ARTICLES

 

Internal Migration, Socio-Economic Status and Remittances: Experiences of Migrant Adolescent Girl Head Porters in Ghana

 

 

Kennedy A. Alatinga

Department of Community Development Faculty of Planning and Land Management University for Development Studies, Ghana. Email: kalatinga@gmail.com

 

 

---

ABSTRACT

This paper examines the relationship between internal migration, socioeconomic status and remittances, drawing on the experiences of migrant adolescent girl head porters in the cities of Accra and Kumasi in Ghana. Through an explanatory sequential mixed methods research design involving 503 individual surveys and 24 in-depth interviews, the paper established that 61% of adolescents studied migrated from the north to the south to escape poverty while 29% migrated to work and raise money for school fees. The kayayoo business does not seem to capacitate the adolescent girls to live the kind of lives they have reason to value because only 36% of them remitted to their families. A logistic regression model showed that older adolescents, 1519 years, (AOR=7.32, p<0.05, CI= [1.999-26.802] number of years spent working as head porter, 3-years, (AOR=3.97, p<0.05, CI= [1.633-9.677] and socio-economic status - not poor (AOR=8.63, p<0.001, CI=[4.761-8.435]) significantly influenced remittances. Remittances capacitated recipient families to invest in human capital development and also improved household food security. This study recommends that, in the short-term, adolescents working as head porters to raise money for school fees must be identified and enrolled in schools, based on the Ghanaian Free Senior High School Policy, while the establishment of factories and industries in Northern Ghana to create employment opportunities could be a long-term measure.

Keywords: Capabilities-aspirations, human transport, kayayoo, females, remittances, adolescents, Ghana.

---

 

 

Introduction

Migration remains a critical part of human existence. Migration, whether internal or international, is a public policy concern and is perhaps the most visible sign of globalization in the 21^st century (Goldin and Reinert, 2012; IOM, 2018). In developing countries, internal migration is largely a rural-urban phenomenon. This trend is justified on the hypothesis that urban residents have access to better infrastructure and services, including employment opportunities, health care, education and are generally more prosperous than rural dwellers (Rice, 2008; White, 2016). Urbanization is often equated with modernization, and socio-economic development (Cyril et al., 2013). Contemporary migration trends in developing countries, including Ghana, reveal that the migration process is dominated by youth aged 18-24 years (White, 2016), especially adolescent girls (Temin et al., 2013). Migration may offer adolescent girls who migrate voluntarily the social space for socioeconomic advancement through access to education, access to quality health services, and remittances to reduce household poverty (Mac-Ikemenjima and Gebregiorgis, 2018; Nyasulu, 2017).

Migration trends in Ghana in recent times reveal the movement of predominantly adolescent girls aged 10-19 years largely from the poorer northern regions of Ghana to the richer urban markets of Accra and Kumasi with the aim of improving their socio-economic status (Awumbila and Ardayfio-Schandorf, 2008; Tufeiru, 2014). Opare (2003) claims that most families and households mobilize financial resources, and some even borrow money to finance the adolescent girls' travel to Kumasi and Accra. In the cities of Accra and Kumasi, most of these adolescent girls engage in the head-load carrying (kayayoo) business. The kayayoo 'business' is the situation in which adolescent girls literally use themselves as human transport by carrying loads of goods on their heads for unregulated fees to save money for later investment and remitting to their families back home (Agarwal et al., 1997; Tufeiru, 2014). It appears thus, that the kayayoo business in the urban market centers is a household survival strategy in response to poverty (Awumbila and Ardayfio-Schandorf, 2008). Thus migration becomes a conduit for the adolescent girls to aspire to change the poor socio-economic status of their families for the better.

However, there is currently little evidence on whether these adolescent girls' aspirations of migration are met. For example, the adolescent girls' ability to remit, the extent to which the kayayoo business improves the adolescents girls' socio-economic status, financial situation, and their overall poverty situation and that of their families appear not to have received adequate scholarly attention. From a capabilities perspective, the extent to which the kayayoo business increases the capacity of the adolescent girls to be more secure and live the kind of lives they have reason to value, is also not known. In terms of income potential, head porters are reported to have low levels of earnings. For example, Awumbila and Ardayfio-Schandorf (2008, p.177) and Oberhauser and Yeboah (2011: 22) found that female head porters are among the poorest of urban dwellers in Accra, with an estimated average daily earnings ranging from US$ 1.20 to US$ 2.20 and US$ 3.20 to US$ 5.20 on a good day. The female head porters live and work under deplorable and harsh conditions including poor accommodation, which have implications for their health (Nyarko and Tahiru, 2018; Opare, 2003). It is in these trying circumstances that the adolescent girls aspire to make money to finance their living costs, save and remit to their families in order to make a positive difference.

Examining whether or not the kayayoo business capacitates these female youngsters to achieve their aspirations is of crucial public social policy relevance because they are of school-going age and should normally have been in school. The Sustainable Development Goal Four (SDG 4) is loud and clear as it calls for inclusive and equitable quality education for all (UNGA, 2015: 17).

The purpose of this paper is two-fold. First, it interrogates the adolescent girls' ability to remit, their socio-economic status, financial situation, and their overall poverty situation and that of their families in the context of the kayayoo business. Secondly, it makes a novel methodological contribution by deploying a mixed methods research design, drawing on the centers of aggregation sampling technique and purposive sampling, which are suitable for hard-to-reach populations such as the female head porters (Reichel and Morales, 2017). Earlier studies on the subject matter have been qualitative in nature, relying largely on intuition and convenience to draw their samples (Baah-Ennumh et al., 2012; Nyarko and Tahiru, 2018). Section two presents a clear picture of the kayayoo business in Ghana, followed by the theoretical and conceptual literature in section three. The research setting and methodology are presented in section four. Section five exemplifies the study results and discusses them. Section six concludes the paper and makes recommendations.

 

Adolescents Girls on the Move: The Case of Ghana's Head Porters

Ghana's population is increasingly becoming urbanized. With an urban growth rate of 4.25%, Ghana's current urban population is estimated to be about 56%, making Ghana the third most urbanized country in West Africa (UN, 2018). The proportion of the urban population is further projected to increase to about 63% by the year 2025 (Awumbila et al., 2014: 6). Scholars have pointed to several reasons that explain Ghana's adolescent girls' movement from the north to the south to engage in head-load business - including high levels of poverty in the three northern regions, conceptions of female occupational career structure, and Ghana's urban light goods transportation structure.

High levels of poverty compel people to migrate from the three northern regions of Ghana (Upper West, Upper East and Northern Regions) to the two largest urban cities: Accra - the national capital, and also the capital of the Greater Accra Region, and Kumasi - the capital of the Ashanti Region, in search of jobs and other economic opportunities (Osei-Boateng 2012). These three northern regions also have the highest incidence of poverty in Ghana. The latest Ghana Living Standards Survey (GLSS7) reports that poverty in the three Northern regions worsened, from 70.7% in 2012/13 to 70.9% in 2016/17, in the Upper West Region, from 44.4% in 2012/13 to 54.8% in 2016/17 in the Upper East Region, and from 50.4% in 2012/13 to 61.1% in 2016/17 in the Northern Region respectively (GSS, 2018). During the same period, the poverty situation improved in the Ashanti and Greater Accra Regions: poverty declined from 14.8% in 2012/2013 to 11.6% in 2016/17 in the Ashanti Region, and from 5.6% in 2012/13 to 2.5% in the Greater Accra Region, compared to the national average of 23.4%.

The conceptions of female occupational career structures, particularly in Northern Ghana, appear to fuel the migration practice of adolescent girls. For example, in Ghana, kayayoo is considered a petty form of trading which is culturally understood as being women's work, and so head-load carriers are self-employed, informal sector workers (Agarwal et al., 1997). The authors indicate that the engagement of female children in kayayoo is considered as part of the female occupational career structure. Agarwal et al. (1997) argue that adolescent girls and women engage in kayayoo in order to build up capital to invest later in assets such as sewing machines and more rewarding occupations. Besides, it is less expensive to invest in kayayoo than other informal sector businesses.

The use of human transport in the form of adolescent girl head porters can also be understoond as part of the light goods transport structure in Ghana's urban markets. Commerical traders and some shoppers in Ghana's urban markets engage the services of adolescent girl head porters to transport their goods from one point to another (Agarwal et al., 1997). Inappropraite urban physical planning and design resulting in overcrowding and congestion make the use of human transport within markets more convenient as compared to motor vehicles (Agarwal et al.,1997; Ardayfio-Schandorf et al., 2012).

 

The Theory of Migration: Towards a Conceptual and Theoretical Framework of Aspirations and Capabilities

Historically, colonial development policy was essentially urban-biased and Northern Ghana has long served as a labour reserve for the south (Thomas, 1973). Investments in modern health care and educational facilities were undertaken in Southern Ghana to the detriment of the north. After independence, Dr Kwame Nkrumah, the first President of Ghana, embarked on a rapid industrialization drive to usher in egalitarian development and eliminate the north-south development gap created by the colonialists (Adarkwa, 2012). However, Nkrumah's industrialization policy concentrated particularly in the south, in the 'golden triangle' - Accra-Tema, Kumasi and Sekondi-Tarkoradi - resulting in an unsuccessful policy (Adarkwa, 2012).

The migration of adolescent girls to Ghana's urban cities can be understood within the context of the north-south development gap. Existing research such as Ravenstein's laws of migration (Ravenstein, 1885), Lee's push and pull factor theory (Lee, 1966), Todaro and Smith's rational choice theory (Todaro and Smith, 2012), and Stark's migration decision-making model (Stark, 1984), amongst others draw largely on development and economic theories or models to explain migration. The gist of these theories is that economic reasons fuel migration. For example, Ravenstein (1885) argues that migrants drift from areas of low economic advantage to more endowed areas. Lee's (1966) 'push and pull' theory suggests that 'push' factors such as poverty, the lack of job opportunities, poor social services and inadequate infrastructure in rural areas, and 'pull' factors such as better job opportunities and improved infrastructure in the urban areas are responsible for rural-urban migration. Explaining the economic motives for migration further, Todaro and Smith (2012) suggest that the decision to migrate is calculated rationally based on expected urban-rural real wage differentials. Migrants expect higher wages in urban areas than in rural areas, which are usually agriculturally dominated. According to Stark (1984), rural-urban migration is a survival strategy where migrants under favorable conditions accumulate wealth and send remittances to their families. Remittance is a key concept in the migration literature because available evidence suggests that migration and remittances reduce poverty, improve living conditions, improve household welfare and contribute to human development in origin communities (Abdulai et al., 2017). Other researchers (see for example, de Haas, 2014; Kleist and Thorsen, 2017,)) also argue that poverty and economic factors alone cannot explain migration since migrants need a certain minimum of financial, human and social resources in order to migrate. The need for resources to migrate speaks to two important issues. First, it speaks to the fact that migration is selective because migrants have different abilities and respond differently to sets of circumstances triggering their movement (Lee, 1966). Carling (2002) argues that migration is restricted by poverty, illiteracy and lack of planning in the lives of the poor and disadvantaged. Secondly, if poverty alone were the trigger for the movement of the adolescent girls, one would expect that the volume of movement would be reduced because of the resources and the initial financial costs needed to migrate. On the contrary, Opare (2003) reports that the number of adolescent girl head porters keep rising every day in the urban markets. This scenario foregrounds other issues, such as aspirations and opportunities, in addition to poverty, that may occasion the movement of the adolescent girls (Mac-Ikemenjima and Gebregiorgis, 2018). The different views espoused by researchers and the multifaceted nature of migration suggest that migration cannot be explained by a single theory.

This paper therefore, draws on the capabilities-aspirations theoretical model to complement the main economic theories with a capabilities-aspirations theoretical conceptualization to explain the migration of the adolescent girls.

This conceptualization aims to advance a comprehensive understanding of the north-south migration of the adolescent girls in Ghana. Drawing on Sen's capability approach, human capabilities are regarded as the substantive freedoms people have or enjoy to live the kind of lives they have reason to value (Sen, 1999). The range of human capabilities is diverse and varies from elementary freedoms encompassing being free from hunger to complex abilities such as achieving self-respect, irrespective of one's socio-economic status and gender (Sen, 1989). From this perspective, migration entails people's capability or freedom to choose where to live or to stay (de Haas, 2014). The concept of capabilities used in this context demonstrates the ability of individuals to achieve meaningful outcomes for themselves and their families through migration (Preibisc et al., 2016). Aspirations are more attitudinal or psychological in nature, including migration desires, needs and intentions (Carling and Schewel, 2018). Aspirations depend on people's life preferences and perceptions about opportunities and life elsewhere, shaped by culture, personal disposition and information (de Haas, 2014). According to Kleist (2017), aspirations present an excellent starting point for appreciating the link between migration and young people's sense of the possibilities they have for their lives. This capabilities-aspirations framework is justified because migration first involves the wish to migrate, and second, the realization of that wish (Carling, 2002: 6). In this sense, some people may aspire to migrate but may never realize their dreams because they lack the ability do so, thus becoming involuntarily immobile (Carling, 2002; Carling and Schewel, 2018).

Furthermore, some scholars conceptualize migration as a function of aspirations and capabilities within a given set of opportunity structures (de Haas, 2014). This means that structural conditions in society may shape and offer opportunity structures that enable individuals and groups to pursue and achieve their personal or communal aspirations and capabilities (Merton, 1968; Preibisc et al., 2016). In societies where people are constrained by the lack of, or inadequate opportunity structures, adventurous individuals or groups may migrate to perceived opportunity-abundant locations in order to achieve their ambitions and aspirations (Opare, 2003). Put differently, where local opportunities do not allow people to live the kind of lives they have reason to value, they may migrate once they have the capability to do so. Thus, the lack of these opportunity structures constitutes impediments to individuals' aspirations and their capacity to live the kind of lives they have reason to value. Mac-Ikemenjima and Gebregiorgis (2018) report that the migration of Eritrean youth is largely due to lack of appropriate institutional arrangements to achieve their aspirations and limited opportunities for employment.

Deploying the aspirations-capabilities framework, de Haas (2014) observed that despite significant increases in income and the improvement in general living conditions between 1998 and 2000, rural-urban migration in Morocco had continued unabated. Based on this observation, de Haas (2014: 16) argues that neo-classical push-pull economic models failed to explain migration adequately because it is an integral part of social change, and "to understand society is to understand migration, and to understand migration is to better understand society". Thus, relying on economic push-pull factors alone reduces migrants to objects that lack capabilities, perception and are deprived of social relations (de Haas, 2014). Theoretically, the aspirations-capabilities framework explores the fact that migration may be valued, and yearned for in its own right. Adventure, curiosity or experience and the need to be independent from parental control may drive teenagers to migrate (Carling and Schewel, 2018).

Viewing migration beyond economic factors also allows one to distinguish clearly between the instrumental (where migration is a means to achieve an end, such as earning higher incomes, higher social status, better education) and the intrinsic (value attached to the migration experience, the joy and pleasure derived from exploring new societies) nature of migration (de Haas, 2014). People may migrate not just to achieve material and social resources and lifestyles but because of wanderlust and a desire to discover new horizons (de Haas, 2014). Consistent with Carling and Schewel (2018), de Haas (2014) suggests that the youth demonstrates an innate desire to migrate for varied reasons, including the psychological need to separate from their parents, to prove their independence and satisfy their curiosity. Adolescents and youth may also aspire to migrate to meet future partners and acquire basic assets in preparation for marriage. For these reasons, migration may be considered a fundamental capabilities-enhancing freedom in its own right (de Haas and Rodríguez, 2010). Opare (2003) argues that the current phenomenal rise in the rate at which teenage girls and young women migrate southwards from the north is in fulfillment of their aspirations.

 

Research Setting and Methodology

This research was conducted in the Accra and Kumasi metropolitan areas. Accra and Kumasi are the biggest and most urbanized cities in Ghana where the adolescent girls usually go to engage in the kayayoo business. While Accra is located along the coast, Kumasi is centrally located and serves as a traversing point from all parts of the country. Major financial institutions and government ministries also operate in these cities. Accra contributes about 2030% of the national Gross Domestic Product (GDP) and has the highest average per capita income in Ghana while Kumasi has the second highest average per capita income (Oberhauser, 2011). These two cities are also endowed with modern health care and educational facilities.

Prior to data collection, the author visited the research sites and met with the leadership of the Kayayoo Association in Accra and some key informants in both Accra and Kumasi who have been working with the head porters as part of a formative study. Thereafter, research instruments - a questionnaire for a cross-sectional survey and an in-depth interview - guide were developed.

An explanatory mixed methods research design was deployed in collating and analyzing data. Because hard-to-reach populations and migrant populations are characterized by the difficulty in sampling from them using standard probability methods (Gile and Handcock, 2010), quantitatively, time-location sampling was considered appropriate for generating random estimates for the migrant adolescent girl head porters (Reichel and Morales, 2017). Additionally, there are no accurate statistics on female head porters in Ghana, hence no sampling frame is available (Lattof, 2018). Time-location sampling assumes that it is possible to adequately reach and cover the target population at predefined locations through random sampling of those locations, using the 'centers of aggregation sampling technique' (Reichel and Morales, 2017). The centers of aggregation sampling technique produces a list of non-residential locations where the target population concentrates (Reichel and Morales, 2017). The rationale behind this strategy is that the random selection from this list provides interviewers with the locations where they also randomly select the final respondents (Reichel and Morales, 2017).

In Accra, the following locations (markets) were listed: Agbogboloshie, Tema Station, Mataheko, Rawlings Park, Nima, Madina, Mamobi and Tudu Market. Subsequently, Agbogboloshie, Tema Station, and Madina were randomly selected. In Kumasi, the following locations were listed: Aboabo Station, Central Market, Adum, Kejetia, Roman Hill, Asawase, Alabaa, and Dr Mensa. Of these, Aboabo Station, Kejetia, and Roman Hill were randomly selected. In each of these selected locations, centers of aggregation, where the head porters frequently met or visited, were created (Salentin, 2014). The centers of aggregation sampling technique was then used to select the study respondents from these locations. Adolescent girls between the ages of 10-19 years, and who had been in the kayayoo business for six months or longer were included in the study. This time frame was considered long enough to determine the adolescent girls' remittance ability, their socio-economic status, financial situation, and the overall poverty situation of the girls and their families. These processes resulted in the administration of 503 questionnaires - 253 in Accra, and 250 in Kumasi. The questionnaire was 9 pages long and comprised 87 questions including: the socio-demographics of the respondents, income and savings levels, remittance history, their socio-economic status, and their financial and poverty situations. For purposes of international comparisons, the socio-economic status of the adolescent girls was determined using the World Bank's poverty line of US$ 1.90 a day. Adolescents whose daily incomes fell below US$ 1.90 were categorized as poor and those whose daily incomes were above the US$ 1.90 poverty line were considered not poor. Data collection was done mostly on Sundays when the head porters conglomerated, resting or washing their clothes. The survey data was entered into a data base using Epidata, and then exported into STATA Version 14.0 for analysis. Double entry was done to minimize errors. Cross-tabulations, chi-square test and logistic regression models were used to analyze the data. Qualitatively, purposive sampling was used to select some of the girls for in-depth interviews on specific issues, in order to assess the validity of the quantitative findings. Twelve in-depth interviews were conducted in both Accra and Kumasi. All the interviews were conducted in the local dialect (Twi), tape-recorded and later transcribed verbatim into English. Thematic content analysis was employed to analyze the qualitative data manually. After listening to the audio recordings several times and comparing them with the visual transcripts to ensure that the recordings and the visual transcripts were consistent, the data was then categorized and summarized into themes based on the research objectives. The themes were then compared across the various respondents to establish any similarities and differences in their responses (Leech and Onwuegbuzie, 2008), with the aid of the verbatim quotes.

 

Results and Discussion

Age, Educational Status and Length of Stay as Head Porters

As illustrated in Table 1, nearly 93% of the adolescent girls (92.6%) fall in the age categories of 15-19 years. This age bracket is the prime time for the adolescents to be at senior high schools or at tertiary schools. It was thus not surprising that one-third of the girls had no formal education while 31% of them attained junior high school status, with only 8.2% of them attaining secondary/technical education status. Regarding the duration of their stay in the kayayoo business, more than half (59%) of adolescent girls had worked as head porters for 1-2 years, 28% of them had worked for less than a year, while 13% of them had worked for 3-4 years or longer. These variables - age, length of stay in the kayayoo business, and the educational satus of the adolescent girls - relate to their aspirations for migrating, their earnings potential, their socio-economic status and their ability to remit. For this reason, these issues are discussed in detail in light of the existing literature and the theoretical framework in the ensuing sections. As previously stated, all the migrant adolescent girls in this study came from Northern Ghana. Their reasons for migration are presented next.

 

 

Figure 1 clearly demonstrates that over 60% of the adolescent girls migrated to escape poverty. For this reason, poverty remains the core reason for migrating. Poverty has reduced nominally in Ghana from 56.5% in 1992 to 23.4% in 2017 (GSS, 2018: 10-18). The reduction in poverty appears uneven -the GSS report estimated that the number of poor persons living in poverty in Ghana increased by close to 400,000 people. Even so, these aggregate figures mask the reality because huge regional disparities abound - the reduction in poverty has largely not benefitted the northern Savannah ecological zone. The incidence of poverty has been an endemic rural phenomenon. The highest poverty head-count in 2016/17 was found in the rural Savannah zone at 67.7% (GSS, 2018: 11). For example, the GSS (2018: 14) illustrated that more Ghanaians were living in extreme poverty in 2017 than they did in 2013: the number of people living in extreme poverty increased from 2.2 million in 2013 to 2.4 million in 2017. The implication of this situation is that about 2.4 million people do not have access to the minimum daily requirement of 2,900 calories per adult equivalent of food per day, even if they were to spend all their income on food. At this rate of progress, and without any radical policy direction, achieving the first Sustainable Development Goal (SDG) of ending extreme poverty by 2030, is questionable in Ghana. The results support earlier research that reported that poor households living below the poverty line and who have high unemployment rates are most likely to send out members of their families as migrants to supplement family income (Redehegn, et al., 2019). The results resonate perfectly with mainstream development and economic theories, especially Lee's (1966) push and pull factor theory, because in the context of the adolescents, poverty is the push factor propelling them to migrate. These results also point to the economic and social values of migration where the majority of adolescents see it as an opportune platform to improve their lives.

 

 

Consistent with the theoretical framework, 6% of the adolescent girls migrated because of peer influence and the need to be exposed to city life. In consonance with de Haas (2014), these results are also aligned to the instrinsic value of migration, where young people aspire to migrate not merely to accummulate material and financial resources but also, importantly, for the joy and pleasure of doing so. As de Haas (2014) explains, the social prestige attached to moving to the city to see the 'bright lights' coupled with the freedom to migrate increases people's life satisfaction particularly in the context of rural-urban migration. Four percent of the adolescents migrated in an effort to escape from outmoded socio-cultural practices such as forced marriage and female genital mutilation (FGM). These practices affect adolescent girls' school attendance and performance as well as their psychological well-being. For instance, studies in Tanzania showed that girls who underwent FGM missed school for several weeks to allow for a recuperation period to recover from their wounds, and the pain associated with FGM created a phobia for some girls who became so psychologically disoriented, that they could not concentrate on classroom learning (Pesambili and Mkumbo, 2018).

Importantly, 29% of the girls migrated to raise money from the kayayoo business in order to pay school fees. This finding dovetails into the findings in Table 1 because about 31% of the girls attained junior high school status and would probably seek to raise money to at least, attain senior high school status. One respondent corroborated this finding during an in-depth interview as follows:

My aspirations for coming here have not been met. I came here to work and get money to pay my school fees because my parents can't afford it. I am sorry for coming here, since up to now, I am left with 14 days for us to go back to school, and today my money is not even up to GHS 50 [US$ 9.31 ] (18-year-old adolescent girl, in Agbogbloshie Market, Accra).

Another respondent remarked:

The main reason why I migrated to Kumasi is that my sister has gained admission into senior high school but there is no money to finance her education and there is no help coming from anywhere. If I do not do something about it, my sister will not further her education, so for this reason I have migrated to Kumasi to work to support her education (19-year-old adolescent, Roman Hill Market, Kumasi).

The aspiration of the adolescent girls to migrate and work as head porters in an effort to raise money to pay their school fees is a function of poverty because if their parents could afford to pay the fees, the girls could be in school. At the 2019 Women Deliver Conference in Canada, the President of Ghana, during a panel discussion suggested that Ghanaian women were not equitably represented in the political landscape because women lacked activism and dynamism (GhanaWeb, 2019). But the adolescent girls - future women leaders - have abandoned the classroom to transport heavy loads of goods in the urban markets to support their families, and in some cases, to support their brothers' education.

Adolescent girls missing out on formal schooling opens up spaces for social and political inequalities between women and men and produces entrenched disadvantages, including poverty for women (Kabeer, 2015). This study acknowledges the empowering potential of education and asserts that these adolescent girls must return to the classroom if the activism and dynamism that the President demands of women to propel them into the political limelight is to be achieved. Foucault (1977) postulated that knowledge is power because knowledge in the form of education acts as a catalyst for both individual civic participation and for higher levels of household well-being. Education leverages knowledge to individuals in the form of acquired skills and innovative ideas to make them more employable in better paid jobs (Alatinga and Williams, 2015; 2019). Kabeer (2008) suggests that women's paid work has a transformative potential, leading to empowerment and agency; it also engenders social change. However, to ignite the desired social change, women's paid work must be fulfilling - work that offers a sense of self-actualization - rather than being alienating - work taken up under extreme forms of economic compulsion, distress, sale of labor entailing hard physical labor under unfavorable conditions (Kabeer, 2008). The kayayoo business fits the description of alienating work because it undermines the self-esteem and dignity of the adolescents as females, yet they take up this work because they have no choice of getting a fulfilling one. This study believes that repositioning the adolescent girls to take advantage of education is possible through the Ghanaian Free Senior High School Policy (FSHSP). For example, if the 29% of adolescent girls in this study who were determined to raise money for their school fees were identified and supported to return to school through the FSHSP, the multiplier effects of their contribution to community development would be enormous.

 

Earning Potential, Socio-Economic Status and Remittances

The estimated average daily income of the adolescent girls was (GHS 13 or US$ 2.41) for those adolescents who worked only as head porters and (GHS 42.5 or US$ 7.89) for those who did additional jobs. Regarding their socio-economic status, Figure 4 below illustrates that 35% of the sampled population was categorized as poor and 65% was categorized as not poor based on the World Bank's US$ 1.90/day poverty line. Of the 35% poor population, only 10% managed to remit. It is likely that this category of adolescents engaged in other activities deemed as illicit in Ghana, such as prostitution in their quest to earn income to remit to their families. The following quote supports this claim:

Some people practice prostitution. From midnight around 12am-1am most people in my room go out and do that. They do it [prostitution] themselves without force (18-year-old adolescent, in Agbogbloshie Market, Accra).

Another respondent also claimed:

Those who are into prostitution only do that to support themselves financially, and this is because we do not get the money we need from the kayayoo business. They are easily influenced by men to meet their needs (19-year-old adolescent, Kejetia Market, Kumasi).

Interestingly, only half of those who were not poor were able to remit while the other half were unable to remit as shown in Figure 2.

 

 

Based on the results in Figure 2 above, a chi-square test was performed to determine the association between socio-economic status and the amount of money remitted annually, as shown in Table 2 below.

 

 

The chi-square test produced a chi-square value of 9 and a P-value of 0.010 at 95% level of confidence. The P-value of 0.010 shows a significant level of association between socio-economic status and the amount of money remitted annually. In other words, those who were not poor were better able to remit even higher amounts compared to the poor. Nearly 98% of those who were not poor were able to remit GHS 300 (US $55.58) or more annually to their families. From the disaggregated data, the highest annual remittance amount was GHS 500 (US$ 92.63). One girl remarked:

I have been able to support my mother and siblings back at home with the little money that I get in their feeding (sic) even though I wish I had the ability to do more or better especially with their education (19-year-old adolescent, Kejetia Market, Kumasi).

Based on these findings, a logistic regression model was run to determine the predictors of remittances among the adolescents as presented in Table 3 below.

 

 

The regression results show that the variables - age, marital status, education, number of years spent working as a head porter and socio-economic status -significantly influence the adolescents' ability to remit at p<0.05, p<0.01, p<0.001 significance levels respectively. Based on the adjusted odds ratio (AOR), adolescents aged 15-19 years were about 7 times more likely at p<0.05 significance level to remit compared to those aged 10-14 years. Similarly, at p<0.05 significance level, the AOR of 2, indicated that adolescents who were never married were twice as likely to remit relative to their married counterparts. The number of years spent working as head porter is significantly associated with the adolescents' ability to remit. For example, adolescents who worked for three years as head porters are nearly 4 times more likely to remit compared to those who had worked for less than a year at p<0.05 significance level. At p<0.01, adolescents who attained junior high school and secondary/technical school levels were 0.43 times and 0.23 times respectively less likely to remit with reference to those who never attended school. This scenario reflects the findings in Figure 1 because as many as 29% of the adolescents migrated to work as head porters to enable them to raise money for school fees. This category of adolescents may thus simply accumulate funds for the purpose of paying school fees; hence, they may not be in the position to remit. Finally, at p<0.001, adolescents who were not poor were nearly 9 times more likely to remit relative those who were poor. This finding is also very consistent with the chi-square test results in Table 3 as nearly 90% of the non-poor adolescents were able to remit various sums of money to their families.

In this study, the kayayoo business had capacitated only 36% of the adolescent girls to remit various sums of money ranging from less than GHS 100 to over GHS 300 (US$ 18.5-55.58 annually. Despite these amounts being lower than expected, the results amply demonstrated that receiving families used these remittances to increase household food consumption and also to pay school fees of children. Atuoye et al. (2017) found that remittance-receiving households reported a high incidence of food security in both rural and urban settings in the Upper West Region of Ghana. Mabrouk and Mekni's (2018) research in Ethiopia reported identical results. Remittances positively and significantly impacted upon the access, stability and utilization dimensions of food security. It does appear that the migration of the adolescent girls has an instrumental value. These favourable stories may make it difficult, if not impossible, to address the north-south migration of adolescent girls without a deliberate and coherent policy agenda that seeks to structurally transform the economy of Northern Ghana through the establishment of factories and industries to create job opportunities to address the yawning development gap between the north and the south.

These positives, however, appear to cancel each other out because the majority (64%) of the adolescents are unable to remit. Because age and the length of time spent working as a head porter are positively and significantly correlated with remittances, adolescent girls would have to spend longer periods out of school or they may never return to school in order to work and raise money for themselves and their families.

 

Expenditure Patterns of Adolescent Girl Head Porters

As indicated in Figure 3 above, the estimated average daily income of the adolescent girls was (GHS 13 or US$ 2.35). This amount was then multiplied by 30 - that is 30 days in a month to obtain the average monthly income - GHS 13*30 (GHS 390 or US$ 70.46). As shown in Figure 3, the total monthly expenditure on the various items amounted to GHS 326. The average monthly income over expenditure was then computed as GHS 390-GHS 326=64 or US$ 11.56. The average daily expenditure on food was estimated at GHS 120/30 =GHS 4 or US$ 0.74. Food, health, and water constituted the largest expenditure, amounting to 31%, 30% and 9% of income respectively. The health care expenditure related to only those respondents who sought and paid for health care in the past month preceding the survey (i.e. 63% of respondents). Electricity constituted the lowest expenditure item, amounting to about 7% of the monthly income. The low expenditure on electricity is understandable because most of the adolescent girls lived in make-shift structures such as shacks, kiosks and containers, which may not be connected to the national grid.

 

 

Notably, the colossal health expenditure of the adolescents is an issue of relevance to the country's social and health policy because such a huge health expenditure is potentially catastrophic to poor families - health expenditure that exceeds or equals 30%-40% of household incomes - and could further push both the adolescents and their families into extreme poverty (Xu et al., 2003). The huge health expenditure is not surprising, given the physical and tedious nature of the adolescents' work. The implication of the huge health expenditure is that the adolescent girls may adopt other coping strategies, such as cutting down on other necessities such as food and clothing. This appears to be the case with the adolescents because their average monthly food expenditure was GHS 120 (US$ 22.14). This figure translates to an average daily food expenditure of GHS 4 (US$ 0.74), but some even spend as low as GHS 1.50 (US$ 0.28). This amount of GHS 4 is too paltry to access any reasonable balanced meal in the urban setting to meet the daily calorie requirements of the adolescent girls in light of the brutal nature of their work. The qualitiave narratives (in this study) reported that feeding was sometimes difficult for some of the head porters. As highlighted earlier, some of these adolescents even indulged in activities such as prostitution, with its associated health risks such as contracting HIV/AIDS, and teenage pregnancy, just to make a living.

 

Prevailing Financial Situation and Overall Poverty Situation

In order to appreciate the prevailing financial and overall poverty situations of the migrant adolescent girls, the following two questions were asked: (1) Based on the kayayoo work, how would you assess your current financial situation? (2) Since you started the kayayoo work, how would you assess the overall poverty situation of yourself and your family back home? These questions attracted mixed responses detailed in the results presented in Tables 4 and 5 and the qualitative narratives below.

 

 

 

 

The results from Table 4 illustrate that over half (52%) of the adolescents saw their current financial situation as inadequate and fewer than 1% of them saw their current financial situation as being more than adequate based on the kayayoo business. These results are not surprising because as shown earlier, the average daily income of the adolescent girls ranged from GHS 13 (or US$ 2.41) to GHS 42.5 (or US$ 7.89). Nearly half of the adolescents did not experience any changes in their poverty status. In fact, as shown in Table 5, remarkably, nearly 12% of the adolescents reported that their poverty situation had worsened. The following qualitative narrative bolstered these quantitative results:

I and my family (sic) are now poorer than before. My father borrowed money (GHS 300) [US$ 55.58]) for me to come to Accra because we thought I could raise enough money to pay that money. But for the past two years I have been here, I have been struggling to raise that money. The little money I get, sometimes GHS 8 in a day, I have to spend GHS 1.50 [US$ 028] on food, GHS 1 [US$ 0.18] to bath, GHS .50 [ US$ 0.090] for toilet use, pay for accommodation and light bill. And the person who gave the money to my father is putting pressure on him (14-year-old adolescent, Madina Market, Accra).

These findings are consistent with previous work. For instance, Awumbila and Ardayfio-Schandorf (2008) and Oberhauser (2011) reported that female head porters are among the poorest of urban dwellers in Ghana.

Even so, as shown in Table 5, 38% of adolescents indicated that their poverty situation had seen some improvement. The quote below lends credence to this finding:

To speak the truth, my finances are okay now. I have been able to send money to my parents when they demanded some. I also buy things for myself. I have been able to gather money to send back home to cater for my younger brother's fees since for the past three years my dad said he didn't have (the money). Just three days ago, I sent money home in preparation towards my brother's school (fees) (18-year-old adolescent, Agbogbloshie Market, Accra).

These mixed results and narratives paint a worrying picture about the ability of the majority of the adolescents to improve their well-being through the incomes generated from the kayayoo business. Disturbingly, from the earlier narrative above, migration had worsened the poverty situation of some of the adolescents and their families because some families borrowed money to finance the trips of their daughters. These findings are consistent with the existing literature; Opare (2003) suggests that some families go to the extent of borrowing money to finance the travel expenses of the adolescent girls. The findings are also consistent with the aspirations-capability theoretical framework deployed in this work. The results aptly imply that the kayayoo business does not necessarily capacitate the majority of the adolescents to meet their aspirations in order to live the kind of lives they have reason to value (Sen, 1999). Awumbila et al. (2014) and Hagen-Zanker et al. (2017) assert that migration does not always achieve its poverty reduction potential because of poor living conditions, exposing migrants to floods and diseases such as typhoid fever and cholera.

 

Involuntary Immobility

In order to appreciate the adolescents' own subjective evaluation of their well-being relative to achieving their migration aspirations, the adolescent girls were asked whether they would like to return home. Overwhelmingly, 93% of adolescents responded in the affirmative, as illustrated in Table 6.

 

 

However, the question to ask is: what prevents the adolescents from returning home if their aspirations for migrating are not being met? Here is why:

What am I going back home to do? The situation at home has not changed, there are no jobs for my parents to do and get money for my school fees. There is no food, there is just nothing, and so I can't return home now because the family expects me to work and get money to help them (17-year-old adolescent, Kejetia Market, Kumasi).

The above scenario highlights Carling's (2002) concept of involuntary immobility initially used to describe the number of people willing to migrate but not being able to do so. Instead of migration being considered a fundamental capabilities-enhancing freedom in its own right (de Haas and Rodríguez, 2010), it rather constrains the freedom of choice of the adolescent girls to return home. Contextually, the concept of involuntary immobility may also be used here to describe the situation of the migrant adolescent girls who want to return home but are not able to do so because of lack of resources thus becoming involuntarily immobile. This finding fits in neatly with the aspirations-capabilities theoretical framework because from the narrative above, the necessary structural conditions in the adolescents' places of origin to shape and offer opportunity structures that would enable them to pursue and achieve their personal or communal aspirations and capabilities are inadequate or non-existent (Merton, 1968). In summary, based on the evidence adduced here, the potential of the kayayoo business to improve the socio-economic status of the adolescents and their families does not seem bright. On the contrary, it looks bleak because 64% of them are worse-off in the kayayoo business. In this process, most of the adolescents become involuntarily immobile, and socially dislocated from their families. Based on the capabilities-aspirations framework therefore, the kayayoo business does not seem to capacitate the adolescent girls to live the kind of lives they have reason to value.

 

Conclusion and Recommendations

This paper examined the relationship between internal migration, socioeconomic status and remittances, drawing on the experiences of migrant adolescent girl head porters in the cities of Accra and Kumasi in Ghana. The evidence presented in the paper suggests that poverty is the chief driver of the girls' migration. For this reason, the migration of adolescents largely appears to have an instrumental value, a kind of social protection mechanism to manage household poverty. Yet, only 36% of the adolescents remit to their families for various purposes, including the payment of school fees and improving household consumption. Factors such as age and the length of time spent working as a head porter positively and significantly influenced remittances. While remittances improved the lives of impoverished families, they carried a high cost for the adolescents, who stayed out of formal schooling, either temporarily or permanently.

This study shows that the practice of adolescents forgoing formal schooling to engage in the kayayoo business, jeopardizes the future manpower development needs of Ghana because women constitute 52% of the population, of which adolescents are an integral part. The process of migration also caused some adolescents and their families to be poorer because some families borrowed money to finance their trips. Thirty percent of the adolescents' earnings were spent on healthcare. Some of these adolescents were unable to return home against their wishes, thus becoming socially dislocated. Based on the evidence, two strands of policy recommendations are proffered - short-term and long-term. In the short-term, the Ministry of Gender, Children and Social Protection must collaborate with the Kayayoo Association to identify adolescent girls who work as head porters, to raise money to pay their school fees in line with the Free Senior High School Policy. This strategy will ensure that such girls are not deprived of education because of poverty. In the long-term, the state must deliberately roll out plans to close or narrow the development gap between the north and the south through the establishment of relevant industries and factories to provide employment opportunities. Northern Ghana is a largely agricultural economy and so the construction of dams for irrigation to ensure all-year farming may help boost the local economy of the area. It is in this context that the government's 'One Village, One Dam' and 'One Village, One Factory' policy, if successfully implemented, could provide a fertile ground for the structural transformation of Northern Ghana.

 

Acknowledgment

The Council for the Development of Social Science Research in Africa (CODESRIA) supported the research for this paper under the African Diaspora Support to African Universities Program.

 

References

Abdulai, A.M., Boakye-Yiadom, L. and Quartey, P. (2017). The impact of migration on the welfare of households in Ghana: A propensity score matching approach. African Human Mobility Review, 3(1): 732-750.         [ Links ]

Adarkwa, K.K. (2012). The changing face of Ghanaian towns. African Review of Economics and Finance, 4(1): 1-29.         [ Links ]

Agarwal, S., Attah, M., Apt, N., Grieco, M., Kwakye, E.A. and Turner, J. (1997). Bearing the weight: The kayayoo, Ghana's working girl child. International Social Work, 40(3): 245-263.         [ Links ]

Alatinga, K.A. and Williams, J.J. (2015). Towards universal health coverage: Exploring the determinants of household enrolment into National Health Insurance in the Kassena Nankana District, Ghana. Ghana Journal of Development Studies, 12 (1 and 2): 88-105.         [ Links ]

Alatinga, K.A. and Williams, J.J. (2019). Mixed methods research for health policy development in Africa: The case of identifying very poor households for health insurance premium exemptions in Ghana. Journal of Mixed Methods Research, 13(1): 69-84.         [ Links ]

Ardayfio-Schandorf, E., Yankson, P.W.K. and Bertrand, M. (2012). The mobile city of Accra. Oxford, UK: African Books Collective.         [ Links ]

Atuoye, K.N., Kuuire, V.Z., Kangmennaang, J., Antabe, R. and Luginaah, I. (2017). Residential remittances and food security in the Upper West Region of Ghana. International Migration, 55(4): 18:34.         [ Links ]

Awumbila, M. and Ardayfio-Schandorf, E. (2008). Gendered poverty, migration and livelihood strategies of female porters in Accra, Ghana. Norsk Geografisk Tidsskrift - Norwegian Journal of Geography, 62(3): 171-179.         [ Links ]

Awumbila, M.; Owusu, G.; Teye, J.K. (2014). Can Rural-Urban Migration into Slums Reduce Poverty? Evidence from Ghana. Migrating out of Poverty RPC Working Paper 13. Migrating out of Poverty Consortium, University of Sussex, Brighton, UK. pp 41. From <http://migratingoutofpoverty.dfid.gov.uk/files/file.php> (Retrieved on 15^th November 2014.).

Baah-Ennumh, T.Y., Amponsah, O. and Adoma, M.O. (2012). The living conditions of female head porters in the Kumasi metropolis, Ghana. Journal of Social and Development Sciences, 3(7): 229-244.         [ Links ]

Carling, J. (2002). Migration in the age of involuntary immobility: Theoretical reflections and Cape Verdean experiences. Journal of Ethnic and Migration Studies, 28(1): 5-42.         [ Links ]

Carling, J. and Schewel, K. (2018). Revisiting aspiration and ability in international migration. Journal of Ethnic and Migration Studies, 44(6): 945963.         [ Links ]

Cyril, S., Oldroyd, J.C. and Renzaho, A. (2013). Urbanisation, urbanicity, and health: A systematic review of the reliability and validity of urbanicity scales. BMC Public Health, 13(1): 513.         [ Links ]

de Haas, H. (2014). Migration theory: Quo vadis? International Migration Institute Working Papers No. 100. United Kingdom, Oxford OX1 3TB: University of Oxford.

de Haas, H. and Rodríguez, F. (2010). Mobility and human development: Introduction. Journal of Human Development and Capabilities, 11(2): 177-184.         [ Links ]

Foucault, M. 1977. Discipline and punishment. London: Allen Lane.         [ Links ]

Ghana Statistical Service (GSS). (2018). Poverty trends in Ghana 2005-2017: Ghana living standards survey round 7 (GLSS 7). Accra, Ghana: Ghana Statistical Service.         [ Links ]

Ghana Web. (2019). How Akufo-Addo killed the spirit, soul of gender empowerment activism in Ghana. From <https://www.ghanaweb.com/GhanaHomePage/NewsArchive/How-Akufo-Addo-killed-the-spirit-soul-of-gender-empowerment-activism-in-Ghana-752831> (Retrieved on the 7^th June 2019).

Gile, K.J. and Handcock, M.S. (2010). Respondent-driven sampling: An assessment of current methodology. Sociological Methodology, 40(1): 285327.         [ Links ]

Goldin, I. and Reinert, K. 2012. Globalization for development: Meeting new challenges (New Edition). Great Clarendon Street, Oxford: Oxford University Press.         [ Links ]

Hagen-Zanker, J., Postel, H. and Vidal, E.M. (2017). Poverty, migration and the 2030 Agenda for Sustainable Development. From <https://www.odi.org/sites/odi.org.uk/files/resource-documents/11743.pdf> (Retrieved on the 10^th September 2018).

International Organization for Migration (IOM). (2018). World Migration Report 2018. Geneva: International Organization for Migration.         [ Links ]

Kabeer, N. (2008). Paid work, women's empowerment and gender justice: Critical pathways of social change. Pathways of Empowerment Working Papers No 3. Brighton: Institute of Development Studies.         [ Links ]

Kabeer, N. (2015). Gender, poverty, and inequality: A brief history of feminist contributions in the field of international development. Gender and Development, 23(2): 189-205.         [ Links ]

Kleist, N. (2017). Introduction: Studying hope and uncertainty in African migration. In: Kleist, N. and Thorsen, D. (Eds.), Hope and uncertainty in contemporary African migration. New York and London: Routledge, pp. 1-20.         [ Links ]

Lattof, S.R. (2018). Health insurance and care-seeking behaviours of female migrants in Accra, Ghana. Health Policy and Planning, 33(4): 505-515.         [ Links ]

Lee, E.S. (1966). A theory of migration. Demography, 3(1): 47-57.         [ Links ]

Leech, N. L. and Onwuegbuzie, A.J. (2008). 'Qualitative data analysis: a compendium of techniques and a framework for selection for School Psychology Research and beyond', School Psychology Quarterly, 23 (4): 587604.         [ Links ]

Mabrouk, F. and Mekni, M.M. (2018). Remittances and food security in African countries. African Development Review, 30(3): 252-263.         [ Links ]

Mac-Ikemenjima, D. and Gebregiorgis, H. (2018). Eritrea's youth migration challenge: The role of aspirations and opportunity. African Human Mobility Review, 4(2): 310-330.         [ Links ]

Merton, K.R. 1968. Social theory and social structure. New York: The Free Press.         [ Links ]

Nyarko, S.H. and Tahiru, A.M. (2018). Harsh working conditions and poor eating habits: Health-related concerns of female head porters (Kayayei) in the Mallam Atta Market, Accra, Ghana. BioMed Research International, (2018): 1-7

Nyasulu, T. (2017). Harnessing economic impacts of migrant remittances for development in Sub-Saharan Africa: A critical review of the literature. African Human Mobility Review, 3(1): 645-670.         [ Links ]

Oberhauser, A.M., and Yeboah, M.A. (2011). Heavy burdens: gendered livelihood strategies of porters in Accra, Ghana. Singapore Journal of Tropical Geography, (32): 22-37

Opare, J.A. (2003). Kayayei: The women head porters of Southern Ghana. Journal of Social Development in Africa, 18(2): 33-48.         [ Links ]

Osei-Boateng G. (2012). Ghana TUC/WIEGO women in informal economy law project: A report on Kayayei in Ghana. Accra, Ghana: Ghana Trades Union Congress.         [ Links ]

Pesambili, J.C. and Mkumbo, K.A.K. (2018). Implications of female genital mutilation on girls' education and psychological wellbeing in Tarime, Tanzania. Journal of Youth Studies, 21(8): 1111-1126.         [ Links ]

Preibisc, K., Dodd, W. and Su, Y. (2016). Pursuing the capabilities approach within the migration-development nexus. Journal of Ethnic and Migration Studies, 43(13): 2111-2127.         [ Links ]

Ravenstein, E.G. (1885). The laws of migration. Journal of the Statistical Society of London, 48(2): 167-235.         [ Links ]

Redehegn, M.A., Sun, D., Eshete, A.M. and Gichuki, C.N. (2019). Development impacts of migration and remittances on migrant-sending communities: Evidence from Ethiopia. PLOS ONE, 14(2): e0210034.         [ Links ]

Reichel, D. and Morales, L. (2017). Surveying immigrants without sampling frames - evaluating the success of alternative field methods. Comparative Migration Studies, 5(1): 1-22.         [ Links ]

Rice, J. (2008). The urbanization of poverty and urban slum prevalence: The impact of the built environment on population-level patterns of social well-being in the less developed countries. Research in the Sociology of Health Care, 26: 205-234.         [ Links ]

Salentin, K. (2014). Sampling the ethnic minority population in Germany. The background to 'migration background'. Methods, Data, Analyses, 8(1): 25-52.         [ Links ]

Sen, A. (1989). Development as capability expansion. Journal of Development Planning, 19, 41-58.         [ Links ]

Sen, A. (1999). Development as freedom. New York: Anchor Books.         [ Links ]

Stark, O. (1984). Migration decision making. Journal of Development Economics, 4(1): 239-265.         [ Links ]

Temin, M., Montgomery, M.R., Engebretsen, S. and Barker, K.M. (2013). Girls on the move: Adolescent girls and migration in the Developing World. USA: Population Council.

Thomas, R. G. (1973) Forced Labour in British West Africa: The Case of the Northern Territories of the Gold Coast 1906-1927. Journal of African History, 14 (1);79-103).         [ Links ]

Todaro, M. and Smith, S. (2012). Economic Development (11th ed.). New York, USA: Addison Wesley.         [ Links ]

Tufeiru, A. (2014). The nexus of female capital and North-South labor migration in Ghana: A potential remedy from microfinance. Journal of Developing Societies, 30(1): 91-114.         [ Links ]

United Nations General Assembly (UNGA). (2015). Transforming our world: The 2030 Agenda for Sustainable Development. New York: United Nations General Assembly.         [ Links ]

United Nations, Department of Economic and Social Affairs, Population Division (2018). World Urbanization Prospects: The 2018 Revision, Online Edition

White, M.J. (Ed.). (2016). International handbook of migration and population distribution. Netherlands: Springer.

Xu, K., Evans, D.B., Kawaba, K., Zeramdin, R., Klavus, J. and Murray, C.J.L. (2003). Household catastrophic health expenditure: a multicountry analysis. Lancet, 362 (9378): 111-7.         [ Links ]