Assuming no biological bias towards male or female children, and no other confounding factors, the one child law would not change the ratio of males to females. If we do assume that a particular couple can have a bias toward one gender or the other in their offspring, then the one child law would actually result in a larger number of female births than male births.
For those not familiar, the one-child laws state that ordinarily, a couple can only have one child, but due to the importance of sons in rural areas, in those areas, if the first child is a girl, the couple is allowed one more child. Now, to illustrate the effect: Suppose we have 1000 couples, each with equal chance for a boy or a girl. They all have their first child. In 500 couples, the first child is a boy, and they’re done. In the other 500, the first child is a girl, and they try again. One their second tries, half of them have a boy, and half have another girl. So we have a total of 750 boys and 750 girls.
Now suppose that some couples have a biological bias towards having boys, and an equal number have a bias towards girls. Let’s say, as an example, that a quarter of all couples are guaranteed girls, a quarter are guaranteed boys, and the remaining half all have a coin-flip on each kid (the bias would never actually be this strong, but it makes the example easier). Now, of our 1000 original couples, 250 have boys because they have to, 250 have boys by chance, 250 have girls by chance, and 250 have girls because they have to. The 500 who had girls, half of them random and half of them predestined, each try again. Of those who were random, half, or 125, have girls, and half have boys. Of those who are guaranteed girls, all have another girl. Now, we have a total of 625 boys born and 875 girls.
It’s possible, of course, that there are an overwhelming number of male births for some other reason. For instance, a couple may be more likely to abort a female than a male. But I’m not sure this is related to the one child law.
