Progressive Gender Attitudes, Time to First Birth and Sex Selection in Delhi

Statistical evidence shows that the sex ratio of India's total population is unnaturally male biased (Jha et al. (2011)). There is however disagreement on whether sex selective abortions are practiced for the first birth (Jha et al. (2006), Retherford and Roy (2003)). Using data from the NFHS-3 this paper first finds out that sex ratios for firstborns in Delhi, Punjab and Rajasthan do have significantly less females than the biologically normal ratio and creates an index of women's progressivism using a multilevel latent factor model. Progressivism is defined in terms of attitudes and perceptions favouring the advancement of women towards better conditions in society. Delhi is then taken as a case study and the effect of progressivism on the firstborn's gender and the first birth interval is assessed using competing risk models. The main findings are: (i) A one-standard deviation increase in the progressivism index increases the likelihood of a firstborn girl by 5.8 percentage points compared to women who have not yet given birth. This effect is robust to the set of covariates used to create the index and increases for different subsamples; (ii) More progressive women do not experience longer first birth intervals which, consistent with the first result, may indicate that they are less inclined to sex-select their firstborn. These results imply that women should be taught about their human rights and gender equality. More generally, regional governments should introduce interventions aimed at subsidizing and empowering girls.