Educational Gender Inequality in Sub‐Saharan Africa: A Long‐Term Perspective

While sub-Saharan Africa has a poor and erratic record of economic growth over the long 20^th century, its sustained expansion of education is beyond dispute.  Average educational attainment increased from 0.2 years of schooling in 1900 to 2.3 in 1970 and 5.7 in 2010 (Barro & Lee 2015). Despite this significant progress, SSA’s educational expansion was highly uneven, with certain regions and particular sections of the population benefiting earlier and more than others.

Gender was a major fault line, as boys benefitted disproportionally from emerging educational opportunities. Sub-Saharan African girls today are the most disadvantaged in terms of access to schooling worldwide. Twelve out of 17 countries in the world that have not yet reached gender parity in primary education are located in sub-Saharan Africa. Equally, out of the 20 countries that have not achieved gender parity in lower secondary education are located in sub-Saharan Africa (UNICEF 2020).

In a recent study (Baten, de Haas, Kempter & Meier zu Selhausen 2021) we analyze educational gender inequality in Africa using a birth-cohort approach, covering most of the 20^th century, connecting the colonial with the post-colonial era. We analyze gender gaps in education at three levels:

1. Comparing sub-Saharan Africa to other developing regions (using data from Barro and Lee 2015)
2. Comparing 21 African countries among themselves.
3. Investigating correlates of gender inequality for 6 decades across 1177 African birth regions (using census data on 15.4 million individuals from Minnesota Population Center 2019).

Sub-Saharan Africa compared to other Developing Regions

At the beginning of the 20^th century sub-Saharan Africa exhibited the lowest gender gaps, measured in years of schooling, compared to the Middle East and North Africa (MENA), South Asia (SA) and South East Asia (SEA) (Figure 1). In the early 20^th century, a bit later in sub-Saharan Africa, average educational gender inequality increased everywhere, peaking between 1.5 and 2.5 years of schooling. By the 1980s, Africa displayed the highest gender disparities, despite significant post-colonial closure of the gender gap.

Figure 1: Gender gaps in developing world regions over time

Source: Authors calculations on Barro & Lee (2015).

The ‘Educational Gender Kuznets Curve’

Absolute educational gender gaps provide insight into the unequal accumulation of human capital between boys and girls, but they are more meaningfully interpreted in relation to overall educational expansion. In Figure 2, we therefore express the gender gap in relation to the number of years of education received by boys. Across world regions, this exercise reveals a consistent inverted-U trajectory, which we refer to as ‘educational gender Kuznets curve’. In this perspective, Africa’s performance in terms of gender equality in education no longer comes out as particularly poor. Consistently across world regions, gender disparities first rose as male education expanded, and then declined at 4-5 years of male education. This is also true for sub-Saharan Africa. In fact, throughout its educational gender Kuznets curve, sub-Saharan Africa performed comparatively well. Educational gender inequality in Africa today is still high not because of an exceptional tendency to favor boys, but because educational progress is slow in general, and it is still in its early phases.

Figure 2: Gender gaps in developing world regions in relation to male years of education

Source: Authors calculations on Barro & Lee (2015).

Results of our African cross-country comparison confirm this pattern of an inverted U-shaped educational gender gap trajectories. Almost all of the observed countries saw an initial increase of their gender disparities until the mid-20^th century followed by a levelling off or decline. However, countries in southern Africa (South Africa, Botswana, Lesotho and, to a lesser extent, Zimbabwe) had remarkably low or even reversed gender gaps throughout the 20^th century. This is a striking similarity, since the educational systems of these countries and the degree of overall educational expansion differed markedly between them, despite their geographical proximity. Male absenteeism due to cattle herding and male labor migration for mining activities are likely candidates to explain this more gender-equal south African pattern, as women were left behind and may have benefited from their relative immobility in terms of (missionary) school attendance.

Understanding Sub-National Gender Gaps in Sub-Saharan Africa

Our world region and country-level analysis builds on an existing literature comparing  educational gender gap trajectories at the country-level (see, especially, Barro and Lee 2015). Much less explored is the variation of educational gender gaps within countries, even though these may be expected to be large. To explore sub-national variation, we dig one layer deeper, exploiting the detail provided by IPUMS-curated census data to analyze gender gaps at a fine-grained sub-national region level (Figure 3). Since administrative boundaries change between censuses, we could not consistently trace the ‘educational gender Kuznets curve’ over time. Instead, we created short birth-cohort panels. We respectively use two decades jointly to analyze three periods (1920-39, 1940-59, 1960-1979) in a multivariate regression framework applying the Least Squares Dummy Variable (LSDV) estimator.

Figure 3: Sub-national gender gaps in education (male minus female years of school) not explained by linear and quadratic effect of male education (1940-49 birth cohorts).

Source: Authors calculations on Minnesota Population Center (2019).

To account for the expected presence of an inversely U-shaped relationship between the gender gap and male educational expansion, we control for the linear and quadratic effects of male educational expansion, which again displayed an inverted U-shaped pattern. After controlling for the educational gender Kuznets curve, we were still left with substantial residual regional variation to explain (see Figure 3 for the 1940-59 birth cohorts).

In depth-analysis showed that better connected areas on the coast, with early access to railroads, and those harboring urban locations achieved greater gender equality in schooling than more remote locations. This supports the idea that increased openness benefited African girls’ education throughout the 20^th century. We also find that in districts which formed the heartland of missionary presence in the early 20^th century (both Catholic and Protestant), gender gaps were lower throughout the entire period that we studied, even long after mission schools lost their monopoly in British Africa in the late colonial period (Frankema 2012; Cogneau & Moradi 2014; Meier zu Selhausen 2019). Conversely, we did not find consistent correlations between educational gender gaps and cash crops, agricultural systems, Islam or family systems.

Closing the Educational Gender Gap?

Addressing Africa’s educational gender gaps today is contingent on a good understanding of their drivers, which in turn requires a long-run perspective. Indeed, educational gender gaps were highly dynamic over the 20^th century, increasing rapidly in the early phases of educational expansion, before stagnating or declining in later phases. The bright prospect then is that, as overall education expands further, girls are likely to benefit disproportionally, even without specific policies targeting gender discrimination.

At the same time, we show that countries and regions move through their educational gender Kuznets curves in different ways, some experiencing much larger gaps and slower convergence than others. Our sub-national analysis suggests that closer integration of marginalized regions in national and regional economies will plausibly contribute to more gender equality in educational attainment. Moreover, the persistent correlation between early colonial missionary presence and education gender gaps illustrates how strong and lasting specific local investments in education can be, while also testifying to their limited diffusive effects beyond specific locations.

References

Barro, Robert J. and Lee, Jong-Wha (2015). Education Matters: Global Schooling Gains from the 19th to the 21st Century. Oxford: Oxford University Press.

Baten, Joerg, De Haas, Michiel, Kempter, Elisabeth and Meier zu Selhausen, Felix (2021). “Educational Gender Inequality in Sub‐Saharan Africa: A Long‐Term Perspective.” Population and Development Review 47(3): 813-849.

Cogneau, Denis, and Alexander Moradi (2014). “Borders That Divide: Education and Religion in Ghana and Togo Since Colonial Times.” Journal of Economic History 74(3): 694–729.

Frankema, Ewout (2012). “The origins of formal education in sub-Saharan Africa: was British rule more benign?.” European Review of Economic History 16(4): 335-355.

Meier zu Selhausen, Felix (2019). “Missions, Education and Conversion in Colonial Africa”, in David Mitch and Gabriele Cappelli (eds.), Globalization and the Rise of Mass Education, Palgrave Studies in Economic History, Palgrave Macmillan.

Minnesota Population Center. 2019. Integrated Public Use Microdata Series, International: Version 7.2. Minneapolis, MN: IPUMS.

UNICEF (2020). Gender and education. UNICEF Data (accessed 20 May 2020) https://data.unicef.org/topic/gender/gender-disparities-in-education/

This blog first appeared on the N-IUSSP blog.

Feature image: Photo by Doug Linstedt, 2016, Lagos on Unsplash: https://unsplash.com/photos/jEEYZsaxbH4