Basic rights and Equality

The illusion of gender parity in education: evidence from Bangladesh

4 min


Tomoki Fujii, Sijia Xu and Abu S. Shonchoy

While conditional cash transfer (CCT) programs have been broadly successful in reducing gender gaps in school enrolment and attendance, bringing more girls to school does not necessarily lead to gender equity in the quality and outcomes of education. This column reports evidence from Bangladesh that providing girls with stipends and tuition fee waivers boosted their enrolment considerably. Nevertheless, girls are disadvantaged in the allocation of educational resources within the household and their educational outcomes lag behind those of boys.

The fourth Sustainable Development Goal (SDG)—quality education—marks a significant shift in the policy focus of education in developing countries from quantity to quality. Thanks to national and international efforts over the last three decades, more girls (and boys) now come to school, and remarkable progress has been made around the globe towards the third Millennium Development Goal: promotion of gender equality and empowerment of women.

Yet looking at gender parity in education through the narrow lens of school enrolment or attendance may lead to an illusion of success, as such quantity measures do not guarantee gender parity in the quality of education that children receive.

Take conditional cash transfer (CCT) programs as an example. These programs create incentives for underprivileged households to send their children to school by providing cash when certain conditions, such as targeted school attendance, are satisfied. They have been found to raise school enrolment significantly around the world, suggesting that CCT programs targeted at girls can substantially narrow the gender gap in enrolment.

In Bangladesh, the Female Stipend Programs (FSPs), which started in 1994 and provided girls with stipends and tuition fee waivers, have been credited with narrowing the gender gap in enrolment. Between 1990 and 2016, the gross secondary school enrolment rates for boys and girls increased from 27% and 14%, to 66% and 72% respectively.

While the achievement of gender parity in school enrolment is commendable, girls may still lag behind boys in other aspects, such as educational quality and performance. One important source of gender disparity could come from intra-household allocation of resources—due to strong parental preference for boys—leading to a gap in quality education for girls.

We show that this preference bias may have led to a systematic gender gap in educational quality and performance in Bangladesh, a predominantly patriarchal country. We observe girls consistently underperforming boys, as measured by the pass rates and shares of top students in the Secondary School Certificate (SSC) examination, a national test prior to completing school (see Figure 1).

Figure 1: Performance in the SSC examination by gender over time.

Notes: The solid lines represent the proportion of boys (blue) and girls (red) who have passed the SSC examination among those who took the examination and the dashed lines represent the share of top students who achieved the highest grade point average (locally known as ‘GPA 5’).

Source: BANBEIS-Education Database.

Some previous studies have tended to explain girls’ underperformance in school with supply-side factors, such as a low share of women teachers, unfavorable gender attitudes of teachers, and the lack of a gender-appropriate school curriculum.

In contrast, our study highlights the importance of demand-side constraints due to the allocation of educational resources within households. This constraint potentially limits the effectiveness of education policies and programs.

Analyzing four rounds of household surveys with detailed data on educational expenditure, we investigate the gender gap in three related household decisions about education: first, enrolment; second, expenditure on education conditional on enrolment; and third, the share of educational expenditure on the ‘core’ component—which includes items that directly affect educational quality (such as private tutoring).

We find a clear pro-female bias in the enrolment decision. On the other hand, the decisions on educational expenditure and core share—conditional on enrolment—are significantly pro-male.

For example, girls were 12 percentage points more likely to be enrolled in secondary school than boys in 2010. Nevertheless, conditional on enrolment, the educational expenditure and the core component expenditure for girls in 2010 were lower than those for boys, by 8% and 12% respectively.

Further investigation shows that girls’ disadvantage in conditional expenditure mainly comes from household spending on tuition fees and private tutoring—important items for quality education.

Consistent with this, girls have been significantly less likely than boys to achieve timely graduation from secondary school, conditional on graduation from primary school. Therefore, even though CCT programs like the FSPs can be effective in bringing girls to school, and in helping to improve or even reverse the gender gap in the quantity of education, they may be ineffective in narrowing the gender gap in the quality, amount, and kind of educational resources given to children.

This finding is applicable not only to Bangladesh, but also potentially to many other developing countries that are struggling to achieve gender parity in education, especially those in South Asia. Complementary policies—such as school quality improvement programs, and vouchers for free supplementary or remedial education—may be needed to reduce the gender gap in the quality of education, providing equal paths for girls to succeed.


Tomoki Fujii
Associate Professor and Associate Dean
Sijia Xu
Assistant Professor
Abu S. Shonchoy
Assistant Professor