The determinants of child mortality reduction in the Middle East and North Africa

Abstract

Although child mortality rates have declined all across the developing world over the past 40 years, they have fallen the most in the Middle East and North Africa (MENA) region. We investigate the causes of this through an econometric model that differs from previous studies in using the change in child mortality, rather than its level, as the dependent variable. We show that the process of child mortality decline has been characterized by convergence, whereby countries with higher levels of initial child mortality have experienced faster declines than those with lower levels. In addition, we find that public spending on health, growth rates of income and levels of caloric adequacy are robust determinants of the change in child mortality over time. Neither initial mortality status nor caloric adequacy is likely to remain as important for the MENA region in the future as they have been in the past. The region has been benefitting less and less from the convergence momentum conferred by high initial child mortality as its mortality levels have declined over time and this will continue into the future. With regard to caloric adequacy, the region is unlikely to experience significant improvements in the future as it has already achieved a high level of food sufficiency. Accordingly, most countries in the region must look to achieving more rapid income growth and higher rates of appropriately targeted public spending on health in order to achieve further child mortality reductions in the future.

Keywords:

• Middle East
• child mortality
• caloric adequacy
• public health spending
• economic growth

JEL Classification:

• H510
• I130
• O150
• O530

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. Some studies also show that initial levels of population health are associated with subsequent economic growth (Bloom, Canning, & Sevilla, 2004) but the interpretation of this link remains a matter of debate (Acemoglu & Johnson, 2007).

2. A similar comparative edge for MENA in earlier periods is reported in Hanmer, Lensink, and White (2003) and Iqbal (2006).

3. Essentially, what Caldwell (1986) did was the equivalent of plotting infant mortality levels against income per capita levels in order to get a sense of performance in the former while controlling for the latter.

4. We have reported standardized coefficients in order to make it easy to compare economic significance across the independent variables. The corresponding elasticity measure for initial mortality is 1.32 which means that a 10% increase in initial mortality in 1970 leads to a 13.2% decrease in child mortality over the 1970–2010 period.

5. More often than not the literature uses female education as the relevant measure since women are the primary childcare providers in most households and thus the more direct channels of transmission of good childcare and nutrition practices. Female education turns out not to be statistically significant in our regressions. We note in possible explanation that even total schooling is not robust in our regressions (see Table 5). Like average years of schooling, female schooling is also collinear with several of the included independent variables and this may be a factor in its being found insignificant.

6. According to the statistical note in the World Development Indicators database, this variable is defined as follows: The depth of the food deficit indicates how many calories would be needed to lift the undernourished from their status, everything else being constant. The average intensity of food deprivation of the undernourished, estimated as the difference between the average dietary energy requirement and the average dietary energy consumption of the undernourished population (food-deprived), is multiplied by the number of undernourished to provide an estimate of the total food deficit in the country, which is then normalized by the total population. See Appendix for a link to the data source.

7. Herrera and Pang (2005) used this procedure to extract the orthogonal component of the ratio of public expenditure to GDP.

8. The underlying idea is that these two instruments can influence our left-hand side variable only indirectly through their effect on the instrumented regressor.

9. A discussion of selected food subsidy programs in MENA countries and their distributional impact is provided in Iqbal (2006, pp. 58–63).

10. It also appears to have mattered for the attainment of education, another area in which the MENA region has done well in quantitative terms (see Iqbal & Kiendrebeogo, 2015).

11. An analysis of trends in undernourishment in MENA with some country-level detail is provided in the following blog by Farrukh Iqbal dated 20 October 2014: Is MENA's Undernourishment Getting Worse?