Migration between Africa and Europe

MAFE PROJECT Policy Briefing No. 1

June 2010

Project overview

African migration has become a major concern for European policy-

makers. New policy measures are under development, but are not
always based on a good understanding of the underlying causes,
nature and consequences of African migration. The Migration
between Africa and Europe (MAFE) Project aims to overcome this
lack of understanding by collecting and analysing a new and unique
set of quantitative data. Underpinning MAFE is the recognition that
migration is not simply a one-way flow from Africa to Europe. Rather,
return migration, circular migration and transnational practices are
significant and need to be recognised in policy design.

This briefing introduces MAFE research and illustrates, using

preliminary findings, how its innovative approach and methodology
will contribute to improved knowledge and understanding to develop
sustainable migration policies.

Migration between Africa and Europe: The need for data

National and international policy-makers have been paying

increasing attention to African migration over the past decade. In
2005, the European Council adopted the Global Approach to
Migration; this was followed both by a regional dialogue, the Africa-
EU Partnership on Migration, Mobility and Employment, which builds
on the Rabat, Tripoli and Lisbon declarations, and bilateral dialogues
with ‘priority’ countries in sub-Saharan Africa. However, data upon
which to develop a good understanding of the factors, trends and
implications of Euro-African migration remain very limited. In the
case of African migration, conflicting accounts about the overall
volume of migration (over 50 million according to the African Union in
 2

2006, whereas UNDP and ILO estimate just under 20 million for
2010) within and from Africa reflect the paucity and poor quality of
data sources. At the international level, large datasets capture
measures of international migration by collating static national-level
census data, such as the Docquier–Marfouk dataset on international
migration by gender and educational attainment based on OECD’s
Database of Immigrants in OECD Countries (DIOC); the UN
Population Division’s Trends in the Total Migration Stock; and the
Global Migrant Origin Database assembled by the Development
Research Centre on Migration, Globalisation and Poverty (Migration
DRC). Whilst these are very useful, such data give poor insights into
the causes and consequences of migration. At the other end of the
spectrum, a wealth of qualitative data has been produced on
migration flows or migrant groups. But data from qualitative studies
often lack comparability and are obtained from non-representative
samples.

However, some relatively rare large-sample micro-data (data relating

to individuals) collection initiatives have generated very interesting
results, both in advancing collection and analysis methodology, and
in generating insightful findings. These include the Mexican
Migration Project (which began in 1982), now being extended with
the Latin American Migration Project (since 1998), and the Eurostat-
funded project on Push and Pull Factors of International Migration
which looked at migration from Africa to Europe (which began in
1997).

The MAFE Project builds on these advances in international micro-

data collection and analysis, and adapts them to the study of
migration between Africa and Europe.

How and what does MAFE contribute? An overview of

MAFE research areas

The first fully comparative results of the MAFE Project will become
available in 2012. Following this, in-depth research will be carried out
on the following themes and research questions:
 Patterns of migration: How selective is migration according to
age, education, gender, etc? Is the selectivity of migration dependent
on context and time? How does the migratory journey take place?
What itineraries do migrants follow? What is the extent of circular and
return migration? Do return and/or circular migrants display specific
characteristics?
 Determinants of migration: What are the key determinants of
migration for Senegalese, Congolese and Ghanaian migrants? What
factors influence return migration and the successful reintegration of
returnees?
 Migrants’ socio-economic trajectories: How do migrant skills,
investment or other transnational practices contribute to development
in their country of origin? In what ways do integration in the country of
destination or (re)integration upon return influence the developmental
impact of migration?
 Migration and families: What are the impacts of migration on
family arrangements? Are transnational families becoming more
common? What triggers families to reunify or remain transnational?
 3

Providing robust quantitative

data on migration trends and
patterns

The MAFE dataset will be the first large-scale dataset on migration

between Africa and Europe of its kind. It will compile approximately
6,000 individual (1,500 each in the Democratic Republic of Congo,
Ghana & Senegal, and 1,500 in total of Congolese, Ghanaians &
Senegalese migrants across six European destination countries) and
4,500 household records (1,500 each in DR Congo, Ghana &
Senegal). Moreover, the dataset has the potential to be extended to
include a wider range of places both in origin and destination.

FIG 1. THE MAFE PROJECT’S MIGRATORY SYSTEMS

In Africa, the samples in the current version of the MAFE dataset are
representative of migrant and non-migrant populations in major cities
(Dakar, Kinshasa and Accra), although surveys were also conducted in
the region of Kumasi in Ghana, which includes rural areas. All these
locations constitute major departure points for international migration
and are therefore particularly useful study sites.

So far, MAFE household data have been used to look at the impact of
migration on cities in the countries of origin, and in particular at migrant
investment on housing (Lessault et al. 2010). This initial research
indicates that households with migrants are generally better-off in
terms of housing conditions than those without family members abroad.
However, the direction of causality between wealth and migration is not
clear. Further investigation tends to support the hypothesis that
migrants usually hail from relatively better-off households in the first
place. Moreover, migrant investments in housing are rarely used to
directly pay for renovation or improve housing conditions. These
investments can nonetheless contribute to the improvement of housing
conditions in Dakar, notably via investments in the rental market.

At the macro level, findings will produce robust quantitative evidence of

the scale and prevalence of African migration between the two
continents, and of specific streams. Analysis of MAFE data to date
suggests that there has been a limited increase in the rate of
emigration to Europe (Flahaux et al. 2010) over the last 20 years.

This goes against the current political discourse describing rising ‘tides’
of destitute African migrants invading Europe. The final results, based
on a large quantitative dataset, will lend greater policy impact to
previous qualitative research on circular migration, and on
transnational practices in particular.
 4

Comparing different
international migration
streams

The MAFE Project was designed to collect data in several destination

countries. Two or three destination countries were selected for each of
the three migrant groups (see Fig 1): One being a former colonial power
(e.g. France for Senegalese, the UK for Ghanaians, and Belgium for
Congolese), while others were relatively ‘new’ destination countries
without colonial ties and with less historical migration. Observation of
preliminary MAFE data supports the view that colonial ties appear to be
weakening and no longer provide a suitable model to explain African
migration. Belgium is no longer the first destination for Congolese
migrants, and France is no longer seen as the destination of choice for
a significant proportion of Senegalese migrants.

Comparative analysis of the data will enable researchers to disentangle

general processes from national specificities. For instance, whereas
overall a greater share of African migrants go to other African rather
than to European countries, clear differences appear when comparing
DR Congo and Senegal. Most Congolese emigrants stay on the African
continent, whereas the Senegalese tend to go further afield to Northern
countries, in particular to Europe and North America.
FIG 2. SHARE OF MEN AND WOMEN MIGRATING TO COUNTRIES
IN THE GLOBAL NORTH

The use of standardised questionnaires for the three African migrant

groups will also allow us to identify and explore distinctive migration
strategies and selection processes. Migration to Europe appears to be
more or less selective at departure and at return depending on the
country of origin. Migrants from the DR Congo are more educated than
those from Senegal. Interesting differences are also revealed in terms
of gender: Congolese women are more likely to out-migrate than
Senegalese women, while the latter are more likely to return (Flahaux et
al. 2010).
Comparing migrants,
returnees and non-migrants
at origin and destination

Most studies and policy to date have viewed migration between Africa
and Europe as a one-way flow and essentially focused on explaining
departure only. Almost all migration research focuses either on place of
origin or of destination, whereas MAFE has collected data at both. Our
 5

research thus recognises the significance of, and explores, the full
spectrum of movements: Out-migration, return migration, and circular
migration (see section on ‘Uncovering return migration’).

Central to MAFE’s investigation is the desire to better understand the

causes and consequences of migration, which requires comparison of
the characteristics, motivations and circumstances of different groups
living in different countries, both in Africa and in Europe. An analysis of
the causes and consequences of out-migration calls for parallel data on
those who have already left (current migrants at destination and
returnees in origin countries), and on those who have never left (non-
migrants in origin countries).

Besides, by tracing migration routes back to the ‘starting point’,

researchers can see how conditions in places of origin contribute to
migration patterns. The factors and impacts of return are also explored
by contrasting data from returnees in origin countries with those of
migrants still in destination countries or non-migrants in countries of
origin.
Understanding the
factors and implications
of migration

In addition to evidence relating directly to migration, the MAFE Project

has collected in-depth biographical data on education, housing, family,
marital status, work, and social and economic histories through a
questionnaire survey conducted with individuals who are both migrants
and non-migrants. Biographical questionnaires include questions about
the family and friends of the individuals surveyed, from which clear
information about the key role of networks in migration can be derived.
This is central to developing a sophisticated understanding of the
migration decision-making process, as migration is now often seen as a
strategy taken beyond the individual level to maximize collective income,
and/or to minimise or spread risk.

Distinguishing between characteristics and outcomes for different

categories of migrants and non-migrants further allows us to explore
drivers and outcomes of migration. For example, MAFE researchers
have started investigating the relation between migration and the
capacity to invest in Senegal (Mezger and Beauchemin 2010).

Initial findings comparing Senegalese migrants and non-migrants

suggest that the migration experience does stimulate investment in
Senegal (see Fig 3). A migrant who is still living abroad or who has
returned home is twice as likely to own or acquire land or property in
Senegal as someone who has never migrated. Focusing on migrant
characteristics shows that migration might play a role in reducing
inequalities in terms of investment capacity. For example, women and
less-educated Senegalese are usually disadvantaged in their access to
assets, but this disadvantage disappears after they gain migration
experience.
 6

FIG 3. RATE OF ASSET OWNERSHIP FOR MIGRANTS IN EUROPE, RETURN

MIGRANTS AND NON-MIGRANTS IN DAKAR, 2008

Exploring the dynamics

between migration and
socioeconomic, demographic
and political changes

A further key strength of the MAFE dataset has been to collect and
collate data that stretch back over a long period of time. The surveys
have collected retrospective data (on a yearly basis) on individuals and
on members of their personal network through life histories. In addition,
MAFE researchers are assembling contextual data which introduce data
on economic, political and demographic variables over a period of time
to match the retrospective survey data collected. These data are to be
used to study the influence on or sensitivity to changes in migration data
of these ‘external’ variables. Introducing the dynamic time element is
especially useful in understanding a number of aspects of migration,
including: How migration and its outcomes change over time; how long-
lasting effects are; and the directions of cause and effect. Preliminary
research using MAFE data from DR Congo has looked at the impact of
political and economic crises on migration flows. The results suggest
that, when controlling for political conditions, deteriorating economic
conditions only increase migration to Africa (Schoumaker et al. 2009). In
contrast, migration to Europe seemed largely unaffected by economic
conditions and varied essentially in relation to the political situation in DR
Congo.
Uncovering return
migration

In the past, estimates of return migration were few and far between, and
often questionable at best. In many cases, in the context of African
migration in particular, return was assumed not to happen at all. The
phenomenon of return has thus long suffered from being statistically
invisible, which automatically translates into being invisible in institutional
and political discourses. MAFE data can contribute a more sophisticated
analysis. Beyond identifying profiles and trends, the collection of detailed
data on migrant characteristics and histories, and contextual data over
time, allows MAFE to investigate the factors and implications of return.
These retrospective data have for example enabled researchers to
construct a ‘survival estimate’ of return. This analytical tool measures the
likelihood of migrants staying in the destination country rather than
returning, in relation to the time elapsed since their arrival. Initial findings
using this tool suggest, for example, that those who migrate to another
 7

African country have a much higher probability of returning and re-

migrate more frequently, compared to those who migrate to European
countries.

Preliminary assessment of the data on Congolese and Senegalese

migrants sheds light on the prevalence of return. It shows that, after 10
years, a third of Senegalese return from Northern countries, and two-
thirds return from other African destinations. Data for DR Congo show a
similar pattern. In terms of characteristics, skilled migrants do not appear
to return more often than less-skilled ones. Also, Senegalese women are
more likely to return than Congolese female migrants. Looking at the
field of employment upon return, using data on age and occupation, it
appears that retirement is not the main motive for return. The average
age of return migrants is 45 years, and over 70 per cent of returnees
take up an economic activity, with most of these going into ‘self-
employment’.

Further analysis in Senegal suggests though that self-employment might

represent a ‘survival strategy’ for those migrants who did not prepare
their return well, or for whom return is not ‘voluntary’, or when migration
may have been too short to accumulate sufficient savings and/ or know-
how. This situation might also reflect the lack of salaried opportunities in
country-of-origin job markets for returnees to use the knowledge and
skills which they acquired via migration.

Key resources

Key MAFE Working Papers:

Flahaux, M-L., Beauchemin, C. and Schoumaker, B. (2010) ‘Partir,
revenir: tendances et facteurs des migrations africaines intra- et extra-
continentales.’ MAFE Working Paper 7. Paris: MAFE Project, INED.

Flahaux, M.L. and Mezger, C. (2010) ‘Returning to Dakar: the role of

migration experience for professional reinsertion.’ MAFE Working Paper
8. Paris: MAFE Project, INED.

Lessault, D., Beauchemin, C. and Sakho, P. (2010) ‘Migration

internationale et conditions d’habitat des ménages à Dakar.’ MAFE
Working Paper 11. Paris: MAFE Project, INED.

Mezger, C. and Beauchemin, C. (2010) ‘The role of international

migration experience for investment at home: the case of Senegal.’
MAFE Working Paper 12. Paris: MAFE Project, INED.

Schoumaker, B., Vause, S. and Mobhe, A.M. (2009) ‘Political turmoil,

economic crises, and international migration in DR Congo. Evidence
from event-history data (1975 - 2007).’ MAFE Working Paper 2. Paris:
MAFE Project, INED.

MAFE working papers and briefing papers are available online at:
<www.mafeproject.eu/publications>

Other useful readings

Black, R. and Sward, J. (2008) ‘Measuring the Migration – Development
Nexus: An Overview of Available Data.’ Migration DRC Briefing Paper
No. 13, Brighton: Development Research Centre on Migration,
Globalisation and Poverty, University of Sussex.
 8

Grillo, R. and Mazzucato, V. (2008) ‘Africa<>Europe: A Double

Engagement’. Journal of Ethnic and Migration Studies 34(2): 175-198.

Lessault, D. and Beauchemin, C. (2009) ‘Migration from sub-Saharan

Africa to Europe: still a limited trend.’ Population & Societies, No. 452.
Paris: INED.

Project identity

Coordinator Institut National d’Etudes Démographiques (France)

Cris Beauchemin

Partners Université Catholique de Louvain, Research Centre in Demography and

Societies (Belgium)
Bruno Schoumaker

Université Cheikh Anta Diop de Dakar, IPDSP (Senegal)

Papa Sakho

University of Kinshasa, Department des Sciences, de la Population et du

Dévelopment (The Democratic Republic of Congo)
José Mangalu

University of Ghana, Centre for Migration Studies (Ghana)

Peter Quartey

University Pompeu Fabra, Department of Political and Social Sciences

(Spain)
Pau Baizan

Forum Internationale ed Europeo di Ricerche sull’Immigrazione -FIERI

(Italy)
Eleonora Castagnone

University of Sussex, Sussex Centre for Migration Research (UK)

Richard Black

Maastricht University, Dept of Technology & Society Studies (the

Netherlands)
Valentina Mazzucato

Consejo Superior de Investigaciones Cientificas, Research Group on

Demographic Dynamics (Spain)
Amparo González-Ferrer

Funding Scheme Collaborative project (small and medium scale focused research project)

Duration 51 months

Budget €1,498,954

Website www.mafeproject.eu

This briefing paper was prepared by the Sussex Centre for Migration Research (UK) in collaboration
with INED (France). For more information on the MAFE Project contact Catherine Daurele at
<catherine.daurele@ined.fr>.