Economic Dependency and Opportunity Cost of Childbearing: Implications for fertility control in sub-Sahara Africa

The dynamics in the household income within the marital unions especially with the emergence of female breadwinners in the last decade is gradually changing the demand for children and the need for child spacing. It is in line with this that this study examines the impact of changing spousal roles on the cost of bearing children in three sub-Saharan African countries. Demographic and Health Surveys data were analyzed using three levels of statistical analysis, while at multivariate level, ve models were constructed to test the formulated hypotheses. The study identied signicant relationship between economic dependency and the control variables (use of modern contraceptives and fertility timing) in the three countries (Pvalue < .0001). The study also armed a positive signicant relationship between economic dependency and labour force participation of women except in Lesotho (Pvalue < .0001). In addition, higher signicant differential was ascertained between labour force participation and fertility in the study area. The three countries also indicate high signicant correlation of the effect of women economic dependency on fertility. The study concludes that empowering women and allowing them to involve in paid activities will not only reduce fertility but signicantly contribute to poverty reduction in Africa. opportunity cost of childbearing among married women in SSA? How related are dependency, opportunity cost of childbearing and in


Introduction
Traditionally in Africa, men are regarded as the main income providers for the family. The patriarchal nature of the African society also accrues powers and authority to men with regards to household decision-making. However, the impact of recent economic recession in most African countries and continued dispersal of gender egalitarianism have weakened the dominance of men as breadwinners for their families. This has increased women's participation in productive and paid activities. Vitali & Arpino , the dual role played by women as breadwinners and home makers has also been found to signi cantly contribute to lower fertility as women may decide either to delay births or prolong cohabitation without childbearing while young women will delay age at marriage or nd alternatives to family building (Brinton and Lee, 2016). Studies in the United States among college students have revealed that easy access to contraceptives affects the timing of marriage, births and incidence of cohabitation. Women will be able to work for pay, invest in on-the-job training and pursue their professional career (Goldin and Katz 2002;Bailey 2006;Guldi 2008; Christensen 2011). The economic theory of women's fertility and work envisages that the competing demands for women's time to work outside the home or to bear and care for children reach equilibrium where the household obtains maximal gain. While human capital theory explained that health and labor force participation are positively correlated, studies have also shown that challenges in individual health will reduce labour force participation, lower productivity and add extra burden to the family income. On the other hand, low productivity as a result of poor health will reduce earning potentials (Zamo-Akono, 2009; Nanfosso & Zamo-Akono, 2010; Cai and Kalb, 2006;Waghorn and Lloyd, 2005).
In addition, studies have con rmed that higher fertility associated with lower participation in the labour force among women during their reproductive life time, while access to family planning and reduction in birth increase the female labour participation , Shareen, 2012). An Indonesian study attests to the likelihood of women participation in the labour force increasing by 20% with the reduction of one birth over the period of 20 years. Women labour participation and increased level of education increase the value of their time, making it more expensive to bear children (Lam and Duryea, 1999 there are still limited studies on the changing gender roles and its association with fertility choices especially in Sub-Saharan Africa (subsequently SSA). It is thus imperative to answer these questions: how do changes economic dependency affect use of modern contraceptives and timing of next birth? Do changing economic dependency in uence opportunity cost of childbearing among married women in SSA? How related are economic dependency, opportunity cost of childbearing and fertility in SSA?
Providing answers to these questions will unravel the contextual factors that explain the impact of changing roles on the cost of bearing children in sub-Sahara Africa. The ndings will serve as a template for policy makers on how to reduce the fertility level in Africa.

CHANGING ROLES AND FERTILITY IN SUB-SAHARAN AFRICA
In pre-colonial traditional African societies, the roles of men and women are complementary. Women had opportunity to participate in most activities, although this varies from one society to the other. It was during colonial period that the idea that women belong to the home for childbearing and food preparation was introduced through Western patriarchal structures. This led to the erosion of traditional social institutions and restricted women into disadvantaged vulnerable positions. In many contemporary societies, gender determines the roles played and the relative power wielded by men and women. It also establishes the opportunities and privileges held. It is this patriarchal ideology which exaggerates biological differences between men and women. This becomes even more obvious when one considers that the family remains the social and symbolic place in which sexual differences is believed to be primary and at the same time constructed. Maakama (2013) is of the opinion that patriarchy is a set of social relations that enables men to dominate women. The social relations are also determined by power attributes which in uence how society describes what is masculine or feminine. Although '[p]ower is relational, it becomes apparent when it is exercised, power is not associated with a particular institution, but with practices, techniques, and procedures, employed at all levels and through many dimensions' (Luna, Gemma & Dorothea 2017) The value which either men or women hold is tied to their traditional gender-assigned powers which vary from one society to another. For example, in most African rural societies, women are engaged primarily in farming, weaving, pottery, dyeing and trading, while still tied to the home, taking care of the children and playing a subservient role. The role of men which is clearly de ned involves providing for the family, and because they occupy the position of head of the family, they make all major decisions including reproductive choices. This creates a division of labour along gender lines and normal family life has come to be de ned in terms of a breadwinning husband and a domestic wife with ideal feminine attributes which include sexual innocence and motherhood with the expectation that the woman becomes a hard working household provider.
Some studies have revealed that culturally, women in south-western Nigeria are signi cantly empowered at the family level, even in responsibilities that are normatively prescribed for men (Ogunjuyigbe and Adeyemi 2005). This implies some degree of autonomy and power in a marital dyad. The powers are usually exercised in the prescribed periods in which women are expected to abstain from sexual relations with their husbands. Such periods include the period before marriage, during menstruation and postpartum period and also on becoming a grandmother. The existence of these prescribed periods reveals that traditionally women have sexual rights. Bogalech and Mengstu, (2007) also explained that in Ethiopia, traditionally, women have little autonomy on most individual and family or household issues, including the option to choose whether to get modern health services during illness, birth, reproductive health services and others.
However, the ability to take such decisions requires a sense of personal autonomy, which develops in tandem with the knowledge that women can provide for themselves and their children. Since the decisions regarding childbearing in most African settings are mostly made by men, it is generally assumed that African men are less inclined to limit their families than women because they operate in a socio-cultural and economic environment that make high fertility bene cial for both men and women.
Ths however places the risks and costs primarily on women. Studies also con rm that one of the reasons for large family size in sub-Saharan Africa was because reproductive decision making was more rmly under the control of men (Caldwell and Caldwell 1990, Olusanya 1989). Despite this, changes in the level of education of women, urbanization, industrialization and economic recession have rede ned the roles of women in the family and consequently affected family size. Lerch (2013) explained that sharp decline in fertility during the post-socialist transition in Europe and the former Soviet Union was initiated by the adverse material and social consequences of the economic crisis. This has however had a long lasting effect because the onset of childbearing was further postponed even after the economic and political situation stabilized. Studies on women empowerment and fertility in Africa have shown inverse relationship between women empowerment (regarding nancial autonomy, household decision making especially reproductive health issues) and fertility (Hindin 2000;Shapiro 2008). Folarinmi (2013) also identi ed that women who have nancial autonomy are more likely to make positive investments in their children thereby increasing their chances of survival during infancy and increasing their likelihood of ever attending school.

Methods
The study used quantitative data, Demographic and Health Surveys data were used for the analysis with the permission from Measures DHS USA. The data sets for women within 15-49 years were downloaded for three countries in sub-Sahara Africa (Congo 2011, Lesotho 2014, and Nigeria 2013). Each of the countries represent regions in sub-Saharan Africa and their selection was based on the use of modern contraceptive methods among married women. The data were weighted for national representation. More speci cally, data for married women and those living with their partners were extracted from women's data and merged together.

Outcome Variables
The outcome variables for this study are opportunity cost of childbearing and fertility. The opportunity cost of bearing children in this study is the alternative forgone which was measured by labour force participation (whom the respondent worked for). These two variables were used as dependent variables. Labour force participation 'whom the respondent worked for' was categorized into 'work for someone else/' =1, 'family member/self-Employed' =0, while fertility was measured by total children ever born (CEB) which is a count data.

Control Variables
Studies have established the relationship between birth postponement (use modern contraceptives and fertility timing) and fertility (Casique, 2001, d'Albis, Greulich & Ponthière 2017). These two variables were used as control variables in the study. Each of these variables was re-categorized into dichotomous variables. 'Currently using modern contraceptives' was dichotomized into 'use modern contraceptives'=1 & 'non-use of modern contraceptives'=0. 'Preferred waiting time for birth of a/another child' was used for timing of next birth (birth spacing). This was also dichotomized into ≤ 2years=0 and >2 years=1

Independent Variable
The independent variable was economic dependency which was measured by whether the respondent earned more than the husband/partner or equal and partners/husband earn more than the respondents. This was re-categorized into three; respondent earn more (full autonomy =1), Both are equal (Partial=2) and husband/partner earn more (None=3) this classi cation was based on previous studies on breadwinners (

Background variables
The background variables identi ed were level of education of respondents, age, place of residence, age at rst marriage, wealth index, age at rst sexual intercourse and partner's level of education

Data Analysis
The data for the study were analyzed using three levels of analysis: univariate to show the frequency distributions of the socio-demographic characteristics of the respondents; proximate variables and explanatory variables. The Bivariate analysis was used to show the differentials in the economic dependency among women and some selected variables. For multivariate analysis, two statistical techniques were used and ve models were constructed to test the formulated hypotheses. Model 1 shows the unadjusted relationship between economic dependency and opportunity cost while model 2 pointed out the effect of the contol variables on economic dependency and opportunity cost of childbearing. Model 3 reveals the relative risk of economic dependency, opportunity cost of childbearing and fertility, Model 4 established the relationship between the independent variables, control variables and fertility while model 5 is the complete model including background variables. Binary logistic regression was used to show the relationship between outcome variables (opportunity cost of childbearing) and independent variable (economic dependency). While Poisson regression was used to estimate the incidence of relative risk (IRR) of economic dependency and opportunity cost of bearing children on fertility (total children ever born) which is a count data (see Table 1).

Results
The study established from Table 1 that majority of the married women were from rural areas except in Congo Brazzaville, while with the pooled data, (38.9%) of the sampled population lived in urban centres.
In all the counties except in Lesotho more than half majority of the respondents had secondary education while Nigeria has the highest women with higher education (6.4%) followed by Lesotho. With respect to wealth index, Congo Brazzaville had the highest women at the poor layer of the wealth index (38.7%) followed by Lesotho (23%) while with the pooled data, two out of every ve women sampled in the study are poor. It is also important to note that Lesotho has the highest percentage of richest women (48.1% followed by Nigeria (38.2%). In relation to the age of the respondents, Nigeria has the highest number of women between age 25-34 years (44.6%) followed by Congo Brazzaville (41.3%). In the pooled data, 38% of the women are between ages 15 and 24 years. In all the three countries, more than half of the women married between age 16-20 years while Nigeria has the highest number of women that married at older age 26 years and above (13.2%). It was also revealed that Nigeria has the highest number of women who had given birth to 4 and above children while Lesotho has highest number of women with no children. The bivariate analysis between economic dependency and birth postponement in Table 3 shows that women with full nancial autonomy have the highest percentage among women who used contraceptives in all the countries including the pooled data. The signi cant relationship between economic dependency and use of modern contraceptives was established in the study (Ӽ 2 Pvalue = .000).
It was also noted that more than two-thirds of women with full nancial autonomy indicated preference to have their next birth after two years. Signi cant relationship between economic dependency and fertility time was rea rmed in the study (Ӽ 2 Pvalue = .000) except in Lesotho In Table 4, Model 1 pointed out the relationship between economic dependency and opportunity cost of childbearing with logistic regression. Signi cant relationship was ascertained between women who have full autonomy and opportunity cost of childbearing in all the countries including the pooled data except  Model 3 in Table 5 shows the Poison regression of economic dependency, opportunity cost and total children ever born in the study area. A signi cant effect of economic dependency (full autonomy and partial autonomy) on children ever born was rea rmed for all the three countries including the pooled data (P < .001), while a signi cant higher effects of the opportunity cost of childbearing was retreated in the study (P < .001) When the data was adjusted with confound variables in model 4, signi cant higher effects was con rmed with economic dependency (full autonomy and partial autonomy) except for pooled data ( Partial autonomy Congo Brazzaville Odd Ratio = 1.141, P value .000; Partial autonomy Lesotho Odd Ratio =-1.12, P value .000; Partial autonomy Nigeria Odd Ratio =-1.242, P value .000, Partial autonomy pooled data .855 P value .000). For opportunity cost of childbearing, a signi cant coe cient was found with children ever born in all the countries including the pooled data (P < .001). With the full model in Table 6, a signi cant effect of economic dependency (full autonomy and partial autonomy) was ascertained in all the countries except in Lesotho while opportunity cost for childbearing remains signi cant with children ever born in all the countries .9 01 ** .9 01 -.9 01 **P < .001, * p < .005 CI = Con dence Interval for Exp (β) @ 95%

Discussion
The results of the analysis have pointed out the impacts of changing roles (economic dependency) on opportunity cost of bearing children in sub-Sahara Africa. The pattern of education shows that majority of the women had secondary education while Lesotho has the highest women with primary education.
This is an indication that there is an improvement in the women's level of education with resultant effect on nancial autonomy. The exposure of women to western education enables them to delay marriage, control family preference and become more knowledgeable about modern contraceptives methods. Odusina (2017) explained that education will afford women the opportunity to control the number of children and they will be well informed about the health services including family planning and the implications of large family size on the well-being of the mothers. It is believed that with higher levels of education, women's expectation of future earnings will be higher and this will eventually increase the paying jobs and commitment to careers outside home will bring about changes in gender egalitarianism resulting to later marriage and fewer numbers of children. Those who engage in career that take them away from home will nd it di cult to combine the responsibility of their job with childbearing which will have impact on the family size.
Despite these however, the age at rst marriage of the full nancial autonomy indicated that almost twothirds of the respondents married before age twenty (See Fig. 2). This is re ective of early marriage practices in many sub-Sahara African countries. Adedokun, Adeyemi & Dauda (2016) identi ed poverty as the core factor that in uences early marriage in low income countries, since they lack resources to support healthy alternatives for girls, such as prolonged schooling and skill acquisition to secure their future. Delaying marriage will increase girls' and women's decision-making power, increase incomes for individuals and advance economic development for nations. In the same vein, the positive signi cant effect of economic dependency on waiting period for another child in the study rea rmed the link between women's labour force participation and fertility timing. Women who have nancial autonomy or control over economic resources within the household will have opportunity to decide when to have the next child and use family planning. Thus, women who are involved in the household decision making are more likely to space their births and use family planning (Malhotra et al 2002). Women nancial empowerment will in uence reproductive outcomes including contraceptive use, family size and births intervals.  ( Lam and Duryea, 1999). It was documented that changes in the household income as a result of unemployment, economic recession and privatization policy will shift the cost of childbearing from society to family thereby resulting in a decline in total children ever born ( Eltigani, 2005). All the same, the dual shifts inside and outside the home may also in uence women not to have additional children, or delay having another child in order to avoid the additional burdens (Craig and Siminski 2010). Similarly, it is expected that where house chores are shared or men are involved in the housework as a result of the contribution of the wife to family income, the number of children the family intends to have will be affected. Studies from Europe revealed a positive relationship between husband involvement in childcare and the risk of having the second child in Spain and Italy (Cooke 2009).
In all, this study has established signi cant relationship between economic dependency, opportunity cost of childbearing and fertility in all the three countries under study. Woldemicael (2009) found out that increased participation in household decision making among women was associated with lower fertility and discussion of ideal family size with partners. Those who are engaged in earnings that contributed to the household income will have the con dence to discuss with their partners concerning fertility preference and family size. Socio-and demographic factors such as employment and economic status affect women's reproductive preferences not only positively but also indirectly by increasing their autonomy, independence and reproductive judgment within their household and community. The study has rea rmed that fertility choices within conjugal unions does not only depend on father's activity but also on the mothers earning status.

Conclusion and Contribution to knowledge
The study has drawn attention to the effect of economic dependency on the fertility behaviour of women in SSA. The dynamics in the household income within the marital unions especially with the emergence of female breadwinners in the last decade is gradually changing the demand for children and the need for child spacing. This will not only reduce the fertility rate in SSA but will also improve the health of women and reduce maternal mortality in SSA. In fact, one of the cardinal goals of Sustainable Development Goals (SDGs), especially goal 5, is to "achieve gender equality and empower all women and girls". Empowering women and allowing them to involve in paid activities will not only reduce fertility but signi cantly contribute to poverty reduction in Africa. It is therefore necessary to educate females up to secondary school level. This will increase their labour force participation and improve their reproductive health status. Government should ensure free education for female up to senior secondary school and checkmate all socio-cultural factors that may lead to girls dropping out of school.

Limitation Of The Study
This study is limited with some factors which are important to mention. Firstly, the study make used of DHS data which are cross-sectional and the causal relationship cannot be mentioned. Secondly, it is also very di cult to calculate the percentage of income contributed by each partner to maintain the family.
Thirdly, the economic situations in those countries may also affect the fertility choices of the respondents which may not re ect the true fertility desired as contained in the survey. Despite the fact that DHS survey Source : Compiled by the author from NDHS data