Luminis Luminis
← For Parents

Even the Lowest-Income Families Are Buying Tutoring Now. The Chance of Moving Up Has Not Improved. Those Two Facts Belong Together.

November 7, 2026

The Economist did the arithmetic across the world. One American child’s main sport cost the average family $1,016 in 2024, up 46% from 2019. Three in four school-age South Koreans took paid lessons, including 57% of the lowest income quintile. The money has genuinely spread downward. And yet Harvard now admits 3% of applicants, down from 30%, while the chance of out-earning your parents has stagnated in France and Spain and collapsed in Singapore and South Korea. This is about why spending more stopped working, and which part of the spending still does.

This is our reading of The Economist's Finance & economics piece of 2026-11-01, «Intensive parenting is no longer the preserve of the rich», which ran in print under the headline «Tiger mothers of the world, unite». Read the original →

The conclusion first, because it is not pleasant.

Over the past decade or so, parents at every income level, all over the world, have been putting more money into their children’s futures. Seriously, and at real cost to themselves.

And the chance of moving up has not improved.

Both of those are true at once. What this piece is for is the mechanism between them, because once you can see the mechanism, you can tell which part of your money and time is still worth spending and which part is simply a bid against other families.

First set of numbers: the money really has spread downward.

The United States. More than half of American children do an organised sport after school or at weekends. In 2024 one child’s main sport cost the average family $1,016, up 46% from 2019, roughly twice cumulative inflation over the same period. Add the driving and the conferring with coaches and American parents are giving more than 15 hours a week to their children’s sport, which is the time commitment of a part-time job.

South Korea. In 2025, three in four school-age children took paid lessons, including 57% of the lowest income quintile. Among five-year-olds, four in five. The OECD estimates average tutoring spend per pupil has reached about a tenth of average household disposable income.

India, around 27% of pupils take private tuition, 31% in urban areas. China, education spending is growing fastest in the poorest regions.

And not only in Asia. One in three French pupils aged 15 to 18 is in private coaching. Britain is the clearest case of all: according to the Sutton Trust, an education charity, access to tutoring among state-school pupils is now nearly level across the income distribution, 37% in the poorest quartile against 39% in the richest. In London the figure for state-school pupils rises to 45%. And among state-school pupils aged 11 to 16, the share who have paid for private tutoring has gone from 18% in 2005 to 30% today.

So the first fact is that tutoring, coaching and organised sport are no longer markers of wealth. They have become the default. Part of why the British gap narrowed is the expansion of state-funded tutoring schemes, which is genuine help, and we will come back to it.

Second set of numbers: the return did not follow.

The likelihood that children out-earn their parents has stagnated over the past decade in France and Spain, and collapsed in Singapore and South Korea.

Getting into a top university is harder than it has ever been. Harvard admits 3% of applicants, down from 30% in 1960. The share of rural students admitted to Peking University fell from about a third between 1978 and 1998 to around a sixth in 2021, faster than the relative decline in China’s rural population. Which means this is not something that "there are fewer rural people now" can explain away.

The most painful numbers are about the credential itself. According to the Sutton Trust, over the past decade the share of adults from non-graduate families who earn a university degree climbed by ten percentage points, while the college earnings premium those same people could expect fell by eleven. More people got the paper. The paper was being marked down at the same time.

Even with the degree in hand: 30% of Americans from non-graduate family backgrounds reach the top fifth of earners, against 41% of those whose parents attended university. Britain gives the cleanest version. According to the Resolution Foundation, a think-tank, graduates who grew up in poverty, defined by eligibility for free school meals, earned 7% less than their peers a decade after graduating from the same university and the same course. Same institution, same course, and the gap is still there.

Why spending more stopped working: three factors.

Factor one. The thing you are buying comes in a fixed quantity.

Bread does not work this way. If every family buys more bread, more bread gets baked and the price holds. A place at a selective university is not bread. It is what economists call a positional good, something whose value depends almost entirely on how much of it other people have and where you rank relative to them.

The number of seats barely moves. So when every family spends more at once, the result is not more seats, it is a higher bar for everyone. Harvard going from 30% to 3% is not Harvard becoming stricter. It is more people applying, better prepared. Every family did the thing that was best for its own child, and the sum of all those sensible decisions is that everyone now pays more for the same object.

This is not a moral failing and it is not any one family’s fault. It is arithmetic. Chinese has a word for exactly this state, neijuan.

Factor two. The poor started spending more. The rich did not let up.

This is the source’s own explanation, and the easiest line in the whole piece to skim past: the reason the investment did not translate into mobility may be that although the poor are spending more, the rich are not letting up.

The absolute gap in spending narrows. The ranking does not move. You go from one unit to three. The family ahead of you goes from ten to twenty. Both households are more tired. The order is unchanged.

Factor three. The credential being bought is marking itself down.

Look again at those two numbers: attainment up ten points, premium down eleven. When more people hold a thing, it does less work as a signal of difference. That is not a conspiracy, it is what a signal is.

And the British 7% tells us something more important still: same university, same course, and the gap survives. Which means what is actually operating is not only the credential. It is the family contacts, the familiarity with how workplaces work, how the first internship arrived, whether you dare open your mouth about salary. A cram school does not sell any of that.

There is a fourth factor, the most buried, and the one parents most deserve to know.

Matthias Doepke of the London School of Economics and Fabrizio Zilibotti of Yale have found a historical regularity: the more unequal a country, the less attention parents pay to soft virtues like independence and imagination, and the more they pay to qualities like hard work.

Western populations believe inequality is rising and that a middle-class life is no longer a given. So parents cannot relax. Notice the direction of causation. It is not that parents became anxious and so the environment tightened. The environment tightened, and the goal of parenting itself was rewritten.

That explains something many parents cannot quite articulate. You know, actually, that your child needs an empty afternoon, needs unsupervised time to mess about, needs to be bored. You filled the weekend anyway.

You are not weak-willed. You made a rational choice inside an environment you judge to be getting less forgiving.

But see clearly what that choice costs. It lands precisely on independence and imagination. Which happen to be the two capacities hardest for a machine to replace. You traded, for a seat ten years from now, the thing that will be most valuable twenty years from now. That is not a reproach. It is a real trade-off, and you are entitled to know which one you are making.

So what do you do: one division that actually works.

Do not ask "should I spend this money." Ask: with this money, am I bidding against other families, or am I adding something to my own child?

The bidding kind. The fourth activity bought for a certificate that everyone else also has. The class taken for a ranking. The programme joined for one more line on a form. Its signature is this: if every family stopped at once, your child would lose nothing. This kind of spending inflates permanently as others spend more, and it can never buy relative advantage, because relative advantage is the one thing a fixed number of seats cannot supply to everybody.

The adding kind. Sleep. The attention span to finish a long book. One real skill, properly held, that travels with them for life. One adult who is reliably present and willing to acknowledge a difficulty before offering advice. And the two things squeezed out above: independence and imagination.

The beauty of this second kind is that nobody is bidding against you. No other family can spend money to make your child sleep less. These things do not inflate. Their price does not move with anyone else’s behaviour.

About those 15 hours.

Fifteen hours a week is roughly 780 hours a year. That is a far scarcer asset than the $1,016.

The question is not whether to spend it. The question is how much of those fifteen hours is actually time together, and how much is driving, queueing, conferring with a coach, and confirming next week’s slot on a phone.

Convert two of those hours into being together with nothing scheduled. Your child’s competitiveness will not fall. Competitiveness was never lost in those two hours.

Finally, the honest part.

We are not saying do not spend. Money works. Part of the reason the British gap narrowed is that state-funded tutoring schemes expanded, and public money genuinely lowered a bar, which is one of the few unambiguously good pieces of news in the reporting. For one specific child, one good tutor can change the fate of one subject.

What we are saying is different: inside the part you control, separate the bidding spend from the adding spend. The first can only ever return you to the same position as everyone else, however hard you push. The second stays in your child.

The Economist ran two companion pieces the same week. One is on why fewer young people are having children, and the calculation inside it, concentrating resources on one child, is the other half of this article. Our reading of it is here. The other is on nannies earning $200,000 a year, where what the richest families buy with the money turns out to be something an ordinary family can also have. Three views of one machine.

Questions to sit with

1. List everything your child is currently signed up for, and go through it one line at a time asking: if every child their age dropped this one, what would my child lose? The ones where the honest answer is "nothing" are the bidding spend.

2. Of the hours you spend on your child each week, how many are in transit and coordinating, and how many are actually together? Are you content with that ratio?

3. Doepke and Zilibotti say that the more unequal the environment, the less parents value independence and imagination. When was the last time your child had a completely empty, unscheduled afternoon?

from The Economist

Reader discussion

You have read it. What do you make of it?

Loading…