India has, on paper, already won the argument about education. The country runs the largest school system on earth. In 2024-25 the number of school teachers crossed 1 crore for the first time in history: 1,01,22,420 of them (UDISE+ 2024-25, Ministry of Education). Enrolment sits near 24.8 crore children (UDISE+ 2023-24). Higher-education access has climbed from a Gross Enrolment Ratio of 23.7 a decade ago to 30.0 today (AISHE 2023-24). By the raw metric that dominated policy for seventy years (are the children in school?), India has done something genuinely historic.

And yet. In the same country, at the same time, 76.6% of Class 3 children in government schools cannot read a Class 2 text (ASER 2024, released 28 January 2025). The government's own national assessment found that mathematics proficiency falls as children climb the system, from roughly 60% correct in Grade 3 to about 37% by Grade 9 (PARAKH Rashtriya Sarvekshan 2024, NCERT National Report). We have, in other words, solved access and left learning behind.

This is not a piece that treats Indian education as a failure. It isn't one. The wins below are real and I will give them their due. But the central argument of this post is urgent and I want to state it plainly at the top: India is enrolling its children without educating them, and it is doing so while the clock on its demographic dividend is already running down. Every year we spend congratulating ourselves on enrolment numbers is a year of a cohort that ages out of school unable to read, reason, or be employed. That is the emergency. Everything below is anchored in verified, dated figures; where a number is an estimate or contested, I say so rather than smuggle it in.

1. The scale of what works — so we argue honestly

Let me anchor the upside first, because critics of Indian education routinely understate it and a case built on half the evidence is a weak case.

  • The teacher workforce crossed 1 crore (1,01,22,420) in 2024-25, up 6.7% over two years, and single-teacher schools fell to 1,04,125 from 1,10,971 in 2022-23 (UDISE+ 2024-25). Zero-enrolment schools dropped nearly 38%.
  • Higher-education GER reached 30.0 (AISHE 2023-24), with total enrolment at a record ~4.5 crore. The Gender Parity Index stands at 1.08: women have out-enrolled men in higher education for seven straight years, and female GER (31.2) now exceeds male.
  • India placed a record 54 institutions in the QS World University Rankings 2026 (released November 2025), the fourth most-represented country in the world, led by IIT Delhi at global rank 123, up from 150.
  • Post-COVID, ASER 2024 found that government schools outpaced private schools in learning recovery, and foundational numbers ticked up: Class 3 children who can read a Class 2 text rose from 16.3% (2022) to 23.4% (2024); those who can do subtraction rose from 28.1% (2018) to 33.7% (2024).

These are not trivial. A system that has pulled a near-universal cohort into classrooms and is now nudging learning upward deserves acknowledgment. The problem is not that nothing works. The problem is that the base those improvements are climbing from is catastrophically low, and the pace of improvement is nowhere near the pace the demographic window demands.

2. A note on reading the statistics

Because this post leans on a lot of official data, a few technical notes on how these numbers are constructed will help read them accurately. India's statistical system (UDISE+, AISHE, PLFS, the NCERT surveys) is substantial and improving, and the point here is not to doubt the figures but to understand what each one measures.

  • Enrolment counting changed in 2022-23. The drop from ~26 crore to 24.8 crore is partly a methodology shift: the Ministry moved to student-wise, Aadhaar-linked unique-ID tracking, which removed duplicate entries. The government notes older figures are "not strictly comparable" with the new ones, so year-on-year enrolment changes across that break should be read with that caveat.
  • "Employment" in PLFS uses the standard ILO definition, under which unpaid family work counts as employment. This is a globally used convention, but it means the headline 3.1% unemployment rate is best read alongside job-quality measures, which is why this post also cites the youth NEET rate and graduate employability from the same period.
  • PARAKH replaced the older National Achievement Survey (NAS). NAS was school-administered, and its scores tended to run higher than independent surveys like ASER. PARAKH, set up under NEP 2020, is designed to be more independent, one reason its 2024 findings, cited throughout this post, are a firmer basis than older learning data.
  • India does not currently participate in PISA, the main international assessment, having opted out most recently in 2025, so direct global comparisons of learning outcomes rely on older or indirect data.

None of this undercuts the figures below; it simply means reading each with its definition attached. With that context in place, here is the picture the data paints.

3. The learning crisis is the whole ballgame

Strip away every other issue and this one would still be a national emergency.

Start with the World Bank's learning-poverty estimate: as of its India brief published April 2024, roughly 56% of Indian children cannot read and understand a simple text by age 10 (this uses pre-pandemic assessment data, the latest the Bank had, and should be read as an estimate, not a live 2024 measurement). More than half of the children in the world's fifth-largest economy hit the end of primary school functionally unable to read.

ASER 2024 makes it concrete and rural: fewer than one in four Class 3 government-school children (23.4%) can read at the Class 2 level. That means more than three in four cannot. By Class 5, a large share still read below Class 2 level. These are not marginal laggards; this is the median child.

Then comes the finding that should reframe the entire debate: the PARAKH Rashtriya Sarvekshan 2024, the largest system-level assessment India has ever run. Conducted on 4 December 2024, it tested 21,15,022 students across 74,229 schools in 781 districts and 36 States/UTs (NCERT National Report, released July 2025). It deliberately sampled the endpoints of the new NEP school stages (Grade 3, Grade 6, Grade 9), and what it found is the single most damning fact in Indian education:

Learning does not accumulate as children age through the system. It leaks.

  • Grade 3: Language ~64%, Mathematics ~60%.
  • Grade 6: Language ~55%, Mathematics ~46-47%.
  • Grade 9: Language ~54%, Science ~41%, Social Science ~40%, Mathematics ~37%, the lowest score of the entire survey.

Read those three lines top to bottom. A child who could handle 60% of grade-appropriate maths at age 8 is managing 37% of it at age 14. The system is not adding capability year on year; the gap between what a child should know and what they do know is widening with every grade. On specific competencies it is starker still: only about 29% of Grade 6 students could represent and compare fractions, and Grade 9 maths competencies like real numbers (31%) and percentages (28%) sat well below 40%.

The equity texture inside that data matters too. At Grade 3, rural children marginally out-performed urban ones, but by Grades 6 and 9 the advantage flips decisively to urban. ST students scored lowest at every grade, with the widest gap in mathematics. And PARAKH's questionnaire data drew a stark socioeconomic line: high-performing Class 9 students were far more likely to have a graduate parent, electricity at home, books, and a toilet. The system is not a leveller; it is amplifying the advantages children walk in with.

This is the ballgame. A country cannot skill, employ, or democratically empower a generation that cannot read a paragraph or handle a fraction. Every downstream problem in this post (unemployment, the coaching bubble, the global-competitiveness gap) is in some measure a consequence of this foundational failure.

4. Access improved. Equity did not follow.

The good news on dropouts is real: the secondary-stage dropout rate fell to 11.5% in 2024-25, down from 14.1% the year before (UDISE+), an absolute reduction of about a million children, from ~53.5 lakh to ~42.7 lakh. Progress.

But secondary school remains the leak point, and the leak is not evenly distributed. It concentrates by geography and by identity:

  • Uttar Pradesh and Bihar together account for a majority of national elementary out-of-school children in independent analyses of UDISE+ 2024-25. Where you are born still substantially determines whether you finish school.
  • Enrolment of SC and OBC children in government schools fell between 2022-23 and 2023-24: SC enrolment down from 4.59 crore to 4.47 crore, OBC from 11.45 crore to 11.2 crore (UDISE+). When overall numbers dip, marginalised groups absorb a disproportionate share of the loss.
  • Girls' dropout continues to exceed boys' at key transition points, especially the move from primary to upper-primary in rural areas.

A note on a figure you will see quoted elsewhere: some analysts put India's out-of-school children at ~46 million (6-17 years) using UDISE+ 2024-25. That is a credible independent computation, but it is not an official Ministry of Education headline number, and I flag it as such rather than present it as government data. The verified official signal is the dropout-rate trend: improving, but with secondary school still shedding one in nine children, and shedding them unevenly.

Layer on the digital divide, which the pandemic exposed and did not close. Two different gaps live here. On the school side, UDISE+ 2023-24 found only 57.2% of schools have functional computers and just 53.9% have internet, meaning roughly half of India's schools cannot put a child online. On the student side, ASER 2024 surprised everyone: about 90% of rural 14-16-year-olds have a smartphone at home, and 82% can operate one. Access, at least to a phone, is nearly universal. But only 57% used it for anything educational in the survey reference week, while 76% used it for social media. The divide has shifted from who has a device to who can turn a device into learning, and that second divide tracks parental education and income with depressing fidelity.

5. The money problem is a choice, not a constraint

Here is the number that underwrites every failure above. India spends about 2.7% of GDP on education (Centre and States combined, 2023-24, per the Economic Survey). The target, 6% of GDP, was first set by the Kothari Commission in 1966, reaffirmed by the National Policy on Education in 1986, and reaffirmed again by NEP 2020. India has never once reached it. Nearly sixty years of the same unmet target across three landmark policy documents is not an accident of any single budget; it is a settled national choice to under-fund the thing every policy document calls foundational.

That is not to say nothing is being spent. The Union Budget 2025-26 raised the Department of School Education and Literacy allocation to ₹78,572 crore, up 16.3% over the prior year's revised estimate. Inside that: Samagra Shiksha at ₹41,250 crore, PM POSHAN (mid-day meals) at ₹12,500 crore, and PM SHRI schools at ₹7,500 crore. These are meaningful commitments. But absolute crore figures flatter the picture. Against a 24.8-crore-child system and a 1-crore-teacher payroll, 2.7% of GDP simply does not stretch to the teacher salaries, the training, the labs, the libraries, and the technology that foundational learning requires. You cannot fix a reading crisis on a fraction of the budget every expert commission for six decades has said you need.

6. The teacher crisis behind the learning crisis

Learning outcomes do not collapse in a vacuum. They collapse in classrooms that are missing teachers, and India's are missing a lot of them.

  • Government schools carry roughly 8.4 lakh teaching vacancies (2024, Education Ministry to Parliament), and a Parliamentary Standing Committee has flagged as many as 9.8 lakh, over 7 lakh at elementary level alone. Bihar, Uttar Pradesh and Jharkhand account for more than half.
  • 1,04,125 schools still run on a single teacher (UDISE+ 2024-25): one adult expected to teach every subject to every grade in the building.
  • In higher education, faculty vacancies run near 28% nationally on AISHE-based analysis, with thousands of unfilled posts reported in the IITs and central universities.

And it is not only about numbers of teachers but their preparation. PARAKH 2024 found that only about 38% of schools had teachers trained to support Children with Special Needs, and its questionnaires surfaced a quieter crisis of engagement: barely 55% of students reported feeling motivated to attend school, and less than half felt emotionally safe. A system cannot deliver competency-based learning, as NEP 2020 demands, through under-staffed schools run by under-trained and often demoralised teachers. The teacher pipeline (recruitment, training, working conditions, dignity) is not a side issue to the learning crisis. It is the learning crisis, viewed from the front of the classroom.

7. The employability paradox: educated, and unemployed

If the system's job is ultimately to prepare young Indians for economic life, the output data is where the alarm should turn to urgency.

The India Skills Report 2026 (released November 2025, based on over one lakh candidates) put overall graduate employability at 56.35%, meaning roughly four in ten graduates are assessed as not readily employable. Engineering does better at 70.15%, and, encouragingly, women's employability (54%) has overtaken men's for the first time in five years. But an economy cannot absorb a demographic dividend when nearly half its degree-holders are judged unready for the work.

The unemployment data sharpens the paradox into something painful. India's overall unemployment rate is a low 3.1% (PLFS, calendar year 2025). But that headline masks the young and the educated. Youth (15-29) unemployment is 9.9%, more than triple the overall rate, and unemployment among those with secondary education or above sits at 6.5%, roughly double the national figure (PLFS 2025). Most damning of all, about one in four young Indians (a ~25% NEET rate) is Not in Employment, Education, or Training (PLFS 2025): a quarter of the country's youth neither building skills nor earning. This is the signature of a system producing certificates faster than capabilities, and an economy producing degrees faster than the jobs those degrees are meant to fit.

The state's answer has been to bolt skilling onto the side of formal education: the restructured Skill India Programme (₹8,800 crore outlay through 2025-26), folding in PMKVY 4.0 and apprenticeship promotion; the ₹13,000-crore PM Vishwakarma scheme for artisans across 18 trades; and DBT stipend support under the national apprenticeship scheme. These are sensible interventions. But skilling bolted onto a foundation that never taught reading and arithmetic is a roof without walls. You cannot up-skill your way out of a literacy crisis.

8. The shadow system, and what it costs in lives

When a public system fails to teach, a private one rushes into the vacuum, and India's is enormous. The country's coaching-institute market was valued at roughly $7.2 billion in 2025 (IMARC estimate, a private research figure, not government), projected to more than double by 2034, with an estimated 27% of students using private coaching in 2025-26. A parallel, paid education system now shadows the free one, and it charges most heavily the families who can least spare it, precisely the households whose children the public system already under-serves.

This shadow system has a human cost that no RPM or ranking captures. In Kota, the coaching capital, at least 17 students died by suicide in 2024 (press-compiled tally), following 26 in 2023. Nationally, the NCRB has recorded student suicides rising sharply over the past decade. Behind the market-size charts is a generation of teenagers carrying exam pressure that the system manufactures and then monetises. An education system that drives its children to despair to clear a cutoff has failed at something more fundamental than test scores.

I flag the sourcing honestly: the market-size figures are commercial estimates and the Kota tallies are press-compiled rather than official NCRB city data (the latest official NCRB year is 2022). But the direction is not in dispute, and the direction is the point. The coaching bubble is not a quirk of ambitious parents; it is the price the country pays for a public system that does not teach well enough for children to succeed without paying twice.

9. India's quiet retreat from the global yardstick

There is one more number that captures the defensiveness at the heart of the status quo: India has not participated in the PISA international assessment since 2009, when it finished 73rd of 74 countries. It agreed to join PISA 2021, then withdrew citing pandemic learning loss, and opted out of PISA 2025 as well. A country cannot improve what it refuses to measure against the world. The retreat from PISA is a tell: an institutional preference for not knowing where we truly stand.

The counter-current is more hopeful: at the elite end, India is opening to the world. Foreign universities are arriving under the UGC's 2023 regulations: Deakin and Wollongong at GIFT City, and the University of Southampton's Gurugram campus (opened August 2025, the first foreign university on the Indian mainland), with 17 institutions approved by the UGC or IFSCA as of December 2025, and more due in 2026. That is genuine progress at the top of the pyramid. But it is worth holding the two facts together: India is competitive enough to attract global universities for its best students, and unwilling to be measured against global standards for the rest. The reform test is whether the excellence at the top can ever be made to reach the 76% of Class 3 children who can't yet read.

10. Reform is underway — give it its due, then measure it honestly

To be fair to the effort already in motion: NEP 2020 is the most ambitious restructuring of Indian education in a generation, and parts of it are genuinely rolling out.

  • The 5+3+3+4 curricular structure is in phased rollout across the 2025-26 session.
  • The National Curriculum Framework for School Education (NCF-SE 2023) is notified, and new NCERT textbooks aligned to it have reached Classes 1-8, with Classes 9-12 due for 2026-27.
  • PARAKH itself, the body that produced the sobering survey above, is an NEP creation, established within NCERT in February 2023, and its willingness to publish uncomfortable data is itself a reform worth valuing.
  • Delhi University launched the Four-Year Undergraduate Programme's fourth year in August 2025 (about 31,000 of ~71,000 eligible students continued into the research year), and the Academic Bank of Credits has crossed 3 crore student registrations.

That is real motion, and it deserves acknowledgment rather than reflexive cynicism. But motion is not the same as arrival, and the honest audit shows the reforms concentrated where they are easiest (curricular frameworks, textbook revisions, structural relabelling) and stalled where they are hardest:

  • The Higher Education Commission of India (HECI), meant to replace the UGC and overhaul higher-education regulation, has been repeatedly postponed and is still not established.
  • The National Curriculum Framework for Teacher Education remains incomplete, the teacher-preparation reform that the classroom crisis most needs.
  • The 6%-of-GDP financing commitment, restated by NEP 2020 itself, remains unmet at 2.7%.
  • And DU's own faculty association called the FYUP rollout "a recipe for failure" on infrastructure and staffing grounds, a reminder that a reform announced is not a reform resourced.

New textbooks and new stage-labels are the visible, low-cost surface of reform. Teachers, money, and regulatory courage are the expensive, invisible core. Six years into NEP 2020, India has done a great deal of the former and too little of the latter.

11. What reform actually has to do now

The urgency is not rhetorical, so the agenda can't be either. If the foundational-learning data above is real (and it is, drawn from the government's own survey), then the reform priorities write themselves, in rough order of what would move the most for the most children:

  1. Fund the promise. Move decisively toward the 6%-of-GDP commitment that three national policies over sixty years have made and none have kept. Every other item on this list is throttled by the financing gap.
  2. Fix foundations first, and measure them. Make foundational literacy and numeracy (reading and arithmetic by Grade 3) the single non-negotiable national target, with PARAKH-style measurement holding states accountable year on year. The Grade-3-to-Grade-9 competency leak is where India's human capital drains away.
  3. Solve the teacher crisis as a crisis. Fill the ~8-10 lakh vacancies, end single-teacher schools, and finish the teacher-education framework. Competency-based learning is impossible without competent, supported, adequately-numbered teachers.
  4. Close the usage digital divide, not just the access one. Getting internet into the ~46% of schools that lack it, and turning the near-universal smartphone into a learning device rather than only a social one.
  5. Align higher education with employment so that a degree stops tripling a young person's odds of unemployment, and build the HECI and regulatory reforms NEP promised.
  6. Take the pressure off. Reduce the coaching dependence and exam-stress machinery that is costing children their wellbeing and, in Kota and beyond, their lives.
  7. Rejoin the world's yardstick. Return to international benchmarking, because a country serious about improvement does not refuse to be measured.

The demographic clock is the deadline

Here is why none of this can wait. India's advantage, the thing every optimistic projection about the century rests on, is that it is young while much of the world ages. That advantage is not permanent. The demographic window in which India has a large, young, working-age population relative to its dependents is open now and will begin to close within a couple of decades. A demographic dividend is only a dividend if the young are educated, skilled, and employable. If they are merely enrolled, sitting in classrooms, ageing out unable to read a paragraph or clear an employability screen, then the same demographic bulge becomes a liability: a vast cohort of under-educated young people entering an economy that cannot use them.

That is the stake, and it is why the gap between India's enrolment triumph and its learning failure is not a technocratic footnote. It is the central question of the country's next thirty years, and it is being decided right now, one un-taught Class 3 cohort at a time. The reforms are half-built. The money is half-committed. The clock is not half-anything: it is simply running.

We enrolled a generation. Whether we educated it is still, for a few more years, a choice. It will not be a choice for long.

Bharat

Sources (every figure verified against the cited originals; estimates and press-compiled data flagged inline above). School data: UDISE+ 2024-25 and 2023-24, Ministry of Education (teacher count 1,01,22,420; single-teacher schools 1,04,125; secondary dropout 11.5%; facility and enrolment figures). Foundational learning: ASER 2024, Pratham (released 28 Jan 2025); PARAKH Rashtriya Sarvekshan 2024 National Report, NCERT (survey 4 Dec 2024, released July 2025; 21,15,022 students / 74,229 schools / 781 districts). Learning poverty: World Bank India Learning Poverty Brief, April 2024 (pre-COVID data). Higher education: AISHE 2023-24, Ministry of Education (GER 30.0; GPI 1.08; ~4.5 crore enrolment); QS World University Rankings 2026 (Nov 2025). Financing: Economic Survey (2.7% of GDP, 2023-24); Kothari Commission 1966 and NEP 2020 (6% target); Union Budget 2025-26 (School Education ₹78,572 cr; Samagra Shiksha ₹41,250 cr; PM POSHAN ₹12,500 cr; PM SHRI ₹7,500 cr). Employability & jobs: India Skills Report 2026 (Nov 2025, employability 56.35%); PLFS calendar-year 2025 (overall 3.1%; youth 15-29 unemployment 9.9%; unemployment among secondary-and-above educated 6.5%; youth NEET ~25%). Skilling: Skill India Programme (₹8,800 cr); PM Vishwakarma (₹13,000 cr). Coaching & wellbeing: IMARC coaching-market estimate 2025 (private research figure); Kota suicide tallies (press-compiled); NCRB (latest official year 2022). Global benchmarking: India's PISA 2009 result (73rd of 74) and withdrawal from PISA 2021/2025. Policy: NEP 2020; NCF-SE 2023; PARAKH establishment (NCERT, Feb 2023); UGC foreign-campus regulations 2023 (Deakin, Wollongong, Southampton; 17 approved by Dec 2025); DU FYUP fourth year (Aug 2025); Academic Bank of Credits (>3 crore registrations, per UGC). Note on flagged items: the ~46 million out-of-school-children figure is an independent analyst computation from UDISE+ 2024-25, not an official MoE headline; coaching market size and Kota suicide tallies are non-government estimates; the World Bank 56% figure uses pre-pandemic data. Each is labelled as such in the text rather than presented as settled official data. Statistical notes: UDISE+ methodology shift to student-wise Aadhaar-linked tracking from 2022-23 (Ministry of Education; prior data "not strictly comparable"); PLFS uses the standard ILO employment definition (unpaid family work counted as employed); PARAKH (NEP-mandated, more independent) replaced the school-administered National Achievement Survey; India's non-participation in PISA (most recently 2025).