The Nini Number Fell, Eight Were Erased — Mexico's Labour Ledger and the Widest Gender Gap in the OECD

The moment the OECD released 'Panorama de la Educación 2026', Mexican newsrooms settled on a single number: 19. A decade earlier that number was 23. Four percentage points of decline were immediately packaged as 'progress'. I went through the report's tables three times, and every time I stopped at the same place: inside that 19 percent — the people Mexico calls 'ninis', who neither study nor work — the ratio of women to men is roughly 3:1.

Twenty years of writing football numbers has left me with a habit. When a statistic starts making everyone comfortable, it is almost certainly hiding something else. Germany's football dynasty ended in 2026, but the world noticed in 2026 in Russia. The ledger entry had been written long before; people simply needed to read it. Mexico's nini number is that kind of entry — and the real story is not 19, it is 3:1.

I will get to the eight people later. First, why one ratio matters more than four percentage points.

Context: A Two-Decade Word and a Report

'Nini' is a Spanish contraction — 'ni estudia, ni trabaja', neither studies nor works. Roughly two decades ago the word entered Mexican political language and news headlines, then slowly became a generational identity. The nearest English equivalent is NEET: not in education, employment or training. The OECD and the International Labour Organization have used this indicator for years because it answers a simple question: what share of a country's young people are outside education, work and training altogether?

Simple question, complicated answer. Being outside all three is not one condition. Some are searching for work and failing. Some dropped out after failing exams. Some are caring for younger siblings, sick parents or their own children — and no ledger records that as work. The OECD definition puts all three into a single room. That is exactly why Mexico's debate resembles football: a team's goal tally never explains the game; you have to see who is shooting, who is entering the box, and who is only tracking back.

In the 2026 report, Mexico sits at 19 percent against an OECD average of 13 percent. Being six points above the average is not good here, because on this indicator higher means more young people outside the system. A less-quoted line says the decline is not exclusive to Mexico — many countries are moving the same way. If Mexico advertises its four points as a national achievement, the fair question is how much of that came from rowing and how much from the tide.

There is one more layer missing from the headlines. In Mexico, 'nini' is no longer a statistical label. It is a verdict. Politicians use it to assign blame, media use it for headlines, and the person carrying the label pays for life — on job forms, at banks, even in family conversations about marriage.

Core Analysis: Inside the Ledger

What the Four Points Contain

Start with the arithmetic everyone skipped. Twenty-three to nineteen is a rate, not a headcount. A falling rate can coexist with a stable or even rising number of people if the youth cohort grows. The OECD summary does not clearly decompose Mexico's cohort size across the decade. So when we say four points fell, we should add a caveat: we know the squad shrank, we do not know how many people actually left the room.

This is where my old habit applies — Root: Germany 2026. That team won the World Cup while its structural indicators were already turning: average age, sprint distance, midfield speed. Results kept arriving because good results and a good system are not the same thing; sometimes good results are the last breath of an old system. I read Mexico's four points the same way: either structural improvement, or a late goal in the final minutes of a decade. The only way to tell is to open the gender room.

3:1 — The Real Mark on the Ledger

By the OECD's own figures, roughly three in ten Mexican women aged 18 to 24 are outside education, work or training. For men of the same age, it is about one in ten. That is the widest such gap in the OECD. A report line may say 'gender disparity remains significant', but the numbers say something stronger: for women the problem is not marginal, it is close to the rule.

Mariana Belló, a researcher at Oxfam México, pushes further: of the people in Mexico who neither study nor work, she argues, roughly eight in ten are women. Her core point is about stigma — the label is pinned on a young woman while the real question belongs to the state.

Here I have to state a data tension plainly, because hiding it would weaken the analysis. Belló's 'eight in ten' and the OECD's 'three in ten versus one in ten' are close but not identical. If male and female cohorts are roughly equal in size, three-in-ten versus one-in-ten implies women are about 75 percent of the total — three out of four, not eight out of ten. The gap may come from different age brackets, definitions or reference years. This is not proof that anyone is wrong; it is a provenance question. And provenance matters, because policy is written on numbers, and numbers taken from the wrong source sit in the ledger for years.

Either way, both calculations point the same direction: Mexico's nini problem is primarily a women's problem, and that is not an accident.

The Label Is Itself an Immutable Entry

Modern statistical systems behave a little like a blockchain: once an entry is written, it propagates to every node and becomes nearly impossible to change. When a young woman is recorded as a 'nini', that entry does not stay in a report. It travels into family talk, into neighbours' eyes, into recruiters' filters, into her own confidence. If the entry is wrong, there is no easy correction, because nobody knows which cell was wrong.

Carla Pederzini of Universidad Iberoamericana points exactly there: the 'nini' definition is too broad. It collapses different realities into one — the dropout, the job-seeker who cannot find work, and the woman doing unpaid household labour all wear the same tag. A broad definition has one advantage: it is easy to count. It has one disadvantage, and it is enormous: policy becomes impossible, because the treatments are entirely different.

The biggest thing hidden inside that broad definition is unpaid care work. In Mexico, housework, elder care and child-rearing fall mainly on women. That labour has no market wage, so national accounts do not call it work. Yet without it the rest of the economy does not run. A woman who runs a household from six in the morning to ten at night appears in the statistics as unemployed or a 'nini' — her work is never counted, only her absence is. That is not an accounting error, it is an accounting bias. And once bias enters a ledger, correcting it is the hardest work there is.

Pederzini's prescription is blunt: the answer is not pressure on young women but childcare infrastructure. Day care, preschool, elder services — with those in place, many women enter the labour market and leave the nini list on their own. The question is not her ability. It is the state's infrastructure.

The Dhaka Mirror

Reading this report, I kept thinking about Dhaka. Bangladesh's version of the indicator has a different name and different numbers, but the structure is strikingly similar. Our labour force survey has a room called 'outside the labour force', and women are heavily over-represented there. They are recorded as not working, while spending most of the day on household labour, cooking, childcare and elder care. In our language we say such a person is 'sitting at home' — no less cruel than 'nini'.

I remember 2026. The pandemic stopped the Bangladesh Premier League, Euro 2026 and the Tokyo Olympics at once. I ran a twelve-part interview series called 'Empty Stands, Loud Voices' with forty supporters' club leaders in Dhaka, including rival Argentina and Brazil fan clubs. When the stands went empty, the voices didn't stop — that is what I learned there. One interviewee told me: 'My brother watches the match, I cook the food, and everyone says I do nothing.' In that one sentence, Mexico's entire nini debate fits.

This is my second working angle — what European football's common sense cannot explain outside its own market. In London or Berlin, 'outside education, work and training' implies a welfare state where at least one door is open. In Dhaka or Mexico City, that door is often shut, and a shut door at home is renamed 'family duty'. An indicator that cannot capture this difference is not measuring. It is distributing blame.

The OECD Benchmark and the Regional Tide

The least-discussed OECD line is that the decline is not Mexico's alone. That means four points cannot be sold as a national policy success. But it means something bigger too: if many countries move the same way at the same time, the driver may be structural — demography, labour market shape, the link between schooling and jobs — rather than any single national policy. When the structural tide is stronger than the policy, indicators cannot capture it.

One more benchmark matters: 19 against 13. That six-point gap is not just 'worse than average'; it is a ceiling. A country above it can use the word improvement, but not the word solution. And in the gender room Mexico is not merely below average — it holds the widest gap in the OECD. On this indicator Mexico is not a laggard, it is an outlier. Being an outlier carries more information than lagging: it says the problem is not a temporary slump, it is written into the structure.

The Contrarian Angle: Where I Could Be Wrong

First, my weakest point. I have placed a ratio at the centre, yet its two sides come from different sources — the OECD's three-in-ten versus one-in-ten, and Belló's eight-in-ten. I am assembling a picture from two frames. If someone argues the join is illegitimate, much of my argument wobbles.

Second, the 'nini' label may not be useless. The state needs an indicator to allocate budgets. Without a name, there is no accounting. The question may not be the word but what gets placed inside it. What I call a wrong definition, someone else may call a usable simplification.

Third, my most important doubt is aimed at myself: if unpaid household labour were simply counted as work, would the number improve while the problem hides? If care work is classified as economic activity, the state may conclude there is no problem, and political pressure to invest in day care or preschool falls. Changing the measurement changes the number — but does it change the power relation? I cannot dismiss that suspicion.

Fourth, timing. Gender gaps are often lagging indicators. Labour market participation for women can rise five to seven years after an economy recovers, because the decisions are made at home, not in the market. If so, today's structural crisis may be yesterday's recession arriving late — and today's picture may improve on its own in five years.

The thread began with one question, and nineteen posts later it arrived at arithmetic. I have not forgotten that lesson: state the claim first, then buy the claim with evidence. Today's claim is that Mexico's nini story lives in the gender gap. The evidence is that 3:1, which never made a headline.

Takeaway: What to Watch

Here is a prediction, timestamped so I can be held to it. If Mexico's gender gap does not narrow across the next two reporting cycles, then 19 percent is not a floor — it is a ceiling. Mexico will stall before going lower, because the decline cannot continue while half the population stays outside the count.

Watch the methodology too. If the OECD or the ILO ever folds unpaid care work into the measurement, Mexico's number will change shape — but the question stays the same: who holds the power to write the ledger entry? A country that records half its citizens as doing nothing is not measuring its citizens. It is measuring its own missing infrastructure.