Syndicate content

A Daughter Deficit in Africa?

Shanta Devarajan's picture

An excellent special issue of the New York Times magazine on Women and Development  had an article on the “daughter deficit”—the phenomenon, observed in India and China, of many fewer girls than boys being born, and surviving to age 5.  Up to now, I had been thinking of this as an Asian phenomenon, associated with cultural values in India and China.  But the finding by my colleagues Jed Friedman and Norbert Schady, reported in this blog , that in Africa the mortality rate from a drop in income is about twice as high for girls as for boys, makes me think that the daughter deficit (or “son preference”) may be coming to Africa. 

As Jed and Norbert show, the difference is unlikely to be for biological reasons (such as that male fetuses that survive to birth may be stronger than female ones), so the explanation is probably behavioral.  The Times article suggests that, in Asia, the son preference increases as incomes grow (the boy-girl ratio is higher in the richer states of India, for instance).  Could it be that, as per-capita incomes rise in Africa, we may start seeing a widening gap between infant boys and girls?


It's entirely understandable that so many observers of India, like Amartya Sen, have linked the growing masculinisation of the population to adverse changes in women's status. Understandable, but I would argue, wrong. I've tried to show elsewhere ("India's Falling Sex Ratios" Population and Development Review Volume 25, No. 2 (June, 1999), pp. 323-343) that at the same time that female life expectancy at birth in India has risen from something over 20 to over 60, the percentage of girls in the population has steadily fallen. One measures real gains, while the other--apparently--retrogression. The same relationship holds with female literacy: as more girls learn to read, apparently, girls form a smaller percentage of the population. The resolution of this paradox in India--and quite possibly in Africa--I believe resides in the large increases in population size which accompany the demographic transition. Very slight differences in life expectancy of boys and girls (often differences of just a few months) lead to the survival of slightly more boys...but in generations which are much larger than those that preceded them because of the increase in life expectancy. It's the aggregation of those slight differences, I believe, which causes the divergent sex ratios. If we want to measure changes in the status of girls in women, we should do it directly by looking at literacy, life expectancy, height-to-weight ratios, etc. Sex ratios are unreliable.

Submitted by muhidin on
many a time i have asked the sex of the children born to my firends and many other people i intract with and all of which have said boy. am not surprised to read this because it appears as though this is the trend now. my thinking to this would be the advancement in technologies that can detect foetus sex at very tender ages might mean the termination of the pregnancy or carry on. the advancement in science and medicine has had an implication of the ratios because abortion is no longer a big deal as a woman can carry out an abortion and walk home and ready for other activities.

Submitted by Karina Trommlerova on
I would disagree with the Times article that suggests son preference to be growing with increasing income. Similarly, I don´t agree with Peter Mayers comment in this blog that "at the same time that female life expectancy at birth in India has risen from something over 20 to over 60, the percentage of girls in the population has steadily fallen". This comment may be correct, but it does not provide any evidence whatsoever that there are no adverse changes in women's status. The response to both statements is very simple: availability of gender-specific abortions. @ Times: In my opinion, the phenomenon of increasing income and increasing sex ratio (male-female-ratio) in Asia as found by Times is largely related to the availability of abortions. With a higher income, it becomes easier to find out the gender of a fetus and to abort a girl-fetus before it is born. Poor households don´t have resources for this. Thus, son preference is generally in the culture of these countries, and high income together with new technologies makes it possible to apply son preference in the real life. The preference does not change with income, only the possibility to apply it. @ Peter: Consequently, the baby-girls that are born (are not aborted) are really desired and their life expectancy will be higher. However, the proportion of women in society is falling because girl-fetuses are being aborted. If you don´t call this adverse position of women in the society then I don´t know... In sum, if there is a son preference in African societies (which in my opinion is), the sex-ratio (male-female-ratio) will grow with economic growth just like it happened in Asia. At the same time, the number of doctors doing baby-gril abortions will be on the rise.

Add new comment