How does a ban on sex-selective abortions affect child health?

Widespread access to sex selection technologies has further skewed the male-biased sex ratio in India. However, this article suggests that the ban on prenatal sex determination may have adverse consequences on the health and mortality outcomes of the surviving children. Looking at children born between in the decade before and after the ban, it finds that the ban led to an increase in the proportion of unwanted girls, and increased competition for resources among children in large families.

The problem of “missing women” (Sen 1992) is a severe manifestation of gender bias. The availability of ultrasound technology and widespread access to prenatal sex selection technologies has been instrumental in aggravating this problem of son bias in patriarchal societies with deep-rooted social norms such as patrilineal and patrilocal institutions (Jayachandran 2015). As a result of prenatal sex selection and excess female mortality relative to men, the number of missing women is projected to increase to 150 million by 2035, and is reflected in male-biased sex ratios at all ages. Enforcing bans on sex screening and selection has been one of the common policy responses to address skewed sex ratios in countries like China, India, and South Korea (See Dasgupta and Sharma (2022a) for a comprehensive review).

Theoretically, a ban on sex-selective abortions, typically implemented to make sex ratios more equitable, may have adverse welfare consequences for surviving children. It can lead to worsening health outcomes through the channel of a trade-off between quantity and ‘quality’, as couples might continue to have more children until they achieve the desired number of sons. Additionally, the gender gap in health outcomes can worsen if the proportion of unwanted female births goes up following the enforcement of the ban.

The impact of these restrictions on child health outcomes has received limited attention in the literature. In our study (Dasgupta and Sharma 2021), we investigate the impact of a ban on sex-screening technologies and sex-selective abortions on health outcomes in surviving children. Using a natural experiment in India, we exploit variations in exposure to the ban and the sex of the firstborn child to analyse its effects on child mortality, health investments, and anthropometric outcomes.

Ban on prenatal sex determination

The introduction of ultrasound technologies revolutionised access to prenatal sex selection in the 1980s. Ultrasound had advantages over other methods such as amniocentesis, such as lower cost, better accessibility, and portability. This widespread use of sex detection and subsequent sex-selective abortions resulted in a skewed sex ratio, with an estimated 480,000 girls being selectively aborted annually from 1995 to 2005 (Bhalotra and Cochrane 2010). The Indian government took measures to address the issue in 1978, when the identification of foetal sex was banned in public healthcare facilities. Maharashtra became the first state to enact a comprehensive ban on prenatal sex determination in private clinics in 1988. This success prompted the introduction of the national Pre-Natal Diagnostic Techniques (PNDT) Act in 1994, which came into effect in 1996, covering all states except Jammu and Kashmir. The PNDT Act imposed strict punishments on both medical practitioners and clients, including fines, imprisonment, and license revocation.

Data and empirical strategy

The ban – implemented in 1996 across Indian states (except Maharashtra and Jammu and Kashmir) – prohibited the use of technologies like ultrasound for sex determination and criminalised sex-selective abortions. Using the household composition we categorise families into two types: those with a firstborn female child and those with a firstborn male child. Previous studies (Bhalotra and Cochrane 2010, Anukriti et al. 2021) have shown that families with a firstborn female child are more likely to engage in prenatal sex selection to meet their son preference. Thus, a ban on prenatal sex selection disproportionately affects firstborn female families. These families are more likely to resort to son-preferring fertility stopping behaviour¹. Consequently, children born into these larger families face increased sibling competition for resources, leading to higher mortality rates, poorer health outcomes, and lower parental investments.

To examine the impact of the ban on prenatal sex selection, we use retrospective birth histories from the National Family Health Survey (NFHS) conducted in four rounds between 1992-93 and 2015-16. We focus on children born between 1986 and 2005 – we drop children born before 1986 in order to consider birth outcomes of children who were born after the early diffusion of ultrasound technology in 1985 (Anukriti et al. 2021). We also omit children born after 2005 so as to ensure that our treated group remains comparable to the control group. Our analysis includes over 4.56 million mother-year observations across the years in the sample period, with almost 320,000 unique women and 1.02 million child observations for various outcomes.

Using a difference-in-differences estimation framework² we study whether the ban on sex-selective abortion worsens the relative health outcomes – such as antenatal visits and anthropometric measures – and mortality outcomes for children in firstborn female families. Our empirical strategy uses intertemporal and within-family variation to assess the impact of the ban on health outcomes and the gender gap. We also examine if the gender gap in health outcomes worsen across families post the implementation of the ban.

Findings

Our findings indicate worsening of health outcomes along both these margins. The results operate through two main channels: an increase in fertility in intensively treated families (that is, those with firstborn girls), leading to greater competition among siblings for resources; and an increase in the proportion of unwanted girls.

We find the birth of a child in a given year after the ban is 0.7 percentage points more likely in firstborn female families than in firstborn male families. However, while the ban was associated with a rise in the proportion of female children in higher-order births, we find a significant increase in mortality and a reduction in health investments for children in firstborn female families after the ban. We find in firstborn female families, after the ban, neonatal mortality is higher by approximately 25.7%, infant mortality by 26.6% and under-five mortality by 25.5% compared to the mean levels of mortality. We find a significant reduction in health investments for all outcomes – after the ban, the number of antenatal care visits, the number of tetanus injections received by the mother, and breastfeeding duration decline by 28%, 34% and 29.1% respectively in firstborn female families relative to firstborn male families.

The ban resulted in larger family sizes and reduced investments in each child, leading to worsened health outcomes. We find an increased mortality for both girls and boys in firstborn female families, suggesting that boys also suffer due to their higher birth order as couples continued with the fertility stopping rule in the era of the ban.

Conclusion

Our study contributes to the growing literature highlighting the unintended effects of imposing top-down fertility restrictions without addressing underlying social norms. While the ban on prenatal sex selection had adverse consequences for surviving children, we cannot rule out the possibility of explicit discrimination against girls within firstborn female families. We emphasise the importance of considering increased fertility as a factor in worsened health outcomes for both genders in families heavily impacted by the ban. These policies alone may not be effective in changing son preference driven by deep-rooted social norms.

Future research could focus on evaluating complementary policies that incorporate demand-side measures to shift social norms and address gender discrimination effectively (Dasgupta and Sharma 2022b). Interventions such as gender sensitisation programmes in schools and media campaigns may encourage families to value daughters and invest in their well-being. Persisting solely with punitive measures against sex-selective abortions increases female births, but does so at the expense of child health.

Notes:

Fertility stopping behaviour, or differential stopping behaviour, is when couples continue to have children until they reach their desired number of sons (see Basu and De Jong (2010) for more details).
A difference-in-differences strategy is used to compare the evolution of outcomes over time in similar groups, where one changed their behaviour as a result of an event (in this case, the ban on prenatal sex selection) while the other did not.