What Individual Parents Can Do
A new series against my better judgment.
When I was pitching my book proposal for The Parent Trap I had a lot of conversations with people at publishing houses. Some of these people had developed books that sold millions of copies and won big awards, books I myself had read and loved and bought at airports. It was exciting.
In the end, all but two publishers decided to pass on the book for roughly the same reason: they wanted me to embrace self-help as a core theme for equalizing opportunity across families.
I couldn’t do that. I remember one conversation in which I explained that if I pushed the book more toward self-help, I’d feel like I was lying.
And I felt my actual message was energizing enough!
Yes, unequal opportunities for skill development were costing millions of kids a fortune in future income – along with all the health and fulfillment associated with that foregone income.
And yes, self-help wouldn’t fix the main problems facing kids because these problems were too complicated, expensive, and time-consuming for most individual parents to address on their own.
BUT if we could make parents more aware of that fact, and more angry about it, and then channel their anger into a certain type of political strategy – things could change. Just like things changed for older people when they joined forces to pass and protect Social Security and Medicare. And emerging evidence was showing that if parents did achieve policy wins on that scale (e.g. “Familycare”), then something shocking would happen: regular kids would reach adulthood ready to work in the same kinds of jobs as rich kids. Generations of Americans would grow up healthier, richer, happier. America would feel more … American, in the best kind of way.
Most publishers didn’t see dollar signs in this kind of book, and they were right. I have not made a lot of money on The Parent Trap, and neither has my wonderful nonprofit publisher, MIT Press (which incidentally also published a couple of my favorite economics books of all time … great books that also probably didn’t make any money).
But the book did do well enough to get me into a lot of conversations with people around the country. And perhaps the single most common question I got was this: how can individual parents use your book for self-help?
So ok, I want to be a better sport and say something about that. I’m hoping what I say is helpful to some parents, but also makes it clear why it’s asking too much for most families to attempt on their own what rich, highly educated parents do, and why that matters so much for inequality and the overall collective well-being of nations.
Let’s start with tutoring.
Tutoring Works
If your kid is struggling to learn a valuable skill such as reading, writing, algebra, or statistics – or if your kid is learning things “too quickly” and getting bored – a good tutor can probably help a lot.
How much? Perhaps 0.3 “standard deviations” in test scores per year of tutoring for say 2-3 hours per week during a school year. How much is 0.3 standard deviations? My guess – based on the best research linking test score gains to income gains – is that a year of good tutoring would typically increase a kid’s total future lifetime income by about $120,000, or $20,000 in “discounted present value” which accounts for the notion that $1 in the distant future is worth a lot less than $1 today, because in theory someone could invest a dollar today and earn interest right away.
The cost of this intervention for parents paying, say, $50/hour for a private tutor would be around $4,000.
That is for one year of tutoring. If you’re willing to extrapolate to multiple years, these numbers get even more impressive. If your kid can benefit from 5 years of tutoring, say, then declining to provide it might reduce their future income by $600,000 (or $100,000 in present discounted value). That seems hard to believe… but we don’t have research showing linear extrapolation is wrong in this case, and honestly I don’t find it that hard to believe that falling further and further behind in school for 5 years could shift a kid’s life trajectory toward far less lucrative occupations.
To me, this is compelling and my wife and I will probably hire private tutors if our kids ever need them and our finances allow it.
As I have to say over and over again – it’s not that we’re obsessed with maximizing our kids’ future incomes. We’re not. It’s that income is correlated with a lot of other great stuff such as health and job satisfaction and stable families, and if something tends to increase future income by a large amount, it therefore seems likely to increase overall quality of life as well. That is why these numbers both motivate and scare me.
Tutoring: Self-Schmelp
So why don’t I like the self-help angle? “Tutoring is a big opportunity bargain – more families should do it!”
Well, remember this is just the first installment in my series about things that individual parents can do to benefit their kids – and already the private cost is way too high for most families. After a family with median income accounts for taxes, housing, transportation, food, and insurance, $4,000 might be 30-50% of what they have left. So a more realistic “self-help” message might be: “Don’t go on vacation or buy Christmas presents – hire a tutor!”
Even if families can find a way to pay for tutoring, many parents have little clue if their kids are ahead or behind overall, much less in particular subjects (see Chapter 4 in my book). Even if families sense a problem, most won’t know how effective tutoring can be. And even if they become interested in tutoring, they may have trouble finding and procuring the service. If you look up “tutors near me” online you’ll have to choose from hundreds of tutors across dozens of platforms quoting prices anywhere from $0 (“free tutoring” at your local library) up to $500 per hour (“I scored 1600 on my SAT and graduated from Harvard and can teach anyone anything”).
So I believe the only real way to expand access to tutoring is to make a big public investment, and that’s why I include $12 billion of annual federal spending for this purpose in my “Familycare” policy proposal. According to the calculations above, each year this national investment would increase total future GDP by $360 billion, or $60 billion in present discounted value.
In a world with Familycare, public schools would pay experienced, professional tutors to work with kids who need help at school, rather than asking millions of individual parents to pour time and money into navigating an alien marketplace on their own. This isn’t really that new. Tutoring just means very small class sizes for some kids some of the time, and we’ve all chosen to gradually reduce average class sizes in public schools for 100 years. Large-scale tutoring would just continue this trend.
But I digress! Right now we have no Familycare and parents are stuck. What can we do?
Tutoring: DIY
First, you can try to tutor your kid yourself. Many parents, especially highly-educated parents, do this all the time. A little help on math homework here, a little constructive feedback on an essay there – year after year this is like grease in the wheels of child skill development.
Until recently this has only been possible for parents who know or can learn the material, have a knack for teaching, and have a certain type of relationship with their kids. Right now, when I try to teach my toddler anything that feels “learny” he closes his eyes, leans his head back, and says “I fell asleep.”
But AI tools such as ChatGPT can help many more parents tutor their kids effectively. These tools are already unbelievably powerful as tutors, but most kids and especially kids who are falling behind will need a lot of coaching to play along. Many parents who couldn’t be effective tutors can be that missing coach.
Here’s an example. Suppose your son is obsessed with football and struggles with his algebra homework. A parent could sit down with him and start asking ChatGPT some questions.
Here is a quadratic equation from my homework: 5x^2 -20x + 15 = 0 … can you please walk me through how to solve this, step by step?
What does it mean for an equation to have 2 solutions – how is that possible?
Can you please explain how this equation gives insight into something real and important?
The key is to articulate sincere confusion and curiosity as precisely as you can, over and over. By the end of this conversation, ChatGPT explains that “x could represent time, and the equation itself could model the height of a ball at different times. The roots (1 and 3) might indicate the times when the ball is at a specific height from the ground.”
“Ok, fine,” says your teenager, “who cares?” Great question! AI can empower parents to respond with an educational tool built for this exact attitude: project-based learning. A parent who knows nothing about math might continue:
Can you please provide real, specific examples of people or businesses that derive value from describing a football pass with a quadratic function?
ChatGPT now teaches us about the work of football’s “quarterback whisperer” Tom House who capped his professional career as a baseball pitcher by getting a PhD and writing 22 books, virtual reality training company STRIVR that lets people do insane video-game style simulations to improve their real-life skills, sports analytics company Stats Perform (which focuses on soccer but come on it’s amazing — see below), video game giant EA Sports that uses quadratic equations to model football passing, and a bunch of other cool examples, all of which provide opportunities to explore weird, unanticipated corners of reality, as well as careers (all these companies post job openings describing qualifications) in which real people use quadratic equations.
As a bonus, teaching kids how to interact with AI is now also an essential skill in itself – sort-of-almost as important as teaching kids how to interact with “HI.” So these coaching sessions really serve two purposes.
But this kind of AI coaching is still hard, and it won’t work for many parents. So back to finding a tutor.
Diamond in the Rough
Parents can ask teachers at their school if they know of any high-quality, affordable options – local nonprofits, experienced peer tutors and local college students, and school-sponsored tutoring platforms such as Saga. Be aggressive and ask more teachers or the principal if you don’t get any tips at first. Crowdsourcing recommendations from local parenting groups on Facebook can also help (beware scammers).
What should parents look for in a tutor? Experience and happy prior clients. Other things equal, I’d try to find a tutor your kid likes. Motivating your kid to embrace learning is part of what a good tutor can do, so don’t take any reluctance on your kid’s part for granted. Maybe you need to try a few different tutors. Maybe you need a tutor who’s good at teaching math, and loves football, and loves the Detroit Lions, and couldn’t get enough of HBO’s Hard Knocks Season 18…. It goes without saying a good tutor should never make your kid feel stupid or discouraged. In some studies same-gender and same-race teachers can be more effective, probably because they facilitate trust, or because they feel like more realistic role models to kids who might worry they’re not cut out to master certain skills.
In-person or online? Researchers still don’t know enough about this question, but one great study in Italy by economists Michela Carlana and Eliana La Ferrara suggests online tutoring can work very well. I’d still assume – other things equal – that in-person tutoring works better, but if your best option is reputable online tutoring, I’d try it out. Maybe over time online tutoring can become the Toyota Camry of tutoring — it totally gets the job done.
So that’s my curmudgeonly self-help for tutoring, Parent Trap style. I feel one impact of this type of advice is to increase already-high interest and stress around tutoring among highly-advantaged parents. That is why I focused my book on the collective political strategy required to make public investments that benefit all kids and save parents a ton of frustration, rather than on feeding America’s insatiable (and in many ways endearing) appetite for self-help. But then again — maybe some lower-income parents can leverage rich-parent tactics such as tutoring on their own if they gain confidence it really might help their kids succeed.
So that’s tutoring! Up next: child nutrition.
On income being correlated with good outcomes: Francesconi and Heckman (https://onlinelibrary.wiley.com/doi/abs/10.1111/ecoj.12388) assert that it’s mostly due to confounding with SES. I haven’t dug into the citations, but here’s the key bit:
Carneiro and Heckman (2003) and Cunha et al. (2006) present evidence that child cognitive and non-cognitive skills diverge at early ages across families with different levels of permanent income during childhood.19 Levels of permanent income are highly correlated with family background factors such as parental education and maternal ability, which, when statistically controlled for, largely eliminate the gaps across income classes. The literature sometimes interprets this conditioning as reflecting parenting and parental investments, but it could arise from any or all of the correlates of permanent income associated with parental preferences and skills.