Global family change: persistent diversity with development
Évolutions de la famille liée au développement économique : peu de convergence en vue
Luca Maria Pesando
Decades of sweeping demographic, economic, and social change have radically transformed structures, gender roles, and intergenerational bonds of families worldwide, initially in high‐income countries (HICs), and more recently in low‐ and middle‐income countries (LMICs). Luca Maria Pesando and the Global Family Change (GFC) group show that families in LMICs have transformed in multiple ways with socio-economic development, with a wide degree of heterogeneity observed across family domains and world regions, mostly driven by sub-Saharan Africa (SSA).
In a recent study (Pesando et al 2018), we provide a broad empirical overview of the relationship between family change and socioeconomic development drawing on 30 years of Demographic and Health Surveys (DHS) from 3.5 million respondents across 84 low and middle income countries (LMICs) (Figure 1). Our goal is to advance the understanding of the changes in families that are occurring in many economically less-developed nations by focusing on a set of agreed-upon domains that are key to the notion of the “family,” namely fertility, timing of life-course events such as first sexual intercourse, marriage and birth (for men and women, separately), union formation – marriage and cohabitation – household structure, and intra-couple decision-making.
To this end, we conduct two sets of related analyses. We first document associations between the Human Development Index (HDI) – a summary measure of development used by the UNDP to monitor broadly-defined development goals – and novel indicators reflecting multi-dimensional family change. Second, we examine whether there is evidence of cross-country convergence in these same indicators over levels of development and, if so, in which domains. The idea behind the second set of analyses is to assess whether countries have become increasingly similar to each other in specific dimensions of the family, such as fertility or union. Our study draws on a growing literature on whether fertility is converging across contexts (Dorius 2008; Wilson 2011), and extends this literature to broader family domains.
Family change in LMICs
The transformation of the family that has occurred across high-income countries since the 1960s is likely to be entering its peak in LMICs. Several potential drivers of family change have been identified during recent decades, all of which are highly relevant in LMICs: the largest-ever cohort of youth currently entering adulthood; dramatic technological change; rising economic uncertainty; longer lives and lower fertility; narrowing gender gaps in schooling and the labor market; and globalization forces affecting the flow of information, goods and people across the globe.
A stylized summary of our findings is provided in Figure 2, which combines both sets of analyses in a unique diagram reporting the association of family indicators with HDI on the x-axis (weak to strong association, left to right) and convergence over HDI on the y-axis (divergence to convergence, bottom to top). The four quadrants are hence defined by different combinations of associations and convergence patterns, or lack thereof. Our representation points to each quadrant being occupied by at least one family domain – conceptualized as a compendium of family indicators – suggesting that families in LMICs are distinct in many possible ways, and changes in families with development occur differently across domains. A strong association with HDI (x-axis) is observed for fertility, intra-couple decision-making, and women’s life-course timing indicators, yet cross-country convergence over HDI (y-axis) is limited to the latter domain. The remaining domains are more weakly associated with HDI, and cover a broad spectrum of convergence dynamics ranging from divergence (marriage) to modest convergence (men’s life-course timing, cohabitation, and household structure indicators). With reference to the timing indicators, we identify clear gender differences whereby countries are converging towards delayed first sexual intercourse, first marriage, and first birth for women, but less so for men.
Overall, the picture that conforms best to our findings is one of “persistent diversity with development,” whereby families in LMICs have transformed in multiple ways, with changes occurring differently across domains, world regions, and sexes.
A point that is clear from the analysis is that development is not a powerful driver of convergence for all family indicators, suggesting the need to take into account additional factors that might contribute to explaining the observed heterogeneity, such as geo-historical legacies and long-standing differences in social and economic institutions (Esteve and Lesthaeghe 2016).
Implications for policy and the key role of sub-Saharan Africa (SSA)
The emerging picture of “persistent diversity with development” has important implications for understanding the social and economic consequences of global development and globalization, and should be taken into account in policies designed to promote sustainable development and increased individual and family well-being. For instance, although development is associated with fertility declines and marriage postponement – which in turn reflect more gender-equal dynamics within couples – some questions remain unresolved, such as that of whether, globally, higher empowerment within the household translates into increasing female independence outside of unions, and that of how existing institutional support enhances female agency by reconciling family life and labor market opportunities.
Closely related to these policy considerations is the role of SSA that emerges from this study, which differs from that played by the other four regions of Figure 1. Countries in SSA – the region with the lowest levels of HDI – are those that contribute the most to the observed global heterogeneity in family change. Figure 3 reproduces Figure 2 showing how our findings would change if we excluded SSA from the convergence analysis. After removing SSA countries, we move from covering the whole plane (i.e., four quadrants – light grey shapes) to two polarized quadrants (red dotted shapes = world without SSA; bottom-left: low association/no convergence; and top-right: high association/convergence), thereby suggesting a decrease in “persistent diversity with development.”
The peculiar nature of SSA is particularly apparent in the area of fertility (Bongaarts and Casterline 2013; Shapiro and Hinde 2017) and intra-couple decision-making, where estimates show that the world would converge towards lower fertility and increased women’s empowerment in the absence of SSA. We take this evidence as suggesting that further advances in dimensions such as education, literacy, income, and health within SSA might contribute to reducing disparities across regions and put the developing world on a more defined convergence trajectory. Although a convergence trajectory is not desirable per se, it is beneficial to the extent that it is conducive to higher human capital investments, higher female labor force participation rates, and increased compatibility between family life and economic success.
Bongaarts, John and John Casterline. 2013. “Fertility transition: Is sub-Saharan Africa different?” Population and Development Review 38 (1): 153-168.
Dorius, Shawn F. 2008. “Global demographic convergence? A reconsideration of changing intercountry inequality in fertility,” Population and Development Review 34 (3): 519–37.
Esteve, Albert and Ron J. Lesthaeghe. 2016. Cohabitation and Marriage in the Americas: Geo-historical Legacies and New Trends, New York, NY: Springer.
Pesando, Luca Maria and the GFC team. 2018. “Global family change: Persistent diversity with development,” Population and Development Review [online], 1-51.
Shapiro, David and Andrew Hinde. 2017. “The pace of fertility decline in sub-Saharan Africa,” N-IUSSP, 18 December.
Wilson, Chris. 2011. “Understanding global demographic convergence since 1950,” Population and Development Review 37 (2): 375–88.