No wait stop it matters how you raise your kids

This is co-written with Oana Borcan, a fellow economist at the University of East Anglia. It has appeared in The Psychologist. I thank the editor and Oana for allowing me to post it here.

To parents besieged by expert advice, a new scientific idea offers relief: it’s no big deal how you raise your children. Originally proposed by Judith Rich Harris in The Nurture Assumption, and promoted by popsci big shots like Matt Ridley and Stephen Pinker, the idea got another push from Robert Plomin in his 2018 book Blueprint. The latest proponent is economist Bryan Caplan on the 80,000 Hours podcast. Their argument is backed by heavyweight research, but simple at heart. Social scientists, seeing how some parents had nice, clever, hardworking, successful children, assumed the parents must be doing something right. They spent ages hunting for that secret sauce. But they had forgot about genetics! Nice parents might have nice children because they pass down genes which encourage niceness. The same for brains or hard work. In fact, once you account for genetics, there’s little evidence that parenting plays any systematic role. Identical twins are about twice as similar on these dimensions as non-identical twins; since they are also twice as similar genetically, we don’t need “shared environment” or “nurture” to explain anything. While being nice to your children matters for its own sake, improving them is a lost cause. As Plomin puts it, “parents matter, but they don’t make a difference”.

With Plug, it’s probably genetics. With Fatty, let’s not rush to judgment.

This news would be big if true. Since Erasmus, parents have obsessed about how to educate their children. Methods have changed from Puritan rigour, to 18th century experiments (the young Charles James Fox was encouraged to dive head first into a bowl of cream), to earnest Victorian Dads, to today’s helicopterers; but the level of effort is a constant — all based on the idea that children need the right upbringing, and according to this theory, all wasted.

So, is it true? After all, as the Guardian review of Blueprint said: “I am not in a position to question the science”. Well, we are in a position to question the science. The “no difference” theory sounds unlikely because it is. In fact, it’s wrong.

Great scientists can also have blind spots

Harris and Plomin are great scientists, so what did they miss? First, the twin studies behind the “no nurture” thesis are based on key technical assumptions. In particular, they assume that genes and environment are independent. For example, parents can’t treat identical twins more alike than they treat fraternal twins. If they do, then a twin study will wrongly blame the results on genetics.

But it seems very plausible that parents react to their kids’ nature! If your child likes the piano, then you might find a piano teacher. If she prefers the drums, then you have a moral dilemma. Suppose most parents tailor their upbringing like this. Then twin studies can be misleading.

Second, scientists are herd animals. Twins researchers study some things more than others. They’ve focused a lot on IQ and personality — both constructs which were designed to be robust, i.e. hard to change, over time. Psychologists such as Lucy Maddox point out that life experiences and opportunities come in different shapes that are not captured by these rigid standardised measures. Indeed, parents might not be surprised to hear that they can’t change their kids’ deepest personality. Perhaps they just want them to wipe their feet, do well in exams, and not murder anyone.

But twin studies’ most important problem is: they can only pick up the variation that’s already out there. If at present all parents provide very similar environments for their children, then they won’t make their children very different. This matters, because late twentieth century Western childrearing was probably more uniform than ever before or since. Almost all children in these studies went to state-provided schools with nationally-shared curricula and policies; when they got home, they sat in front of the TV, watching the same programs as everybody else.

Before this unique period, families couldn’t be so similar. After it, they no longer needed to be. In previous centuries, families simply had to do things themselves, or with their neighbours. Play and education both took place in the home, and homes were as diverse as people. And most education was apprenticeship: a miller’s apprentices learnt milling, a typesetter’s typesetting. Today, parents and children can access any media they prefer, from Peppa Pig to South Park, or darker, more disturbing content like Teletubbies. With phenomena like charter schools and online education, variety is slowly infiltrating the old uniform education systems, too.

In this era, a hundred flowers can bloom. We don’t yet know how much our children can do, given the right environment! We could be entering a new age of experimentation in upbringing, where parents from West Coast tech elites to Amish farmers can learn from each other, while educational institutions range from traditional to Montessori to homeschooling or even unschooling.

Or we could trust in data from the most homogenous education era ever, and assume there’s no way to do better. But even in that mass-production world,  is it really true that family environment makes no difference? Actually, no. Recent studies have shown that upbringing has an impact on children’s later life success.

Parenting makes a difference in systematic and measurable ways

This is known to economists from the clearest possible scientific tests: randomized experiments. We can’t experimentally reassign children to different parents — we’re not monsters, and please don’t call to offer us your teenager — but sometimes real life does that anyway. Here’s an example: some Korean adoptees were assigned to American adopters by a queueing system which was essentially random. So there was no correlation between adoptees’ and parents’ genes. Yet, adoptees assigned to better educated families became significantly better educated themselves. Adopters made a difference in other ways too: for instance, mothers who drank were about 20% more likely to have an adoptive child who drank. This can’t be genetics. It must be something about the environment these parents provided. Other adoption studies reach similar conclusions.

More evidence comes from the grim events of death and divorce. If your parent dies while you are very young, you end up less like that parent, in terms of education, than otherwise. Again, that can’t be genetics. And children of parents who divorce become more like the parent they stay with. In other words, when parents spend time with their children, their behaviours and values rub off.

So, if you care about your children, you’re going to have to put the effort in. Dead is the dream of locking them in the cellar, browsing Netflix, and trusting your genes will set them right. More seriously, Plomin’s argument is that you should just be nice to your kids in the here and now: “Parenting is not a means to an end… our relationship with our children should be based on being with them, not trying to change them.” The idea goes back to Jean-Jacques Rousseau, who argued that since 18th-century children often died before adulthood, parents should focus on making them happy in the present, rather than improving their future self. If parenting makes no long-run difference, setting tough boundaries, or withholding treats until the homework gets done, is pointlessly harsh. But as we’ve seen, that isn’t necessarily true, and while you may love being with your children, we think you will love it more after teaching them not to throw food at you.

So, how can parents put their children on the path of life-long achievement?  A lot of what we know comes from psychology research on attachment and economics research on “early life investments”. The mind is at its most malleable in early childhood, and infants imitate everything they see.

Since the 1960s psychologists have talked about the importance of early attachment. The theory is that if a child has a strong emotional bond with its parents or carers, it develops a sense of security and trust, and that relationship models the way it feels and interacts with others later. So, if you give an infant undivided attention, spend quality time and respond promptly to their needs, they will acquire social competencies and better emotional regulation. One can debate about how universally valid this model is, and whether it applies equally to Western nuclear families and farming communities in Africa where caregiving is an extended family affair. Child-led parenting, especially when you’re doing it alone, often feels like Sisyphean work. Western parents may dream of playing pass-the-parcel with a choir of benevolent aunties expressing their delight in jolly, high-pitched tones. But the benefits of persevering are clear. A 2021 systematic review of 102 randomised controlled trials across 33 countries, from low- to high-income, concluded that parenting interventions in the first 3 years of life, especially those that emphasised responsiveness to the child’s needs, do indeed improve children’s socioemotional development and cognitive skills.

Economists obsessed for a long time with how children gain standard cognitive skills such as reading and counting, because we know those skills translate into better livelihoods later on. It has recently dawned on us that soft skills, such as resilience, patience, and initiative are equally, if not more helpful in life. Parents teach all these skills to some extent, but by putting in the time and using basic props like books and toys, they can accelerate their children’s learning. Recent studies — again, based on rigorous randomised trials — show that enriching parent-children interactions can produce important gains in cognitive and socio-emotional development. For instance, a program in Colombia showed mothers how to spend quality time with their babies, including reading and playing with home-made toys. Those babies quickly got better at verbal and non-verbal communication, and became more responsive and easier to pacify in social interactions.  The gains were larger for children from deprived families.

A counter-argument is that many environmental effects are short run. You can force your boy to study, but as an adult, nature will out and his idiocy will reassert itself. Maybe, but “short run” outcomes like exams matter to a child’s future! A halfwit with a degree earns more money than one without. He may even rise to political prominence. And some environmental effects linger on. When mothers of growth-stunted children in 1980s Jamaica were taught how to engage their children with home activities to provide cognitive and psychosocial stimulation, their children caught up with their peers. Twenty years later, those children were still earning more than the control group.

The bottom line is this: how much and what you say to your child from their first few days literally carves new paths in their brain. We know this from research on speech development. When mothers responded to their babies’ cues with the most basic vocalisations, they accelerated their children’s language development. So go ahead and babble along with your toddler.

Parents matter and they make a difference

The record of parenting “experts” is inglorious. In the 1930s, mums were told that cuddling their children would raise a dependent sissy. In the 1950s, autism was blamed on “refrigerator mothers”. Harris and Plomin offer a refreshing, libertarian alternative. Relax! Parenting makes no difference. Just be nice. Unfortunately, this siren message is no more accurate than the previous ones. We all have good reasons to try very hard to bring our children up right.

It can be hard to extract advice from science. Science is great! But upbringing is complicated. We don’t understand it very well. Scientists are one-eyed. They focus on their particular research methods, bypass what doesn’t fit, come up with an elegant theory, and foist it on the world. The “no difference” thesis is the latest. Like the “nothing works” thesis in crime prevention, it is, despite its surface cheeriness, a counsel of despair. Like that thesis, it’s false. The whole debate has probably made little impression on most parents. But many people — maybe you, dear reader — think hard about parenting, and want to follow the best, latest evidence. Don’t be misled. Parenting has consequences.

---

If you liked this, then I would love you to do three things:

• Subscribe to this newsletter. It’s free. If you’re not sure, read more.

• Share this post on social media. By telling your friends and/or followers, you’ll be doing me a huge favour.

• Read about the book I’m writing. It’s called Wyclif’s Dust, too. You can download a sample chapter.