) DR Congo (b) Ghana (c) Senegal Trends in returns Figures 2a to 2c show trends in return migrations by destination (Africa, Europe, and, for Ghanaians, North America). These trends notably show that returns from Europe to DR Congo and to Senegal have significantly decreased since the 1980s. In contrast, returns from Europe to Ghana have fluctuated widely. In all three countries, returns from Africa are higher than returns from Europe. Figure 3. Probability of returning with 10 years of first departure from DR Congo, Ghana and Senegal, by destination of migration. Annual estimates smoothed by 5 year moving averages. (a) DR Congo (b) Ghana (c) Senegal 5. Quality of estimates The quality of the estimates will be evaluated using both MAFE data and other sources. Several tests will be done using the MAFE data. First, using the household data and taking advantage of the record of deceased children, we will test the extent of a classical bias of retrospective data, namely the fact that they are usually based on people who are still alive at the time of the survey. Second, using the biographic data that allow to take into account the whole migration history of the interviewees, we will measure to what extent trends computed only with data on first departures and returns provide a biased measure of migration trends (because of the omission of repeated migration). Third, by comparing household and individual MAFE data, we will finally test to what extent data collected through proxy respondents in household surveys are accurate regarding the data of first departure and return.

5 A second approach will consist in comparing migration trends computed from household survey data to migration trends from other sources, especially migration statistics in some destination countries (number of asylum applications, visas and residence permits issuance). In the end, all these tests will allow to discuss the opportunity to replicate (or adapt) these five questions (or only 3 questions if we restrict to departures) in other household surveys routinely carried out in developing countries, such as the DHS, LSMS or MICS. References Beauchemin C. (2011), Rural urban migration in West Africa: towards a reversal? Migration trends and economic situation in Burkina Faso and Côte d'ivoire, Population,Space and Place, 17(1), Beauchemin C. (2012). Migrations between Africa and Europe: Rationale for a Survey Design, MAFE Working Paper, 5, INED, Paris. Donato, K. (1993). Current Trends and Patterns of Female Migration: Evidence from Mexico, International Migration Review, 27, 104, OECD (2008). Perspectives des migrations internationales, OECD, Paris. Massey, D. S. (1987). The Ethnosurvey in Theory and Practice. International Migration Review, 21 Special Issue Winter, Parsons C., R. Skeldon, T. Walmsley and A. Winters (2007), «Quantifying international migration: a database of bilateral migrant stocks», Policy Research Working Paper, 4165, The World Bank, Washington D.C. Piché, V., J. Gregory, et al. (1984). "L'analyse historique des migrations: la pertinence de l'analyse longitudinale à partir des histoires rétrospectives." Genus XI(3 4): Poulain M., N. Perrin N. and A. Singleton (eds) (2006), Towards Harmonised European Statistics on International Migration, Louvain la Neuve, Presses universitaires de Louvain, 744 p.