A study of the underlying determinants of return migration of international return migrants to the Berekum Municipality , Ghana

Different migration theories generate competing hypotheses regarding the underlying determinants of return migration. To unravel some of these claims, data from a survey involving 120 return migrants in the Berekum Municipality, Ghana were used to assess the underlying determinants for the return migration of Ghanaian migrants. The study adopted a quantitative approach to research involving simple random sampling technique. The instrument for data collection was an interview schedule, made up of both open and closed-ended questions. The results have shown that the main determinant for their return migration was family related followed by their desire to invest in Ghana. There was networking among the returnees with non-migrants providing information on jobs and investment opportunities in Ghana and the majority of them indicated that their decision to return was as a result of the information received. The study, therefore, recommends that government should make use of the linkages between migrants and their families as well as friends to evolve a policy to attract productive Ghanaian nationals in the diaspora to return home to assist the nation’s forward match towards development.


INTRODUCTION
Although the focus of migration studies has traditionally been on emigration towards Western countries, the issue of return migration has recently been receiving increasing attention in the migration literature (Asiedu, 2005;Rodriguez and Egea, 2006).It was a major theme of the G7 summit of the major industrialized countries held in London in July 1991 (Teitelbaum and Weiner, 1995).Much academic writing on return migration, however, has been focused on the contributions which return migrants can make to economic development in countries of origin (Diatta and Mbow, 1999;McCormick and Wahba, 2001;Thomas-Hope, 1999).There has been comparatively little research on the individual and contextual factors which determine return migration.It is, therefore, not very surprising that most policy measures taken to encourage return migration have failed (Dustmann et al., 1996).The underlying behavioural mechanisms of return migration is an important topic to explore further empirically, since different migration theories offer radically opposed interpretations of return migration (Constant and Massey, 2002).
While emigrations are easily explained by simple static models where the driving force is wage differentials between regions, return migrations occur despite persistently more favourable conditions in the host countries (Stark, 1991;Mesnard, 2004;Zakharenko, 2008).But the question is, in the face of substantial wage differentials, why would migrant in rich countries return E-mail: elijahyendaw@yahoo.com.to their countries of origin which are generally poor?It is, therefore, not clear whether the increasing number of returning migrants to Ghana as observed by the European Commission (2000), Twum-Baah (2005), World Bank (2006) and IOM (2009) is due to the influence of other non-monetary factors other than just economic variables alone as has been presented in simple static models such as neo-classical and the new economics of labour migration theories.
This study, therefore, attempts to fill a gap in the Ghanaian migration literature by providing some quantitative evidence regarding the underlying determinants for the return migration of Ghanaian migrants using the Berekum Municipality as a case.The observations made so far raise the following questions including: Why do migrants return to their countries of origin?And what is the nature of their return decisionmaking process?Based on the objective of this study, it was hypothesized that there is no significant relationship between the background characteristics of returnees and their main reasons for returning home.

Conceptual and theoretical issues
Migration may be defined as a temporary or permanent change in the usual place of residence across space in a given time period (Weeks, 1999).It has time dimensions which are often used to classify migrants.International return migration, which is the focus of this study, refers to the act of a person returning to his or her country of citizenship after having been an international migrant in another country and who is intending to stay in his/her own country for at least one year (United Nations Statistics Division, 1998;IOM, 2004).Within the context of this study, a return migrant has been depicted as any person who has returned to the Berekum Municipality in the course of the last five years after having been an international migrant in another country for a minimum period of five years.Similar studies in Ghana had employed the five-year benchmark in their study samples (Ghana Statistical Service, 2002;Collinson et al., 2009).The essence of this time frame is to allow for comparability of findings in different historical and environmental settings and to enable the returnees recount their migratory experiences without any serious memory relapse.
In the 1970s, the theories on return migration viewed the returnee as a migrant who returned home because of a failed migration experience that did not accomplish the desired outcome (Cassarino, 2004).For instance, the neoclassical migration model viewed the return decisions of migrants as the outcome of a failed migration project which did not yield the expected benefits.In other words, in a neoclassical stance, return migration exclusively involves labour migrants who miscalculated the costs of migration due to imperfect information before departure and who did not reap the benefits of higher earnings.Return occurs as a consequence of their failed experiences abroad or because their human capital was not rewarded as expected.
However, by the 1990s, the focus regarding migrants' reasons for return shifted.Return then was understood as a successful experience abroad where the migrant accomplished the goals of higher income and the accumulation of savings while remitting part of their income to the household; acquisition of higher education, skills, and foreign work experience; as well as the accumulation of social capital in the form of networks, values and attitudes.From the perspective of the new economics model, international migration and return is viewed as a calculated strategy that aims to mitigate credit market imperfections at origin in which migration serves to accumulate sufficient savings to provide the capital, or at least the collateral required to obtain a credit for investment at home, in particular in business activities.Once they have achieved the target level of savings, migrants return to their home countries (Stark, 1991;Mesnard, 2004).One of the most debated issues has been that of human capital gains for emigration countries through the return of migrants (Ammassari and Black, 2001;Hunger, 2004).The human capital model of socioeconomic attainment views migration as a form of investment whereby the individual initiates a geographical move with the expectation of drawing net cumulative gains over his/her working life (Wilson, 1985).
This study adopts the return decision model (Figure 1) developed by Black et al. (2004).The model was adopted because the variables embedded in its analysis were found insightful for the study.That is, the return decision model by Black et al. (2004) articulates most of the factors which motivate the return decisions of migrants.According to the model, the factors influencing the decisions to return include both 'push' and 'pull' factors that are economic, social, personal, and political in scope.It has, however, been observed in the return decision framework that on balance, family and life cycle factors might be more important for returnees than for initial emigration.The main assumption underpinning the model is that the influences on the decision to return come in the form of both information about options and inputs that structure how these options are viewed.Concerning the element of information, the decision to return-as long as it is voluntary is typically made after comparing information about conditions and prospects in the host country with those in the country of origin (Koser, 1998;King, 2000) as well as information about policy interventions in the form of extra incentives or disincentives to stay or return (Bloch and Atfield, 2002).For example, a number of Hong Kongers who migrated to Canada in the 1990s to pursue better economic opportunities re-migrated to Hong Kong after weighing conditions of Canada and that of Hong Kong.That is, while Canada at the time offers a low-risk environment to spend money, Hong Kong offers a better environment to earn money: better jobs, better pay, quicker promotions, fewer and lower taxes, and a better environment for entrepreneurship (Dana, 1996).
Meanwhile, Malmberg (1997), Fischer and Martin (1999), however, observed that nobody is a perfectly 'rational' decision-maker, and different people come to different conclusions even on the basis of the same evidence.One reason, according to Reichnert (2002), relates to individual attributes such as age and gender.Another reason relates to the broader context of social relations including peer pressure to stay or not in the diaspora and the perceived influence that the individual migrant will have on the home country (Black et al., 2004).These are reflected in the model as 'inputs' to the return decision (Figure 1).As shown in Figure 1, the variables required for the study had been broadly catalogued into structural, individual and policy interventions.
Within the context of this study, structural factors include conditions in the country of origin (e.g. more jobs, tax incentives, better wages etc) and the host country (e.g.discrimination, unemployment, difficulty in integration, etc).Individual factors include migrants' sex, age, marital status, education, religious affiliation and social relations.Policy interventions are also composed of extra incentives such as social benefits or disincentives to stay or return such as non renewable of visas.The main advantage of this conceptual framework is that it is simple and easy to understand.It also adequately addresses most of the issues involved in the study.

Study area
Geographically, the Berekum Municipality is located in the Western part of the Brong-Ahafo Region in Ghana.It lies between latitude 7° 5' South and 8.00° North and longitudes 2° 25' East and 2° 50' West.The Municipality shares boundaries with the Wenchi Municipality and the Jaman Municipality to the Northeast and Northwest respectively, the Dormaa Municipality to the South and the Sunyani Municipality to the East (Figure 2 Given the favourable physical characteristics of the area (that is rainfall, temperatures, humidity and soils), the dominant economic activity in the Berekum Municipality is agriculture It employs about 57 percent of the working population (Ghana Statistical Service, 2010).Aside agricultural activities, the people are also engaged in non agricultural occupations such as trading, small and large scale businesses, service related occupations, artisan and a few are into construction and manufacturing.The Berekum Municipality's close proximity to Cote d'Ivoire is one remarkable feature which promotes economic and commercial activities between the Municipality and Cote d'Ivoire.The Berekum Municipality was selected for the study because it is has been noted for international migration and return (Anarfi et al., 1999).That is, international migration is generally considered as an integral part of livelihood and advancement strategies for most families in Berekum (Berekum Municipal Assembly, 2007).

DATA AND METHODS
A survey involving 120 return migrants was undertaken in the Berekum Municipality through the use of an interview schedule.An interview schedule was found suitable for the data collection because it afforded the researcher the opportunity to interpret the questions in the local languages understood by respondents, which otherwise would have been next to impossible using questionnaire.The instrument was developed based on a review of related literature (Thomas-Hope, 1999;Tiemoko, 2003;Black et al., 2004;Asiedu, 2005).However, since migration studies are unique in terms of historical and geopolitical landscapes, the necessary adjustments were made to the instrument in order to situate the research within the local context (which is Ghana).The instrument consisted of five (5) main modules.Module 1 touched on the background characteristics of the respondents, such as sex, age, marital status, occupation, and level of education while Module 2 explored the returnees' migration history which includes their destination countries and duration of stay.Module 3 examined some of the push and pull factors that motivated the respondents to return back to Ghana.Module 4 discussed the return decisionmaking process of the respondents such as their pre-return mechanisms while the final Module which is Module 5 analyzed their future migration intentions.Further, the instrument was structured to include both closed and open-ended questions.
The instrument was pre-tested at Dormaa Municipality between 6 th and 10 th January, 2011 to check its reliability and validity.Dormaa was selected for pre-testing because it has similar socioeconomic and demographic characteristics as Berekum Municipality (Ghana Statistical Service, 2002).The actual data collection was conducted with the aid of three field assistants (trained in data collection and interpretation of questions) between March and April 2011.
The total number of returnees in the Municipality as at the time of the survey was 204.This was obtained through a list compiled during a reconnaissance survey using snowballing.The aim was to generate a sampling frame for the study.Out of the 204 returnees identified, about two-thirds (120) were randomly selected for the study using the lottery method of the simple random sampling technique.A number of studies have arbitrarily used various percentages to determine sample sizes (Agyei-Mensah, 1997;Ghana Statistical Service, 2010).The units of analysis for the study were individual returnees who were aged 18 years and above and who had ever travelled abroad.The rationale for interviewing migrants aged 18 years and above was that in Ghana 18 years is the age of maturity and ability to give consent.
The data analysis was carried out with the use of SPSS version 17.Specifically, descriptive as well as inferential statistical techniques such as frequencies tables and chi-square were employed to process and analyze the data.For instance, the chisquare test statistic was engaged to determine the level of influence of the respondents' background variables (such as sex, age, marital status, education, occupation) on their main reasons for returning to Ghana.The background characteristics of the respondents were regarded as the independent variable while the main reasons for return were treated as the dependant variable.One main challenge encountered in the study was that there was no database (sampling frame) on return migrants in the Municipality.

Socio-demographic characteristics of the respondents
The results (Table 1) indicate that the respondents were mostly males (83%), who were young (68 per cent were 20-39 years) and were married (50%).This is consistent with previous findings by Anarfi et al. (2003) who opined that most return migrants to Ghana were young and were in their active ages who could be useful for the socioeconomic development of the country.The fact that half of the respondents were married was expected in view of the observation that a large proportion (68%) of them were aged between 20-39 years, the age at which it is considered ideal for people to marry (Anarfi et al., 2003).The analysis, however, appears to be at odds with Zlotnik (2003) and Twum-Baah (2005)'s observation that feminized migration is increasing in Africa as a result of higher levels of education for women and changing social norms.The disparity in male-female ratio could, however, be explained by what Anarfi et al. (1999) had observed that as custom requires, most females prefer to stay behind while their male partners emigrate and remit home.
The results further showed that a higher proportion (42%) of the returnees had attained Senior High/Vocational/Technical education, while about a quarter had Tertiary level education.The respondents were mostly Christians (91%) which is in consonance with results from the 2010 Population and Housing Census report of Ghana which indicated that majority of Ghanaians were Christians (69.5%) (Ghana Statistical Service, 2010).The results suggest that about six out of ten return migrants were more likely to be traders or artisans.Some returnees did not stay long at their various destinations while others did.Table 1 indicates that about eight out of ten returnees mentioned that they stayed at their destination for between five and nine years while 25 per cent said they stayed for a period of ten years or more.

Destination country and duration of stay by sex
Table 2 shows that the most preferred destination of the return migrants was Libya (27%) followed by Germany (17%).Almost the same number of respondents travels to Germany (39%) and the UK (38.7%) but a higher number of males (29.3%) than females (14.3%) travel to Libya (Table 2).The fact that majority (27%) of the return migrants from the Berekum Municipality travelled to Libya might be the case where most young people from the Brong Ahafo region sojourn through the Sahara desert and the high sea under harrowing conditions with the sole aim of entering European destinations such as Italy and Spain through Libya.The above finding is also consistent with the observation that the Brong Ahafo region is one of the most affected localities in Ghana noted for irregular migration to Libya (Awumbila, 2007;Tanle, 2012).However, the results have shown that female returnees were numerous than male respondents who travelled to the UK (28.6%), Germany (23.8%), Canada (4.7%), and Spain (4.8%).This might be explained by the fact that most economic activities reserved for labour from developing countries are menial and social work in nature which mostly favours female migrants such as cleaning and hotel businesses.Meanwhile, comparing the number of respondents who returned from European and American destinations to that of African destinations, the analysis (Table 2) indicates that cumulatively most emigrations from the Berekum Municipality came from European and American destinations.This confirms Twum-Baah's (2005) observation that recent political crisis and changes in the fortunes within the sub-region have reduced the significance of the intra-regional migration streams, in favour of American and European destinations.Some returnees did not stay long at their various destinations while others did.Table 2 indicates that about eight out of ten return migrants mentioned that they stayed at their destination for between five and nine years while 25 per cent said they stayed for a period of ten years or more.A higher percentage of males (28.3%) compared with females (10%) stayed for 10 years or more (Table 2).The current observation might be due to the conjugal and reproductive roles of females where they are sometimes compelled to return home, for instance, to get married or join a spouse at home.

Reasons for return migration
The influence of return migration on socio-economic development in the country of origin, according to Ammassari and Black (2001), varies significantly depending on the motives for return.This section of the thesis highlights the respondents' reasons for return to Ghana and how the return decision was arrived at.Background variables such as sex, level of education and marital status are used in the discussion to show how they influence the return decisions of the respondents.

Main reasons for return by sex
Table 3 shows that family related factors (27%) were the main determinants for the return migration of the respondents followed by their desire to invest their savings (19%).But across the various reasons for return, the analysis revealed that a higher percentage of females (39.4%) than males (24%) returned home because of family related factors.The differences between males and females regarding their main reason for return (family) could be that naturally women's maternal, domestic and conjugal roles are such that they are sometimes compelled to return which explain why more females than males returned for the purpose of family reasons.The fact that family reasons featured prominently among their return reasons was consistent with the return decision model by Black et al. (2004), Tiemoko (2003), King (2000), Black and Ammassari (2001) where family reasons and life cycle factors were amongst the most important reasons influencing return including family ties and the wish to rejoin family and  friends.It also buttresses the notion of the new economics of labour migration that not individuals but rather families or households are the main decisionmakers with respect to migration and return (Stark, 1991).Also, concerning respondents who reported that the decision to return was because they had accumulated sufficient capital for investment, 20.8% were males while 9.1% were females.The disparity between male and female respondents regarding their desire to invest their savings in Ghana could be attributed to the fact that naturally males are expected to fend for themselves, which explains why more males than females migrated in search of financial capital for investment and who returned to invest their savings and to work at home.The above evidence conforms to the new economics of labour migration literature where migration may serve as a means to accumulate sufficient savings to provide the capital, or at least the collateral required to obtain a credit for investment at home, in particular in business activities (Stark, 1991;Mesnard, 2004).The fact that a significant proportion of the returnees returned to invest their savings and to work at home is a good opportunity for the nation as it could raise the productive capacity of the country and generate further capital.

Main reasons for return by marital status
Marital status could influence a person's decision to return home.Table 4 reveals that marital status plays a major role in the respondents' decision to return home.For instance, among respondents who indicated that their return decision was motivated by family related factors (27%), the results showed that about six out ten returnees were widowed, nearly 30% were never married, 26.3% were divorced while about 22% were married.The fact that a significant number of those who were widowed cited family factors as their main reason for return could be due to reasons such as bereavement, remarriage and their desire to care for the family at home, which explain why most return migrants who were widowed returned because of family related issues.For the never married category, the return might be possibly due to their desire to get married and settle down for good.
The results further showed that among those who returned because of investment (19%), about 24.4% were married followed by those who were widowed (20%) and those who were never married (14.1%).The fact that married return migrants were numerous among respondents who said they returned to invest their savings at home could be attributed to their determination to insure the future of their families such as their children and spouses.Also, concerning respondents who reported that they returned because they ended their training/study, the results indicate that 7.0% were never married, 6.1% were married, while none was divorced or widowed.The above revelations were expected because people who are single are more likely to be less restricted  or less burdened to travel out for educational purposes.That is, the never married group are mostly considered to be mobile and do not have spouses or children to think about when they intend to emigrate for further education.

Main reasons for return by level of education
Table 5 indicates that level of education plays a significant role in the return decision of the respondents.For instance, among the return migrants who returned because of their desire to work at home (17%), the results showed that 23.1% were returnees with tertiary level education; and of respondents who returned because they had ended their training or study abroad (5.6%), the analysis revealed that most of them (15.4%) had tertiary education.This implies that migrants with higher level of education are more likely to return and work at home than those with lower level of education possibly because of their expectation of higher earnings back home after their return.The above evidence finds credence in the neoclassical approach to return migration where the individual migrant returns because of the human capital accumulated in the host country achieves higher (relative) returns at home than at the destination.An example, according to Dustmann (2000), is student migrations, where the level of human capital obtained would situate the individual within the average group in the host country, but would place the returnee among the educational elite at home and may increase expected income at home sufficiently to trigger return.Furthermore, those respondents who returned because of unemployment (9.6%), integration difficulties (7.3%) and government restrictions (13.6%), the analysis signifies that returnees with higher level of education were less affected by those factors.Concerning unemployment as a reason for return for instance, 15.4% were those with primary school education, 12.2% had junior high school education, 7.9% had senior high school education while a few (7.7%) had tertiary level education.

Decision to return
One of the objectives of this thesis was to assess how the decision to return was arrived at.For all return migrants, a decision was taken at one point in time to return home based on certain considerations.The decision to return may be influenced by households, friends or initiated by the individual migrant (Tanle and Awusabo-Asare, 2007).The decision to return, according to Black et al. ( 2004), may also be triggered depending on the type of information received about existing conditions at the country of origin.

Information and source before return
Results from the study showed that about 85% of the return migrants said they had information about Ghana before their return while 15% said otherwise.For those who received information, Table 6 indicates that their main sources of information was from friends (42.2%), followed by parents (24.2%).This evidence implies that on a whole, family and friends were the main agents of influence for returning home which supports the notion of the new economics of labour migration approach that not individuals but rather households and friends are the main decision-makers with respect to migration and return.The findings are also in consonance with other evidence by Tiemoko (2003) and Cassarino (2004) who opined that families as well as friends are the main sources of information on the decision to return home; providing insights on jobs, legal matters, social tensions and security.Overall, the analysis implies that most of the respondents while abroad showed interest in what was happening in the country as they knew they would return home one day.

Type of information received and influence on decision to return
Regarding the type of information received before their return, Table 7 indicates that more than a quarter (35%) of the respondents had information on investment opportunities followed by those who had information on job opportunities (25.5%).On the question of whether they were influenced to return by the type of information received about Ghana, the results showed that 85% of the returnees indicated that they were influenced by the information received while a few (15%) said their return was not motivated by the kind of information obtained about the country.This implies that migrants while abroad rely on information about the origin country to make decisions on whether to stay abroad or return home.The above findings support the observation made in the conceptual framework (Figure 1) which indicates that migrants' return decisions are made after comparing information about conditions and prospects in the host country with those in the country of origin.

Main reasons for return by background characteristics of respondents
As observed in the return decision model (Figure1), background variables (such as sex, age, education, marital status etc) may play a significant role to the return decision of migrants living in host countries (Black et al., 2004).To unravel this claim, a hypothesis was set that there is no significant relationship between respondents' background characteristics (such as sex, age, education, marital status and religious affiliation) and their main reasons for returning to Ghana.A chi-square test of the above hypothesis showed some interesting statistical results as shown in Table 8.While some of the background variables such as sex, age and level of education indicated a significant relationship between the respondents' reasons for returning to Ghana; others such as marital status and religious affiliation showed no significant relationship with the respondents' reasons for returning home (Table 8).
The hypothesis that there is no significant relationship between the sex of returnees and their main reasons for returning home was rejected.The chi-square statistic indicated a significant relationship (X 2 =11.735,P=0.042) between sex and reasons for return migration.Regarding the sex of the respondents and their main reasons for returning home, the results (Table 8) showed that 16.1% males as against 10.3% females came back home because they had ended their study/training, 15.2% males as against 4.0% females returned because of unemployment, 20.3% males and 47.2% females said their return was because of family related factors, 10.1% males and 3.7% females mentioned government restrictions at destination while 12.1% males as compared to 2.4% females indicated their return decision was because they have accumulated sufficient financial capital for investment in Ghana.
The hypothesis that there is no significant relationship between the age of returnees and their main reasons for returning home was also rejected.The chi-square (X 2 =  37.882, P=0.013) at an alpha level 0.05 indicated significant relationship between the age of the returnees and their main reasons for returning to Ghana (Table 8).This implies that return migrants' reasons for returning to Ghana can be predicted or explained in terms of their age.For instance, among those respondents who reported that their return decision was due to family related reasons, a significant proportion (59.5%) was aged between 20-39 years, the age at which marriage is considered ideal (Anarfi et al., 1999).Also, regarding the respondents who indicated that their return decision was due to government restrictions at the destination, about a quarter (26.7%) were young (20-39) while those aged 50 years or above were not affected by government restrictions.This was expected because young people are more likely to engage in migration related offences than elderly people.The educational level of the respondents has also influenced the return motivations of the returnees (Table 8).For example, a chi-square test on the hypothesis that there is no significant relationship between the educational level of returnees and their main reasons for returning was rejected.At an alpha level of Yendaw 305 0.05, the chi-square statistic (X 2 = 33.662,P=0.039) indicated a significant relationship between their level of education and the main reasons for returning to Ghana.This implies that return migrants' reasons for returning to Ghana can be explained in terms of their level of education.For example, among respondents who reported that their return to Ghana was because they had ended their training/study, a large number of them (23.1%) were those with tertiary level education followed by those with senior high/middle school education (4.0%).Meanwhile, none of the respondents with primary level education returned home because they had completed their study or training abroad.The above observation was further expected because migrants with lower level of education might find it extremely difficult to further their education abroad.Also, with regard to returnees who cited integration difficulties as their main reason for returning home, a higher percentage (15.5%)had primary level education followed by those with junior high/middle school education, implying that migrants with lower level of education are most likely to be affected by integration difficulties at destinations.With respect to returnees who also came home because of unemployment, a large proportion (49.3%) were returnees with primary level education followed by junior high/middle school education (16.7%).This was further expected because migrants with higher education have employable skills than those with no or lower level education.
The results (Table 8), however, did not reveal any significant relationship between marital status, religious affiliation and the returnees' reasons for returning home.For instance, among respondents who reported family related factors as their main reason for returning home, a higher percentage were widowed (65.5%) followed by respondents who were divorced (35.7%).The chi-square results (Table 8) gave X 2 = 24.223, P=0.282 at an alpha level of 0.05 indicating no significant relationship between the respondents' reasons for return and their marital status.The current observation, however, might be due to factors such as selection bias and other confounding in the data collection and analysis.In terms of their religious affiliation and their main reasons for return, no significant relationship was depicted.This shows that the religious affiliation of migrants do not influence their return decisions.Meanwhile, the above results should be extrapolated with caution and replicated with care.

Intention to travel again in future and reasons for intending to travel again
For various reasons, some return migrants may decide to migrate again while others may not.The respondents were asked whether or not they intended to migrate abroad again and if yes the reason for intending to travel again.Results from Table 9 revealed that about eight out of ten respondents (80%) intended to travel again while a small number (20%) said otherwise.Out of the 95 respondents who said they intended to travel abroad again, more than half (57%) of them said the decision was economic followed by those who cited adaptation difficulties (19%).The fact that a higher proportion of the returnees intended to re-emigrate again might be consistent with what Cassarino (2004) and Colton (1993) had observed that returnees may not be able to reintegrate if the 'gulf' between norms, values, and opportunities at origin diverges from that of migrants' expectations about their activities after return.This, according to Cassarino (2004), occurs when migrants are not able to remain well informed about the economic, social and political situation at home during their stay abroad.

DISCUSSION AND CONCLUSION
The study examines the underlying determinants of the return migration of international migrants to Ghana using Berekum as a case.Specifically, the study analyzes the factors responsible for their return and the nature of their return decision-making process.The study has shown that about 83% of the returnees were males who were mostly young (20-39 years) and were married (50%) and a significant proportion of them (36%) were engaged in trading or business activities.The fact that a higher proportion of the return migrants were young suggests that they are not returning home for retirement but are actively working and therefore could be useful to the country's socio-economic development.
The main destination country of the returnees was Libya (27%) followed by Germany (17%).This was expected because migration to Libya in the Brong Ahafo Region in general is a common phenomenon as observed by Awumbila (2007) and Tanle (2012).Regarding the length of time spent abroad, the results revealed that a higher percentage of males (28.3%) compared to females (10%) had the longest duration of stay abroad (Table 2).This observation however, might be due to the conjugal and reproductive roles of females where they are sometimes compelled to return home, for instance, to get married or join a spouse at home.
Seven main underlying reasons accounted for their return migration to Ghana which included family related reasons (27%), desire to invest their savings (19%), the wish to work at home (17%), government restrictions at destination (14%), unemployment (10%), end of study (6.0%), and integration difficulties (7.3%).It was however, revealed that female returnees who were widowed were more likely to respond to family related factors for return.The fact that family related reasons featured prominently among the respondents' reasons for returning home supports the basic assumptions of the conceptual framework for the study (Figure 1) where family and life cycle factors were emphasised as the most important determinants for return migration.The results are in line with studies by Ganguly (2003) and Hazarika et al. (2011) who observed that family-related factors for return are of particular importance to the decisions of most migrants including family issues such as bereavement, strong family ties, homesickness, the wish to rejoin family and friends, the wish to get married from home, wanting to have their children brought up in their home culture and going home to care for elderly parents.It was, however, observed that the current evidence appears to be at variance with findings by Dana (1996) who stated that the return decisions among his study samples were purely motivated by severe political and economic forces, including adverse entrepreneurial environment, which at least partly explained the occurrence of the boomerang phenomenon observed at the time.The variation, however, in the two findings could be due to factors such as differences in the two countries, the focus of the two studies, the target population involved and the methods used.
It was further discovered that a significantly higher proportion (85%) of the respondents had information about the economy of Ghana before their return and friends (42.2%) as well as parents (24.2%) were their main sources of information for returning home.This to a large extent indicates the respondents' lack of independence in taking their own return decisions.

Individual return decision
Conditions in the host country  Concerning the nature of the influence, more than a quarter (35%) said they were attracted by investment opportunities in Ghana followed by employment opportunities (26%) which goes to confirm findings by Dana (1996) who observed that most Hong Kongers entrepreneurs in Canada returned back to Hong Kong to take advantage of better economic opportunities such as better jobs, better pay, quicker promotions, fewer and lower taxes, and a better environment for entrepreneurship.On the issue of whether the decision to return was influenced by the type of information received, over 80% indicated that their return decision was the result of the information obtained about the nation.Results from a chi-square test statistic (Table 9) showed that background variables such as age, sex, and the educational level of the returnees were significantly related to their main reasons for returning to Ghana.Meanwhile, no significant relationship was, however, found between marital status, religious affiliation and the returnees' reasons for returning home.For instance, among respondents who reported family related factors as their main reason for returning home, a higher percentage were widowed (65.5%) followed by respondents who were divorced (35.7%).What this implies is that age, sex and educational level of migrants abroad should be considered seriously when designing policies to attract migrants to return home.It was, however, discovered from the results that about eight out of every ten return migrants (80%) intended to travel abroad again in future and the main reason cited was economic related (57%) followed by difficulty adjusting in Ghana (19%).

POLICY RECOMMENDATIONS
The study has revealed a number of interesting results based upon which useful policies could be developed.In line with the main findings of this paper, the following policy recommendations are made.Among other things, the study has revealed that a high proportion of the returnees were in their active ages (young).This suggests that they are in their most economically active ages and could therefore put their experiences to useful purpose.It is therefore recommended that the country's policy makers should factor return migrants into the scheme of things in Ghana's resolve in achieving the various Millennium Development Goals.
Furthermore, the study has shown that for the majority of the returnees, contacts were regularly made with their family members and friends in Ghana providing information on jobs and investment opportunities and an appreciable number of the returnees reported that their return to Ghana was as a result of the information Yendaw 309 obtained on investment and employment opportunities.The possibility is, therefore, there for government to use the linkage between the migrants and their families as well as friends to evolve a policy to attract a large number of productive Ghanaian migrants as possible back home to assist the nation's forward match towards a modern society.In particular, special governmental effort should be made to attract investors from among Ghanaian emigrants abroad.For instance, programmes such as access to credit facilities, tax incentives, social assistance, and sound investment climate are but a few that could be pursued.Despite the developmental potential which the returnees represent, the study has revealed that for eight in every ten return migrants, the decision was to return home temporarily and to go back and the main reason cited was economic related followed by difficulty adjusting in Ghana.The long term implication is that some of the returnees are likely to become part of the brain drain currently experienced in the country.The study therefore recommends that all NGOs aimed at providing economic support for migrants should implement such projects at their places of origin and not at the destination because it is the economic deprivation at home, more than anything else, which compels them to migrate abroad.More so, all stakeholders in the field of migration should disseminate through nationwide mass information campaigns the risks and realities of international migration.This is important because it will help reduce future brain drain in the country.
Moreover, future research regarding the underlying determinants of return migration of Ghanaian migrants should be replicated in other cities of the country in order to draw comparative analysis and generalization for the whole nation.Meanwhile, more detailed research could be carried out on each of the identified element that contributed to the return migration of migrants.This will help determine the relative importance of each of the factors which motivates the return of migrants to their countries of origin.

Figure 2 .
Figure 2. Geographical location of Berekum Municipality.Source: GIS unit of the Department of Geography and Regional Planning, UCC.

Table 1 .
Socio-demographic characteristics of the respondents.

Table 2 .
Destination country and duration of stay by sex.

Table 3 .
Main reasons for return by sex.

Table 4 .
Reasons for return by marital status.

Table 5 .
Main reasons for return by highest educational level.

Table 6 .
Information and source before return.

Table 7 .
Type of information received before the return.

Table 8 .
Main reasons for return migration by socio-demographic characteristics of respondents.

Table 9 .
Intention to travel again in future and reasons for intending to travel again.