After about a decade of civil war, Liberia still has a fertility similar to other countries who have not had such similar experiences. The civil war was expected to lead to a decrease in fertility given the expected increased usage of contraceptives and decrease in the desire for children. So, the question about the role of education, contraception and desired fertility was asked. To answer this question, four main questions were asked. First, whose education is more significant in reducing fertility, husband or wife? Second, does usage of contraceptive lead to low fertility? Also, does the desire for more children or another child lead to having higher fertility? Lastly, does the woman’s education (wife’s) have only a direct effect, only an indirect effect or both direct and indirect effects? Unfortunately, there was no available data that was collected during or a few years after the civil war. Trends for Liberia from the 1970s to the 2010s have shown that education and contraceptive usage shows an upward trajectory while the average ideal number of children (proxy for desired fertility) shows a downward trajectory.
Using three waves of the Liberia Demographic and Health Survey data along with a Poisson regression analysis and the Structural Equation Model, this study found; First, for 1986 wave, husband’s education, rather than wife’s education, was more significant in explaining fertility. In the 2007 wave however, wife’s education, rather than husband’s education, was more significant in explaining her own fertility. In the 2013 wave, both wife’s education and husband’s education were found to be significant in explaining fertility. Secondly, usage of modern contraceptives has a significant positive effect on fertility for the three waves while usage of traditional methods had significant positive effect on fertility for only in 2013. This finding is consistent with empirical works that have found pacing of birth to be the reason for increase demand for contraceptives. Overall, contraception, like wife’s education, was significant in explaining fertility in Liberia. Also, desired fertility had a
significant positive effect on fertility. While education, especially wife’s education, has contributed to the fertility reduction in Liberia, contraceptive usage and desired fertility have played no role in Liberia’s fertility reduction. Age of a woman, age at first birth, child deaths and urban residence have also contributed in explaining fertility differential in Liberia. Lastly, wife’s education has both direct and indirect effects on fertility, which means the effect of wife’s education on fertility is partially mediated by contraception and desired fertility. Whereas in 1986 the indirect effect is positive, in 2007 and 2013 the indirect effect is negative.
Fertility levels have declined across the globe, but Africa’s fertility is still above its replacement. Beginning the first phase of the demographic transition in the early 1950s, Sub-Saharan Africa experienced high population growth rate due to high birth rate, but began to experience decreasing fertility in the next phase of the transition in the 1980s (Owoo et al., 2015). This declining fertility according to some scholars was led by crisis which was either economic or political in nature (Agadjanian & Prata, 2002; DeRose et al., 2002; Eloundou-Enyegue et al., 2000). Decades of high fertility (compared to other regions like North America, Europe among others) complemented by increasing survival of children in most developing countries, including Sub-Saharan Africa (SSA) has led to increasing dependency burden through an increase in young population, making economic growth and development agendas through human capital investment a daunting task (Ashford, 1990; Kollehlon, 1989; Shapiro, 2017).
Africa’s youthful population is expected to increase by 40% in 2030, doubling in 2055 from its current level and at same time continue its growth trajectory for the remaining part of this century (Population Devision, 2015). With this, Africa cannot realize much improvement in standards of living of its people (World Bank, 1989). The high demographic pressure resulting from increasing population with a fertility rate of about five children has led to unemployment especially for the youth, food insecurity and pressure on basic social service like education, health, water, electricity among others in the sub-region (Economic Community of West African States, 2012). With the relatively higher youthful population and current fertility rates way above the replacement rate, SSA still needs to control its population (Williams, 2012; Youthpolicy.org, 2010).
Increase usage of contraceptive, higher level education and a decrease in desired fertility have greater potentials of reducing fertility. For instance, according to de Bruijn (2006), research has shown that contraceptive practice is among the five-strong proximate factors in determining the level and differentials of fertility. Also, according to Pritchett (1994), a greater portion of the fertility differential can be explained by desired fertility. Whereas a report by The Africa-America Institute (2015) has shown that there has been an increase in both primary and secondary education across Africa, research by Sharan et al. (2009) and Tsui et al. (2017) have shown that, Africa has since 1995 experienced an increase in the usage of modern contraceptives and for that matter contraceptive usage.
Again, with an increase in the participation of women in the labour market complemented with increased earning opportunities available for women, desired fertility, as measured by the ideal number of children, is expected to decrease (Bhattacharya & Haldar, 2012; Kreyenfeld, 2010; Ortiz-Ospina & Tzvetkova, 2017; Snopkowski et al, 2016; Verick, 2014). Yet the space at which population is declining in SSA is quite alarming. Research has shown that, for a substantial decline in fertility to occur in the region, a significant drop in fertility desires are required (Beatty, 2015). According to Bongaarts (2008), if Sub-Saharan Africa (SSA) continue with its recent slow pace of fertility transition, the size of SSA’s population may reach 2.02 billion in 2050. According to him, this will have an adverse effect on the economies of SSA in terms of development, food security and resource sustainability.
Knowledge about education-fertility, desired fertility-actual fertility and contraception- fertility relationships will inform policymaker on how to design and implement fertility driven educational policies; on the extent to which family size ideals should be lowered to achieve the desired drop in actual fertility; on the role of contraceptives in reducing fertility. Also, it will inform policymakers about the unmet need for contraception and how best to implement family planning programs.
Statement of Problem
Although Sub-Saharan Africa (SSA) is perceived as a possible source of population threat in the future, a number of countries such as Liberia and Sierra Leone among others have received little attention as far as fertility studies are concerned. Whenever issues of population are raised in Africa, large population countries like Nigeria, the Democratic Republic of Congo among others receive the greatest attention. Small-population but high- fertility countries like Liberia, Sierra Leone, Togo, Guinea Bissau, Republic of Congo among others receive little or no attention. Also, among the sub-regions in SSA, the west has always been among the sub-regions with relatively high fertility (Bongaarts & Casterline, 2012; Bongaarts et al., 1984). Against this backdrop, this research seeks to inquire about the role of some variables in Liberia’s fertility, a country located on the west coast of Sub-Sahara Africa (SSA). The population of Liberia has more than doubled within 33 years, with more than 60% of its population under age 25. Existing work on Liberia has focused more on cultural, religious and proximate determinants of fertility (Kollehlon, 1989, 1994; Nichols et al., 1987), with Kollehlon (1984) inquiring into the employment status and occupation of women and their fertility behaviour.
Although Liberia has a relatively small population and has experienced fertility decline since 1981, it continues to have an average fertility rate of 4.6 children per woman with a population growth rate of 2.44%. The estimated population is measured at approximately 4,689,021, which represents a little more than a 100% increase from its 1984 population (2.3million). With more than a decade of civil war experience (between 1989-2003) fertility rates should be considerably low, with a signification portion of any increase coming from the younger cohorts, since they have experienced little or no crisis (DeRose et al., 2002; Eloundou-Enyegue et al., 2000). War and its resultant economic downturn are expected to
have a negative effect on fertility through delayed marriage, increased incidence and duration of marital separation, lower frequency of intercourse and impaired fecundity and gestation as well as the discouragement resulting from uncertainties. These will in turn force the affected individuals to postpone birth or delay entrance into the next party by using contraceptives even if the affected people have a high taste for shorter birth interval and larger family size. (Agadjanian & Prata, 2002; DeRose et al., 2002; Eloundou-Enyegue et al., 2000). Fertility decline during and some few years after the war is similar to that of other countries in SSA like Ghana, Zambia and Madagascar who had not gone through such crises.
Furthermore, with an increase in education, the value of time increases and is expected to cause a decrease in fertility (Diebolt & Doliger, 2005). Increase in the value of time should cause childbearing age couples (or women) to desire fewer children. Also, if because of education, the cost of having another child outweighs the cost of controlling birth, then more contraceptives will be demanded. Educational enrollment for the most part of the war period saw a gross increase from 36.40%, 9.36% and 0.93% in 1970 to 102.38%, 45.16% and 9.30% in 2011 respectively for primary, secondary and tertiary educations; contraceptive usage saw an increase as well (Liberia Data Portal, 2016; LDHS Report, 2008, 2014). Desired fertility (as measured by the average ideal number of children) saw only a slight decrease from about an average of 5 children per woman in 2007 to an average of 4.8 children per woman in 2013. Increase in educational enrollment and contraceptive usage, coupled with decreasing fertility desires is expected to cause a significant decrease in the fertility of Liberian women.
Although there is no information to investigate the effect of the civil unrest on fertility, what role did education and contraception play in the fertility changes observed in Liberia? Has the level of education, necessary for a drop-in fertility, increased over time? What was the
reason the increasing contraceptive demand? Better still, is the decreasing effect of desired fertility so small to complement the decreasing effect of the civil war, education and contraceptives usage have on fertility? Therefore, the question about the role of contraceptive, education (of both women and men) and desired fertility in determining fertility may shed more light on this puzzling situation in Liberia.
Also, the question about the mechanisms through which education affects fertility has not been answered for most countries, so in trying to know more about the roles of education, contraception and desired fertility, inquiring about whether education’s effect on fertility is fully mediated by contraception and desired fertility will contribute to the fertility literature. This analysis of mediation will give a more in-depth knowledge on the mechanisms through which education affect women’s fertility in SSA.
Justification for Mediators
Mediation is where the effect of an independent variable on a dependent variable passes through a chain of effect(s), thus, the independent variable (the causal variable) affect some variable(s) called the intermediate, intervening or process variable(s), which in turn affect the dependent variable also called the outcome variable. The mediating variable accounts for the relationship (thus, the why and how) between the independent variable and dependent variable. The need for mediation test is to broaden our understanding of the processes through which one variable affects another variable (Namazi & Namazi, 2016). For instance, education is known to affect fertility negatively, how does this effect takes place? First of all, education increases the productivity of a person which in turn increases his/her chance of gaining employment in a better-paying company or working environment. Consequently, employment in a higher paying job will increase the income/wealth of the individual and for that matter his/her household. This will intern cause the individual to
demand more quality children over quantity. (Becker & Lewis, 1973; as cited in Ahene- Codjoe, 2007). This implies that, education leads to the employment of a person, which in turn leads to the increase in the income or wealth of his/her household and then eventually leads to a decrease in fertility. Also, highly educated women can take good care of their children by providing them with better nutritional diet and health care, which then increases the survival rate of their children. A woman of such calibre would be certain of achieving her desired fertility. Thus, the education of women, decrease the mortality rate of her children which in turn decrease her fertility. In these two examples, education is the independent or causal variable, employment, decreased child mortality and increased household income are the mediating variables and fertility is the dependent or outcome variable.
Education, especially women’s education, is known to decrease own child mortality, increase knowledge and use of contraceptive and increase the socio-economic status of a woman among other variables. All these factors also affect fertility (Jejeebhoy, 1995; Snopkowski et al., 2016). As to whether these mediating variables fully or partly mediate the effect of education on fertility remain unanswered in most countries and is part of the objectives of this research to determine whether contraceptive usage and the fertility desire of a woman mediate the effect of education on fertility in Liberia. For instance, Tawiah (2017) pointed out a possible relationship between education and contraceptive usage, living conditions and age at marriage in a study he carried out in Ghana, whereas a paper by Snopkowski et al. (2016) estimated the relationship between women’s education and working status, local mortality, husband education, social class and contraceptive use. In the three sites (locations) selected by the authors, the estimation produced the expected relationships between education and the five mediating variables, but the relationship between these variables and fertility is where the difference lies. Most of these mediating
variables were mediated by another set of variables. In a nutshell, while some yielded the expected relations in two or all the sites, others showed no relationship between them and fertility.
In Liberia, while one research talked about a break in the link between education and employment (Kollehlon, 1984), other empirical works have shown that education increases the use of contraceptive in Liberia (Nichols et al., 1987; Nicholas, 1995). Since, enough work has not been done on fertility and other variables, it is difficult to conclude without any evidence about variables that will mediate the effect of education on fertility and so using data from Liberian Demographic and Health Survey (LDHS), more can be said about the variable that mediates the effect of education on fertility.