The Impact Of Economic Empowerment On Household's Decision On Fertility,education And Social Connectivity

The prime objective of this doctoral dissertation was to explore the impact of economic advancement on the decision making process of the household in Ghana.The household was the primary focal point of the study due to its significance role in national building and economic development.Specifically,the research aimed at investigating the direct and indirect effects of wage on fertility decisions.Additionally,it inquired into the impact of intra-household income inequality on gender disparity at secondary education.Lastly,it probed into the transmission mechanism through which income influenced some social cohesion activities namely trust level and political participation.To achieve the above objectives,secondary data from the sixth round of Ghana Living Standard Survey(GLSS 6)were used.The GLSS 6 is a national representative data which covers demographic characteristics of respondents as well as their health,fertility,educational attainment and other socio-economic information.Three different methodologies were employed to answer each of the research questions.The study at chapter three,which decomposed the total effect of wage on fertility decisions into the direct effect(quantity-quality trade-off hypothesis)and the indirect effect(opportunity cost of wage),used the Marginal Mediation Analysis(MMA).Previous studies on fertility avoided analyzing the two effects simultaneously in a single model.This is because of differences in the measurement metrics of dependent variables in the mediation model.Two dependent variables used in such model are number of children per woman,which is a count variable,and hours of work,which is a continuous variable.The equation with the count variable as its dependent variable was computed by a poisson estimation hence its estimated coefficients have multiplicative properties.The other equation whose dependent variable is hours of work was estimated by an Ordinary Least Squares(OLS)making the properties of its coefficients additive.The differences in the properties made it difficult to calculate the indirect effect.However,the MMA transformed the different measurement metrics into uniform property making computation of the indirect effect easier.But before the equations were estimated,the study dealt with the endogeneity inherent in the model.Issues of omitted cofounding variable(s)and simultaneity among some of the variables makes the results from direct estimations of the equations invalid and null.The instrumental variable(IV)estimation was used to correct for the endogeneity.Respondent's belongingness to a trade union was used as an instrument.A Two-Stage-Least-Squares(2SLS)estimation was therefore applied to the equation whose dependent variable is hours of work.On the other hand,a Two-Stage-Residual-Inclusion(2SRI)approach,was utilized in the equation with number of children as its dependent variable.Using 791 Ghanaian women,it was revealed that a percentage increase in wage directly diminished the number of children per woman by 0.85.Moreover,a percentage rise in wage increased hours on the labor market which consequently reduced fertility by 0.016.One importance of fertility decisions is its impact of the education of children in the family.It is demonstrated in literature that there is an inverse relationship between sibship size and educational attainment as resources are highly diffused in larger families.It is not surprising why enrollment in education has shot up amidst low fertility rate in Ghana.In spite of this gender inequality continues to plague the higher levels of Ghana's educational sector.At chapter four,the research evaluated the effect of intra-household income inequality on gender disparity in secondary education.There exist a substantial difference in the enrolment rate between boys and girls at secondary schools in Ghana.A contribution of the study to both theory and policies is that it postulated that one reason for this menace is the differences in parental expectation of the economic returns of education for the sexes shaped by the divergent income men and women earn on the labor market.The study asserted that if parents expect boys to reap more economic benefit from education,they will invest more resources in the education of their male children as against girls.Such expectation is highly influenced by the differences in the income of fathers and mothers in the household.Another significance of the study is that it employed the Propensity Score Matching(PSM)technique to investigate the above assertion.The advantage of this technique over other estimation approaches is that it deals with the problem of self-selection biasedness inherent in the use of observational data.PSM imitates experiments which is the gold standard in causal estimation.It reconstructs the data in observational studies so that samples in the control group will act as counterfactual for those in the treated class.In PSM,the pre-treatment characteristics of the control and treated classes are similar so the mean difference in the outcome variable between the 2 groups after the treatment has been applied is attributable to the treatment variable.Separate Propensity Score models were estimated for boys and girls.1368 boys were used in this study while the number of girls stood at 1111.To accelerate the convergence of the pre-treatment characteristics between the control and treated groups,the model for boys were stratified into 5 classes while that of girls had 6 strata.A 1:1 matching as well as matching without replacement were chosen as these approaches do not inflate the variance of the model.Covariate diagnosis tests were conducted to find out if indeed the balance in the baseline characteristics between the controls and treated has been achieved.The standard difference test as well as the box plot and cumulative frequency tests all attested that there was a balance in both the boys' and girls' models.The average treatment effect on treated(ATT)confirmed that in households where fathers earn more than mothers,parents tend to have higher expectations for boys(than girls)and hence invest more resources in their secondary education to the disadvantage of girls.Another crucial factor,together with fertility and educational decisions among others,which influence the welfare of the individual is how well s/he is connected in the society.The social dimension of the utility function is mostly underestimated though its relevance is now emerging in literature.The final estimation analysis,carried out in chapter five,sought to unearth the trajectories through which income influence social cohesion activities in Ghana.This aspect of the dissertation has added to literature by exploring the channels through which key indicators of social cohesion such as trust and political participation are influenced by income.The mediating variables used were working hours and accessibility to technological gadgets.This study,like the one in chapter three,was plagued with endogeneity arising from omitted cofounding variables and simultaneity.Here too,respondent's belongingness was used as an instrument to resolve the endogeneity.Using 1677 respondents,the Structural Equation Modeling(SEM)unveiled that income has positive direct and indirect effects on political participation.Trust level,on the other hand,is adversely influenced by income both directly and indirectly.The findings showed that the indirect effect of income on both political engagement and trust is through accessibility to technological devices.The trajectories through hours of work were,however,found to be insignificant.