It is probably fair to say that the World Bank’s latest report on intergenerational mobility – Fair Progress? Economic Mobility across Generations around the World – is the first-ever attempt to paint a truly global picture of how achievement – or the lack thereof – is transmitted across generations. Though there are results for income mobility for a subset of countries, most of the analysis focuses on educational attainment across 148 economies, representing over 95% of the world’s population.
As nicely summarized here by the main authors, the picture that emerges from the report is decidedly mixed. Absolute intergenerational mobility in education – measured by how many children end up with more schooling than their (most educated) parent – is on average lower, not higher, in developing countries than in high-income countries. This is despite the fact that parents are typically less educated there to begin with. And although the gap is falling, that is entirely driven by lower mobility in the richer nations, not by rising mobility in poor countries. Relative mobility – one minus the correlation between parents’ and their children’s education – is also lower in poorer countries, suggesting opportunities are more unequally distributed there.
One silver lining among these mixed findings seemed to be the unambiguous decline in traditional gender gaps. Comparing people born in the 1950s with those born in the 1980s, absolute mobility (with respect to their parents) rose faster for women than for men in developing economies. Among those born in the 1950s in South Asia, Africa or the Middle East, the share of men with tertiary schooling was roughly triple that of women: 14% versus 5% in MENA, and 7% versus 2% in both sub-Saharan Africa and South Asia. When we look at the cohort thirty years younger, the gap fell to 14% versus 9% in Africa, 18% versus 16% in South Asia, and disappeared completely in MENA (24% even). Surely, this is unadulterated good news, right? The world must be moving towards gender equality in educational attainment!
Except that it isn’t. It is moving right past it… By three measures that the report looks at – (i) the share of people with more schooling than their parents; (ii) the share of people born in the bottom 50% of the schooling distribution that reach the top 25%; and (iii) the share of people with completed tertiary education – girls aren’t only catching up with boys. They are overtaking them. In fact, Africa, MENA and South Asia are the only three regions where men born in the 1980s are not behind their women contemporaries. In all other regions of the world, though there was a “traditional” gender gap for the 1950s cohort, the 1980s cohort sees a reverse gender gap in tertiary education.
And the size of that gap tends to be larger the richer the country or region. In East Asia and the Pacific, 30% of women in that cohort have completed tertiary schooling, against 29% of men. In Latin America, it is 21% versus 18%. In Eastern Europe and Central Asia, 42% versus 38%. In high-income countries, 54% versus 44%. That’s a ten percentage-point gap, and it has been growing steadily for each decadal cohort since the 1950s. Today, that ten percentage-point gap is larger than the traditional gap (where men are more educated than women) in any region of the world for this cohort.
In Figure 1 below, I take the ratio of those two shares for the cohort born in the 1980s: women with tertiary education relative to men, and plot it against GDP per capita (in logs). Obviously, those who think of gender equality as the goal would like to see all countries lined up at 1.0. There is a clear pattern that traditional gender gaps – measured here only for tertiary education – are smaller for richer countries. One should never really take cross-sectional associations as indicative of what happens “as countries develop” but, if you momentarily suspend disbelief, it seems plausible to think of this as suggesting an association between “development” and lower traditional gender gaps.
Regrettably, however, the convergence is not to the line of equality. The non-parametric regression line crosses that line at a GDP per capita of around $5,000/year, and is statistically significantly above it from $7,000 or so onwards. To zoom into the world’s richer regions, Figure 2 shows the differences (no longer the ratios) between women’s and men’s completed tertiary education for the 1980s cohort, in all countries for which we have data in three of our country groupings: Europe and Central Asia, Latin America and the Caribbean, and the high-income industrialized nations. (High income countries in ECA and LAC are grouped with the industrialized nations.) In this figure, a gap of 0.1 indicates a country where female tertiary education completion is ten percentage points higher than the same rate for men – roughly the high-income country average reported earlier, and also approximately the gap in the US and France. In countries at the extreme right of the figure, such as Slovenia or Estonia, men have completion rates around 20 percentage points lower than women. (In countries on the left, traditional gender gaps persist.)
Surely, the emergence of these massive reverse gender gaps is cause for concern. We now live in a world where women and girls remain clearly disadvantaged in terms of educational opportunities in most countries in Africa, the Middle East and South Asia (and many other countries scattered around the world) but where men and boys experience the same situation in Japan, most of Europe and the Americas.
Even in these countries, there is no denying that women still experience discrimination and disadvantage in most aspects of their lives. In the labor market, women earn less than men for the ‘same work’ in almost every country. Political representation is typically biased against women. Domestic violence victims are overwhelming female. And so on. Rightly focusing on all these forms of female disadvantage should not, however, prevent those of us who care about equal opportunities from recognizing that something very worrying is happening to our boys and young men in the educational systems of affluent societies.
It has been suggested, in the US context, that boys are more sensitive than girls to the disadvantaged circumstances around them, beginning with in-utero exposure to stress. So perhaps growing inequality – of incomes and opportunity – in richer countries can help account for some of the reverse gender gap. Others have pointed to gender differences in gang membership and exposure to crime and violence – a factor relevant in much of Latin America and the Caribbean, as well as in richer countries. The report does not really pursue the causes behind this trend, and there is no room to speculate further here. Although research on the educational disadvantage of boys is certainly going on in the United States, the findings in this report point to a much more widespread problem, including in much of the developing world. Surely much more work is needed on understanding the causes of this rising reverse gender gap. There is now wide consensus that gender inequalities are unfair, and lead to wasted human potential. That remains true when the disadvantaged are boys, as well as girls.