Internal and International Migration
The discussion surrounding how countries in sub-Saharan Africa can reap the benefits of a demographic dividend needs to pay closer attention to the effects of internal and international migration flows. Surprisingly little attention has been paid to the ways in which migration alters the relative size of the working age population, the ratio of children to the total population, and the education level of a population, all of which are key variables in the discussion of the demographic dividend. This chapter aims to address this gap in the literature by drawing on two new datasets on internal and international migration flows as well as an innovative visualization of migration systems. The data suggests that, during the period 1990–2010, the spatial patterns of migration flows in sub-Saharan Africa were distinctive from those observed in the more developed world. Economic factors seem to play a less important role in shaping migration compared to violent conflict and political unrest. The results point to three clusters of countries between which migration flows are concentrated, underlining the importance of the regionalization of migration within sub-Saharan Africa.
The surge of African migrants and asylum seekers attempting the risky voyage across the Mediterranean in 2014 and early 2015 has brought international migration to the front pages of newspapers, fuelling public anxiety about a looming European migration crisis caused by large uncontrolled inflows from sub-Saharan Africa. Future population growth is often associated with an imminent increase in emigration from the African continent to the Western world. How the European Union should respond to the rising influx and whether to relax or restrict immigration are hotly debated issues, but, unfortunately, political rhetoric and populist media dominate at the expense of fact-based discussions. In the countries of origin in sub-Saharan Africa, debates tend to focus on ways of reducing the emigration of skilled Africans to Western countries and on how to combat human trafficking and smuggling. Although the scientific community is well-equipped to inform public and political debates on migration, little attention has been paid to establishing a comprehensive picture of migrationflows within and away from sub-Saharan Africa. Research on migration around the globe and variations across countries has been hindered by a lack of comparable statistics on internal and international migration flows. The dearth of understanding regarding where people move in the world has inhibited research on the links between the demographic transition and migration.. Hence, the existing literature on the demographic dividend has been rather cautious about grappling with the complexities of migration.
This chapter summarizes the available evidence on the patterns and trends in internal and international migration in sub-Saharan Africa over the period 1990–2010. We take advantage of two newly available datasets on internal and international migration around the globe, allowing us to compare these migrationflows across the region and place them in a global context (Abel and Sander 2014; Bell et al. 2014). The forces which shape migration are summarized in the next section, followed by a discussion of the available data on internal and international movements. Thereafter, there are two sections which provide analyses of spatial patterns of migration in sub-Saharan Africa in a global context. Firstly, key trends occurring in global international migration flows are presented together with relationships between within-sub-Saharan African movements and levels of economic development. Secondly, patterns of internal migration are examined for selected countries in the region, followed by a discussion of links between internal and international migration. By way of conclusion, we provide an outlook on the future of migration in sub-Saharan Africa and its potential impact on future population change in the coming decades.
2 Why People Move
Migration as a means by which people aim to improve their financial and social wellbeing is not a new phenomenon. In fact, the migratory paths of our ancestors can be traced as far back as 200,000 years1. The reasons why people move are often complex and typically involve a mixture of motives related to income, family, housing, and education. The determinants of migration have traditionally been studied with reference to Lee’s push-pull migration model (Lee 1966). Based largely on the determinants of migration that have been observed in the developed world, the push-pull model suggests that uneven processes of development are at the core of factors that explain migrationflows between regions as well as between countries. According to Lee (1966), wage differentials are a key determinant of migration, although people living in the poorest countries are usually not the most likely to migrate due to a lack of resources necessary to realize a move. The notion that (economic) opportunities at the destination and deficiencies at the origin trigger migration is admittedly a rather simplified view of the migration decision-making process, as it overlooks a number of non-economic factors that play an important role in shaping migration in sub-Saharan Africa, most notably violent conflict, environmental change, and the demographic transition.
Violent conflict is associated with vast population displacements out of conflict zones, often into neighboring countries, and thus causes a shock to the migration system that is extremely difficult to foresee. The United Nations High Commissioner for Refugees (UNHCR) estimated that by the end of 2013 there were 51.2 million people ‘of concern’, which included 33.3 million internally displaced persons, 16.7 million refugees, and 1.2 million asylum seekers2. As reliable data on the spatial patterns and trends in internally displaced persons is extremely sparse, the focus of this chapter is on internal and international movements of legal migrants and refugees, although internal displacement in many conflict situations is far more common than migration captured by national statistics. Nevertheless, the analyses presented here highlight the refugee movements triggered by the numerous conflicts seen in Central and East Africa in the Republic of the Congo, Rwanda, Burundi, Uganda, South Sudan, and Sudan since 1990. In addition, the economic collapse in Zimbabwe caused substantial emigration to South Africa. Outside Africa, the violent conflicts in Iraq and Afghanistan are also noteworthy.
The ways in which environmental change shapes migration are complex and closely linked with socio-economic, cultural,, and political factors determining migration decisions. For example, drought-stricken populations may be constrained in their mobility due to the associated lack of capital (e.g., livestock, harvest), which is necessary to realize a move further afield. The combination of factors affecting migration makes it extremely difficult to identify the causal relationship between environmental change and human mobility. Nevertheless, a general pattern has emerged of short-distance migration in response to slow-onset environmental change being more common than longer distance movement triggered by sudden environmental disasters, such as tsunamis and hurricanes (Stojanov 2008).
The level and direction of migration is also shaped by demographic factors, most prominently the size of a population, its growth rate, and age structure (e.g., Preston et al. 1989). The combination of strong population growth and a young age structure releases an abundant labor supply. Regional and cross-national variation in labor supply and labor demand often act as drivers of both domestic and international labor migration. As current patterns of labor migration from South to West Asia demonstrate, movements tend to originate in countries experiencing strong population growth and target destinations with rising labor demand. Will future population growth in sub-Saharan Africa result in equally large movements towards the oil-rich Gulf countries and traditional destination countries in the developed world?
Predicting the likely future trajectory of migrationflows to, from, and within sub-Saharan Africa is difficult given the scarcity of data on migration and the dearth of research on the topic. It is unlikely that the trajectory of sub-Saharan Africa will closely follow that of many Asian countries, given the profound differences in socio-demographic characteristics between populations, most notably rates of literacy and educational attainment. Hence, it is unclear whether a growing young adult population that exerts pressure on domestic education systems and labor markets will trigger substantial migration flows from countries south of the Sahara to more developed, rapidly aging countries.
3 Migration and the Demographic Dividend
When considering the role of the demographic dividend in shaping migration trends, one may speculate that a growing labor force may be associated with higher volumes of movement within sub-Saharan Africa, rather than increased emigration to more developed countries. In this scenario, regions or countries that have more effective policies in place to ensure the adequate provision of services and incentives for economic growth are successful in attracting labor migrants from neighboring countries with less favorable conditions. A key question underlying this scenario that has not yet been adequately addressed in the literature is whether the trend of migration streams going up the income ladder, which is observed across the more developed world, also holds in sub-Saharan Africa. Provided that income differentials trigger migration, developing countries might be able to reap fiscal and economic benefits from their sizable young adult populations and perhaps even transition from a migrant sending to a receiving nation by placing emphasis on solid political institutions, expansion in education, and strong economic growth. This chapter aims to shed some light on the effects of cross-national variation of gross national income on migration trends in sub-Saharan Africa.
The demographic dividend not only shapes migration trends, but in turn it is also shaped by migration. Important in this context is the selectivity of migration by age, with young adults being the most mobile age group. The dividend begins with change in the population structure, which is commonly attributed to falling fertility. At the same time, the relative size of the working age population and the number of children per adult of working age can also be influenced by migration among younger adults. In the oil-rich Gulf countries, for example, immigration has led to an increase in the relative size of the working age population, whereas in Nepal the working age population has shrunk substantially due to emigration. At the sub-national level, similar effects are caused by migrationflows up the urban hierarchy. Around the globe, internal migration tends to result in a growing labor force in cities and a shrinking labor force in rural areas The movements of young adults, especially if they leave their children behind, can significantly alter the demographic prerequisites needed to capture a demographic dividend.
Besides the tendency to select younger adults, migration is also selective of more highly educated individuals. Rising education levels within a country’s population are commonly attributed to rising investments in children, as falling fertility lowers the number of children each couple has to support through education. But the second demographic dividend, which results in the higher economic productivity of a working age population, can also be shaped directly by high-skilled immigration. Highly-educated people tend to exhibit a greater propensity to move up the urban hierarchy towards metropolitan centers rich in job opportunities, and are more likely to move over long distances than their less educated peers. Therefore, highly-skilled migration has the potential to affect not only the first demographic dividend through alterations in national and regional population age structures, but also the second demographic dividend through shifts in the education level of populations.
Migration makes it much more difficult to establish how accelerated growth in a country’s economy can be achieved. Besides strategic investments in health, education, and good governance, managing migration effectively through adequate policy development will be crucial for countries in sub-Saharan Africa. Such policies should facilitate both the flow of unskilled labor between neighboring countries, and circular migration of skilled labor between origin countries and destinations in North America, Europe, and Australia. International circular migration programs can be an effective means for reducing a labor surplus while at the same time reducing the risk of losing human capital to brain drain (Hugo 2013). However, the science-based development of circular migration policies has long been hindered by a lack of comparable data on migrationflows and migrant characteristics at regional and global scales, a topic to which this chapter now turns.
4 Data on Migration in Sub-Saharan Africa
In contrast to other components of demographic change, comparative statistics on migration have long been absent from national and international statistical collections (Bell et al. 2014). This is rapidly changing, with the emergence of new estimates of international migrationstocks (UNPD2013) and more recently, estimates of international migrationflows between 196 countries (Abel and Sander 2014). The IMAGE project (Comparing Internal Migration Around the Globe) has sought to advance comparative statistics on internal migration by conducting an inventory of data collection practices and developing a suite of methods and metrics for the purpose of cross-national comparison (see e.g., Bell et al. 2014, 2015). Notwithstanding, there remain severe impediments to cross-country comparisons due to differences in data collection instruments, the types of migration data collected, and the spatial and temporal framework employed. This is more marked for the analysis of internal migration than international moves, which have benefitted from significant recent advances in data collection and estimation procedures (see for example Abel and Sander 2014).
Globally, three main instruments are conventionally used to collect migration data: population and housing censuses; population registers and administrative data sets; and surveys. Population censuses are the most common source of data, with 142 of 193 UN member states collecting some form of internal migration data in the 2000 Census Round (Bell et al. 2014). The number of countries collecting data on the size of their immigrant population is broadly similar. Data from population registers and administrative sources is used in at least 50 countries around the globe, however, these collections are largely limited to Europe and Asia. Surveys are much more common, but vary widely with respect to sample size and coverage, severely limiting their utility for migration analysis. Compared with other parts of the world, data on migration in sub-Saharan Africa is relatively scant. Of the 48 UN member states, 28 collected internal migration data via a census in the 2000 Census Round, 36 collected internal migration data via a survey, while no countries employed a population register or administrative data set (Bell et al. 2015).
The questions and criteria used to capture migration within censuses, registers, and surveys vary widely around the globe, further complicating regional or global comparisons. The most common form of migration data collected by population censuses is place of birth, with 26 African countries collecting these data in the 2000 Census Round. This question provides information on the number (or stock) of internal migrants currently living outside their region of birth, as well as the size of a country’s immigrant population. Lifetime data is usually collected on a relatively coarse spatial scale, limiting its utility for the analysis of spatial patterns. A more serious limitation of this data is the lack of information regarding the timing of moves, making it difficult to identify changes in migrationintensity and patterns over time. Census questions on recent migration do allow temporal shifts in the intensity and pattern of migration to be more readily explored. These questions take a number of forms. Most common in sub-Saharan Africa are questions asking respondents about their place of previous residence at some time in the past, commonly one or 5 years ago (16 countries in SSA). Data on the duration of current residence is also frequently collected (15 countries) and, when coupled with information on place of previous residence (11 countries), can be used to explore internal migration patterns and estimate recent immigrant stock. Despite the growing availability of census data, differences in the spatial and temporal frames used to collect migration data severely constrain cross-national comparisons and the quantification of international movements around the globe. This is further impacted by a lack of data availability, with collection not always guaranteeing the dissemination of migration data. For this study, census data is used to explore the spatial patterns of internal migration for a small sample of countries.
Surveys present an alternative to census data for countries in sub-Saharan Africa. The Demographic and Health Survey, conducted by USAID, has collected data on duration of current residence across a number of survey waves. While this data provides next to no information on the spatial pattern of moves and is limited to women aged 15–49, its strength lies in the standard approach to collecting data on internal migration, capturing a broadly comparable measure of internal migrationintensity to be calculated with a sample of 35 countries. Together, this data provides some limited insight into the intensity of internal migration across the region. To explore the intensity and pattern of international migration in the region, we draw on bilateral flow estimates developed by Abel and Sander (2014).
5 Spatial Patterns of International Migration
This section draws on new estimates of bilateral migrationflows covering the period 1990–2010 (Abel and Sander 2014). To quantify international migration flows, an indirect estimation methodology was developed to determine the number of movements required to meet changes over time observed in migrant stock data published by the United Nations. The estimates capture the number of migrants who changed their country of residence over 5-year periods, thereby omitting most seasonal and circular types of movements. The estimates include refugee movements that were registered by the UNHCR but fail to take into account undocumented migration, nor do they cover recent migrant streams triggered by the violent conflicts in Syria, Libya, and Yemen.
The visualization shows that most migrants move over short distances within the same region or between neighboring regions, and relatively few move between continents. North America, Europe, and the oil-rich Gulf countries in Western Asia are the destinations of flows that come from furthest afield (most of which go through the centre of the circular graphic). Sub-Saharan Africa recorded movements of about 10 million migrants. The largest flow (3.7 million) was within the region, whereas 2.2 million people moved to Europe and North America. The flow within sub-Saharan Africa was therefore larger in volume than the stream from Central to North America (3.2 million), but smaller than the flow from South to Western Asia (4.9 million).
At the global scale, the intensity of international migration was estimated to be stable at about 0.6% of world population moving over 5-year periods since 1995. In 1990–1995, the violent conflicts in Eastern Africa and Afghanistan, as well as the fall of the Iron Curtain, triggered stronger movements (0.7%). The system-wide intensity of international migration within sub-Saharan Africa was above the global average in 1990–1995 (7.7 million moves, or 1.5% of the population) but declined to 3.7 million (or 0.5% of the population) in 2005–2010. The total net loss of population through migration to countries outside the region was 1.8 million (or 0.24% of population) in 2005–2010.
In summary, the spatial patterns of international migration in sub-Saharan Africa are different from those observed in other world regions. Migration occurs over relatively short distances, chiefly between neighboring countries, and appears to be triggered largely by non-economic factors that, in the case of violent conflict and political instability, are very difficult to predict. Although migration does not resemble the patterns observed elsewhere, its impact on national population size and age structure has been more subtle than in many developed countries, with the exception of refugee movements in Eastern Africa. The findings suggest that international migration may only play a minor role in alleviating or intensifying labor market imbalances created by a large cohort of young adults.
6 Internal Migration in Sub-Saharan Africa
The existing literature on the demographic dividend has paid little attention to within-country variations, although it is well established that cities tend to lead the demographic transition compared to poorer rural areas (Williamson 2013). This neglect is somewhat surprising, given the key role that internal migration plays in shaping the distribution of population within countries. Evidence suggests that there is significant heterogeneity in the intensity, spatial pattern, and impact of internal migration around the globe. Results from the IMAGE project which draw on census data indicate significant regional variability in internal migration intensities across Africa, with pockets of both high and low mobility (Bell et al. 2015). The geographic coverage of this analysis is relatively limited, due to a lack of detailed census data.
In summary, there is significant cross-national variation in the intensity of internal migration across sub-Saharan Africa, with pockets of high internal migration echoing clusters of countries with similar patterns of international migration, suggesting an association between these types of mobility. The limited evidence presented on the spatial distribution of migrationflows across the urban hierarchy, confirms the importance of rural-to-urban flows in many sub-Saharan African countries. These are accompanied by strong inter-urban and urban-rural exchanges, suggesting a diverse range of spatial strategies and individual motives for migration that seem to go well beyond economic factors.
7 Linking Internal and International Migration
International and internal migration are typically viewed as two separate demographic components, although they are clearly intertwined at the individual level, because migrant trajectories often involve a sequence of internal and international moves. Conventional wisdom suggests that internal migration from rural to urban areas acts as a stepping stone for international migration; and that large cities serve as gateways for international immigrants, causing the displacement and replacement of the host population. Neither of these important links has been rigorously studied (King and Skeldon 2010), largely because of the dearth of longitudinal migration data that allows migrants to be traced across national borders. Hence, existing evidence on the relationship between internal and international migration is mostly based on a small number of national surveys.
The spatial representation of the typology of countries shown in the map in Fig. 6, however, echoes the three clusters of countries described earlier and underlines the importance of the regionalization of migration within sub-Saharan Africa. The West African group includes a diverse set of countries with regards to their pairings of migration intensities and income levels. The East African group is dominated by countries with negative international net migration and high intensities of internal migration, whereas the Southern African group includes the only country with high international migration gains, high internal migration intensities, and higher income levels (i.e. South Africa). These findings, however, should be interpreted with caution, given the limited data quality discussed earlier.
This chapter has provided a summary of the existing evidence on patterns and trends in internal and international migration in sub-Saharan Africa. Our aim has been to fill the gap in the literature created by the absence of adequate studies which examine the relationship between migration and countries’ ability to reap the benefits of the demographic dividend. A key obstacle to a better understanding of this relationship is the limited availability of reliable migration data. While the United Nations has made some progress in promoting a unified standard for collecting statistics on migration, national legal frameworks and the divergent interests of nation states limit the degree of alterations which national statistical institutes are willing to implement so as to improve cross-national consistency. In response to the limitations of secondary data sources, the scientific community has focused on the development of harmonization and estimation methodologies as well as the exploitation of new alternative data sources, such as mobile phone data (Abel and Sander 2014; Raymer et al. 2013; Wesolowski et al. 2013; Bell et al. 2015). The collection of adequate data and the promotion of open access to census and register data should be a top priority in sub-Saharan Africa and beyond.
Projecting the likely future trajectory of migration in sub-Saharan Africa and its impact on population size and age structure is inherently difficult, especially in the context of African migration being largely triggered by shocks to the system, such as violent conflict and political regime changes. These difficulties are reflected in the United Nations opting for the simple but unrealistic assumption of net-international migration rates converging to zero by the year 2050. Even fewer studies have been devoted towards projecting sub-national populations within countries south of the Sahara, owing partly to the lack of adequate data. Another source of uncertainty in population projections for sub-Saharan Africa is the speed of expansion of education. A commensurate expansion in education and the growth of a more skilled workforce could lead to higher levels of internal migration as individuals seek to maximize their economic potential, but, unless domestic job markets are sufficiently attractive, it could also result in an increase in emigration from Africa to the more developed world.
- Hugo, G. (2013). What we know about circular migration and enhanced mobility. Migration Policy Institute 7.Google Scholar
- Rees, P., & Kupiszewski, M. (1999). Internal migration and regional population dynamics in Europe: A synthesis. Population Studies No.32. Strasbourg: Council of Europe Publishing.Google Scholar
- Rees, P., Kupiszewski, M., Kupiszewska, D., Bell, M., Charles-Edwards, E., Stillwell, J., Daras, K. (2015). Impact of internal migration on population redistribution: A cross-national comparison. Paper presented at the the eighth international conference on population geographies, Brisbane, Australia.Google Scholar
- Stojanov, R. (2008). Environmental factors of migration. In Development, environment and migration. Analysis of linkages and consequences. Olomouc: Palacky University.Google Scholar
- UNPD. (2013). Trends in international migrant stock: Migrants by destination and origin, the 2013 revision. New York: United Nations, Department of Economic and Social Affairs, Population Division.Google Scholar