Rural-urban migration and household livelihood in the Agona West Municipality, Ghana

Rural-urban migration, a multidimensional phenomenon, is becoming part of the daily reality in Ghana and many other developing countries. For instance, the impact of migration on households whose member(s) migrated is relatively not very clear. A study was therefore undertaken to examine and explain the impact of rural-urban migration on rural migrant’s households’ livelihood in three communities in Agona West Municipality, Ghana. Snowballing technique was used to select and interviewed 121 respondents to obtain information on the motives of migration and the impact of remittances on household livelihoods. Eight male and female heads of households were purposively selected from each village for the focus group discussion. Frequencies, percentages, diagrams and tables were used to explain the data obtained. The study identified two types of households; the migrant and non-migrant households. Also it was observed that males, were often forced to migrate because of their poor living conditions and low educational attainment, the better-off migrant households characterized with higher educational attainment on the other hand, often migrate in order to accumulate more wealth. The better-off migrant households are therefore positively impacted; better access to health-care, more education opportunities, and agriculture than the poor migrant households. The study therefore recommended that government policy on poverty and employment especially for rural areas need to be revisited for better employment opportunities in the rural areas for the young adults and thereby curb the drift to urban areas.


INTRODUCTION
Migration (internal and international) occurred both in developed and developing countries. Internal migration appears as a massive phenomenon, exceeding international migration as there are around the world 740 million internal migrants compared to 214 million international migrants (UNDP, 2011). Many internal migrants originate from rural areas, and many of them are youth, given their higher propensity to migrate than older individuals (World Bank, 2011). Initially, migration patterns in sub-Saharan Africa were mostly maledominated however it is progressively becoming feminized, in other words there is an increasing number *Corresponding author. E-mail: adeogunleye2005@yahoo.com.
Author(s) agree that this article remain permanently open access under the terms of the Creative Commons Attribution License 4.0 International License of female migrants in cities working in informal sector jobs (Oberhauser, 2016). And that female migrants oftensend more remittances to the rural part of their family regularly (Tacoliand Mabala, 2010).
Migration is neither a new phenomenon, nor a failure of development, nor replacement for development. Rather it has been seen as a response of individuals to better economic and non-economic opportunities and an expectation of increased economic welfare in urban areas (Tanle, 2003). The patterns of migration especially in the sub-Saharan Africa are multifaceted. For instance an individual may be tempted or forced to migrate as a result of cultural, demographic, socio-economic, ethnic conflicts, environmental, natural disasters, or political (large scale infrastructure projects and resettlement) (Zoomer and Otsuki, 2016). The decision to eventually migrate is influenced by a mixture of several of these aforementioned factors.
Current trends in migration in Africa also seem to have significant socio-cultural effects on households and communities, because migration is now becoming an important livelihood strategy worldwide (Reda et al., 2012;Ghana-United Nattions, 2017;UNDP, 2011). It is believed to be one of the most important elements that trigger rural-urban migration especially; in developing countries. It is seen as one of the main strategies to diversify, secure and improve livelihood, often in combination with other strategies (McDowell and De Haan, 1997). A livelihood comprises the capabilities, assets (including both material and social resources) and activities required for a means of living (Oberhauser, 2016). Livelihood assets enable households and individuals to produce, participate in labour markets and to collect sufficient household income (Ellis, 2000). Livelihood assets are the stocks of capital that can be used by households to create the means of living or to improve their welfare level. The notion of livelihood is associated with social institutions like family, village and other social networks facilitating and sustaining diversified livelihoods (Jiao et al., 2017;Nyberg-Sorensen et al., 2002).
Livelihood strategies of people changes in response to the constraints and opportunities they are exposed to, and likewise the decision to migrate or not. There are many factors that shape these decisions which correspond to the contextual, socio-economic and policy considerations. A policy like the resettlement program in Ethiopia affects not only people"s livelihood strategies, but their access to assets and every other aspect of their livelihoods (Tacoli, 2010). Adopting migration as a livelihood strategy, may not necessarily lead to improved livelihoods. In other words, migration may lead to either improvement or deterioration in livelihood status of people.
Rural-urban migration in this case can be considered as a household strategy in which economic and social links between the migrant and his or her rural household are maintained because the migration process is a conscious strategy of the whole migrant household inorder to diversify the household income sources and to become less vulnerable to shocks, risks or local constraints in their livelihood (Ellis, 2000). The decision of the household is based on the consideration of the profits of migration (like remittance flows, higher local incomes or the spreading of risks) and the costs of migration (like travelling costs or the lack of labour forces) (Lindley, 2008).
The contribution of remittances in migration processes and the whole migrant household act tactically as one decision-making unit (Braw et al., 2014). Migration can then be conceptualized as a type of livelihood diversification by households, an answer to household's income risks and in this way remittances can form a kind of income insurance of the sending households (Okhankhuele and Opafunso, 2013) Migrants remit or transfer money to relatives in rural areas. Remittances improve chances of access to land and other resources, while the need to access land can be a key motivating factor for migration. Remittances are invested in land and other resources, used to pay for education, health, housing and direct food purchases. Remittances maintain multi/spatial households that combine farm and non-farm activities and rural and urban residence (Tacoli and Mabala, 2010).
The impact of rural-urban migration however, is not a linear process rather it is more complex with a lot of variables coming to play-type and extent of migration, size of remittances and local context. These variables are interrelated with each other and can have influence like the loss of human resources in rural areas and the impact of remittances and earnings for rural households. The major impact of migration and remittances on the livelihoods of rural households therefore depends on the expenditure, investments and labour allocation of the sending households.
Migration in Ghana is not only on the forefront of the demographic and development transitions in Africa, but also at the front of the urbanization trend. In recent years, the migration routes have tended to be more of rural-tourban migration to cities in Ghana, like Accra and Kumasi attracting traders, young adults seeking employment and educational opportunities (Adepoju, 2003;Appiahnig, 2013).
Ghana"s rapid population growth and urbanization also have important linkages to migration. For instance, the 2010 national census in Ghana recorded a population of 22.7 million (2006), 24.66 million (2010) to 29.09 million people (2017) (United Nation Estimate, 2017;Ghana Statistical Service, 2017). At the national level, about 53.4% of Ghana"s population is urban, an increase from the 1984 level of 32% (Ghana Statistical Service, 2017).
Ghana, like most parts of Africa, is still predominantly rural, but it is urbanizing steadily due partly to rural urban migration. More than 80% of Ghanaian migrate with about 70% going to the urban areas (Ballard, 1983;Ghana Statistical Service, 2000, 2010. The greater Accra and Ashanti regions attract more than half of all internal migrants and migrants make up a substantial share of the population in these regions (Tutu, 1995;Tanle, 2003;Ghana Statistical Service, 2010). The southern regions of Ghana; Western, Central, Eastern, greater Accra, Volta, and Ashanti are the destinations for 88% of all internal migrants, while the Northern and the two Upper regions together account for only 5% of the total.
On the other hand, the "pull" hypothesis emphasizes the attractiveness of the urban life and the rural-urban wage gap. In particular, in Todaro (1969) and Harris and Todaro (1970) probabilistic models, migrants are attracted to cities with the expectation of a higher wage than they receive in agriculture. In view of this, migrants are willing to accept the probability of urban unemployment, or lower wages and "underemployment" in the urban informal sector. According to Todaro, a migrant is willing to accept urban employment or lower wages in the urban informal sector as long as he expects to "graduate" to the urban modern sector in the future.

Problem statement
Studies of migration in Ghana exploring its patterns, determinants, and impacts on welfare and poverty date back to the 1960s. Early contributions by Caldwell (1969) used census and survey data and found a negative effect on migrants place of origin/locality"s (rural) in terms of income, but a positive effect of a household"s own income on the probability to migrate. Other important determinants of the likelihood to migrate as noted by Caldwell (1969) include presence of friends or relatives in the destination (migration networks). Tutu (1995) also observed that males are more likely to migrate than females and that younger persons are more likely to migrate than the aged. In the Agona West Municipal, migration of the youth from rural areas to urban areas has been a matter of great concerns (Appiahnig, 2013). This is because those who migrate especially the young adults, return with improved livelihoods or remit their families which in turn improve their family's socioeconomic status in the community. This situation according to relative deprivation theory of migration may encourage other youth in the community to also migrate to urban areas. These developments have much implication on the rural communities.
In addition, much of the influence depends mostly on household size, the larger the households the greater the number of migrants. It is therefore not clear whether the increasing migration of members of households in the Agona West Municipality suggest a simple response to Tanle et al. 3 the persistent relationship of the poor economic conditions in rural areas. Moreover whether the poor and uneducated from the Agona West Municipality like their counterparts from other regions engage in migration to the towns in other to improve their household livelihoods as the findings regarding the relationship between education and the probability to migrate have been conflicting, with estimating a negative relationship and Caldwell (1969) reporting a positive asso, ciation. Castaldo et al. (2012) also highlighted the importance of remittances sent by migrants in urban areas to rural origin communities in raising the welfare of households and narrowing the welfare gap between rural and urban communities. It is also not clear whether those who migrate to the towns remit their household members in the rural areas to increase their income and consumption or even the family"s social status as observe by Tutu (1995). More recently, efforts at establishing a relationship between migration and household welfare generally found that migration tends to increase the welfare of sending households. Using data from the 1991/1992 and 1998/1999 rounds of the Ghana Living Standards Survey (GLSS), Sam et al. (2013) found that migrants have a higher standard of living than non-migrants. But one question that still remains is that, is this phenomenon visible in the Agona West Municipality among the rural folks who send their members to the towns?
The uncertainty and connection between rural-urban migration and livelihood and the effect of rural-urban migration on households livelihoods need to be unravelled in order to enhance our understanding of the socio-economic context in which rural households in the Agona West Municipality live, and make suggestions that may help curb or regulate rural migration. It is therefore important to study and comprehend the motives behind migration in the Agona West Municipality.

Objectives of the study
The main objective of the study was to assess the effects of rural-urban migration on household livelihoods in the Agona West Municipality.
Specifically, the study sought to: outline the main motives for migration; investigate the use of remittances by households; and examine the effects of remittances on household livelihoods.

Research questions
The study therefore sought to answer the following questions: What are the main motives for migration in the Agona West Municipality? What are the uses of remittances by migrants" households? What are the effectsof rural-urban migration on household livelihoods?

Rationale of the study
It is hoped that this thesis will add to the existing knowledge on the perspective of the migrant-sending households and the effects on the households.
Facilitate an understanding of the development potential of out-migration for the migrant-sending households, a force that may increase or diminish inequalities in the sending area particularly in Ghana. The study may also inspire other researchers to conduct further research on the issue on rural-urban migration and the impact on household livelihood especially in Ghana by focusing on the areas of origin.
This will in turn then provide information for planners, policy makers and local agencies like the Ghana National Population Policy, the Growth and Poverty Reduction Strategy in their overall effort to formulate and implement population redistribution or migration policy and related programmes.

REVIEW AND THEORIES OF MIGRATION
The scale and patterns of migration have been of interest to social scientists since the laws of migration were first formulated by Ravenstein (1885) explaining migrationdistance hypothesis, states that migration is inversely related to distance and that most migrations occur over short distances. Adepoju (2003) in reviewing Ravenstein"s migration-differential hypothesis added that economic motives were the most dominant causes of migration and that development in transport and communication would invariably increase the tempo of migration. The impact of migration on the rural household was not captured by the theory.
Many of the subsequent migration theories in contemporary studies are more or less variations of his arguments. For instance, Lee (1966)explains the factors affecting migration could be positive and negative characteristics of both the origin and destination. And added that a framework for analysing the volume of migration; the characteristics of migration and the decision-making process. The framework classified the pull-factors as the attractions and socio-economic opportunities available in other localities and the pushfactors to include the deteriorating socio-economic conditions in the areas of origin (Lee, 1966).
According to Lee (1966), if the negative features of the origin were more powerful, then migrants were pushed out, while at the destination, if the positive features were more powerful, then migrants were pulled in. Since the 1950s in developing countries Ghana inclusive, rural poverty has pushed migrants out, while higher incomes and cultural amenities associated with the large cities have lured migrants. Lee"s theory is more applicable to the changing socio-economic context of contemporary society tried to explain migration as transition from a stagnant rural agricultural sector to a growing modern industrial urban sector due to the labour surplus in rural areas will supplement the labour shortage in urban areas, which serves as the drive for rural-urban migration. This model, assumed that rural economies initially present a specific context in which there is surplus labour in the agricultural sector. Hence, the agricultural sector is able to supply labour force to the modern industrial sector which can grow by accumulating capital and obtaining labour from the agricultural sector (Dugbazah, 2012). Unfortunately, this model"s assumption of zero marginal productivity and remuneration in the agricultural part is highly arguable. Also, it does not adequately describe the rural-urban migration process of many developing countries in contemporary time. This is because even if agriculture productivity and wages may be low, they are not completely non-existent as rural people are able to make a living at the subsistence level. This model might have been applicable in the late 1960s when urban areas experienced high levels of unemployment (Dugbazah, 2012). Ravestein (1885) and Lee (1966) were able to establish the relationship between migration and socioeconomic development, emphasizing that people will always move when confronted by better opportunities.
The neoclassical economic theories were developed on principles of individual optimizing behaviour treat migration as an economic phenomenon in which the migrant weighs the costs and migration (Free, 2010). Todaro (1969) one of the fore runners of the neoclassical economists in terms of classic rural-urban migration theory postulated that migrants respond mainly to economic incentives, earnings differentials, and the probability of getting a job at the destination is the major influence in the migration decision. In other words, the decision to migrate is a function of the wage differentials that exist between urban and rural areas and the probability of finding a job in the city. This model thus highlights the importance of the probability of finding a job in cities along with the popularity of higher wages there, which motivates a potential migrant to finally migrate. This theory assumes that rural-urban migration will take place whenever the urban expected wage exceeds the rural wage (Todaro, 1969: 73). Also, the Neo-classical economic explanations, assume a homogenous individual who is undifferentiated by gender, class or other factors, to be making rational decisions to maximize economic interests and ignores other factors, like marriage, dependency relations (social factors), floods, river erosion, drought, etc. Despite its limitations, the model/theory identifies a very vital link between perceived employment opportunities and migration. A lot of studies support the argument that the basic motive behind the decision to migrate from rural to urban areas is guided by a search for employment opportunities in urban settings (Tacoli and Satterthwaite, 2003). This is evidenced by the fact that after migrating, most of the people usually enjoy higher income relative to agricultural income (Adepoju, 2003).
Current models of internal migration, like the household theorists by Clarke and Drinkwater (2001) explained that people act jointly not only to maximize expected income, but also to minimize risks for the members of the kinship unit (Farkhanda et al., 2014). The main focus of this approach is that migration decisions especially in developing countries are not made by isolated individuals, but by families or households (Krantz, 2001;Farkhanda et al., 2014;Tacoli, 2002). Network theoryby Bakshi (2008) another relevant theory to this study (migration and livelihood), attributes migration to personal, cultural, and other social ties. And argues that in migrant-sending communities, information about jobs and living standards are efficiently transmitted through an arrangement struck between personal networks such as friends who emigrated (Agesa and Kim, 2001). This is because, the family eventually benefits from the migrant through remittances which enable them to cope with adverse economic shocks.
Evidently, traditional migration theories have not addressed the household aspects of migration. For instance, in the neoclassical economic models and the push-pull demographic models, migration was seen as the outcome of individual decisions (Krantz, 2001). The new economic concepts and theories emphasized the importance of the family or the household as the primary site of decision-making however could not explain the fact that household decisions and actions do not necessarily represent unified and equally beneficial outcomes for all members (Tacoli, 2002;Potts 2009).

Livelihood approaches
The concept of livelihood is associated with social institutions, as family, village and other social networks facilitating and sustaining diversified livelihoods (Carney and Clark, 2008). Therefore, the livelihood approach became popular. The livelihoods of people are ways and means through which they make their living with resources available to them (Ellis, 2000). Chambers (2008), however stressed that livelihood is sustainable when it is able to cope with, and recover from stresses and shocks, able to keep its assets and capabilities both now and in the future, while not depleting the natural resource stock. The fundamentals of the livelihood concept are livelihood resources, livelihood strategies, livelihood outcomes and vulnerability context (Chambers, 2008).
Livelihood approach is used to identify the main constraints and opportunities faced by poor people, as articulated by them. It builds on these definitions and then supports poor people as they address the constraints, or Tanle et al. 5 take advantage of opportunities. The framework is neither a model nor a universal solution but rather a means of stimulating thought and analysis, and it needs to be adapted and elaborated depending on the situation. Chambers (2008) argue that the ability of households to have access to sustainable livelihood strongly depends on whether or not they have access to five forms of capital assets: natural, physical, human, social, and financial assets. The idea of a livelihood framework as a tool for analysis is simply to capture the main elements, which comprise the complex livelihoods of people at a given point in time, as well as the course and dynamics of change in livelihoods (Carney and Clark, 2008). The natural capital water, land and natural resources in the environment of people are used to create means of survival. The profits of these sources can be direct or indirect and they are related with user regimes and property. Human capital: labour, health, education, skills and everything that is required for able-bodied and talented labour forces. Furthermore, physical capital: machines, transportation vehicles, buildings, roads, electricity, communications, etc.; all assets that are produced by industrial production processes. Financial capital: all the financial resources in the form of accessible stocks and regular inflows of money that people use to attain their livelihood outcomes, like savings, loans and credits. Finally, social capital, the social resources through which people are able to achieve their livelihood objectives (Ellis, 2000). The livelihood framework offers no explanation for the role of power relations and politics. However, these power issues like institutions, laws or policies affect the choices that people make with their livelihood assets. Not enough emphasis is given to the informal structures and processes that affect access within the community. Krantz (2001: 25) noted that though the frameworks taking note of gender considerations attempts to increase the voice of women which is difficult to achieve successfully in practice. It is therefore important to include political capital in livelihood research. And a modification of this framework to suit the study being undertaken became necessary.

CONCEPTUAL FRAMEWORK
The concept of a livelihood strategy has become central to development practices in recent years. However, the uncountable possible proportional mixes of activities undertaken by a households, is not always clear regarding the constituents of a distinct livelihood rather than just a slightly different mix of activities within the same general livelihood (McDowell and Hess, 2012). A precise operational definition of livelihood remains vague, as does an associated method for identifying livelihoods in quantitative data (Francis, 1999). This probably helps explain why the more quantitative development scholars (e.g. economists) have been slower to adopt the concept as compared to the most qualitative ones (e.g. anthropologists and sociologists).
The ability to operationalised the concept of a livelihood strategy becomes especially important when examining the "improvement" of livelihood to paraphrase much current development discussion (Scoones, 2009). Thus it is quite important in choosing distinct livelihood strategies that earn high returns for households, especially in rural agricultural areas.
As explained earlier on, thisstudy adopted the"migration as a livelihoods strategy" as its conceptual framework which was originally adapted by Thiemes (2005) from the Sustainable Livelihood Framework. This framework is seen to be the appropriate framework for this study because it clearly shows the relationships between migration and livelihood outcomes taking into consideration how vulnerable households are to external effects ( Figure 1).
The basic elements of most livelihood frameworks are: livelihood resources: what people have, variously referred to as stocks and stores, assets and capital (both tangible and intangible); livelihood strategies: what people do (e.g. agriculture, wage labour, migration); livelihood outcomes: what goals they are pursuing, and the living that results from their activities.
As a component of the framework (Figure 2), livelihood strategies differ with regard to whether people have to deal with gradual changes or crises Carney (1998). These are agricultural intensification, livelihood diversification and migration (Carney and Clark, 2008). Agriculture is notably the most important economic activity of rural households in most developing countries (Tsegai, 2005). In the past, it has often been assumed that increasedin farm productivity would create more nonfarm income earning opportunities in the rural economy via linkage effects.
However, this assumption is no longer tenable for many poor rural families as it is obvious that farming on its own is unable to provide a sufficient means of survival (Potts, 2009).
As a result of inadequate income from agriculture, most households are compelled to embark on livelihood diversification strategies such as migration, in order to vary the sources of household income (Afshar, 2003). In sub-Saharan Africa, most household level diversifications are not just non-farm, but also non-rural in character, such as rural-urban migration. It is widely agreed that the capability to diversify livelihood is more beneficial for poor rural households (Toulmin et al., 2000;Mack et al., (2005) however argues that having alternatives source of income generation can make the difference between sustainable livelihood and destitution. Diversification does not necessarily have an equalising effect on overall, rural incomes as families, that are relatively better-off are typically more able to diversify their livelihoods than poorer rural families (Ellis, 2000).
For this study, the theory of sustainable livelihood, migration as a livelihood theory are adopted even though it may raise some debate in terms of rural livelihoods sustainability, because the concept lay emphasis on the fact that people's livelihood may depend on migration. Migration is said to reduce poverty in the migrant's home of origin. Zoomer and Otsuki (2016) however argues that it is the migrants themselves who's level of poverty reduces as they find themselves with good jobs and therefore contribute more to the area of the destination than the origin.
According to Knowles and Anker (1981), there are uncertainties surrounding income of families, savings and investments are reduced by migration. For instance, remittances from migrants have for a long time provided generations with an inseparable source of income among the Bihari migrants in Calcutta (McDowell and De Haan, 1997). It has also been found in Kenya that income from migrants plays crucial role in the lives of the less endowed as compared to the wealthy ones since the less endowed have little livelihood opportunities. Most of the remittances from migrants are geared towards payment of households' debts (Francis and Hoddinott, 1993).
Even though migration is often linked to insecurity in the literature because of break of family ties, labour repatriation and retrenchment due to economic failures and political reasons, reduces risks, seasonality and shocks (Francis, 1999). For instance, migration of young ones from a household reduces the household"s tendency to face food insecurity in times of hardships. And young women who migrate tend to save and invest their earnings in preparation to their marriages and in so doing may not contribute physically to the household but they also do not rely on their households (Krantz, 2001). These contribute vitally to the reduction of insecurity, shocks and seasonalities. Migration also reduces tensions that exist within households and keep families longer in the long run.
Furthermore, migration as a means of diversifying livelihood will also favour households, due to their higher level of education relative to that of poor households, which subsequently translates into their higher propensity to obtain employment (Toulmin et al., 2000). The evidence is mixed regarding the gains and losses of household diversification strategies to agriculture (Toulmin et al., 2000). Negative effects are associated with the withdrawal of critical labour from farming activities, while positive effects include the alleviation of credit constraints and a reduction risk to income (Ellis, 2000;Banerjee and Duflo, 2007).
Efforts made to make the standard of living better for the migrants than before are known as enhancement strategies . It is hoped that these efforts will make households better in comparison to their situation before leaving as it is believed to make them wealthy.
The outcome as theutmost of the components in Figure 2 could either be improved, worsened or unchanged. Because the framework is human centred, it identifies opportunities and how people use them but also how externalities influence their usage. The outcomes in most cases are in the form of income, well-being, vulnerability, and food security which could either be improved, worsened or unchanged due to migration. Taylor (1999), states that although individuals migrate, they do not sever ties with their source households because at times source households may pay migration costs and support migrants until they become established at their destinations. Family members who remain behind may reorganize both their consumption and production activities in response to the migrant"s departure and migrants (often children) share part of their earnings with their household of origin through remittances.Continuing interactions between migrants and rural households suggest that a household model would be more appropriate than an individual level model of migration decisions. Consequently, migrants become involved in the economic development of their places of origin and therefore are considered as agents of development. They can contribute to development not only through remittances, investment and entrepreneurial activities but also through the transfer of newly developed skills and knowledge, or through fostering democratisation and the protection of human rights in their countries of origin (Taylor, 1999). The current study adopted a modified version of the sustainable livelihood framework considering migration as the only livelihood strategy. Migration as a Livelihood Framework is shown diagrammatically in Figure 2.
A keen study of the statistical report of the Agona West Municipality indicates that, economically, the resource base of the municipality is determined by the natural resources of the area (Agona West Municipal Aurthority Profile, 2010). The implication is that, the municipality is predominantly agricultural. Whether the factors influencing migration are "push" or "pull", migration is not a one-off phenomenon that occurs in a specific place at a fixed time. Studies on internal migration clearly demonstrated that greater job opportunities in the cities and metropolitan areas "pulled' migrants from rural areas (Afsar, 20003). Normally, rural dwellers have little incentive to remain in agriculture. Instead, they chose to migrate to nearby towns or district towns or cities in search of more remunerative non-agricultural jobs. So, considering the fact that agriculture is the main occupies of most of the land use of the Agona West Municipality. It may therefore be expected the active work force move from their rural dwellings to the urban areas in search of industrial or factory employment. The need to find out explain whether they get more remunerative jobs after migration or end up swelling the ranks of unemployed in the urban labour market as predicted by Todaro (1969) is necessary. Given the fact that poor migrants can hardly afford to remain unemployed, one must go far beyond unemployment data to examine the impacts of migration on poverty. However, for those who live on the edge of extreme poverty in rural areas, migration to nearby towns may give them temporary relief from unemployment.

Study area
The Agona West Municipality is situated in the eastern corner of the Central Region and is made up of six subdistricts or Town/Area Councils. The area falls within the moist tropical and semi-deciduous forest with a lot ofvaluable timber trees like mahogany, sapele, silk cotton, wawa and odum (AWMA Profile, 2010). Forest food crops like plantain, banana, cassava, cocoyam and maize are also cultivated. The major soil type found in the municipality is classified as forest ochrosols. These soils are alkaline and richly supplied with nutrients which make them suitable for cultivating varied agricultural produce like cocoa, citrus and coconut. Vegetable and sugar cane cultivation are widespread (AWMA Profile, 2010). The majority of the people in the municipality are into the production of cocoa and cassava. Agriculture is the major economic activity engaging more than 64% of the municipal population (AWMA Profile, 2010). Being a nodal town, the municipal capital, Swedru, is a nodal town and the availability of markets in most of the major towns in the municipality, trade and commercial activities are promoted. The municipality has a total population of 115,358, out of which females constitute about 61,199 (53.1%) and males are 54,159 (46.9%) (Ghana Statistical Service, 2010). Sex structure of the municipality shows that the population is predominantly female 53.1%. According to the Ghana Statistical Services (2017), youths in Agona West Municipality are well known for out-migration; this might have been made easy because of the location being a nodal with a trunk A and other roads.

Research philosophy
Based on the positivist and the interpretivist research paradigms, the mixed method approach (quantitative and qualitative) was used. The data collected from both the quantitative and qualitative will add more credence from the result that will be obtained from the field.

Data, sources and target population
Data for this study were collected from primary sources from the respondents on the field. The secondary sources include information from GSS, literatures and other relevant records.
The target population for the study consisted of households" heads from the three selected study communities: Nkranfo, Nsuansa, and Nsonan in Agona West Munipality. The selection of these communities was based on the fact that the rate of youth out-migration in these communities is more frequent and higher (Arthur, 2009;Agona West Municipal Authority, 2014).

Sampling technique and procedure
The sample size for the study was 169 made up of 121 household heads who had some household members as out-migrants and 48 key informants from the study communities to partake in the focus group discussion. About 171 household were identified in the three communities during a reconnaissance survey but only 169 respondents accepted to participate in the actual survey.
Purposive sampling technique was used to select the three communities for the study. Purposive because it was observed that the rate of youth out-migration in these communities are more frequent and higher, as such the likelihood of obtaining the needed information for the study could be easy (Arthur, 2009; Agona West Municipal Authority, 2014). Snowballing technique was then used to randomly identify and select 175 migrant households. Quota allocation based on number of migrants" was used to select Nkranfo (39), Nsuansa (50), and Nsonan (32) in the study areas. Snowballing technique was then used to identify households that have family members that have migrated through the help of the community leaders who identified some migrant households who in turn identified other migrant households until the required sample size was obtained for each community (Table 1)

Research instruments and data processing and analysis
Interview schedule, questionnaire and focus group discussions (FDG) were used to obtain the necessary information for the study. The responses from the interview schedule was then numbered serially, edited, coded and summarised in tables, charts and graphs, frequencies and percentages. The FGD"s were cleaned and manually transcribed verbatim to supplement the quantitative result. Table 2 shows that majority (42.1%) of the respondents were in the young adults aged between 30 and 39 19.9%, below the age bracket of 20 to 29 years (19.8%). The demographic implication is that a large number of the respondents (62.0%) were in the economically active age group of 20 to 39 years. With respect to sex, the analysis revealed that 57.0% (69) of the household heads were females; these results is consistent with Brydon (1987) findings in the same study area that the gender composition of household heads has been altered due toearly migration patterns that made more women to become heads of households. This conclusion is also inagreement with Lee"s theory of migration that postulated that more males migrate than females. The males in these localities travelled to the cocoa growing areas in search of new employment opportunities. Moreover, about 38.8% of the household heads were married followed by respondents who were never married (23.1%). This finding is in line with Amanor"s (2001) conclusion that more people in rural areas are likely to be married compared to the urban area population perhaps due to their farming activities. With regards to household size, the study showed that the largest household size in the study areas was between 6 and 7 members (39.6%) and closely followed by households who had 10 and above (36.5%).

Socio-demographic characteristics of migrants
One of the main objectives of this study was to find out the underlying causes of frequent out-migration of youth in the Agona West Municipality of the Central Region of Ghana. Hence, heads of migrants' households were used as a proxy to obtain the needed information migration in the study areas. This was necessary because the literature on internal migration indicate that the migration decisions of individuals have the propensity to influence their migration outcomes, in particular, issues relating to resource accumulation (Anarfi et al., 2003;King, 1998).
Results in Table 3 showed that more than half (54.8%) of household migrants in the study area were predominantly males which go to validate what has been observed earlier in Table 2 where the majority of those left behind were females and became the household heads. This finding is in agreement with what Clarke and Drinkwater (2001) observed, that culturally, men everywhere in the world Ghana inclusive, are expected to fend for their households and therefore migrate and remit to their households was an alternative source of income. Also, according to Dugbazah (2012) in rural traditional society, women are often expected to stay at home caring for children and cultivating the household farm, while the men migrate. The results from Table 3 showed that most of those who migrated in the study areas were within the age cohort of 20 to 30 (43.8%) and that was followed by the age cohort of 31 to 40 (30.6%). This result implies that not only are the majority of those who migrate from the study area males, but are also in the economically active age group (74.4%). This finding supports the literature on internal migration dynamics in Ghana which suggest that most rural out-migrants are mainly males (GLSS, 2005;GSS, 2012). It was also discovered that about 28.9% of migrants who migrate out of study area secure jobs in the service and sales sectors followed by 28.1% who are involved in agricultural related occupations. This later conclusion also gives credence to Brydon"s (1987) findings that the males migrated to cocoa producing areas in Ghana to look for employments by engaging in agricultural activities

Motivations for migration in Agona West Municipality
The study revealed that the main reasons for youth outmigration in Agona West Municipality were economic (40%) followed by the desire to pursue further education (31.2%) and escape from conflict was the least reason why people migrated (5.0%). This finding confirms the fact that the most important determinants of migration from the study area are the search for better economic opportunities. This conclusion agrees with Todaro"s (1969) andLipton (1997 argument that areas that cannot absorb their own employable labour tend to become sending areas as its people seek better livelihood options. In the FGD session, the discussants agreed that due to the vulnerability in farming activities resulting from rainfall failure in the Agona West Municipality, most young people involved in agricultural production consider migration as a risk-averse livelihood strategy (Table 4). This was further confirmed by a female discussant in the FGD: "Rainfall has not been consistent here, whereas it fails in a current year, there will be excess the following year and hence farmlands are flooded. To prevent the yearly agricultural problems, some members should try other sources of livelihood. This, in the long run, will work to perfection because we believe that when the farming fails, the non-farm migrants will bring something home" A 60-year-old male FGD participant added: "The decision to migrate is seen as an economic investment and that household members raise funds for transportation cost and use networks to help their household members settle in the city. You see in the city they are able to work and help the family in so many ways".
The aforementioned evidence lends credence to what Harris and Todaro(1970) observed that the main reasons for migration among many people is to secure urban industrial jobs and the perceived existence of higher wages and differences in expected earnings. It can, therefore, be said that perceived income differentials between rural and urban areas contribute significantly to internal out-migration of youth in Agona West Municipality.

Migration destinations of migrants
As observed in the literature, the destination choice ofmigrants depends on their motives for migration (Agesa and Kim, 2001). The results in Figure 3 showed that the Greater Accra region (38%) was the preferred destination of most youth from the study areas followed by the Central Region (24%). The respondents" preference for Greater Accra as their main destination could be attributed to the fact that most opportunities for employment and facilities for modern living are concentrated in the Greater Accra region, which functions as the national capital as well as the headquarters of most multi-national companies and government institutions (Ghana Statistical Service, 2017). It is thus not surprising that most migrants (38%) from the Agona West Municipality migrate to the Greater Accra region to take advantage of the aforementioned opportunities. This to an extent confirms the fact that about 40.0% of the migrants from the study area migrated to the cities for economic reasons (Table 4). In a FGD held with household heads in the study areas, it was revealed that employment in farm and non-farm sectors in the rural areas of Agona West District was by far less attractive in terms of income than employment in Accra and elsewhere. On this same issue this was what a 45 year old female household head said about the precarious nature of working at home (Agona West Municipality): "Even if migrants find jobs in the municipality, the relatively low wages would make it not possible for them to meet their financial responsibilities such as the payment of school fees, buying uniforms for their children and meeting emergency health expenses" [A female household head aged 45].

Another female indicated that:
"The migrants who go to Accra are often better off than the others, so we even advise them to go to Accra. Moreover, in Accra, there are much more opportunities than anywhere in Ghana" [A 35-year old female participant]. With respect to Central region as the second leading destination for most migrants (Figure 3) from the study area, it could be due to the role Central region (Cape Coast) played and continues to play as the major hub of most educational institutions nationwide. In relation to the Central Region, a male discussant stressed: "Why on earth will I send my child to school in any part of Ghana when Central Region boast of the best secondary schools in Ghana coupled with tertiary institutions, I cannot even bear the extra cost of sending him/her afar when we have the best here" [A 49-year old male FGD participant].

Employment status of migrants before and after migrating
The literature on migration suggests that migration of people in search of greener pastures in urban settings is largely influenced by the employment status of the people involved in the migration process (Thieme and Wyss, 2005;Conway, 1992). In this regard, the present study assessed the employment status of migrants in the Agona West Municipality before migration and after their migration to their current destinations. Results in Figure 4 shows that before migration, 51.2% of the migrants in the study area were unemployed but after migration the percentage of migrants employed at their respective destinations increased to 70.2% and the number of migrants that were not unemployed after migration at the their destination place being as low as 29.8%.
These revelations suggest an apparent existence ofmore job opportunities at the destination than at their places of origin and this is in tandem with findings by Oberhauser (2016) who opined that migrants tend to have access to employment opportunities at their destinations than their hometowns or places of origin. In addition, urban areas offer many economic opportunities to rural people for changing jobs and becoming upwardly mobile even with a low asset base and few skills (Okhankhuele and Opafunso, 2013;GSS, 2010). Even if urban wages are not higher, work seems to be available more regularly than in subsistence agriculture. Hence, although rural-urban migration requires more capital and contacts, a general advantage is that work can be found all year round independent of the season.

Remittance behaviour of households' migrants
Remittances can be a valuable source of income for a household livelihood and can also serve as a means of risk diversification as it compensate for a loss of labour (Tsegai, 2005). These remittances can significantly assist the purchase of consumer goods, and in some cases, raise household savings and may, in turn, change the local household income distribution in a positive direction (de Haas, 2007;Dugbazah, 2012). To unravel some of these nuances, this study analysed the uses of remittances by households in the study areas.
To begin with, the respondents were asked to indicate whether they have ever received any form of remittances from migrants. The result showed that majority (85.1%) of households indicated that they receive remittances from migrants with only a few (14.9%) respondents saying otherwise. This result is supported by the findings that majority (80%) of migrants remit (either in cash or goods) their households back home after migrating. And this remittance help in compensating for the labour lost due to migration (Africa Development Bank, 2008;Castaldo et al., 2012;Tacoli C, Mabala, 2010). This finding however, contrasts Adams (2007) and Anarfiet al. (2003) findings in Ghana that about 49 to 24% of internal migrants remit to their households. This could probably be due to the fact that the studies were undertaken in different study periods and areas though in Ghana.
The frequency of remittances was also measured in order to understand the remittance pattern of household migrants in the study areas since the frequency of remittances from migrants has implications on the livelihood status of migrants" households. Table 5 shows that out of a total of 121 household heads surveyed, 60.3% confirmed that they received remittances from migrants every month while 7.4% said they received remittances on an annual basis. It was also found that nearly 14.9% (17) of household heads said they received remittances once a while from migrants. This implies that most migrants from Agona West Municipality remit to their household members on monthly basis perhaps due to the kind of occupations the migrants are engaged in.
This was summarized in the FGD where participants explained that: "This is our government work where we also receive our monthly salary (

Types of remittances received and amount
The literature on remittances revealed that migrants often remitted to their household with cash than any other material items (Afsar, 2003). This study sought to explore the various forms of remittances migrants send home. About 95.0% received cash as the main form of remittances sent by most migrants. These findings support Afsar"s (2003) observation that most migrants' households receive cash as remittances compared to other forms of remittances. The findings also parallel that of Sam et al. (2013) that majority of the remittance receivers prefer cash remittances rather than goods or combination of both.
In a focus group discussion, a male household head discussant affirmed the significance of cash among migrants" remittances in the Agona West Municipality: "Here if you travel and want to send anything back home it must be in the form of cash because we value cash than goods. How can you be in the city working and sending only material things home when you know we pay school fees?" [A 51-year-old male household head].
"Even if you send so many goods home without money we will sell the goods for money because it is money we use to do everything, ahh! … Why do you send me goods when you know I have to provide housekeeping money in the house?" [A male FGD participant aged 44].
With respect to the amount of remittances sent, it was discovered that about a third (31.4%) of the respondents said the most frequent amount received as remittance was between 201 and 300 Ghana cedis followed by 23.1% who indicated that they frequently receive between 101 and 200 Ghana cedis (Table 6). Only 6.6% households got remitted with 500 Ghana cedis and above. Though not much is known from literature on the average amount remitted to a household from internal migrants, Sam et al. (2013) found that 44.5% of households receive between GH¢4000 and GH¢7000 annually with others (22.8%) receiving GH¢7000 and above as remittances from international migrants. Adams (2007) added that the mean per capita total remittances received from internal migrant is only about 30% compared to the amount received from international migrant.
During the FGD session, a 44-year-old male participant who often receives cash from his daughter had this to say:   Table 7 presents the recipients of migrants" remittances in the study communities. This is important because a study by Morales et al. (2002) showed that though the flow of remittances through migration has the ability to improve people"s access to other capabilities but this however largely depends on the receiver of migrants" remittances. From the analysis, the main recipients of remittances were household heads (44.6%) followed by migrants" spouses (20.6%). The present evidence was expected in view of the fact that household heads are regarded as bread winners in most Ghanaian cultural systems and are therefore expected to receive any form of remittances that flow into the household. However, this finding contrasts that of Adams (2007) that though 53% of all migrants in Ghana remit, 99% of them rather remit to relatives and friends for accountability and possible legal actions.

Recipients of migrants' remittances
Concerning the recipients of migrants' remittances in the study area, this was what a 45-year-old female discussant had to say during a FGD: "My husband used to send the remittance through hiselder brother but I was not getting the full amount of what was meant to be mine. This continued for about five

Uses of remittances by households
This part of the study assessed the uses of migrants" remittances from the perspective of household heads. Many households use the money to provide for their basic needs such as food, shelter and clothing. According to IFAD (2008),households can choose to spend remittances on basic needs like increased consumption and investments in housing, health and education or on agricultural production or for investments in commercial activities, like small enterprises or business. As such the study sought to find from respondents the main uses of remittances from migrants. Table 8 shows that 27.3% of the remittances received were used to purchase food items for households" consumption; this was followed be agriculture (24.8%) which is the primary economic activity of the villages. Meanwhile, remittance was least (11.6%) spent on health. It is, therefore, clear from the findings that remittance was spent largely on households' up keep. This agrees with the observations of de Haas (2007) and Reda et al. (2012) that remittances are mainly spent on households" daily consumption rather than investments in agriculture and entrepreneurial activities. But it can be inferred from Table 8 that remittances were largely spent on basic needs as 62.8% (27.3 + 23.9 + 11.6) except housing and investment. The result also supports the claim that remittances led to increase in food consumption and food security, better access to health and increase in educational opportunities (Dugbazah, 2012).
Literature further added that the uses of remittance are subject to the consumption pattern of the households; this is because some households will invest, others will spend on food items or invest in agriculture, health or education of household members ( de Haas, 2007;Dugbazah, 2012).
In the FDG, a female household head said: "Prior to migration, there was no help from anybody and hence at times we were devoid of three square meals per day, on the contrary, due to the remittances, the household can boast of three square meals. At times when the household is yet to remit, we borrow from friends just to ensure we meet our normal daily consumption pattern as has been since the remittances started flowing" [A 37-year-old participant].

Effect of remittances on household living conditions
Moreover, a male household head indicated that the household is now enjoying things they could not enjoy prior to the migration of a member: "We now have television and refrigerator in this house which we could not previously acquire, had it not been this money life would have been same, indeed it is good to travel elsewhere to make a living. My happiest moment with the remittances is that the rest of the children are no longer facing school fees challenges as he (the migrant) did, anytime he remits to the house there would be a special allocation for school fees" [A 45-year-old FDG male participant].
Even though the current findings are similar to an earlier study by Caldwell (1969), the main use of migrants" remittances was to meet household basic needs such as food requirements for migrants" households. This according to the respondents improved their food security and nutritional status. It was observed that in the case of Agona, the investment in productive sectors was limited, but consumption on basic needs was greater. This situation, however, is not necessarily negative, as improvement in livelihood is broadly defined to include access to basic needs, which in turn have positive effects on the well-being of household members (Schiff, 1999;Ballard, 1983). Findings from this research suggest that remittances make a strong contribution to the provision of households" basic needs in the Agona West Municipality. About 80.2% of the respondents admitted that the social status of their households have improved only about 14% of the respondents said that there had not been any change in their social status even though they have some of their family members migrated. This was, however, expected because the type of jobs the migrants are engaged in at the destinations and other factors could play a critical role in the ability of migrants to cause a change in the status of their families. The fact that most of the respondents said their social status has improved could be what de Haas (2007) had observed that only when economic status (poverty level has reduced) of households improves that societal status is seen to also improve.
In the FGDs, it was found that there was some prestige associated with being a member of a migrant household and that the prestige increased with the quantum of remittances.
This was the response a male discussant gave during a FGD session: "When your household migrant remits to your household enough and as frequent as possible, your status in the society is gone up, people tend to associate with you the more and you become highly respected and honoured" [A 50-year old male household head].

Conclusion
A large proportion of rural out-migrants from Agona West Municipality are mostly young adult males within the age category of 20 to 30 years.
Generally, the migrants have various motives for migrating, but most people move from the study areas because of economic reasons such as employment. This is because their level of employment increased after migration since unemployment among the migrants was higher after migration as compared to before migration.
Remittances were frequently sent by most migrants to their households members on monthly bases and mainly in the form of money. The highest amount of money remitted among most migrants to their household members is between 201 to 300 Ghana cedis per month.
A large proportion of remittances received by migrants" households are largely used for household consumption such as food and this lead to improvement in the livelihood status of the migrant"s households in the study areas, the Agona West Municipality.

Recommendations
The following policy recommendations are made based on the findings from the field work, Agona West Municipality Authority should encourage the young people in the study area to take advantage of the numerous social interventions programmes established by government. The National Youth Empowerment Programme aimed at providing jobs for young peopleeither living in the rural or urban areas in the country. The Ministry of Agriculture and Agona West Municipality should ensure effective and efficient support Tanle et al. 17 for rural farmers to make farming attractive to the young people.Future research should be carried out to compare the differences between the livelihood status of migrants" and non-migrants households in other municipalities study areas.