Before we tackle the question of why parents prefer sons over daughters, let’s review some fundamental concepts of evolutionary biology and psychology.
You need to have an understanding of these concepts before proceeding and if you’re already familiar with them, a nice little review won’t hurt.
It is the number of children an individual can produce in his or her lifetime. In humans, males have a higher reproductive potential than females simply because they produce much more sperm in their lifetime than females produce eggs.
While males tend to have higher reproductive potential, females tend to have higher reproductive certainty. This means that almost all females reproduce whereas a significant number of males may not get a chance to reproduce at all.
Phrased in a different way, we can also say that human males have a higher reproductive variance than females.
Our psychological mechanisms are wired to seek reproductive success i.e. successfully passing on as many genes to the next generation as possible (having kids who can successfully reproduce).
A good way to measure a person’s lifetime reproductive success is by counting how many children and grandchildren they leave. The more the number the higher their reproductive success.
Keeping these concepts in mind, let’s delve into the question of why human parents sometimes prefer sons over daughters…
More sons = greater reproductive potential
Since human males have a higher reproductive potential compared to females, having more sons means more of your genes have the chance of making it to the next generation.
When it comes to reproductive success, more is better. Having a head start is always preferred. If conditions turn out bad later and some genes die, others can survive. Therefore, parents prefer sons over daughters in average conditions.
Average conditions mean that the factors that influence reproductive success are not extreme.
Now, there can be many factors that can influence reproductive success but one of the most important of them all is ‘availability of resources’.
Hence, in this case, ‘average conditions’ would mean that the resources that the parents can invest in their children are neither too much nor too less- they’re average. But what if the resources are not average? What if parents have less or more than average available resources to invest? Will that affect their preference for sons versus daughters?
Reproductive certainty matters too
Reproductive success is both a function of reproductive potential and reproductive certainty. It’s just that under average circumstances, reproductive potential becomes more important because there’s already a good degree of reproductive certainty.
But when the available resources are meagre, the balance of the equation shifts. Now, reproductive certainty becomes more important. In other words, when available resources are less, reproductive certainty becomes a more important determinant of reproductive success.
As you might have guessed, in such a situation daughters become more preferable than sons because they have greater reproductive certainty.
When you don’t have a lot of resources to invest, you can’t run the risk of producing sons whose reproductive certainty is low. They may not get a chance to reproduce at all, especially when their parents are able to invest very little in them.
There is a direct relationship between the reproductive success of males and their resourcefulness. The more resourceful a male, the higher he is on the socioeconomic ladder and the greater his reproductive success tends to be.
Therefore, when there’s a resource constraint, parents can’t simply go for the possibility of passing on a greater number of genes to the next generation. They’ve got to aim for certainty. As they say, ‘beggars can’t be choosers’.
It isn’t surprising, therefore, that women without a long-term partner or married to low-status men tend to produce an excess of daughters while women married into resourceful families tend to produce an excess of sons.
Known as the Trivers-Willard effect, research has shown that humans in the highest economic bracket (Forbe’s list of billionaires) not only produce an excess of sons but also leave more grandchildren through sons than daughters.1
The logical conclusion that we can make from all that we’ve discussed above is that parents who have slightly less than average resources should show no preference towards either boys or girls. They should prefer boys and girls equally.
The slight decrease in resources cancels the reproductive benefits that having additional male sons could generate. However, should economic conditions worsen, they’re likely to prefer girls over boys.
An interesting study conducted by researchers from two business schools showed that parents who had both daughters and sons spent more on daughters in bad economic times.2
These parents seemed to unconsciously understand that in tough economic conditions reproductive certainty became more important than higher reproductive potential.
Here’s a short animation by MinuteEarth throwing more light on this phenomenon:
Consistent with what we’ve learned so far, a study conducted in polygynous Northern Kenya showed that economically sufficient mothers produced richer milk (with more fat) for sons than daughters while poor mothers produced richer milk for daughters than sons.3
Note that in a polygynous society, a male with higher socioeconomic status has a greater chance of attracting multiple wives and having multiple children and grandchildren with them.
- Cameron, E. Z., & Dalerum, F. (2009). A Trivers-Willard effect in contemporary humans: male-biased sex ratios among billionaires. PLoS One, 4(1), e4195.
- Durante, K. M., Griskevicius, V., Redden, J. P., & White, A. E. (2015). Spending on daughters versus sons in economic recessions. Journal of Consumer Research, ucv023.
- Fujita, M., Roth, E., Lo, Y. J., Hurst, C., Vollner, J., & Kendell, A. (2012). In poor families, mothers’ milk is richer for daughters than sons: A test of Trivers–Willard hypothesis in agropastoral settlements in Northern Kenya. American journal of physical anthropology, 149(1), 52-59.