Missing women are not where you think
March 5, 2012 2 Comments
Typically there are more male births than females, an imbalance which evens itself out by the time a handful of testosterone-fuelled adolescent males have gotten themselves into sticky situations. In some parts of the world, however, there are substantially more men than women – notably in India and China, with tens of millions of ‘missing women.’ This is often put down to selective abortions/infanticide, but the data do not entirely support this theory.
Sources: Review of Economic Studies (2010) 77, 1262–1300, Missing Women: Age and Disease (pdf link); My analysis.
Around half the missing women in both countries are down to the mortality rate from mundane things such as cardiovascular disease being much higher amongst women than in the west (where fewer women than men die from such things). Over half the excess deaths occur within the adult population.
In China, circa 40% of the missing women are down to selective abortions and first year deaths, but this is not the case in India, where the largest cause of excess death is “injuries,” accounting for some 0.25m excess deaths per year, mostly occurring between the ages of 15 to 29.
Related Post: Increased chances of dying if you don’t get on well with your colleagues
Maths question: If a country were to implement a policy so that families had to stop having children once they had a boy, but could have as many daughters as they like until then, what would the male-female ratio be? Answer under the fold.
Still 50%-50%, presuming male and female births were equally likely.
Marc Gawley 
Maths question nit-pick.
Your answer is only correct if you assume that women can have (and are willing to) an infinite number of girls before they get the desired boy. Which they simply can’t. Asymptotically approaches 50/50, however …
Could you explain it more to me? In my simple mind I imagine there are 128 women. In the first round 64 boys and 64 girls are produced. 64 women now stop having children. Next year, 32 boys and 32 girls pop out, and now only 32 women want to have another try. So 16 and 16, then 8 and 8, 4 and 4, 2 and 2, then 1 and 1, and now we’re down to just one women, with everything even in terms of daughters and sons so far. In an entire population this would be a rounding error. Actually, having just typed all that out, I see what you mean now (!) In a normal size population there would be many more rounds of pregnancies, and not all women would be willing to have that many children…hence as you say, it asymptotically approaches 50/50, from the side of there being a few more daughters…
Thanks!!