We finally have clear empirical evidence for what dads have known for a long time: it is harder to raise daughters than it is to raise sons. Although social scientists have known since the 1980s that the parents of girls are more likely to get divorced than the parents of boys, why that is the case has always been a puzzle.
One hypothesis is that parents have a strong preference for boys. In countries such as India and China there are many more men than there are women. Decades ago the Nobel Prize winning economist Amartya Sen explained the phenomenon of 100 million ‘missing girls’: a strong preference for boys means selective abortions and even infanticide. Researchers believed that the differences in divorce rates in developed countries such as America between parents of boys and parents of girls is just another example of this strong preference for boys.
But a new study by economists Jan Kabátek and David Ribar in The Economic Journal (and reported in The Economics) presents another hypothesis. They use detailed family data from the Netherlands and show that there are no differences in divorce rates between the parents of girls and boys in the first twelve years after the birth of the children. This makes it difficult to believe that there was an inherent preference for boys: if there was, it would surely have come to the fore in the first twelve years. But it is only when children are older than thirteen (and younger than eighteen), that the economists find a significant difference. They calculate, for instance, that the parents of fifteen-year-old girls are 10% more likely to divorce than the parents of fifteen-year-old boys. They attribute it to the increased relational stress in families with teenage girls.
But what does this have to do with economics? Some academics mockingly refer to economics as the imperial science. They are right. When most students attend their first economics class the expectation is that they will learn about how financial markets work or, for the more uninformed, how to ‘make money’. But this is not what economists do. Like a well-known lecturer once said: ‘if I knew how to make money, I wouldn’t be here right now’.
It is better to think of economics as the study of human behaviour. Economists want to understand why people make certain decisions – and whether those decisions are better or worse for their wealth and well-being. We don’t, therefore, just study money matters, but also all the beliefs, habits and other things that determine human behaviour.
Like folklore. Think of folklore as the collection of stories, traditions and ways of doing things that are unique to a group of people and that are passed down over generations. The younger generation of Afrikaans speakers might not sing folk songs, but there are still things that are transferred between generations: from children’s stories and tolletjiebrei (a form of knitting using a cotton reel) to recipes for rusks and what to do if a lion charges at you.
In a new article in the Quarterly Journal of Economics (QJE), economists Stelios Michalopoulos and Melanie Xue create a new dataset of folktales from more than 1000 societies from around the world. They then use advanced machine learning techniques and come to three conclusions. The first is that communities that tell traditional stories depicting women submitting to men are also more likely to have laws today which subordinate women to men. One can’t help but think of the daring selflessness of Racheltjie de Beer, an Afrikaans folk tale of a young girl protecting her brother from the cold night only for herself to die, in contrast to this submissiveness.
Michalopoulos and Xue also find that communities that tell folktales where competition between characters is presented as harmful as opposed to beneficial are more likely to be risk-averse today – and also to have fewer entrepreneurs. I think back to the Tortoise saying to the Hare: ‘slow and steady wins the race’, a favourite among Afrikaans-speaking parents. Maybe such stories inspired the founders of Firstrand, Capitec and Atterbury to build successful firms.
Some communities have many stories where fraudsters are caught out and punished. Michalopoulos and Xue find that such communities have higher levels of trust and are also more prosperous. I remember some of Pieter Grobbelaar’s ‘Jakkals en Wolf’ (Jackal and the Wolf) stories from by childhood where the villain, Jackal, got away with murder. Maybe we should see the fraud of Steinhoff and others as having their roots in the Jackal’s shenanigans…
It is not only old folktales that influence our behaviour. Many of us don’t make decisions on the basis of good information, but rather on the grounds of what others around us think. One of the most important decisions of our lives – and one that thousands of prospective students will have to make before March – is the choice of their field of study. How do students make this important decision?
That is the question that nine authors ask in another new QJE article. Their answer: brothers and sisters. They use data from Chile, Croatia, Sweden and the USA to show that younger children often only follow the example of their older siblings – irrespective of whether it will be a good decision for them. The authors rely on cut-offs in admission requirements to confirm that these results are causal, in other words, that we can be sure that the younger children are simply imitating their older siblings’ decisions. Interestingly enough, this relationship disappears completely when the older sibling does not finish university; suddenly the big brother or sister is not such a fantastic role model anymore.
The fact that people even outsource the most important life decisions naturally has implications beyond the individual. Universities try hard to be accessible to as many people as possible, but the importance of the advice of older siblings in the decisions of prospective students means that there are some groups that are going to be over- and underrepresented. This makes one wonder: should universities give bigger bursaries to firstborns because of the great chance that their younger siblings will follow in their footsteps?
Universities are just one place where this role-model effect can be measured. Another study comes from my own discipline. Although the trend is busy changing, especially in South Africa, men are overrepresented in economics. Two American economists, Catherine Porter and Danila Serra, wondered if they could encourage women to study economics by exposing first year students to more female role models. They find a big effect: while only 9% of women had economics as a major before the intervention, this increased to 17% after it.
What we learn about female role models also has implications for the study about divorce. In spite of the fact that couples parenting teenage daughters have a higher chance of divorce, there is one exception: this does not occur where the father grew up in a family where there were also girls. Put differently: the 10% higher chance of divorce for couples with fifteen-year-old daughters disappears if the father of that girl had a sister growing up.
It is unsurprising that people are social creatures. We learn from listening to stories and seeing how others do things. This does, however, mean that the stories we listen to and the people we look up to matter. ‘You are what you read’ says one of my favourite movie characters. Add to that: ‘what you do depends on who and what you listen to.’
*A shortened version of this article was originally published in Rapport (in Afrikaans) on 14 February 2021.