Internal migration is ‘an important route out of poverty in places where conventional development efforts have had limited success’ (IOM, 2005). To what extent do you agree with this statement? Excerpt
The International Organisation for Migration (IOM) (2005, p.23) maintains that ‘internal migration is turning out to be an important route out of poverty in places where conventional development efforts have had limited success’. This is a reflection of the growing recognition in the development literature that internal migration can be a conduit out of poverty for vulnerable and rural households in low-income countries (IOM 2005). However, in this paper it is counter-argued that the relationship between migration and poverty is a function of contextual specificity and is neither linear nor simple.
First, it cannot be assumed that all poor people have the wherewithal to move. Empirical research suggests that the poorest of the poor or chronically poor people are less likely to be able to move due to spatial remoteness, high costs of moving and ‘risk related to foregone domestic product’ (Skeldon 2003, p.1). The ability to adopt migration as a livelihood strategy is dependent on interactions between institutions of the household, community, state and market (Moore 2001), as well as access to and control over resources or capital (Kothari 2002). Secondly, when the poor do migrate, it is not clear whether they are able to enhance their livelihoods and whether outcomes are influenced by the initial deprivation levels of the migrant (Wheeler, Sabates and Castaldo 2007). These deliberations are underpinned by the idea that there are no a priori factors which cause people to migrate and which lead to a particular set of outcomes (Kothari 2002; Amin 1995).
One may thus argue that internal migration is not always development-induced, as the IOM asserts. This argument is premised on the notion that it is difficult to delineate between the causes and consequences of migration (Amin 1995) because migration is both a cause and consequence of poverty (Skeldon 2003; Kothari 2002). Significant is the idea articulated by Amin (1995, p.32) that ‘migrations are not only the consequences of an unequal development, which could in itself be the result of “natural” causes such as the different natural potential of different regions. Migration is also in itself a part of the unequal development, as it serves to reproduce the conditions that aggravate these’. This suggests that people may migrate out of poverty in order to improve their livelihoods and/or migrate into more vulnerable conditions and concomitantly become further impoverished through the act of moving (Kothari 2002; Skeldon 2003).
The dialectics between migration and poverty have spawned polemical debates in the development discourse. In some quarters, internal migration is associated with adverse or little distributional effects as measured by urban demographics and conventional economic indices of poverty and inequality (Stark and Yitzhaki 1982). For example, Nord (1998) found that the migration patterns of the poor reinforce pre-existing spatial concentrations of poverty. De Haan and Rogaly (2002) on the other hand, highlight the contextual specificity of migration–poverty interactions, while Kothari (2002) and Skeldon (2003) illustrate that migration can both alleviate and perpetuate poverty.
These ‘mixed reviews’, according to Sabates-Wheeler, Sabates and Castaldo (2007, p. 308), are rooted in the complexity of drawing empirical distinctions between cause and effect. This paper further suggests that because poverty is a multidimensional construct, the outcome of migration on poverty is a function of how the latter is measured and conceptualised. Thus the crux of this paper is to explore the manner and extent to which migration contributes to ‘a process whereby the longer term prospects for moving out of poverty are better, or worse, than before’ (Kothari 2002, p.10). In this regard, this paper posits two overarching questions: first, under what conditions can migrating be a successful livelihood strategy by which the poor can improve their living conditions and move out of deprivation? Secondly, to what extent and in what ways does poverty shape the decision to migrate and what are the consequences of movement?
To address its empirical enquiries, the paper will focus primarily on the specificities of internal migration in Ghana. There is a strong policy orientation in the country in support of what Grant and Nijman (2004) have termed a regional push or government bias towards spatially concentrated growth regions, under the assumption that those regions will propel the larger national economy to higher levels of development. A corollary of such policies in Ghana has been the exacerbation of unequal development, poverty levels and spatial differentiation between the country’s North and South regions. These differences have engendered linkages between and even within the two regions, including migration, remittance flows and trade (Boakye-Yiadom 2008). For these reasons, Ghana presents an interesting case study in relation to the research objectives of this paper. While the paper acknowledges the multidimensional quality of poverty, the impacts of migration on poverty are analysed at the meso-level in terms of monetary remittances due to data constraints. In Ghana, urban settlements are defined as a population threshold of 5,000. All localities with a population size of less than 5,000 are designated as rural.
Migration constitutes a significant if not dominant role in the livelihood strategies of the poor (Kothari 2002). In much of the literature, poverty and unemployment are believed to fuel large scale population movements from deprived areas to those which are perceived to offer better economic prospects (Kothari 2002; see Todaro 1976). For example, it is assumed by academics and policymakers that internal migration usually entails flows from poor rural areas to richer urban areas (Tsegai and Le 2011). This is Bakewell’s (2007) idea of a sedentary bias whereby migration is perceived as ‘development-induced’ (McDowell and de Haan 1997) or symptomatic of unequal development — as the IOM suggests in its assertions. In the case of Ghana, such assumptions are insufficient to explain population movements as studies illustrate that the propensity to move is shaped by different factors which are both internal and exogenous to the household.
Data from the Ghana Living Standards Survey (GLSS) 5 illustrate that there is less migration in the North than in the southern regions of the country (GLSS 2008, p.50). Non-migrants, defined as ‘those born at their current place of residence who have never stayed away for a year or more’, made up over half the populations of the Northern, Eastern and Upper West regions at 58.5 per cent, 69.6 per cent and 65.2 per cent, respectively, in 2005/06 (GLSS 2008, p.50). Migrants in the Upper East and Northern regions made up just three per cent of their populations, while the Upper West had a somewhat higher migration rate of eight per cent (Ackah and Medvedev 2010, p.5). The South, particularly Accra, had the largest proportion of in-migrants with approximately four in every ten residents of the region in this category (GLSS 2008, p.50). By contrast, the level of in-migration is generally low in the North, particularly in the Upper East (6.7 per cent) and Upper West (6.1 per cent) (GLSS 2008, p.50). This suggests that regions with higher rates of in-migration have lower levels of poverty, and vice versa.
Over two-thirds of internal migrants in Ghana come from the relatively better off southern regions–Ashanti, Central, Eastern and greater Accra (Ackah and Medvedev 2010). The evidence contradicts common perceptions that most migrants originate in the North; this area accounts for only ten per cent of internal migration in the country. Though out-migration from the North is low, migrants from the more urbanized regions mostly originate in rural areas (Ackah and Medvedev 2010). Thus, the GLSS data denote that the broad dynamic entails spatial movements from poorer to richer areas (McKay et. al 2011).
Rural–rural migration, on the other hand, is primarily undertaken by poorer, less educated groups with fewer assets. Moving from one place to another entails economic and social costs and requires a certain level of capital (particularly social and economic) which may preclude some (Kothari 2002). Still, Caldwell (1968a) notes that rural–rural migration bears many similarities to rural–urban patterns in that spatial movements are generally towards better off regions. Drawing from this proposition, rural–rural flows are not antithetical to rural–urban movements (ibid.). Barrett et al. (2001) have shown that Ghanaian men travel (seasonally and temporarily) from the North to the more fertile cocoa growing areas (Ashanti, Brong Ahafo, and northern parts of the Volta region) as labourers seeking employment during the cocoa season. The ‘period of inactivity’ in the North coincides with the time of peak agricultural demand in those regions so the men can work on a cyclical basis and return home for the single growing season (Anarfi et al. 2003; Abdulai 1999).
The GLSS does not encompass the various differentials in rural out-migration. Data are not disaggregated into levels of poverty and therefore do not indicate whether those who migrate are the poor within the poorer regions (McKay et al. 2011, 2). Broadly, it is difficult to determine whether the propensity to move is nuanced by initial deprivation levels of the migrant. McKay et al. (2011) have, however, done so by drawing linkages between population quintiles and levels of out-migration (ibid.). They found significant levels of internal migration in all quintiles but slightly lower levels in the poorest fifth of the Ghanaian population (ibid.). These findings, they suggest, are indicative that the poorest of the poor cannot afford to move (ibid., p.3). Caldwell (1968b) records similar observations. That a cross-section of the population are excluded or cannot migrate to certain areas due to poverty raises questions about the possibility of increased spatial inequality and concentrations of poverty as a result of out-migration by better off groups.
Not only is inequality a possible outcome of migration, it may also be a primary determinant (Waddington 2003). At the community level, Ackah and Medvedev (2010) found that communities with higher levels of literacy, higher rates of subsidized medical care and better access to water and sanitation are less likely to produce out-migration while the lack of access to these services increases the likelihood of out-migration (ibid.). Ewusi (1986) showed the same findings. Beals, Levy and Moses (1967, p.486) found negative effects on migration in communities with low incomes, but the income of individual households has a positive effect on the probability of migrating. Similar findings have been documented by Caldwell (1968b). However, despite the inequalities and lack of development concomitant with the rural areas, Caldwell (1968a) demonstrated that some people do not want to move because of a preference for village life, satisfaction of farming, the imperative to care for the old and infirm and laziness or a lack of desire to move. Hence, while migration remains central to many households, it is a response to the conditions of poverty for some (Kothari 2002) but not all.
Men are more likely to migrate than women, which Bukh (1979) attributes to the ‘breadwinner’ concept (Bukh 1979) whereby men are culturally expected to seek the proverbial bread for their wife and children (Dugbazah 2007, p.78). Also younger, better educated persons between the ages of 15 and 25 (Caldwell 1968b, p.368) are more likely to migrate to urban areas (Tutu 1995; Caldwell 1968b; Ackah and Medvedev 2010; Boakye Yiadom 2008). According to Ackah and Medvedev (2010, p.10) these are usually ‘the best of the worst’ within disadvantaged communities. Family size also plays a significant role in out-migration–the propensity to migrate increases exponentially with family size and the probability of an individual moving to a town increases with the total number of siblings of either sex (Caldwell 1968b). There is also an inverse relationship between the propensity to migrate and the distance from the area of destination (Caldwell 1968b; Ackah and Medvedev 2010). Hence, it is clear that there are different motivators influencing the decision to migrate that are internal and external to the household (Kothari 2002).
‘Perhaps the critical issue in the migration and poverty equation is whether remittances can help to alleviate poverty’ (Skeldon 2003, p.11). In this regard, Dugbazah (2007) has argued that migrant remittances can be a valuable source of income and livelihood for poor households in Ghana; they can serve as an avenue for risk diversification and, concomitantly, alter local household income distribution. Indeed, a substantial portion of the literature on Ghana has focused on the impact of internal migration on household income/consumption and welfare gaps between rural and urban communities (e.g. Tutu 1995; McKay et al. 2011). It is estimated that annual internal remittances in Ghana are approximately US$324 million. However, Skeldon (2003) has cautioned that more important than the idea or amount of income are the uses to which monies are put and the probable impact of remittances on communities of origin. In this case, the idea of investment as opposed to consumption becomes another yardstick by which the migration–poverty nexus is measured.
A significant number of econometric studies have sought to establish a correlation between migration and household welfare (cf. Ackah and Medvedev 2010; Boakye Yiadom 2008; Dugbazah 2007; Beals, Levy and Moses 1967); these have generally found a positive association between the two variables. Using a cross section of households (3,987) from the GLSS5 sample, Ackah and Medvedev’s (2010, p.15) analysis shows that migration is beneficial for only a subset of households who send migrants to urban areas. Though their study does not establish a causal link between out-migration and household welfare, there is the indication that migrants sent to urban areas are more likely to send remittances and remit larger amounts (ibid.). This denotes a positive correlation between the two variables provided migration is directed towards urban areas (ibid.). Though households that send migrants to rural areas also receive remittances, payments often cannot recoup the expenses incurred in moving, resulting in decreases in household welfare (ibid.). This shows that the dialectics between migration and poverty is not linear; migration is not necessarily better for all households as some can become further impoverished.
These dynamics are underpinned by differentials in initial deprivation levels and characteristics of the two migrant groups. Migrants destined for urban areas tend to have better initial deprivation levels at micro- and meso-levels and they are generally educated individuals from communities with lower average levels of education, which increases their earning potential (ibid.). Additionally, migration to urban areas entails a larger number of family members moving together or chain migrating, suggesting that their combined earning ability is greater and they are thus more likely to receive remittances of greater amounts (ibid.). Thus in Ackah and Medvedev’s study, household size, education, migration experience/direction and social capital are important determinants in shaping the welfare impacts of out-migration on poor households.
Findings from Boakye-Yiadom’s (2008, p.273) research using data from the 1998/99 round of the GLSS (GLLS 4) also depict a positive correlation between remittances and poverty at the meso-level, with remittances reducing the incidence of poverty by 6.4 per cent. Research findings also indicate that more than 60 per cent of rural households in the sample, which would have been poor in the absence of remittances, were able to escape poverty. Concurrently, however, it is indicated that some migrants who are poor would have been non-poor had they not migrated (ibid.). Tsegai (2005, p.24) explored income differentials between migrant and non-migrant households in the Volta region, using data from a household survey conducted in the region and found that migrant households earn more than their non-migrant counterparts. Examining the welfare outcomes of migrants and non-migrants in Ghana in 1991/2 and 1998/9 (using nationally representative survey data), Litchfield and Waddington (2003, p.10) found that migrant households have statistically significantly higher living standards than non-migrants. Notably, however, when non-monetary indicators of living standards, such as education enrolment, are incorporated into the study, there appears to be no differences between migrant and non-migrant households (ibid.). This highlights the imperative for more inclusive, multidimensional approaches to poverty that is currently lacking in the literature.
Qualitative studies have also shown that remittances ‘make a strong contribution to reducing economic vulnerability at least at the household and local community level’ (Dugbazah 2007, p.383). Dugbazah (2007) notes that remittances can significantly ease the purchase of consumer goods and, in some cases, increase household savings. Looking at the welfare impacts of migration on the Abutia villages in the Volta region of Ghana, Dugbazah illustrates that remittances serve as the most essential source of family income for most households surveyed and even supersede agricultural income in some cases (ibid). The centrality of remittances is encapsulated by the fact than an overwhelming 98 out of 107 households receive remittances, whilst only nine households do not (ibid.). At the same time, however, the quantity of annual remittances are generally low. Though the average annual remittance in the area is approximately GHC 180 ($99.23), almost two-thirds of households receive less than GHC 100 ($55.13) (ibid., p.379-380). Some households receive as little as GHC10 ($ 5.51) throughout the year, and others receive up to GHC 3,600 ($1,984.61) (ibid.). These remittances are generally sent on either a monthly or bi-monthly basis and although the sum is small on the aggregate and sometimes infrequent, Dugbazah maintains that the remittances are significant for meeting basic needs and reducing household poverty and vulnerability (ibid. p.381). Thus, this paper posits that while remittances may not reduce the incidence of poverty, they can reduce its severity. One may draw linkages with this perspective and Skeldon’s (2003) imperative of focusing on the impact, as opposed to the amount, of remittances in discussions on migration–poverty dialectics.
Abdulai, Awudu. 1999. International migration and agricultural development in Ghana. Scandinavian Journal of Development Alternatives 18, no.1: 61-74.
Ackah, Charles and, Denis Medvedev. 2010. Internal migration in Ghana: Determinants and welfare impacts. Policy Research Working Paper 5273, World Bank.
Amin, Samir. Migrations in contemporary Africa: A retrospective view. In The Migration Experience in Africa, edited by Jonathan Baker and Tade Akin Aina, Uppsala: Nordiska, Afrikainstitutet, 1995.
Anarfi, John, Stephen Kwankye, Ofuso-Mensah Ababio,and Richmond Tiemoko. 2002. Migration from and to Ghana: A Background Paper. Migration, Globalisation and Poverty Working Paper C4.
Bakewell, Oliver. 2007. Keeping them in their place: The ambivalent relationship between development and migration in Africa. IMI Working Paper 8, James Martin 21st Century School, University of Oxford.
Barrett, Christopher, Thomas Reardon, and Patrick Webb. 2001. Nonfarm income diversification and household livelihood strategies in rural Africa: Concepts, dynamics, and policy implications. http://inequality.cornell.edu/publications/working_papers/Barrett-Reardon-Webb_IntroFinal.pdf (accessed April 1, 2012).
Beals, Ralph, Mildred Levy, and Leon Moses. 1967. Rationality and migration in Ghana. The Review of Economics and Statistics 49, no. 4: 480-486.
Boakye-Yiadom, Louis. 2008. Rural-urban linkages and welfare: The case of Ghana’s migration and remittance flows. PhD diss., University of Bath.
Bukh, Jette.1979.The Village Woman in Ghana. Upsalla: Scandinavian Institute of African Studies.
Caldwell, John. 1968a. African Rural-Urban Migration. Canberra: Australian National University Press.
Caldwell, Jonh. 1968b. Determinants of rural-urban migration in Ghana. Population Studies: A Journal of Demography 22, no. 3:361-377. http://dx.doi.org/10.1080/00324728.1968.10404945 (accessed March 17, 2012).
De Haan, Arjan, and Ben Rogaly. 2002. Migrant workers and their role in rural change. Journal of Development Studies 38: 1–14. http://www.tandfonline.com/doi/abs/10.1080/00220380412331322481 (accessed April 1, 2012).
Dugbazah, Justina. 2007. Gender, migration and rural livelihoods in Ghana: A case of the Ho district. PhD diss., University of Birmingham.
Ewusi, Kodwo. 1986. The Dimension and Characteristics of Rural Poverty in Ghana. Accra, ISSER Technical Publication.
Ghana Statistical Service. 2008. Ghana Living Standards Survey–Report of the Fifth Round (GLSS5).” Accra: GSS.
Grant, Richard and, Jan Nijman. 2004. The re-scaling of uneven development in Ghana and India. Journal of economic and social geography 95, no.5: 467-481.
IOM. 2005. Migration, development and poverty reduction in Asia. Geneva: IOM.
Kothari, Uma. 2002. Migration and chronic poverty. Working paper no 16, Institute for Development Policy and Management, University of Manchester.
Litchfield, Julie, and Hugh Waddington. 2003. Migration and poverty in Ghana: Evidence from the Ghana Living Standards Survey. Sussex Migration Working Paper 10.
McKay, Andy, Adriana Castaldo, Saskia Gent, Priya Deshingkar, and Matteo Sandi. 2011. Internal migration and poverty: Evidence from Ghana and India. Briefing Paper, Migrating out of Poverty Research Programme Consortium, briefing paper no.1, University of Sussex.
Moore, Karen. 2001. Frameworks for understanding the intergenerational transmission of poverty and well-being in developing countries. CPRC Working Paper 8, International Development Department, School of Public Policy, University of Birmingham. http://www.chronicpoverty.org/uploads/publication_files/WP08_Moore.pdf (accessed April 21, 2012).
Nord, Mark .1998. Poor people on the move: Country to country migration and the spatial concentration of poverty. Journal of Regional Science 38, no.2: 329–351. http://onlinelibrary.wiley.com/doi/10.1111/1467-9787.00095/pdf (accessed April 15, 2012).
Sabates-Wheeler, Rachel, Ricardo Sabates, and Adriana Castaldo. Tackling poverty-migration linkages: Evidence from Ghana and Egypt. Social Indicators Research 87, no. 2: 307-208. DOI: 10.1007/s11205-007-9154-y (accessed April 25, 2012).
Skeldon, Ronald. 2003. Migration and poverty. Paper presented at the conference on African Migration and Urbanization in Comparative Perspective, Johannesburg, South Africa, June 4 – 7.
Stark, Oded and, Shlomo Yitzhaki. 1982. Migration, growth, distribution and welfare. Economics Letters 10, no. 3: 243-249.
Todaro, Michael. 1976. Internal Migration in Developing Countries. Geneva: International Labour Organization.
Tsegai, Daniel and, Quang Bao Le. 2011. District-level spatial analysis of migration flows in Ghana: Determinants and implications for policy. Regional Science Policy and Practice 3, no. 2: 87-100. http://onlinelibrary.wiley.com/doi/10.1111/j.1757-7802.2011.01031.x/abstract (accessed April 25, 2012).
Tsegai, Daniel. 2005. Migration and Household Income Differentials. Paper Presented at the 11th EADI General Conference on Insecurity and Development, Bonn 21 -24 September.
Tutu, K. A. “Determinants of Internal Migration.” In Migration Research Study in Ghana, by K.A. Twum-Baah, J. S. Nabila and A. F. Aryee. Accra: Ghana Statistical Service, 1995.
Waddington, Clare. 2003. Livelihood outcomes of migration for poor people. Working Paper T1, Sussex Centre for Migration Research.