Countries with Slow and Irregular Fertility Transitions

Abstract

Data from Demographic and Health Surveys and the UN Population Division present fertility trends in countries where fertility decline has been slow and irregular, i.e., Cameroon, Comoros, Côte d’Ivoire, Guinea , Mozambique, Senegal, and Tanzania. Using the decomposition method, the chapter highlights the influence of female education as well as the role of some proximate determinants of fertility. The results of the analysis show the predominance of the impact of educational policies and programs on fertility trends. Improving female educational levels appears to be the driving force of fertility decline, while the absence of progress in female education , or its deterioration, are stumbling blocks. Slow fertility declines or their stagnation are compounded by relatively high levels of unmet need for family planning and high mortality levels, which are prone to reversals. Hence the need to reinforce female educational programs, better address the unmet need for family planning, keep up with efforts in improving maternal and child health , and allocate more resources to the health, education, and employment sectors. These conditions all appear to be necessary for an accelerated fertility decline, the opening of a demographic window of opportunity, and the capturing of a first demographic dividend.

Appendix

1.1 Methodological Note on Decomposition Method

We use the decomposition method to determine the sources of change at the national level of fertility measured by the Total Fertility Rate (TFR). Two decomposition types are used: basic decomposition and advanced decomposition to specify the various sources of change in the TFR.

1.1.1 1 Basic Decomposition

The basic decomposition assumes that the TFR at the national level at time t (Y_t) is a weighted average of TFR of different levels of education (y_jt) and the proportion of the female population aged 15 to 49 in each category of the level of instruction j (w_jt).

$$ {Y}_t={\displaystyle \sum }{w}_{jt}\ast {y}_{jt} $$

As a result, the national change in the TFR can be decomposed into the composition effect (change over time in the distribution of women in different categories of the level of education) and the performance effect (changes in women’s behavior within the various categories of the level of education) as follows:

With ${\overline{y}}_j=\left({y}_{j{ t}_1}+{y}_{j{ t}_2}\right)/2$ and $\varDelta {w}_{\mathrm{j}}=\left({w}_{j{ t}_2}-{w}_{j{ t}_1}\right)$ the same formula is applied for${\overline{\mathrm{w}}}_{\mathrm{j}}$ and Δw_j.

1.1.2 2 Advanced Decomposition of the Performance Effect

Here, the performance effect is broken down as follows:

$$ {\mathrm{y}}_{\mathrm{j}}=\alpha +\beta\ {\mathrm{x}}_{\mathrm{j}} + {\mu}_{\mathrm{j}} $$

Where:

• α is the basis performance when x = 0 (x here is the level of education of women).

• β is the increase of the TFR associated with a unit increase in the level of education x

• μ _j is the residual effect of other variables that are not considered in the model.

The total change of the TFR is as follows:

Thus, the performance effect is broken down into the three following components:

• B1: Basic performance effect . This is the effect of education policies and programs implemented in the country (schooling, literacy, and information and communication for women’s behavioral change)

• B2: the effect of the differentiation of the TFR by level of education of women (risk related to level of education)

• B3: the residual effect of other variables not considered.